This is because they cover the end sections of the sill panel, which runs from the back of the front wheel arch to the front of the rear wheel arch, and there is only a gap of about 1/8" or so between the two panels. The narrowness of this gap means that even if they are treated with Waxoyl or similar by spraying from above (by removing the splash panel in the front arch and the trim panel behind the doors) it will almost certainly just form a bridge across the top of the gap and not penetrate it. So even if the sill itself has been painted after fitting the back of the cover panel cannot be, and with the oxidisation caused by arc or MIG welding corrosion starts eating away at the new panel almost immediately. Spraying Waxoyl or similar as described above quite possibly accelerates this process as it will prevent circulation of air, so any moisture that gets into the cavity from below e.g. as a result of road spray cannot dry out easily.

Eventually the corrosion will become visible on the outside of the cover panel as bubbles with rust streaks running down. This usually happens near the bottom of the cover panel where it curves round underneath, which makes it easy to miss and even easier to ignore for a while, Until it gets bad enough to be worth doing something about.

Once the cover panel starts to corrode inside the cavity it actually becomes thicker - the products of corrosion having a greater volume than the original metal - and this gradually fills up the gap between the cover panel and the sill. Flakes from further up also fall into the bottom of the gap, eventually bridging it, and the rust debris comes into contact with the sill panel. The debris gets wet from the aforementioned road spray and takes a very long time to dry out - possibly never in a daily driver in anything other than a desert environment - and this starts rotting the sill panel too. If you leave replacement of the cover panels for too long, when you finally 'get a round tuit' you may find considerable corrosion or possibly even holes in the sill too even though the sill may have been protected by spraying the inside of the box sections.

In the first of the pictures here you can see that the lower part of the cover panel that was adjacent to the sill is heavily corroded bare metal, whereas the area immediately above and in front of the sill is completely covered with a waxy protective coating and has no rust at all. Although I left replacement perhaps longer than I should have, the sills were almost completely rust free with just one or two very small patches of surface corrosion, otherwise being painted a curious metallic silver-green as can be seen in the second picture. I was lucky.

I don't intend to go through the full description of replacing body panels, just a few points specifically concerning these two. I chose to cut the existing panels just above the top of the sill in an area that was rust-free, in order to keep the repaired area to a minimum. If replacing the sills at the same time you will probably have to cut higher for access. I used a nibbler in a drill and flanged the repair panel to fit behind the bottom of the existing panel to keep distortion of that panel to a minimum - that is your datum. The rear panel is easier to deal with as it is welded to the rest of the car on all bar one of its edges and this should give you the correct fitting relatively easily. The exception is the leading edge that curves round the sill. This should have a right-angle flange to it but on all of mine it was too deep and even though it fits into a recess on the sill panel it was still proud. Rather than grinding it down I choose to fold it further onto the back of the panel, carefully so as not to distort it, trial fitting it until a straight edge runs cleanly across the sill and the repair panel with no gaps. Test the coating on the repair panels with cellulose thinners. If it comes off easily then it is just a shipping coating and must be removed before welding in place. If not it is a primer that can be left, although the edges that are to be welded should be cleaned back to bare metal. In both cases coat the whole of the back of the repair panels with a zinc-rich weld-through primer before fitting. After welding, the curved join between the sill and the repair panel should be treated with seam-sealer, but do not flush it over, leave a shallow vertical groove.

With the front panel the front edge is unsupported so you have to be careful to get the correct curvature. As you want to keep a gap between the sill and repair panel I placed a strip of wood about 1/8" thick along the top edge of the vertical face of the sill for the repair panel to butt up to. Screwing along the bottom edge and welding along the top edge, then removing the wooden strip, leaves this gap. Without the temporary wood packing you could end up with no gap between the sill and repair panel which will cause problems later. But before welding careful work with angle-grinder, hacksaw and files is needed to get a good join where the horizontal welded joint meets the vertical door opening. The join between the rear edge of the repair panel and the curved part of the sill is not treated with seam-sealer but left open.

After welding, shaping, and painting we want to limit the possibility of corrosion occurring again, if no action is taken the welded edges will start rusting almost immediately. Unless you are in a very hot climate and the body panels are equally hot a product like Waxoyl, although you may be standing it in a bowl of hot water to liquidise it, will return to its normal 'tacky' state immediately when sprayed and not run down the 1/8" gaps between sill and cover panel but simply lie along the top of the gap. Either thin the Waxoyl with white spirit to a liquid and spray it in a jet rather than a fan until it runs out of the bottom of the sills (onto newspaper for example) or spray clean engine oil into the gap which will similarly run out. Get a strip of card (not metal which will scratch) and poke it into the gap to ensure the Waxoyl/oil is fully distributed along the length and height of the gap, then spray some more in. With Waxoyl you may find that it drips again in hot weather, whereas with oil it shouldn't. Don't be tempted to put the Waxoyl that has dripped out back into the pressure can for reuse, it may contain particles that will block the jet sprayer, but you can use it for brushing on behind the front arch anti-splash panel and elsewhere.

Don't be tempted to seal the bottom edge of either front or rear repair panel with seam-sealer or equivalent. You must allow any water or moisture to run out and not gather, and you will notice that the sills and castle rails have drain holes (see 'Sill Construction and Drainage' below) for the same reason. It is better to protect the cavities and allow some water from road spray to get in then drain freely out, than to try to prevent any getting in and end up trapping it which will accelerate corrosion.

Note: Even if your 1/4 panels do not need replacement I strongly recommend you make sure the top of the gap is clear then squirt a jet of clean engine oil into the gaps and distribute it with a strip of card. Do this even if the panels above the gap show signs of treatment to ensure it has got right to the bottom of the gap. Although Vee had been comprehensively treated these areas had been missed (although I squirted oil in them some years ago), and they aren't even mentioned in Lyndsay Porters Guide, a significant omission in an otherwise excellent reference book.

Behind each front wheel there is a 'splash' or closing panel which seals the gap between the front outside edge of the footwell and the front wing, having a rubber seal between it and the wing. There is also a small panel and seal for the gap between the top of the box-section visible from the engine compartment at the rear of the inner wing and the top of the wing.

The main splash panel is bolted to the footwell with seven bolts, the lowest one being up into the front of the floor pan and not visible in these pictures. Particularly the lower ones can corrode into the nuts welded to the panels, and shear off. In bad cases the corner of the footwell can crumble away.

With the splash panel removed you have access to the aerial, and also the rear nuts for the door hinges. You can also examine the front section of the sills and the lower edges of the front wings for corrosion, and dribble oil down the very narrow gap between the two as described in 1/4 panel replacement above.

Added October 2008:
One thing to consider when adding an aerial to a front wing is the box-section that is immediately below it and behind the inner wing - see the accompanying thumbnail. There should be one or two large holes in the inner wing to enable you to see inside this box section, and the main harness and accelerator cable go to the cabin via it. If you drill the hole in the outer wing too close to the seam between the wing and the panel at the base of the screen you will also have to drill a hole in this box-section. Even then if you have a manually fully retractable aerial it may foul the bottom of the box-section, and a motorised aerial probably won't be able to be installed at all. The box section goes all the way forward to the splash-panel, and then there is a tapered 'trumpet' section in front of that, so you would have to go all the way forward to just about the front axle before you get a clear space under the wing if you are too close to the inner edge.

However by positioning the aerial close enough to the outer edge of the wing there is plenty of room between the side of the box-section and the side of the outer wing for any likely aerial. This means mounting the aerial on a curved and angled surface rather than a flat and horizontal surface, but all the aerials I have seen have a fitting that can cope with this kind of location.

These pictures from Paul Tegler show a motorised aerial in the process of being installed behind the splash panel.

Fortunately it is quite easy to turn the peg and clip through said 90 degrees. The peg is easiest as it has a square fitting, just squeeze the lugs on the back with a pair of pliers and push it out, and it will snap back in after turning (it's actually a fraction off-square but will go in the new way with just a bit of a push). The clip requires a bit more work. It's a little bit rectangular, which means the sides of the slot have to be filed out slightly. The overall width of the clip is greater than the length of the hole in the console, so will cover the sides of the hole if you are careful in your filing. Only file the minimum width needed for the clip, and in the middle of the sides of the hole, which will stop the clip sliding fore and aft and revealing the edges of the old hole (as well as nullifying the point of the exercise by putting the peg out of alignment with the clip again). Also only file enough to just be able to get the clip pressed into the new hole, if you file too much the clip will be pulled out of the console on the end of the peg.

However even if you did that (as I did on the roadster many years ago) all is not lost. There is a depression in the console that looks as if it were designed to take a mounting plate for the clip, complete with two blocked-off holes for fixing screws. It's a simple matter to cut a plate to fit this depression, drill through the plastic console and use countersunk screws to attach it, and it is much easier to get a tight-fitting hole for the clip in the plate than it is in the plastic.

• MGB roadsters originally had two body numbers.
• One with no prefix, starting from '1', stamped into a steel strip, spot-welded to the inner wing behind the radiator on the dynamo side. This number was written into the Abingdon records.
• The other with an 'MGB' prefix, starting from '101', stamped into an alloy strip, screwed to the inner wing by the carbs, not recorded by Abingdon.
• On Mk2 roadsters (but see below) the MGB tag moved in front of the radiator, still on the carb side, rivetted in a vertical position.
• Mk2 roadsters for North America had body tags prefixed MGBU, and those for Germany and Sweden from 69 to 74 were GBSN, and both had their own series of numbers.
• From 1969, roadsters with Swindon/Cowley bodies just had the MGB tag, which was now the one recorded by Abingdon until 1972, after which body numbers ceased to be recorded.
• GTs, MGCs, and V8s had different prefixes and each had their own number series, one for roadsters and another for GTs in the case of the MGC.
• Until GT production started body numbers stayed pretty close to chassis numbers, for example Clausager dates chassis number 66580 to July 1965 and body number 66808 to August 1965. With the separate series for GT, North American and Germany/Sweden, but all cars using the same chassis number series, they got further and further apart.
• Rubber bumper cars had the body number on a larger strip welded into a depression on the bonnet slam panel.
• Australian kit-built cars were very different in several aspects.

