Nov 132014

Time pressures delayed the full electrical shakedown testing until after the engine was up and running. Power is only required to the ignition and fuel pump circuits to start the engine so all the other fuses were removed. Having said that, I did end up fitting the fuse for the instrument voltage regulator in order to obtain oil pressure and temperature readings while the engine was running.

The other fuses were now fitted in turn, stopping to test each of the components they fed, before moving on to the next fuse. The resulting snagging list was encouragingly minor:

  • The main beam can’t be flashed from the indicator stalk
  • The original hazard flashing unit is on the blink!
  • The brake fluid warning light isn’t coming on
  • The wiring to the rear brake and sidelights have been crossed over
  • One of the cooling fans is a little noisy and spins the wrong way!
  • The heater fan is making contact with the housing
  • Neither the washer or wipers work
  • Only one horn worked

Overall I was quite pleased with that for a first test. Especially as it was the first time the cooling fans and wiper motor had been tested since I rebuilt them and the washer, wiper, brake fluid warning light and horn were all simply missing earth connections. So easily solved.

Indicators and hazard lights
There was a fair amount of head scratching when wiring up the hazard warning light switch. The part number identified it as a hazard switch from an XJ6 and its pin connections conflicted with those in the Illustrated Parts Manual!

Hazard wiring diagram Hazard switch wiring

So I attempted to go back to first principles (or my understanding of them!) to work out the correct wiring. The one thing that was puzzling me was why separate indicator and hazard flasher units were used. They are essentially performing the same role – converting a constant DC supply into a square wave, with a suitable duty cycle to provide the correct flash duration and frequency.

The light bulb moment came when reading the requirements to pass an MOT. The hazard lights must be able to operate without a key in the ignition while the indicators are only powered when the ignition switch is on. The other difference is the power rating as the hazard unit needs to draw more current drive both sets of indicator lights at the same time.

The duty cycle of the unit is achieved by a bimetallic strip which expands and contracts depending on whether current is flowing. When the indicators are on, the current generates heat in the two metals, which expand at different rates. This causes the strip to move away from its contact and break the circuit. Without a current, the strip rapidly cools and the circuit is re-established allowing current to flow once more and the process to repeat.

Therefore the flashing frequency or ‘duty cycle’ is directly related to the rate of expansion/contraction of the bimetallic strip, which is a function of the current flowing. This is why modern LED indicator bulbs often do not work on classic cars as they draw far less current and therefore may not generate sufficient heat to switch the traditional units. Modern transistor based units are available to overcome the problem.

Therefore the switch needed to have the following connections:

Hazard switch in OFF position

  • Indicator flashing unit is introduced into the circuit (the two green wires are connected)
  • Hazard flashing unit is cut out of circuit (connections between the LGN, GR & GW wires are broken)
  • The indicators can then be operated via the switches built into the indicator stalk

Hazard switch in ON position

  • Indicator flashing unit is cut out of circuit (by removing power to its green wire)
  • Hazard flashing unit is introduced (by connecting its LGN output to both the GR & GW indicator feeds)

The only problem was the random frequency of the hazard flashing which was easily solved by fitting a new flasher unit.

Cooling Fans
The S2 cars have an otter switch to turn on the cooling fans at its rated temperature and requires a special connector, which has been on back order for ages. A spare dash switch was rigged up in its place so the fans could be switched on and off at will, which would be useful when testing and tuning the engine. The first fan spun up and ran smoothly and quietly, which was very pleasing, as they hadn’t been bench tested after their refurbishment.

However the operation of the second fan was noisier because the blade was not quite square on the armature shaft, causing vibrations. Removing it was a fiddly process with the engine bay completed so I was cursing not having bench tested it!

But more importantly, the motor was rotating the wrong way, pushing air forward towards the radiator rather than drawing air through it. Its effect would be worse than having no fan at all. When travelling at speed there would be a natural flow of air through the radiator. Having a fan blow against this flow would reduce the cooling ability.

