Mar 212013

The fuel tank has been another area that hasn’t gone as smoothly as I had expected. At first glance the tank appeared to be fine however the problems were only revealed once it had been removed.

Over time the untreated tank had rusted through which is a common problem. It had been repaired by welding several replacement patches but the weld seams weren’t the best and had started to corrode quite badly.

My concern was the thickness of the tank, or lack of it, either side of the weld. I think water that had entered the tank must have been settling in the various troughs in the weld so it was again rusting from within.

As luck would have it, there was an advert in a Jaguar magazine for an unused later S2 fuel tank. The S1 owner had purchased the tank unaware that the design had changed from a single breather pipe to three from chassis number 1R1393, some 28 cars before mine! In fact I wasn’t aware of a difference until I’d read his advert.

The three breather pipes were introduced with the addition of the expansion tank into the fuel system, late in the production run of the S2 cars. I think this is why it’s not referenced in any of the manuals.

A number of owners on the E-Type forum have reported running problems as a result of fuel starvation. Insufficient venting of the fuel tank was enabling a vacuum to build in the tank, acting against the fuel pump. A common remedy is to drill a small hole in the fuel filler cap. I can’t see how the addition of an expansion tank would alleviate this problem. The S1 and early S2 tanks were just vented directly to the outside rather than via an expansion tank. So fuel starvation issues may still be a problem …. but more of that later.

The replacement tank been already been treated with the white PVC type tank sealer. However, during the time it had been kept in dry storage, the sealer had started to crack and come away in sheets. It would all have to be removed otherwise it would quickly start to clog the fuel filters.

I started to research the different types of tank sealers on the market, mainly to find out the best method to remove the sealant used in my tank. The conclusion was that the PVC type sealers have very good coverage but don’t fill holes or seams well and have poor film strength. I’d found that out to my cost!! Most recommended using a paint stripper to remove it.

I had a large tin of Nitromors lying about so I gave it a go and poured it into the tank along with some nuts and bolts. The latter providing a mechanical method of dislodging loose bits of sealant. The problem with modern Nitromors is that they have changed its properties to a more gel like consistency rather than liquid. As a result it wasn’t very good at getting into the baffled areas.

I then switched to Por-Strip which was better but didn’t reach much of the internal baffle surfaces. I needed a new approach.

One of the local powder coating firms offers a burn-off service but they didn’t want to do it as they were concerned the temperature would melt the brazing. Eventually I took it to a furniture restorer who was able to dip the tank. However the borescope revealed that there was still quite of a lot of sealant stubbornly attached.

I think paint stripper just softens the sealant which then becomes a sticky goo. Unless it is successfully removed immediately it simply sticks back to the tank surface as it dries out.

Like most troublesome issues, I put it to one side to have a ponder. I would tackle it at a later stage!

I recently found out that the PVC coatings can also be removed by dissolving in either Acetone or MEK (Methy Ethyl Ketone). Spurred into action once more, I ordered a small quantity of Acetone and added some sealant flakes to confirm it worked. 20 litres of Acetone have now been ordered and the tank is going to receive a thorough soaking at the weekend. Fingers crossed!

It still leaves the dilemma of how to treat the tank to avoid it rusting from the inside out. Not only that but it dawned on me that swilling around sealant inside the tank might cure the rusting problem but would be guaranteed to cause another. I’d realised that the PVC sealant had done its job and sealed the ends of all the internal pipes for the expansion tank connections.

Unless the Acetone can unblock them, I might have to resort to removing the brazed joints and withdrawing the pipes. I’m now wondering whether tank sealants might be the cause for some of the people suffering from fuel starvation problems.

 Posted by at 7:27 am
Mar 202013

A fuel expansion tank was added during the production run of the Series 2 cars, located in the boot space on the LHS rear wheel arch. The expansion tank is vented to the outside and so it’s internal pressure is always equal to the current atmospheric pressure.

Thermal expansion of the fuel increases the pressure in the fuel tank relative to the pressure in the expansion tank. As a result, fuel passes from the main fuel tank to the expansion tank until the pressures are equalised. As fuel is consumed or when the fuel contracts due to cooling temperatures, the pressure in the fuel tank decreases relative to the expansion tank. Any fuel in the expansion tank is then returned to the fuel tank as the pressures equalise once more.

Note: I have subsequently seen a Jaguar Service bulletin which indicated that fuel passes back from the expansion tank to the fuel tank by gravity, not pressure. Although with such a small bore for the breather pipes, I would have thought it’s probably a combination of gravity and pressure to overcome airlocks.

When the expansion tank was removed, I carefully labelled the various pipes between the two tanks with masking tape. For some reason I wrote the descriptions on the masking tape in pencil, which didn’t stand the test of time and were illegible by the time it was ready to be refitted. A lesson learnt!

A piece of 6mm jute was bonded to the tank as a protective layer between the tank and the wheel arch. However the bond was greater than the strength of the jute and so would need replacing. It’s available from most of the re-trimming firms and was bonded using the same AF178 contact adhesive used for the heat insulation.

I now had to work out how to make the connections between the two tanks as it’s not covered in any of the manuals available. The expansion tank has four outlets and the fuel tank only three. The additional outlet on the expansion tank is for the vent which exits to the outside via the boot drainage pipes. Although I believe the vent is connected to the emission control system for cars supplied to the USA.

The first task was to work out the difference between the four internal pipes in the expansion tank. I used a length of garden wire with the end bent over to form a hook. This enabled the wire to be jiggled so that the hook engaged with the end of the pipe inside the tank and therefore could determine the internal length of the pipe.