However going by actual examples it was sometime between April 65 and October 66 that the 'MGB' tag moved to what Clausager says is its Mk2 position, and almost certainly at the same time the unprefixed tag also moved, to the very front edge of the inner wing on the dynamo/alternator side, orientated vertically, between two bolts for the outer wing. GT production started in September 65 so it's possible that was the actual change point. There does appear to be an example of a one-owner, unrestored (going by its present condition) 63 roadster that has the MGB number but not the other one.

Why two body numbers? Some cars (MkII roadster bodies built at Coventry) - could have four tags - two body number plus the Commission and Chassis numbers. MGBs were built in at least three plants, possibly four, being shipped between them so some means of keeping track at each stage was required. The spot-welded, un-prefixed, steel tag was almost certainly attached during body assembly, certainly before painting. As the MGB-prefixed tag was originally screwed, but also painted, it was probably attached at the paint and trim plant i.e. at the next plant in the sequence. This could well explain why the two numbers could drift apart, for example a 1966 example has its two numbers over 500 apart. The Commission number plate was rivetted to painted bodies i.e. at the trimming stage before shipping to Abingdon. The Chassis number plate was rivetted by Abingdon after final assembly. This still doesn't fully explain two body numbers, but there must have been a reason, they wouldn't have done it just for the hell of it, it's simply lost in the mists of time.

I've never seen a picture of a Chassis number stamped into the body of a UK-built car, and Clausager indicates this is a very confused and uncertain feature. However Australian kit-built cars definitely did. Their special (BMC Australia) plate had the body number as a suffix to the chassis number, as well as the engine number and body colour. On an example of a Mk2 this plate was on the inner wing in front of the radiator by the oil cooler pipes. The body number was stamped into the actual shell, just behind the opening for the pedals on the carb side in one case, beside it in another, and the chassis number was stamped into the top of the carb-side chassis rail at the very front. There also appears to have been a 'dicast' plate behind the wiper motor giving the date the kit of parts was put together, presumably supplied by Abingdon, the actual build date being later of course.

The rubber bumper car is not too difficult to get into - as long as you know what you are doing. It may be possible to gain access in a similar way through the recessed grill of 70-72, it isn't through the honeycomb grill of 73 and 74 (but see below), and it probably isn't in the pre-70 chrome slatted grill.

On an RB car like my 75 factory V8 one can push a long rod or screwdriver through the inner mesh grill, in a particular position and at a particular angle, and push the end of the lock lever so releasing the bonnet pin. The positions and angles are shown in the following pictures.

The first picture shows a long rod inserted and bearing on the end of the lock lever where the cable is attached. What you are trying to do is use the rod to push on the end of the lock lever to release the bonnet/hood pin.

The second picture shows the horizontal angle of the rod, which is inserted through the mesh, right at the very top, the 4th hole from the centre mounting of the mesh. On mine the mesh makes 'diamond' shapes and the very top row only has the bottom half of the diamond, the top half is covered by the frame to the mesh.

The third picture shows the vertical angle of the rod. The end outside the car is angled down as far as it will go it is in contact with the bottom edge of the bumper.

The thing to bear is mind is that the rad is only a couple of inches behind the lock. So if the end of the rod goes too far it could puncture the rad. On mine, with the end of the rod just touching the rad at the angles shown, there is 10 inches of the rod sticking through the mesh. So if you wrap some tape around your rod, say, 9 inches from the tip, then at the angles shown as long as you keep the tape your side of the mesh you should be clear of the rad. The lock lever will require quite a push to release the bonnet pin, so be careful the rod doesn't suddenly slip and go too far.

Honeycomb Grille: (Updated August 2007) The only way I can see of doing it with these grilles is to push the out the grommet in the left-hand inner wing that the lighting wiring goes through, then insert a length of stiff wire with a loop twisted in the end, feeding it through straightening it as you go (you can't feed it straight in because of the wheel arch) then hook it round the release lever and pull. You *can* see to do this through the grille, but it will be fiddly. Should work for the earlier slatted and recessed grilles as well, if there isn't an easier method from the front.

Preventing the problem in the first place - all models

Once (hopefully) open, the first thing to do is give yourself a 'second string' in case it happens again. I have used a length of curtain pull cord, which is a very strong braided nylon, with one end tied round the lock lever and the free end pushed through the hole in the left-hand inner wing where the headlight wiring goes through. Tie a loop in the free end big enough to get your finger through and that will stop it working its way back out again.

Some say they have removed a reversing light then used a length of stiff wire to hook into the lock mechanism, which sounds pretty tricky, and I can't see how you would get the leverage to operate the latch from the side.

Denise Thorpe has said that if you twist the lock to one side or the other as far as it will go it reveals enough space to drill a small hole through the skin that will be covered when the lock is centralised again. Inserting a narrow probe through this hole, and fiddling about, should enable you to push the latch out of engagement and so release the lid. Someone has said their lock twists nearly 90 degrees, but that sounds like the key-way has been cut away (or maybe they are thinking of the push-button and not the lock as a whole!), both of mine barely twist enough to drill the 1/16" hole so that it could be covered by twisting the lock back the other way.

The accompanying photos (click a thumbnail) show the features of the release mechanism including what stops it twisting very far, what comes loose/drops off to cause the problem, and where to pull/push to open the latch. These are taken on a GT for ease of access but the lock on the roadster is identical.

A future project is to look at the feasibility of using a bonnet or choke cable mounted on the rear bulkhead and permanently connected to the lock mechanism. Whatever you use as a one-off or a more permanent 'emergency release' needs either to pull the upper part of the release lever (the part the cam bears on) forwards (i.e. towards the front of the car) or the lower part of the release lever (that hooks under the bar on the rear panel) backwards to clear the bar. On a GT a piece of timber with cross-section up to 3/4" high and 1 1/2" wide can be pushed through the loop of the bar to push the lower part of the release lever backwards and release the tailgate. It take surprisingly little force to do so, so there shouldn't be much pressure on any permanent 'emergency release' cable used on a roadster. Such a cable obviously introduces a security risk since it is easy for someone to gain access to the cabin of a roadster, but hiding the handle should reduce that risk. Watch this space.

In the meantime, get out there and put some Locktite on the screw and tighten it up!

Added November 2009:
Still watching? I'm still waiting to get a round tuit before looking at this but Bruce Mills has written to me with what he has done. As well as adding an emergency boot-lid release he has gone a lot further and added an electric release triggered from a key fob for both lid and doors. A 'popper' spring on the doors then pushes them open just a bit for you to open manually the rest of the way. The boot lid had one as well to start with but it exerted too much pressure on the catch so the solenoid was unable to open the latch and had to be removed. These changes have allowed Bruce to remove the locks and handles to get a smooth finish.

Two strange pieces of carpet (amongst many) fit immediately in front of each rear light cluster. One flat squarish piece glues to the inside of the rear wing in the space that is bounded by the back of the rear wheel arch, the boot floor, the rear light cluster, and under the boot lid rim (roadster) or C-post trim (GT). The other is two pieces stitched together such that the natural inclination of the 'hinge' is for the pile-sides to fold together. The wider half lies on the side piece of the boot floor and the narrower half covers the back of the rear light cluster. Both pieces are handed so try them both sides for the best fit.

The narrow part that covers the back of the light cluster is held in position by a plastic trim fastener pushed through the carpet into a bracket, but neither the bracket nor the fastener are shown in the official parts lists I have. The roadster didn't have these brackets so it was always a bit of a puzzle how the 'hinged' pieces were supposed to fit. It wasn't until I got the V8 which happened to have just one of these brackets and fasteners that I realised, and I was able to make three more brackets using the existing one as a pattern. I was able to get suitable fasteners at Halfords from their range of generic trim fasteners at the time. These are black plastic, have a 15mm diameter head, 17mm overall length, 15mm shank length. The shank has serrations and a maximum diameter of about 7mm.

I made the brackets out of some bits of body panels I had lying around (Hunts Eighth Law: "If you haven't found a use for something yet, you haven't kept it long enough"). All measurements are approximate but these are taken from the one original bracket I have. It isn't critical, it isn't visible, and I doubt even a concourse judge would be looking at them.

For each bracket cut a flat strip 19mm wide and 55mm long. Mark a line across the width 15mm from each end, this is where a bend of approximately 40 degrees is made. Mark and centre-punch the centre of each 19mm x 15mm area for the hole to accept the plastic fastener. Get the fasteners and drill a hole of suitable size for them but in my original they are 6.5mm diameter. Finally make the bends so you end up with something like in the picture at the left.

The bracket is secured on the top stud for the rear light cluster and held by an additional spring washer and nut. It is angled down and towards the boot/hatch opening, as shown in the picture on the left. The plastic fastener pushes through the carpet and into the free hole in the bracket.

Battery 'shelf'
The wheel arch and battery shelf pieces leave the chassis rails exposed, so I cut and glued some additional pieces for a neater appearance.

The 'Crack of Doom'

When I bought 'Bee' the drivers door had the 'crack of doom' which is split at the top of the door skin near the front door glass channel. After much study I decided there was a design weakness in the door skin, in that the flange that supports the outer rubber seal ends just before the front door glass channel. I welded a small piece in such that it extended the flange alongside the channel and round the front, meeting up with another flange. The first two pictures from Lyndsay Porters Guide shows the area quite well. The first shows the area that splits (arrowed), and the second where I have extended the flange in yellow and blue (false colour). Subsequently I saw a Heritage door that seemed to be constructed in just this manner. I MIG-welded the crack, and used a large oval plate behind the door skin to support the door mirror. It has a slightly larger radius than the door skin, so bracing the mirror over a large area. I also made sure that there was a gap between the side of the quarter-light frame and the thin part of the rubber seal on the windscreen when the door was closed (next three pictures) to give some room for scuttle-shake over rough surfaces.