It is rather odd because reversing the supply polarity of series wound DC motors (and indeed shunt wound motors) does not reverse the direction of rotation. The only way to rectify this would be to reverse the connections on either the field or the rotor windings (but not both!).

The rotating force (torque) on the armature shaft is the result of the interaction of the magnetic fields of the armature and the field winding – opposite poles attract, like poles repel. The directions of these magnetic fields are dependant on the direction of the current following through them.

Therefore changing the polarity of the power supply does not reverse the direction of rotation because both the field and armature currents will be reversed and therefore both magnetic fields. A double negative if you like, that cancels each other out!

The only practical option for these Lucas motors was to reverse the field winding, which was a very simple job.

Field wires just needed swapping All sorted – an easy soldering job

The fan now rotates in the correct direction to draw air rearward through the radiator. The fan was also re-seated to stop the fan vibrating so much.

I was trying to fathom an explanation as to why the motors were rotating the wrong way and then recalled that I had difficulty dismantling the two motors I’d acquired many moons ago, at the start of 2012. They were of the same design but had the fans mounted the wrong way round.

The person had indicated they were selling them as they had upgraded to modern fan units. However I suspect they had sourced motors from a different vehicle and noticed they were turning the wrong way. They had then mistakenly concluded mounting the fans the wrong way round would rectify the situation, rather than just improve their efficiency in making things worse!!

Wiper motor & Intermittent module

Intermittent control above
brake warning light

I’d decided to fit the Hella intermitted wiper module that other E-Type forum members had advocated. It works independently from the two speed wiper switch on the dash and the modification is reasonably discreet and reversible. The intermittent control uses the hole in the dash for the rear window heater on the FHC, which is blanked off on the OTS cars.

Like the hazard warning switch, the part number on the wiper motor switch also suggested it had originally come from an XJ6. The problem was the connected terminals were different to the correct switch (OFF 5-7, Normal 4-5, Fast 2-4) and so it was another case of trying to work out how it should be wired.

The combination of common sense and trial and error eventually produce the correct operation, which differed from the wiring suggested by the forum post. I’ve wired intermittent module into the ULG feed rather than the suggested YLG wire, which is for operating the motor at high speed.

At the same time, I also decided to add a jumper lead (with an inline diode) from the washer switch to the slow speed supply to the motor. Therefore the wipers now operate automatically while the washer switch is pressed. The diode is needed to stop it working in reverse, with the washer operating when the wipers are switched on!

Heater Motor
A complete new heater unit had to be fitted as the original had largely rusted away. The resistor, providing the dual speed operation, is riveted to the base plate of the heater motor. I’d foolishly assumed the new units would be supplied complete with the resistor already attached – dream on!

The resistor is sold separately and, due to the lack of clearance, requires the base plate to be removed from the motor in order to fix it. The wires are then soldered to the resistor rather than via spade connectors. As a result, the loom has bullet connectors built in, approx 6″ from the end of the wires, to allow the motor to be separated from the loom if it needs to be removed for servicing.

Resistor provides 2-speed operation Loom wires are soldered in place Fitting was tricky due to
the engine frame

However the looms didn’t come with the rather crucial two wires needed from the resistor to the motor terminals! The other issue came trying to re-fit the motor to the heater housing which had already been attached to the bulkhead. Removing it wouldn’t be an easy option as the cooling system had now been filled.

The reason the motor and fan cage couldn’t be fitted onto the housing was the base plate fouled on the engine frame. It was necessary to detach the fan cage from the motor by undoing the clamping grub screw. This allowed the cage to be fed up into the housing and then reconnected to the motor once the motor was clear of the engine frame.

The fan just had to be re-attached ever so slightly nearer the motor to cure the fouling problem found during the initial shakedown tests.

Alternator testing
The electrical component that I’d been putting off testing was the re-wired alternator. The modifications to the alternator were more far reaching than any of the other electrical work and so there was more scope for things to go wrong. Once all the other electrical issues had been resolved, I fitted the alternator belt and prepared to start the engine. Would it work or would it blow any of the other components?