Three of the pipes ran from the bottom to the full height of the tank while the fourth terminated as soon as it entered the tank, as depicted in the photo. The short red pipe (A) is at the bottom of the expansion tank and therefore must be for returning fuel to the main fuel tank. So I would expect this to be connected to a pipe which terminates fairly high up at the top of the fuel tank.

The other two blue pipes (B and C) terminate at the top of the expansion tank and so would be for the pipes passing fuel from the main tank to the expansion tank, when the main tank is full to the brim. The fuel entering the expansion tank via B & C would then fall to the bottom, to be returned via pipe A when the pressure in fuel tank reduced.

I would therefore expect B & C to be connected to pipes which would normally only be submerged in fuel when the tank is full, ie terminating at the very top of the fuel tank. It was now time to get the USB borescope out to investigate the fuel tank as ends of the pipes are hidden due to the internal baffles.

In fact the corresponding red pipe A in the fuel tank was easy to determine as the end of the pipe end can be felt via the large oval opening. The borescope did confirm that the other two pipes terminated at the top of the tank. The end of pipe A turns downwards for about an inch. I assumed two pipes are needed to pass fuel to the expansion tank due to the baffles. It shouldn’t matter how the B & C outlets are connected, ie B to B or B to C, as they are both performing the same role.

Anyway, assuming my logic is correct, I think I’ve worked out the correct connections!!

 Posted by at 10:35 am
Mar 042013

The final task to complete the heat insulation was to install the Koolmat in the cabin area. My main concern had been the ability to be able to bond the underfelt to the silicone side of the Koolmat. A number of adhesives were tested using off cuts of Koolmat and underfelt. Normal silicone sealant was found to be the best product by a considerable margin, which alleviated my fears.

However I’m now leaning towards replacing the underfelt with some 1/4″ Dynaliner when the final trim goes in. The combined thickness of the Dynaliner and Koolmat would be closer to the original bitumen sound insulation with underfelt and so there shouldn’t be clearance issues with the central console. Also the underfelt was destroyed when it was removed during the adhesive tests and so would probably cause more headaches if (when!) the trim needed lifting in future.

The Koolmat instructions suggest starting the installation with the toe box and then work backwards. I decided to do the two panels under the seats first, rather than dive straight in, as I’d not used the ALPHABOND AF178 high temperature contact adhesive before. This would provide a few easy panels to become familiar working with the adhesive before having to assume contortionist positions to install the toe box area.

Holes were cut in the Koolmat for the seat belt mounting and seat runner fittings, the latter would be screwed in place during the gluing to ensure good alignment. However, when the Kootmat panels were trial fitted, they would not lie flat on the floor as the ends of the Huck bolts for the radius arm mounting protruded above the floor pan. To overcome this an off cut of brake piping was used as a punch to cut out suitably size holes in the Koolmat, just about visible in the photo above.

The AF178 contact adhesive needs to be applied to both surfaces and then allowed to become touch dry, when it no longer lifts away when touched. The Koolmat can then be pressed in place. I chose to apply the adhesive using a brush and purchased a small wallpaper roller to apply pressure once in place, which was a good buy.

The contact adhesive certainly lives up to its name as there’s almost no opportunity to reposition the Koolmat once the two surfaces have made contact. As a result, the alignment of the first few panels was acceptable rather than perfect. I was glad I had started with the easier under-seat panels first. I then started adding alignment markers on the Koolmat and bodyshell during the trial fitting which help enormously during the final fitting, when you needed to get it right first time.

Martin Robey replacement floorpans had been fitted during the bodyshell rebuild. These are manufactured to cater for both the non-flat floor Series 1 and 2 cars and therefore have two separate sets of seat runner mounting points. The unused mounting points and floorpan holes were blanked off with short bolts and blanking grommets (9mm & 19mm) respectively to ensure the floor was watertight. I’m assuming the various floorpan holes are to provide drainage should it ever be required.

The instructions on the AF178 tin are conspicuous by their absence. I wanted to take my time and planned to install the Koolmat over several days, and therefore needed to clean the brush. The manufacturer’s ‘technical’ department wasn’t particularly helpful with their suggestion to use ‘a solvent’. I would never have thought of that! When pressed on the type of solvent, they didn’t know, and it took trial and error to determine it needed to be cellulose thinners.

I wasn’t too impressed with the accuracy of pre-cut kit supplied by Koolmat and if I were to do it again would probably purchase a roll of Koolmat and cut it from templates I made myself.

I think the person cutting the kits must have been working from memory and from a car they’d only seen once. I very much doubt they had ever fitted one of their kits!!

The issues that needed to be worked around were:

  • They suggest making incisions at each corner for the piece covering the lowered floorpan area. A much more effective way to mould it to the floorpan and avoid crumpling was to cut out narrow wedges perpendicular to the length of the floorpan as shown in the photo above.
  • They provide a piece to cover the area between the lowered floorpan and the toe box. This piece bears no resemblance to the area it is intended to cover. In the end I cut this into two small pieces for the floor, in front of and behind the lowered floorpan, and two very narrow strips to run along the outer edge.
  • From their installation photos, Koolmat simply cover over the gearbox cover. Therefore the two lower side pieces for the transmission tunnel around the gearbox were far too high and required trimming. They now finish just below the transmission/gearbox cover so the cover could be removed for maintenance.
  • The same transmission pieces overlapped the rear transmission tunnel piece by such an extent, that it was almost pointless fitting the latter.

To be fair to Koolmat, at least they provided a reasonable amount of off cuts to be able to complete the job where gaps existed.

The gluing of the Koolmat to the gearbox cover will wait until the trial fitting of the interior trim, including carpets, just in case there are issues with clearance. Finally a bead of silicone sealant was run along all the joints in the Koolmat.