Update January 2005: During a thread on this subject in a mailing list Rich Chrysler posted the text of a BMC (Canada) Technical Service Bulletin from 1963 showing this problem and saying a production change was being made. My first thought was "well, they never did" but by the date of the bulletin the MGB had only been in production little more than a year and less than 20,000 had been made, so the problem must have been serious. 30 years and 500,000 cars later we know it still occurs, but most cars don't have it, so I'm thinking that they did make a change which considerably improved things, but did not completely eliminate it, hence the further change some time later (i.e. after 1973 when mine was made) which can be seen on the Heritage doors. The drawing in the bulletin shows three edges joining at a single point, whereas the Porter picture shows a slightly different arrangement, and the former would be more prone to cracking in my view. For interest a scan of the complete document can be seen by clicking here.

Hanging and Adjustment

I see the objective as four-fold:

1. To get the crease where the chrome strip goes running in a straight line from the rear wing, through the door and into the front wing when the door is fully shut.

2. For the door to achieve that from open *without* the rear of the door being raised or depressed by the action of the lock on the striker-plate or binding on the pin and catch.

3. For the wing sill and door panels to be flush with each other all the way round.

4. For the gaps to be even all the way round.

There are no less than twelve separate steps to correct alignment of the doors and 1/4-lights to aperture front wing and windscreen.

On my drivers door the front and rear wings were not in line, so lining up the rear of the door with the rear wing and the front of the door with the front wing resulted in a zig-zag. I had to slacken the bolts holding the front wing, put a piece of timber from the top of the wing to the bottom of the garage roof, and jack the car up to press the wing down into the correct position! Sounds drastic but it worked.

Adjusting the overall height of the door is set by sliding the hinges up and down wrt the A-post. As well as the four visible Philips screws there is a nut behind the front wing which is accessed by removing the splash-plate aft of each front wheel.

Getting the leading edge of the door flush with the front wing is set by sliding the hinges in and out (relative to the car as a whole) on the A-post, and getting the top of the leading edge adjusted relative to the bottom is achieved varying how much each hinge is moved relative to the other.

September 2009: There has been some discussion of this area recently on a mailing list. John Davies nearing the end of a restoration has found that the bottom of his drivers door needs to be moved out by about 1/16" or so to clear the wing. Opinion was that the hinges can't be adjusted on the A-post, even though they have sliding threaded plates like in the door and for the striker plate. Even if the screw holes for the hinge in the A-post don't give enough adjustment I can't see any reason why they can't be filed out a bit to give more movement, but of course that would leave a mark, if not a gap in the paint, around the hinge and is why this adjustment really needs to be sorted before painting. There was talk of putting a shim in the door, but the threaded plate the hinge screws go in to is boxed in so may not be accessible. Then John described a method of bending the hinges his local garage showed him, which would achieve much the same thing without actually moving the hinge relative to the A-post. Basically you put a wedge into the hinge and close the door onto it. Pushing against the door carefully should tend to bend the part of the main hinge body that the pivot is attached to (which is much thinner than the tongue itself) and will have the effect of moving the pivot and hence the door out from the body slightly. Of course you can go too far, and it is not possible to bend it back again without removing the hinge from the body and door, or revert to Plan A and move the hinge on the A-post with all that entails. Care required!

Adjusting the height of the rear of the door relative to the front is the second-easiest of the adjustments to perform. Each hinge-to-door fixing has three Philips screws that go through the door panel, through the hinge and into a sliding threaded plate inside the door, I think, although this plate doesn't seem to be a listed part. If you slacken all of these you should be able to imagine being able to slide the whole door forwards and backwards on the hinges, and this sets the basic fore-and-aft position of the door in the aperture and hence the front and rear gaps. With the same screws loose if you let the door go the rear edge will drop right down, the top hinge will slide out of the door and the bottom hinge will tend to slide into the door. Likewise if you lift the rear edge of the door the top hinge will tend to slide into the door and the bottom hinge will slide out and that is how you set the height of the rear of the door relative to the front. But before you slacken the screws you need to decide whether you want the top of the vertical gap between door and front wing to get smaller or the bottom of that gap to get larger. If you want a smaller top gap you slacken the top screws fully but just slightly loosen the bottom screws. Then when you lift the rear edge of the door it will pivot about the bottom hinge and the top hinge will slide further into the door. If you want the bottom of the gap to get larger you just slightly loosen the top screws and fully slacken the bottom screws, then lifting the door will pivot it about the top hinge and the bottom hinge will slide further out of the door. However if the hinges themselves are worn or bent you there may not be enough adjustment left to raise the rear of the door to the correct position. In that case you may be able to get by with swapping top and bottom hinges over (there seems to be no difference in top and bottom hinges, only RHS and LHS) or another possibility is to shim the lower hinge to get more 'lift'. However both these will disturb the 'up and down' and 'in and out' settings of the front of the door.

With the height of front and rear and the panel gaps set there are now four more adjustments left to do - yes four. You may have to twist the whole carcass of the door in order to get the rear edge flush with the rear wing for the whole of its length. The factory used a large bar that clamped onto the top and bottom edges of the door to achieve this.

Finally you adjust the striker plate in and out so that the rear edge of the door is flush with the rear wing and up and down so the door does not lift or drop as the lock engages, and add/remove shims between the striker plate and the B-post so that the lock opens and closes cleanly and doesn't bind.

However if the sills and rear wing have been badly fitted, or the door skin, it will be very difficult to adjust the door to compensate - or even fit properly. For that reason you should remove the minimum of front and rear wings and door whilst fitting new sills, even though the door will be in the way. They are vital to get correct alignment.

Happy hanging.

Update September 2003. I was never happy with the fit of the drivers door which had always used the striker plate to lift the door into position slightly, which is something it shouldn't have to do. This makes a bit of a clatter when closing, and unless the lock and striker plate are kept lubricated with a smear of grease it can need a push to open and the grease can get on one's clothes. I knew this from the outset but I'm sure I originally spent ages trying to get the door-to-hinge adjustment correct so this wasn't required but was unable to do so. Finally I decided I had to try again and blow me if slackening the top hinge to door screws alone allowed me to raise the rear edge of the door into the correct position, it only taking a few minutes including removing and replacing the trim. However that moved the lock slightly forward in relation to the striker plate and the two were now binding a little. There were already three shims under the striker plate, so I used one as a pattern to make another out of even thicker plastic, and one thick and two thin have now put the striker plate in the correct position, and the door now shuts with a nice click and springs open instead of needing a push.

Key Lock

One of the first things you should consider doing, and certainly if repainting the doors or working on the locks, is to replace the very poor arrangement for holding the external lock in the door skin. This is nothing more than a flimsy spring clip which at best allows the lock to rotate in the door skin slightly, wearing away the paint around the hole and allowing rust to set in. At worst the locating 'pegs' in the door skin that stop the lock rotating too far can wear and allow the whole lock to be rotated far enough to unlock the car just with a screwdriver in the key slot.

There is a much better 3rd party lock fitting kit available from many of the usual suspects which secures the lock positively in the door and giving it no chance to move and chafe the paint. It is quite easy to fit once you have the door trim off, you just have to remove the existing spring-clip, ease the lock out of the door slightly to be able to fit the circular washer over the end (chamfered side facing the door skin) of the forked lever that operates the locking rod sticking out of the latch assembly, relocate the forked lever over the rod, then the C-shaped washer just slides into the groove vacated by the spring-clip. Tighten the grub-screws equally, and Bob is your Mother's Brother. Note that to remove the door trim you will need to remove the door-pull/armrest, the window winder, and whatever arrangement you have for opening and locking handles, then unscrewing or carefully levering up the concealed snaps. On Mk2 cars with the rectangular safety handles simply slide the top half of the plastic finisher upwards and the bottom half downwards, being careful not to tear the vinyl. The screws in the window winders seem prone to seizing, but if you remove the waist rail as well then once you have released all the trim panel snaps you can rotate the panel downwards about the winder handle and that will give you sufficient access. Remember to refit any polythene sheets over the holes in the inner skin to protect the hardboard trim from rain-water, and if you don't have any then fit some!

Added November 2009: Cars used in all weathers and particularly those kept outside can suffer from the locks getting stiff, as can garage locks and padlocks. You can ignore it until the key snaps off in the lock, or do something quick and easy sooner rather than complicated and expensive later. Vee was kept under a car-port for a long time with the prevailing weather against the drivers side, and that lock periodically used to stiffen-up, rather than the little-used passenger side. I've seen various various recommendations for oil, silicone and teflon sprays, and graphite powder each usually stating that only they should be used! This advice from a locksmith and safe engineer says silicone and teflon can be used, but graphite is better, and oils like WD40 should definitely not be used. I see silicone and teflon advertised but not specifically mentioning locks, so I'd steer clear of those. Graphite powder specifically for locks is available in small quantities, but for small quantities like a lock you can make your own (as recommended on the above site). Take a graphite pencil, and cut back the wood to expose 1/2" or so of the graphite core. Fold a sheet of paper in half and cut a diagonal, so that when opened out again the crease forms a point, like an arrow or a paper dart. With the pencil held over the crease scrape down the exposed graphite with a sharp blade held at right-angles to the graphite until you have a little pile of powder. Fold the paper and tap the pile into the crease. There should be little shutter in the lock across the key entrance, wedge this open with a matchstick, put the tip of the crease in the folded paper into the slot, angle the paper and tap it gently so the graphite powder trickles into the lock. Don't do this in windy conditions or it will simply blow away! Remove the matchstick and use the key in and out and turning it to distribute the graphite and it should soon ease up. Normal use of the lock will distribute it further. In exposed conditions like Vee's you may need to do this once or twice per year.

My first thought was to mount it above the sill in the footwell somehow, but that would probably have meant screwing through the trim panel which I didn't want to do. Then I investigated hanging it from the square-section bracing tube behind the bottom edge of the dash. Possible, but it would have to hang several inches down and not swing about. Looked along the tube towards the centre console to see the two brackets for the LHD steering column, which were a definite possibility, but again the holster would need to be lower and not swing about. Then my eye fell on the lower screw in the side of the centre console, and Bingo, the ideal location.

Repaired the frame by welding a strip between the two broken holes, and re-drilling the top one, did a test fit and it looked fine. However the water bottles are quite a bit smaller diameter than the washer bottle and if at an angle in the holster would fall through the frame. Started pondering welding more strips around the bottom, but then I seemed to remember I had replaced the water bottle as well when I replaced the frame because it was holed and leaking. Another root through the box of bits and Lo and Behold! Cut the top of the washer bottle to just below the top of the frame, perfect. Then it was just a matter of stripping and painting the frame, and installing.

The Leyland Parts Catalogue shows drawings and part numbers for five parts - the main board, the front and rear header rails, and the left and right cant rails. The drawings appear to show that the front header and cant rails are of a similar size and thickness to the rear header, for all eras of GT, but on my 75 V8 although the rear header was obviously a covered and padded removable board, the others were solid when tapped and could have been hard plastic. I wasn't able to find any reference to the front and side rails from any source, new or used, via Google, and only one to the rear section. Cutting a panel from sheet is definitely feasible for the main board, recovering kits are available at the rather shocking price of £40-£60. The only other possibility seemed to be breakers and abandoned projects. In jest my pal suggested I take mine out and have a look at it, and was horrified when I said I had already decided to do just that! I then remembered the sun-roof and realised I wouldn't be able to remove the main section to use as a template, but at least I could remove the rear section and have a look around.

To do that the C-post trim panel really needs to be removed, which involves removing the screws from the rear window catch and another screw right at the top. I then discovered five spring-clips which need careful levering if they are not to rip the hardboard. The plastic covering on the trim panel had pressure bonded to that around the rear window, but peeled apart without ripping with a bit of careful pulling.

The rear header is attached with two screws through the load-space light plus four spring clips. There appear to be two holes at the ends of this rail for more clips, but only four are shown in the Parts Catalogue and the top of the C-post trim panels support them anyway.

With those out the way it became apparent that far from the cant rails being rigid strips, they are part of the body structure and simply covered with headlining material glued on. The front header rail is the same, as the rear-view mirror screws into it with short screws, the edges of the holes being at earth potential when tested with an ohmeter, i.e. part of the body. These rails form a 'shelf' running along the front and sides of the roof. My pal managed to get a set of old side and front pieces, which amazingly someone had been able to peel off the body with ripping. Comparing these to the strips in the recovering kit reveals the new pieces to be extremely thin and flimsy by comparison, even taking into account that the originals will have hardened over time and have a coating of glue on them. However I suspect the originals were heat-formed to fit round the box-sections to make attachment easier, and the ends of the front section have stepped ends (also heat-formed) for the ends of the side pieces to sit on, to make a smooth transition from the one to the other.

The main board just pushes in from the rear above this shelf, the rear panel covering the back edge. The back edge on mine is tucked into a metal recess which is part of the body structure, that may be as a result of having slid back a bit when the sun-roof was fitted, and would need prising forwards and down for removal. As far as cutting a panel to fit goes you wouldn't want to go the full width available above the shelf as I felt various protrusions i.e. reductions in width at various points. Neither would you want to cover less than half the width of the shelf, because if the panel moved over in use it could leave a visible gap, so maybe covering 2/3 or 3/4 of the shelf is the way to go. The amount of board that is 'hidden' above the shelf is surprising - 2" to 2.5" at the sides and just over 3" at the front!

However! The maximum size of the tailgate opening at the top is 39", and at the bottom is 41.25". As can be seen above the maximum width available for the board is 43", so some bowing of a board this wide would be required to get it in. Even with the suggested width of 41" it would still have to fed in from quite low down or bowed along its length, and because the highest point of the tailgate opening is lower than the shelf the board would have to be bowed from front to back at the same time. The other issue relates to the upward curvature of the roof. To follow this curve either the board has to have a 'natural' curve produced by steaming or damping and then drying in a mould, or it has to be slightly wider than the maximum available so that the board can then be forced to curve upwards. But that would require a precisely cut board - too wide and it wouldn't go in even bowed up, too narrow and it will flop down again. I don't actually know whether it is a natural bow in the board or a precise width that forces it to bow up, as I say at the beginning my V8 has a glass-panel sunroof the surround of which holds mine up against the roof as well as preventing removal without interference to the sunroof which I'm not prepared to do at this point.

Then one trip away I had the idea of putting the load-space carpet over the luggage instead of underneath it and that together with the tinted windows meant you had to look pretty closely to see anything. The only drawback with that was that the luggage then slid about all the time on the shiny spare wheel cover which was irritating. However it gave me the idea of getting a piece of carpet to do the same job. Next Stoneleigh MG Spares Day I spotted someone selling load-space carpet quite reasonably at about £15, but also selling sheets of carpet a couple of metres square for just £10, so I got myself one in black. It was described as 'Volvo quality' but was significantly inferior to the load-space carpet from the same vendor and other carpet squares from other vendors, but as it wasn't for 'walking' on and only a load-space cover I reckon it will do just fine. It will also allow me to make one long enough so that with the back seat lying flat with large loads I can still cover it all.

Exactly the same fasteners to the top of the windscreen as for the soft-top header rail - originally AHH6262 for cars up to chassis number 38387, AHA7709 from then to the end of production for the UK, AHA8491 for North America.

The sides use the standard tonneau bar sockets AHH8391 screwed to the body (with the hood frame where that is fitted).

Bracket AHH7202 is screwed to or through the body of the hard top, with bracket locating AHH7203 between the bracket and the socket providing the adjustment between top and body, from a screw going through a slot in the bracket into a welded nut on top of the locator. Another screw goes through a hole in the tonneau socket into a threaded hole in the locator to resist pulling upward from air pressure or Bernoulli effect.

At the rear bracket AHH7199 also screws through or to the hardtop body, its bar slotting under the same retaining plates 14B1730 screwed to the tonneau panel as used by the soft-top.

But going back to the original strength Abingdon were carrying out impact tests in the 60s and 70s, pages 216/217 of 'MG by McComb' and 142 of David Knowles's 'Untold Story' showing the results of just such impacts into a concrete block at 30 mph by a roadster and a GT. Even now Euro NCAP only tests at 40mph (albeit as a partial frontal and not a full frontal as this is) but the damage is usually much more severe than this. Apart from the crumpled bonnet there appears to be no damage behind the front arch.

There is also this well-documented case of an MGB at Silverstone hitting a concrete barrier at an estimated speed of 90mph and experiencing a deceleration of 75G in not much more than a couple of feet. The driver was unconscious but this was from the effects of deceleration, not impact of his head with any object. And although the driver suffered a broken bone in his foot this was purely from the force of his foot flying forwards and hitting the toe-board or pedals, there was no intrusion into the footwell.

Modern cars have carefully designed crumple-zones at the front and the rear to give a more gradual deceleration than the Silverstone case, and a strengthened passenger 'cage'. However this means that even in relatively minor accidents the crumple zones will be deformed to the point where the car is declared a write-off (although there is a legal cottage industry specialising in the repair of many of these). Also, and more importantly, above a certain impact speed the passenger compartment will crumple like tin-foil. This is obviously an extreme case, and the fact the 'other vehicle' was a truck doesn't help, but remember the next time you are bowling along a single-carriage way, that even at legal speeds your closing speed with someone coming the other way is 120mph.

And nothing is going to protect you (short of a tank) if you pull out of a side turning while talking on a mobile phone.

Folding the hood/top
Replacing the hood/top
Hood/top fittings

Octagonal Origami (or, the art of folding the MGB hood)

I posted the following on the MG BBS in response to an enquiry, and one person was kind enough to contact me and say it had helped him, so I decided to include it here. It relates to the 1970-on Michelotti hood, not the packaway or earlier folding types for which see here.

By the way I keep seeing exhortations not to fold the hood if it is damp. Fair enough, I wouldn't put the car away with a damp folded hood, but are you going to keep driving with the hood up after a shower now the sun is shining again? Nor am I, but it does go up again when I get to my destination. In fact I always erect the hood when in the garage, just in case the roof decides to start leaking.

With all fasteners (front and rear) undone I pull the header rail back just enough for the hood to clear the slotted chrome thingies behind the cockpit, then push the header rail forward again. At this time the rear window is dangling down behind the rear seats, I lift it up and forward and lay it over the central part of the hood.

The quarter-lights are now sticking out at the sides. I lift the rear window part and tuck each rear quarter-light under the rear window so that they lie between the rear window and the main part of the hood. If you get this right the three pieces of clear plastic (rear window and two quarter-lights) will be lying flat, unfolded, uncreased and all the folds will be in the black vinyl. You can see one of the cords that pulls the rear bar into the correct position when the hood is erected.

I now pull the header rail back lift the rear bow by pulling up on the trailing edge of the vinyl (the part that hooks into the chrome thingies behind the cockpit) so that the rear bow doesn't wedge on top of the inner arches, and lower the whole thing down into the space in front of the rear cockpit rail.

Finally I drape the rear window part, complete with tucked-in quarter-lights, forward over the header rail, making sure the header rail clamps are closed. If you fold the black vinyl edges correctly it is these that lie over the screen fasteners and not the clear plastic.

I have seen two pieces of cloth employed at this point - one between the clear plastic and the header rail and the other between the tonneau cover and the hood to protect the clear plastic but personally have not bothered to do this - after nine years and 25k wet and dry miles my windows are still in pretty good condition. You should have two hood straps (or rather two ends of one strap each side) that allow the rear part of the hood to be rolled round the bottom part of the whole folded arrangement and held up out of the way of anything on the shelf, but although I have these straps I use the full-length tonneau cover as a hood cover fastened to the rear heel board so there is no need.

With the rear bow attached to the hood material using the two strips that are provided for this purpose I found the bow wasn't in quite the right place when the hood was erected, but too far to the rear. I now have two cords, one each side, tied between the middle and rear bows, and adjusted so that the rear bow is pulled up right under the seam as the header rail is pulled forward. See the 2nd picture above. Subsequently I discovered the early folding frame had something similar as standard. It was two lengths of webbing joining the front and rear bows and going over the middle bow. One can be seen clearly in Clausager on page 54, 2nd picture down, as cream webbing contrasting with the black hood material and grey frame. Also a glimpse of it in the bottom-left picture, which also seems to show the rear bar sewn or glued into hood material flaps the same as the later hood.

My hood/top doesn't fold low enough Updated July 2008 This relates to something that happened when I restored the car some 19 years ago, so I don't have any photos and it is dependant upon memory to a large extent. I'm only including it now as someone has written to me with what seems to be the same problem, although it doesn't appear to be a common one as I can't recall anyone coming back when I mentioned it on the various lists and BBs some years ago.

When I first had the roadster it came with a full-length tonneau cover with snaps half-way along so it could be used as a hood cover as well, the front half of the cover being tucked up under the folded top on the shelf. However my hood sat so high when folded that the snaps were nowhere near the posts on the heelboard, whereas pictures of cars showed the top of the hood dropped right down into the space behind the seats. But what I could do was fold the hood back with all the rear fasteners still done up, which made it very quick and easy to erect. A friends car of the same age (73) was exactly the same, and he related how he could erect his top while driving along!

When I restored the car I wanted to get the hood lower so I could use the tonneau cover as it does look much neater, and after some pondering realised that a couple of the struts in the frame were folding as far as they could, then jamming, and that was stopping the top and frame folding all the way down. After a bit of beating on these struts (time elapsed prevents me from remembering which) and changing the angles 'lo and behold' the frame and cover now drops all the way down as shown here, with the frame bows resting on top of the inner arches, and I can fasten the tonneau cover heel-board snaps and now have a neat hood cover. But I then discovered I can no longer fold the top back with all the rear fasteners in place as it traps and stretches the material. No matter, I'd rather have the neat folded appearance.

I've got the sticks, but it is a real stretch to fasten the snaps of the full-length cover when used as a hood cover, and it doesn't really make that much difference to the appearance anyway so I don't use them. I think they are mainly intended in that position to shape the separate hood cover which I don't have. Where the tonneau sticks are useful with the full-length cover is when both sides are fastened full-length, when if you position the sticks in front of the seats instead of behind it lifts up the centre of the cover so rain runs off the sides instead of pooling in the middle and dripping through the zip. I've never seen this advice anywhere else.

I don't understand why so few stick up in the air like mine did, unless it was a different manufacturer of the frames for a while, but even then I'd expect more cars to be affected. Having said that at the Stafford MG Enthusiasts run to Upton House in September 2009 I noticed quite a few of each type i.e. some low and some high. These were all the later Michelotti style, and there didn't seem to be any correlation with the year i.e. some earlies and some lates with both types. No pics, maybe next year if I remember.

Replacing the hood

Two things to be aware of here - buying a new hood, and fitting it.

Location of fittings

I've been asked a couple of times now for a diagram and dimensions of the hood fittings, so here it is.

The hole to the left of the 8cm line for the rear-most peg is the tapped hole for the static shoulder belt attached with a quick-release fastener for some markets in various years. The front-most peg is measured from the front of the socket for the hood/top tongue, and has a press stud close behind it.

Having tried a couple of products I recommend Turtle Black Chrome. It is quick and easy to apply, brings tired bumpers back to black and shiny, and they stay that way for ages. A subsequent application several months after I got them to a good shine made hardly any difference, they were still so good. It has the same effect on various items of black trim of different materials on my 89 Celica although I only use it on the bumpers on the V8. It is a black almost creamy liquid, and a word of warning, if you get it on paintwork you can only get it off again with car polish, so I slip sheets of paper behind the edges of the bumpers while I am applying it.

I had previously tried ArmorAll and sure enough with enough applications left to dry in the sun and then finally polished up you could get them back to black and shiny, but the effects faded very quickly, in fact they seemed to end up rougher and greyer than originally, even when not exposed to sunlight for long periods. Some recommend its use on interior vinyl but others say it dries it out and can cause it to crack in the sun. Having seen the effect on the bumpers I can believe it. About the only thing it is good for is tyres for a show finish, being better than that awful high-gloss tyre paint beloved of second-hand car showrooms. But if you are going to drive the car don't bother, after a few dozen miles the effect has worn off.

Another recommendation I have seen is black boot polish, and whilst I can believe it works I can also believe that it will rub off black on light-coloured clothes even after being polished up. I say this after having had to clean a light-coloured carpet where someone with highly polished shoes had been shuffling their feet.

First remove the two large screws from the bottom brackets and the three small screws going through the slam-panel to the plastic sockets in the top brackets then slide the whole grille forwards, keeping it vertical, just enough for the top brackets to clear the front edge of the slam-panel.

Next tilt the grille by pushing its bottom edge back as far as it will go, this should allow you to tilt it even more by bringing the top edge further forward to clear the end of the wing drain channels and the very front edge of the wings themselves.

Now you should be able to lift the whole grille up, keeping it square and tilted, with the outer edges of the grill passing through the small gap between the back of the bumper and the front face of the wing. If your gap between bumper and wing is too small to allow this, slacken the nuts securing the bumper to the chassis legs which should allow it to angle forwards increasing the gap.

This image shows the vertical edges of the grille and the reinforcing bar for the upper part of the rubber bumper. This bar sits in a channel across the top edge of the bumper, then bends back for a short distance and then downwards. There should be about 1" clearance between the edges of the grille and these bars i.e. they should not interfere with removal at all.

GT Screens

Roadster Screen

Seat Foams and Covers Updated October 2006

When planning to replace seat foams and/or covers pay particular attention to the shape of the foams and the covers. Despite reading three sets of instructions, including Lyndsay Porters otherwise excellent 'Guide to Purchase and DIY Restoration of the MGB', I only discovered that the foams and covers for the 4-synchro cars, at least, are handed after I had glued on the first set.

The seat foams and hence covers are tapered on the inside edge - the front is narrower than the back - I assume to take account of the shape of the transmission tunnel. Once glued on I did not want to risk damaging my new covers by ripping them off again. Fortunately the difference is not great, and although one can see the mistake if one looks for it, at least it does not stick out like a sore thumb.

I've looked at early and late seat frames at Stoneleigh and they all seem to have the same taper, but the early covers seem to be square.

Seat Webbing/Diaphragm

I changed the webbing on both seats when I restored the roadster in 1990/91 and the drivers side again about 10 years later. A few years later sitting in the passenger side I realised that side needed doing as well. The webbing hadn't broken as before, but had lost its tension. As such it came off relatively easily, but I remember some difficulty in fitting new webbing on all previous occasions, getting enough leverage to stretch the webbing enough to get the 2nd hooks into the frame, and doing it on my own. Previously I had used a length of timber and some rope, levering the timber against the frame with one hand, to pull on the rope looped through the main wire frame of the webbing, using one foot to hold the seat steady on the ground, and the spare hand to press the clip into the frame. It was a bit of a fiddle tying and untying the rope each time and getting the right length of loop, this time I decided to try something different, and it was much easier.

Remove the seat from the car by undoing the front bolts first - a bit long-winded with a spanner. This then allows the seat to be moved right forward so you can use a socket ratchet on the rear bolts, which is much quicker. If you undo the rears first you will probably have to use a spanner on all four. When lifting the seat out of the car watch the inner 'fixed' rail doesn't fall off and hit your bodywork, it is now only hanging on the sliding lip. The outer rail should be held securely by the seat-locking mechanism. Remove all the spring clips holding the seat cover onto the frame and peel the cover back as normal, and remove the old webbing. Don't chuck it yet, one of the hooks on the new webbing was missing when I received it, and another pinged off somewhere, the old ones were fine as replacements. I used a length of wire tied into a loop - the length of the loop is relatively immaterial. Slipped the loop over the hook, braced one foot against the frame of the seat, pulled on a tommy-bar through the wire loop such that the hook was in line with the hole in the frame, and when I had pulled it far enough the end of the hook dropped right into the hole in the frame. Easy-peasy. The spare hand can also be used to steady the seat, and if you sit on the ground you can use both feet on the seat frame! Sometimes the wire got trapped between hook and frame, sometimes it pulled out easily, but even in the former case a bit of levering with a screwdriver released it. Then pull the flaps of the covers round the frame, refit the spring clips, and refit the seat to the car. Only took a few minutes to get all hooks located. I'd imagine the same way with a diaphragm would work just as well.

Refitting the seat is the reverse of removal, again watching the inner rail doesn't fall off. Take this opportunity to clean and grease the sliding parts of the rails, making sure they are straight and flat. Also replace the wooden slats and alloy spacers if they are missing or (in the case of the slats) rotten. If the seat lock wasn't working properly, now is the time to tweak that as well. I've not found it difficult to locate the bolts (rears first as they are easier to see) by wiggling things round until they line up, but some recommend putting bolts up through the rear holes first (you can't get at the fronts) as locating pins.

Seat Rails by Les Bengtson

The MGB seat rails are made of a combination of wooden rails and two aluminium spacers directly under the four bolts (two per side) which hold the seats to the floor. The basic set up for bolting the seats in is thus:

There are two "L" shaped steel tracks with a curved over lip at the top of the upper arm of the L. These are the seat adjustment rails (Seat Slide set in the Moss Catalog). These adjustment rails have an upper section that is bolted directly to the base of the seat frame. This is the part with the adjustment locking arm, located to the front of the seat, which is used to allow the seat to be moved forwards and rearwards. The lower section, which has a series of square notches towards the front end for locking into by the release tab of the upper seat adjustment rail, is bolted to the floor of the car. Both the seats and the adjustment rails are "handed"--there is one for the right side and one for the left side and they are not interchangeable. The lower adjustment rail is bolted to the floor using two 1/4"-28 TPI bolts on each side.. Depending on the year of the car, and whether the bolts are original or not, they vary from 1" to 1 1/4" in length. These bolts pass through the aluminum spacers and, hence, into the floor of the car where they are engaged by captive nuts (nuts welded or otherwise secured to the frame of the car to hold them in place). These aluminum spacers (about 1/4" thick and 3/4" in diameter) are sitting in two holes in a wooden spacer strip (Moss "packing strip") which is a strip of wood 1/4" thick, 1 1/8" wide and 20 5/8" long. It has four holes bored through it, the innermost two of which are used for mounting the MGB seats.

These packing strips and their associated spacers are a problem. The wood tends to attract and hold moisture. This is especially true when rain or other liquid spills get onto the floor of the passenger cabin. This retained moisture in turn, promotes rusting of both the floor and the seat adjustment rails. Under certain conditions (constant humidity and not allowed to dry out), the packing strips can rot and come apart. To do away with these problems, I decided to make up some packing strips out of aluminum bar stock. It is relatively easy and can be done with simple hand tools.

First remove the seat from one side of the car. Slide the seat fully backwards and use a 7/16" box end wrench to remove the two front bolts holding the lower seat adjustment rail to the floor. Then, slide the seat fully forward and use either a wrench or socket to remove the rear two bolts. Carefully lift the seat from the vehicle. The lower adjustment rail which does not have the adjustment locking tab will tend to drop off as you move the seat. Remove it when you can by sliding out the rear of the upper rail. Then remove the other lower adjustment rail by releasing the locking tab (adjustment handle) and sliding the lower adjustment rail to the rear. Examine these rails for rusting. If rusted, clean them with either some form of media blasting, naval jelly or wire brush. Prime and paint them. (Priming helps the paint to stick better.) Set aside to dry.

Then remove the old packing strips and aluminum spacers. Use a 1/4"-28 Unified National Fine (UNF or NF) taper tap to clean up the holes for the hold down bolts. Clean up the bolts as necessary. They can be replaced if too rusty by using standard 1/4"-28 hex head bolts, 1" long. If you can find stainless steel bolts, they probably would be the best, reducing the likelihood of the bolt and captive nut rusting together.

After tapping the holes, lay down the new packing strips, re-assemble the lower adjustment rails to the upper adjustment rails (use a little grease to lubricate the mechanism) set the seat back into the car and bolt down. Then, check to see that the seats move forwards and backwards easily. If they did so when you removed them, they should do so now.

MAKING REPLACEMENT PACKING STRIPS OUT OF ALUMINUM BAR STOCK

You can purchase aluminum bar stock from several sources, including many hardware stores. You want a piece 1/4" thick, 1" wide and at least 20 5/8" long. This bar stock is most commonly found (in hardware stores) in lengths 4' long. One of these strips will serve to make two packing strips. Or, you can choose to purchase your aluminum strips from a metal retailer and have them cut to exact length, saving you from having to do this. Metal retail outlets are listed in the yellow pages under several headings including machine shop supplies.

First, assuming you have a four foot length of bar stock, measure 20 5/8" from each end and make a small mark using a Sharpie, magic marker, etc. Use a machinists square, tri square, etc. to draw a line across the bar exactly 20 5/8" from each end. Cut on the lines. Then, use the uncut end (the end that was cut at the factory) as your reference point for all future work. Measure from the end 2 3/8" and make a cross mark, then measure from the same end 18" and make a cross mark. The end you are measuring from is now the front end of the packing strip. Then, use your square to mark a thin line all the way across the strips at these points. Then, use a set of dial calipers to measure .400" in from one side of the strip and make another mark on top of the cross marks. If you only have a normal rule or tape measure to work with, measure in 3/8" (.375") and this will be close enough. (Note: The original wooden strips seem to have two holes at one end as the mounting points can differ, some need 17" for the second measurement and not 18". Check against the old packing strip, or if that is too badly rotted use the runner itself or floor tapping dimensions themselves.)

At the intersection of the two marks is where you will drill your two holes. Use a center punch to make a mark for the drill bit to start into or, better yet, use a center drill. Drill two holes of about 9/32" or 5/16". These slightly larger holes will allow some slight adjustment of the packing strip when being bolted in. Then, lay the strip into the, car and make sure the holes align properly. If you have done your work carefully, they should. If not, use a round file (chain saw file works well) to enlarge the holes so that they mate properly with the holes in the floor. The end that you originally measured from goes towards the front. The holes are off center in relationship to the centerline of the long axis of the packing strip. Put the short side to the inside of the seat and the slightly longer side outwards. This is easily seen by holding the packing strip against the lower adjustment rail. In the proper position the rail and packing strip will fit perfectly, while in the wrong position, they will overhang each other.

The new packing strips may be painted or left raw. I painted mine black using Rustoleum and allowed them to dry before assembly and installation. All of this work can be done with hand tools. It will result in a packing strip that is stronger and longer lasting than the factory wooden strip. It will also allow you to examine the condition of the seat adjustment rails and repair or replace them as necessary.

This article is copyright 1999 by Les Bengtson and may be reproduced for personal use as long as the copyright and authorship is acknowledged. Please direct any questions to: ragnar@aztec.asu.edu. Happy MGing.

Seat Mounts

There have been a couple of questions recently about measuring up new floor pans for the seat rail mounting points. I'd expect Heritage panels to come with the holes and captive nuts already attached, after-market panels certainly don't or at least not always. The following measurements were obtained from my two cars, both sides, which although showed they slight variations from panel to panel there was a consensus.

The distance between the hole centres for the two rails, i.e. across the width of the car is 12 5/8".

The fore and aft distance between the two holes for each rail should of course be governed by your rails!

The front holes are above the fixed cross-member, about 1" behind the centre-line of the jacking point in two raised pads. These pads are in a flat area between the front and rear flutes, which coincides with the cross-member underneath to give a flat surface for welding. For what ever reason the pads seem to be neither central to the flat area, nor central to the seat mounting points, the holes are displaced rearwards. If you use 1" behind the centre-line of the jacking point you should be OK, there is plenty of fore and aft adjustment of the seat after all!

The distance from the holes to the inner sill panel is 2 5/8" this should be enough to give a small clearance for sill carpets and not rub. It is important this spacing isn't too great or it moves the seat closer to the tunnel which causes two problems: 1) The seat can't be moved as far forward as it will foul the tunnel, and this also has an impact on accessing the rear bolts. 2) Depending in what type of belts you have the seat back may foul the buckle, which will also limit how far the seat can be moved forwards as well as how far the seat-back can be tilted forwards.

The distance from the holes to the tunnel is about 4 1/4" at the rear and 3" at the front, the reduced front measurement being due to the widening tunnel at this point. This occurs on both cars but may be different on 3-synch to my 4-synch cars.

August 2009: There are often questions about what to do about the front holes/nuts when replacing the floor or part of it. I had the same question after welding in a part floor and failing to mark, drill and weld the fronts before welding the floor in! I opted to weld the nut to a 1" strip of metal the width of the nut, cut a slot in the floor slightly more than the thickness of the nut plus strip, then with a bolt in the nut I was able to fiddle the nut through the slot, turn the strip at 90 degrees to the slot, and weld. But after reading other suggestions of a long bolt that goes right through the floor and crossmember (distorting the pan/crossmember if overtightened), or welding a drilled and tapped plate to the top of the floor pan (OK if it replaces the alloy spacer, but doesn't give enough thickness in my view), and how drilling holes in the rear crossmember to allow insertion of a socket to make installation of the gearbox mounting rubbers easier, it suddenly occured to me that surely the easiest method is to drill such a hole in the front crossmember, plugging it afterwards with a rubber bung (never mentioned in the case of the rear crossmember) to prevent water ingress and corrosion, although that box-section is hardly water-tight anyway, and drainage and air circulation might be beneficial.

Clausager reports that MGBs always had mounting bosses for three-point lap and shoulder belts, but they were dealer-fitted to North American Mk1 cars (and possibly locally manufactured) and until 1st April 1967 on UK cars when they became compulsory. The lap belt mounting points were always on the tunnel and sill either side of the seat, but the shoulder belt mounting point moved about a bit, and it is this that causes the confusion.

Originally as this picture from Clausager shows they were on the rear arch, low down and about mid-way from front to back.

On Mk2 North American roadsters, and cars for other markets from January 1971, mounting points on the tonneau panel were used which necessitated a quick-release device so the belt could be detached each time the hood was raised or lowered or the tonneau cover fitted or removed! The fastening utilised a plastic button, most have probably broken by now and been replaced by inertia reels. When I received my roadster it had inertia reels - but fitted to the tonneau panel. Very inconvenient as these had to be unscrewed to raise and lower the hood! This picture was taken on the 2007 New Forest run and shows a metal button, possibly a mod by the owner, and very neat.

It seems that when the tonneau mounting point was provided the wheel arch boss was deleted - but only from roadsters, so these cars from November 67 to December 71 only have the tonneau panel mounting point. However this may only be North American roadsters as UK roadsters didn't get the tonneau mounting point until January 1971. To retro-fit inertia reel belts to these cars, or static belts with a fixed shoulder bracket (the original style with the keyhole fixing having long gone), or if the original quick-release fittings on the tonneau panel are broken or missing, you will have to provide a mount on the inner arch. Probably the easiest way to do this is with 1/8" steel plates either side of the arch panel and bolted through. The belt should come with a spacer/boss that the bracket will rotate around even when the bolts are fully tightened.

Only when North American cars got inertia belts as standard in December 1971 was a mounting point on the inner arch reinstated, however now it is right at the back of the arch and just above the chassis rail, although the tonneau mounting points remained. Even though UK cars didn't get inertia belts until May 1977 they got the arch boss in this new position as well. The outstanding question is, did UK cars lose the arch mounting point from January to December 71? Or did they keep it but it moved position in December 71? This shows Bee's inertia reels fitted to this point, using an appropriate right-angle bracket to keep the orientation of the inertia reel correct (otherwise it can be difficult to pull the belt out of the reel for fastening).

GT's however kept the boss in the original position, there was no alternative mounting position for the shoulder belt until inertia reel mounting points were provided on top of the arches in December 71, and for whatever reason they decided to leave the arch points as they were. This from my 75 V8.

Another fabled gadget - the seat belt hook, possibly only fitted to 1980 models. This is on a GT, mounted on the rear waist rail, with an enlargement of the actual hook inset. On a roadster the hook was mounted much lower down on the trim panel, below the folded hood frame.

Once you have the nuts and washers off the studs you can lever the trim off the other fastenings. Start at one end with a screwdriver levering against a wood block resting on a cloth to protect the paint, then free the other end. You should be able to take one free end and lift the trim off the next fastening in line, sometimes lifting off one edge first then the other works best. By doing both ends first you avoid digging the end of the trim into the paint when taking it off the last fastener.

Drill out the old fasteners, which are held on with aluminium pop-rivets of about 2mm in diameter. You may have to hold the cup-shaped washer in a pair of pliers to stop it spinning round with the drill.

Fitting kits are available containing the full complement of pop-rivets with cup-shaped washers attached and six studs and nuts, sometimes washers too sometimes not, as shown in this picture.

However I have found that if a panel has been replaced from the trim down there can be a double-thickness of metal plus filler, and the pop-rivets as supplied are not long enough to go right through. Consequently they expand inside the hole and can work loose, sometimes when trying to refit the trim. Also if a panel has been skimmed, painted and drilled a couple of times or more the holes can become large and uneven, so again the pop-rivets don't expand behind the holes but inside. I bought some long-reach rivets and transferred the cup-shaped washers over, then used an appropriately-sized washer behind the panel, so not only did the pop-rivet go through a skim of filler two panels and a washer, but it also secured firmly behind the washer no matter how large and irregular the hole was. The cup-washer removed from the original pop-rivet, a long-reach rivet, and a suitable washer can be seen in the picture.

When preparing to fit new fasteners to the body hopefully the painted holes will be large enough to accept them. But if not, slightly overdrill the hole and put some paint on the bare edges. In any case dip the fasteners in Waxoyl before putting them in the holes in the panels. Even if you have a long-nose pop-rivet gun its nose may be too big to go inside the cup-washer. Don't be tempted to fasten them like this as the rivet gun may expand the washer and make it much more difficult to get the trim on, slip a small nut over the pin on the rivet so the gun bears on the nut and the nut on the cup-washer. Have a supply ready as they jump off and get lost as the pin breaks.

Before attempting to fit the trim to the car go carefully over the edges and make sure there are no sharp flashings that will cut into the paint.

Added October 2008

Drive train component weights
Engine, dry with clutch, 18G/GA	358lb
Gearbox, 3-synch (OD or non-OD not specified)	78lb
Rear axle, banjo, disc wheels	117.5lb
Rear axle, banjo, wire wheels	123lb

As far as polish goes Mer is quite popular in the UK, especially with the concourse brigade it seems, but I find it leaves white powdery traces in the crevices of my less than concourse but rust-free black-tulip roadster and these only seem to appear over time several hours or days after polishing, so I keep finding new bits, and keep polishing them off makes me look like an anorak. The benefit of Mer is that it doesn't take any colour off onto the cloth. Turtle Wax does take a little off onto the cloth, which implies it is a mild abrasive, although I have been using it two or three times a year for 12 years or so and haven't worn through yet, but is much easier to apply and polish off and gives a good deep shine. T-cut and other 'paint restorers' should only be used to remove very fine surface scratches, as part of the process of repainting a panel, or on a body that has not seen washing or polishing for many a long day. More frequent use will wear paint down to the primer.

Tree Sap: I haven't needed it on either MGB but my Son's Celica that had been street-parked in London for several years was covered in thousands of very tiny clear raised spots which no solvent I had including WD40, Plus Gas (penetrating fluid), white spirit, petrol, meths plus cutting paste, bug wipe tissues etc. would touch. The only thing that did was 1000 grit wet and dry, but I didn't fancy doing the whole car in that! In the end I found another kind of 'abrasive' in Halfords which was clay-based and like a harder than normal Plasticine. This took the lot off without that much effort and left the surface in perfect condition, and although you could see the 'dirt' from the sap (or whatever) in the clay and had to keep turning it to a clean surface, it removed no paint. Two years since then and it has been fine. Subsequently I get my Son's BMW Z3 M Coupe to look after as well now they have a son and a 'sensible' car and the BMW was languishing unused and getting bashed on the streets of London. This had the same spotting, and the same clay block removed that as well.

Update 2007: The packaging and formulation of Turtle has changed. Whereas it was a creamy liquid it is now almost like water. It doesn't give the depth of shine on the roadster that it used, and incidentally takes on no colour either. It seems fine on the ZS so is probably formulated more for today's water-based paints than the older cellulose. I have some fairly recently purchased Autoglym Super Resin Polish and that seems to have the creamy consistency that Turtle used to, and gives a better shine than Turtle does now, so that is my product of choice for the moment. Autoglym also sell Paint Renovator (amongst more than 40 products just for retail and cars, plus others for trade, motorcycles, PSV and others!) which is a mild abrasive. This should only be used if the resin polish doesn't remove any surface marks, but is good for removing over-spray on a partial panel respray. An even stronger abrasive, but still less than 1000 grit wet-and-dry is Halfords Cutting Compound. Neither this nor Paint Renovator contain silicones like the Resin Polish do so can be used before painting as well as after.

Update 2008: Looks like what I have is Turtle High Gloss Wax, but Turtle Wax Original is also still manufactured so maybe that is the one to go for - when you can find it.

Update 2009: But then again ... while I was using the Turtle High Gloss I noticed the finish was gradually getting duller and viewed at an angle with my glasses on (never a good idea) the bonnet was showing fine surface scratching. "Oh well" I thought, "I never expected the paint to last this long (20 years) anyway". But when I started using the AutoGlym the shine immediately came back and the scratches went, for several outings anyway. I had always been concerned that the original Turtle showed a lot of colour on the cloth, i.e. it was taking a microscopic layer off each time, although it hasn't worn through (yet!) anywhere. The Turtle High Gloss didn't take any off at all, which on the face of it is better for the paint, but as I say the finish suffered. With the AutoGlym, whilst it does show some paint on the cloth, it doesn't seem to be as much as the Turtle original did, but gives a much better finish than Turtle High Gloss. So on balance I'll be sticking with the AutoGlym.

Removal of the glass itself is quite straightforward. Remove the waist-rail strip with the brush seal and the upper stop-peg near the top of the inner skin. Then with the window part-way up remove the bolts securing the regulator to the inner door skin, support the glass while sliding the wheels on the regulator arms out of the channel across the bottom of the glass, and lift the glass up and out. If you need to remove the regulator as well then fold it up and it will come out of the large hole in the skin. I got both mine out easily enough, so much so I didn't really pay attention to exactly how I did it, then had a helluva game getting the first one back in! You have been warned.

Adjustment is really limited to the stop-peg near the top of the inner skin so that the top of the glass just tucks neatly into the rubber piece at the top of the 1/4-light frame. The bottom stop is by way of a plate welded into the door, and so not adjustable. This should have a felt pad gripped by some serrated teeth so the bottom of the drop-glass doesn't drop onto it with a clatter or rattle when fully down. Whilst the adjustment at the bottom of the 1/4-light long leg will of course affect the drop-glass, it's primary purpose is to get the correct alignment to the windscreen upright. The plain channel at the lock end of the door also has a certain amount of adjustment in two directions, but apart from pressing the channel onto the glass gently while tightening it to stop the glass jumping out as it is wound up and down, the other adjustment should be allowed to take up its natural position, which will be controlled by the front channel and the brush and rubber seals against the glass.

Stiffness when winding seems to be the biggest problem, so much so that people break the plastic handles and replace them with after-market metal ones. With several BL vehicles over many years I've never had this problem, but as I say many complain of it. The glass slides up and down in 'flocked' rubber inserts fitted into the metal channels of the 1/4-light frame and another plain channel at the lock-end of the door. These rubber strips can rip and cause the glass to jam, and if the flock wears off the rubber may well make the action a lot stickier. The two channels may also not be in line so are trying to flex the glass as it moves up and down, and the regulator mechanism may also be stiff from rusting or lack of lubrication. The first thing to do is to determine whether it is the regulator or the glass that is causing the problem.

There is quite a bit of lost motion in the regulator and this can be used for diagnosis. Wind the window down, then start winding it up. Now grasp the top edge of the glass and lift - it should come up at least an inch or so. Now turn the winder again in the 'up' direction. It should move easily for an eighth of a turn or so before it starts trying to raise the glass some more, then you know the problem is with the glass and the channels, and it was probably very stiff when you tried to pull it up by hand. But if the glass moves relatively easily when pulling it up by hand but the winder remained stiff while taking up the lost motion, then you know the regulator is the problem. In theory if the glass gets easier to move as it is wound up that could well point to the lock-end channel being the problem, as there is progressively less of that in contact with the glass as it rises. But note that the regulator should move veryeasily back and fore through its lost-motion area, this can be checked in various places between fully up and fully down, but pulling up the glass by hand will always exhibit a certain amount of resistance. Of course you could have both problems! In which case deal with the regulator first, then you will know how much of a problem the glass is. There are quite a few pivots, gears, wheels, and a large spring in the mechanism, any one or more could be siezed or stiff, so you will have to work through them all. If the car is a daily driver and is regularly going to get wet use waterproof (white) grease to avoid rain running down through the doors washing it off. For 'dry' cars a general purpose grease should be fine.

I've seen silicone lubricant spray recommended for easing the glass channels, but what long-term effect it has on the rubber, and whether any short-term benefits are then replaced by a worse long-term situation I don't know. Much better is a suggestion from Mark Robinson to use graphite powder, put some on your finger-tips and you can feel how slippy it is, but it never attracts dirt or gets sticky, which is why it is used for lubricating locks. Making your own like for door locks probably isn't feasible in this case due to the quantity that will probably be required. It doesn't seem to be widely available, locksmiths may sell it but in quite large and expensive quantities, Tooled-up and Graphite Store are two possibilities for small quantities. However the point is that like metal winder handles, it shouldn't really be necessary, and it is better to fix the root cause once and for all rather than bodge it and have the problem arise again later. Examine the condition of the rubber inserts and if in any doubt as to their condition, replace them. Physical wear and tearing of these can be caused by incorrect adjustment of the lock-end channel or buckling or crimping of either channel so this should be checked before using new inserts. It's possible new non-OE rubber inserts are too fat or too sticky, which leaves you with a bit of a problem. It will be possible to open out the lock-end channel a little which should at least halve the problem, more care will be needed for the 1/4-light frame, especially the upper half.

But first I wanted to see how effective they were. I bent a couple of brackets from aluminium curtain track to slot in the tonneau bar sockets and wired a piece of old plywood to them. Couldn't see behind me using the interior mirror of course, but it was good enough for testing on a quiet road. I was surprised how effective it was. At 40mph there was no buffeting at all, whereas you can feel it start from as low as 10 or 15mph without, and at 70 mph there was still hardly any, so the solid ones certainly are effective. But I had in mind a fabric one so I could at least roll it up and tuck it away when I was not using it.

A neighbour had given me some aluminium sheet many years ago - her husband had been in the car industry in design even further back and this was some 3/16" thick stuff he had used for chassis and body frames, it was the ideal thickness for a snug fit in the tonneau sockets. I cut and shaped a pair of brackets that plugged in the sockets and more-or-less followed the line of the folded hood frame behind the seats nearly to the horizontal line of the tonneau cover over the folded hood. These ended in a vertical post about 10 1/2" high that was going to be the support for the fabric. I say 'vertical' but they are inclined in towards the centre of the car to follow the line of the screen siderails when viewed from the rear as the adjacent pics.

I drilled holes at the top and the bottom of these posts to thread 'shock cord' like bungy or those spider things used to secure suitcases to roof-racks, but much thinner at about 3/16" diameter - about £3 for 3 metres. These were going to run across the car to support the top and bottom of the fabric. A length of poly tubing to go over the vertical posts to act as a smooth support for the sides of the fabric to stop fretting and wearing through of the fabric - about £1 The bungy runs vertically through these tubes to avoid having any knots at the top and hence clearly visible.

Black veiling material seemed close to what I had in mind, although the holes were a little bigger than I would have wanted and the threads 'holding them together' a little thinner, but for £1 for enough for four single thicknesses, two double, or one quadruple it was cheap enough. Before cutting the material I did test runs with one, two and four thicknesses held on to the frame and bungys with clothes pegs to check effectiveness. As I suspected even four thicknesses was not as effective as the solid board, but was beginning to affect visibility. So I decided to opt for a double thickness with clear polythene in the middle. After the test I noticed that the brackets had tended to work their way out of the sockets a little. I didn't think it would come out altogether but neither did I want to risk it, so I drilled and tapped the tongues of my brackets for a cheese-headed screw, which I then ground off at an angle. This would enable the tongue to slide in without difficulty, then the high edge of the screw head would engage in a convenient hole that was already in the socket. To remove the bracket from the socket one has to pull against bungy tension to disengage the screw head from the hole before the tongue can come out of the socket - impossible in use.

I had trouble finding some polythene that was clear for distant objects (most 'clear' stuff is fine when wrapped around something but opaque when looking down the road through it) but eventually found a large bag that had originally contained a folding garden chair - £0. Rolled the top and bottom edges of this around a length of dowel to form a tube to thread the bungy through, stuck down with double-sided tape. The same neighbour who had given me the aluminium was very interested in how I was planning to use some of it, and used her machine to sew up a double-thickness of the material so that it slotted over the verticals and I could thread the bungy through from side to side to make the finished article.

So for a fiver and some time I had something the Navigator declared herself well pleased with during our recent visit to the New Forest, but there are a couple of areas for improvement. Both concern the polythene which is very thin and was chosen for its visibility: Because it is so thin the buffeting that it is now stopping reaching us is causing it to flap which is quite noisy, it also bows out under the air pressure and this can result in reflections from the sun which reduce the area I can see through, so the hunt is on for a stiffer but equally transparent substitute (hood rear window material perhaps?) for MkII. I might also extend the screen to the tonneau cover to reduce the draft behind the seats ... or is that going to far?

Update: Put a couple of thin metal rods vertically between the top and bottom bungies about 1/3rd the way in from the ends. This keeps the bungies at the correct distance apart and supports the front of the screen, both of which reduce the bowing, flapping and reflections.

Update May 2005: In very blustery conditions on the A14 passing Cambridge the bracket my side broke where the horizontal strut is turned upwards, but at least the remaining half still sheltered the Navigator. I tried welding it but it wasn't going to work so I fabricated a new one out of the large amount of aluminium sheet I still have from when I originally made it. If I get another breakage I'll get another pair of tonneau sticks and modify those.

Update May 2007:

Must be something about May. On the way down to the Tamar Valley Run the Navigators side broke this time. I was able to jury-rig another support out of a jack handle and some wire, but wasn't going to fiddle about with aluminium any more as it obviously can't take the stresses. I'd thought about modifying tonneau sticks, but it was going to be some time before I could pick up another pair at a spares show and needed it before then, and didn't want to chop up the pair that came with Bee. So I toddled off to B&Q to see what they had and was surprised to see quite an array of alloy and steel bar and tubing both round and square, and got a length of bar and two of tubing of different sizes, all in steel. The part of the old strut that goes into the tonneau sticks socket and horizontally rearwards seems strong enough, it is the angle where the horizontal section turns upwards that has broken each time. So I shortened the horizontal section by about half and with a bit of swaging fitted it into a length of the larger diameter tubing, cut to just before the turn upwards. I was going to use the bar for the turn from the horizontal and the vertical section and it had to be small enough to go through the netting and plastic tubing along with the bungy, which made it too small to fit into the tubing used so far. But with more swaging I used the smaller tubing as an intermediate, and the three were welded together. I bent the bar to the appropriate angle, cut it to length, welded a couple of brackets for the bungy where it turned across the car, fitted it and the plastic tubing and netting, and job done. A couple of quid and a few hours fun in the garage, with plenty more bar and tubing left over for when the other side breaks again.

Which it did June 2008 on the Bridgend Vales and Valleys Run. Another couple of hours using the passenger side as a pattern and the drivers side should now be OK. However the internal plastic membrane needs a bit of attention as the corners have come unglued and are wrinkling up, but I'm pretty amazed both it and the material have lasted as long as they have given the battering it gets on long journeys.

Update July 2009: No further problems yet but whilst the mesh fabric is holding-up the plastic inner sheet is getting a bit ratty and opaque. Idly Googled 'windstop fabric' one day and this link to a document explaining how to repair an MGF/TF windstop comes top of the list. Not a lot of use to me in itself, but it includes a link to Ikea for some apparantly suitable fabric at only £1.36 a roll! Not dead handy to me, and not available online, but worth a look the next time I'm passing.

If that isn't enough you can get small, convex stick-on blind-spot mirrors that can be attached to a flat-glass door mirror e.g. as here. They are called 'blind-spot' mirrors but some of them are so convex that you can see almost as much as in a flat mirror, but everything is made exceedingly small. They should be fitted to the outer upper corner of standard mirrors. And of course most cars these days come with the outer edge slightly convex anyway so don't need them. My Toyota came with one on the drivers side, but I took it off as adjusted as above I found I didn't need it and it just reduced the clarity of what was in the mirror.

Recently someone took the rather superior view that Americans always have their mirrors adjusted correctly as they can be passed both sides. For a start what is important is to be sure there is no one in the lane you want to move into before you move into it, which side you can be passed is irrelevant. But Americans have those annoying 'bifocal' door mirrors, which are fitted precisely because they are no better at adjusting their mirrors than anyone else, and nanny state has decided to intervene. These mirrors use a convex lens effect on the outer portion so that even when they are adjusted such that the plain part shows what is behind the car the convex part shows what is on the rear quarter. Hopefully. Of course that gives a distorted view across the mirror as a whole, so they also need to be shouted at with 'OBJECTS IN MIRROR CAN BE CLOSER THAN THEY APPEAR'!