Unfortunately I didn’t have a suitable ammeter to measure the theoretical maximum output of approx 60 Amps and so my testing was limited to measuring the voltage at the battery terminals with a multi-meter:

  • With the engine off, the terminal voltage should be approx 12.7v
  • When running at idle, the alternator should raise the terminal voltage to around 13.9v-14.3v
  • The terminal voltage should reach between 14.3-14.6v running at 2500 rpm

In theory if the alternator output drops from the last two levels then it points to the failure of one or more of the rectifying diodes. If it rises above 14.8v, then the 4TR voltage regulator unit is not limiting the maximum voltage and needs to be replaced to avoid overcharging the battery. However I also have to consider that any lack of output could also be due to the additional diodes introduced to self-energise the field winding.

It was a welcome anti-climax that the initial test appeared to be successful. Although I’ll need to do further checks to ensure it’s charging when the car is run for a longer period.

Apr 162014

The target date for starting the engine has been drifting for some time now. In part because I was being a bit too methodical in my approach but mainly because I was getting side-tracked with non-essential tasks, such as refurbishing a second steering wheel!

With Spring in the air, I couldn’t use the poor weather as an excuse either. So a list of tasks has been written that need to be completed before it could be fired up …. for the first time in 18 years!

One of the tasks was to complete the cooling/heating system and fill with anti-freeze. The radiator had already been fitted when the bonnet was off to install the engine. The hoses had been put in place and just needed a few final jubliee clips tightened. So it shouldn’t be too time consuming to complete ….

Fortunately one of the things I’d been side-tracked with was trying to work out how the lower stone guard, between the radiator and bonnet frame, should be attached. In my haste to install the radiator, I’d forgotten the two rear brackets that fit between the radiator mounting bracket and the mounting grommets.

It wasn’t too difficult to lift the radiator in situ to insert them but the task would have been a lot trickier if the system had been filled and the hoses firmly attached.

I wasn’t too sure of the correct routing for the water pipe connecting the radiator to the expansion tank mounted on the bulkhead. The most logical path was along the top of the picture frame and then underneath the inlet manifold, strapping it to the side of the engine frame to avoid it getting in the way of the carb overflow pipes.

The original anti-freeze recommended by Jaguar was Bluecol, an ethylene glycol based anti-freeze which also contains various corrosion inhibitors. The ethylene glycol anti-freezes are very poisonous to humans and animals if ingested and so they were generally dyed green or blue.

Nowadays most cars use ‘advanced’ OAT (organic acid technology) anti-freezes, which is often dyed pink but, rather helpfully, is also dyed a variety of other colours depending on the manufacturer. Again these have inhibitors but they have been found to cause a number of problems when used in classic cars and are not suitable for systems with copper or brass components, such as the E-Type.

The inhibitors in both types of anti-freeze are consumed over time although more quickly in ethylene glycol. Therefore it’s important to change every two years to avoid corrosion problems. The OAT variety only require replacing every 5 years.

Another additive that some owners recommend is Redline Water Wetter which is designed to improved the efficiency of the cooling system. So 5 litres of Bluecol and some Water Wetter were duly ordered and I was ready to start filling.

Filling the cooling system
Or more precisely how not to fill the cooling system! The instructions suggest checking all the joints for leaks before filling first with 1.5 litres of water, then the Bluecol and finally topping up with water. All perfectly sensible. A quick check confirmed all the jubilee clips were in order and the radiator drainage tap was off, so time to start filling.

Just as the last 1/2 litre of Bluecol was being added came the dreaded sound of rapidly dripping water, which had quickly formed a pool underneath the heater area. Grabbing a handful of rags I dashed round to contain the leak. Hmmm …. bone dry. However, just to the left, water was gushing out of the engine drain tap. Muppet!!

With the tap closed, filling continuing without incident until about 12 litres had been added. Once again a torrent of water began to cascade over the EDIS coil pack.

Aaaargh! Water was pouring out around the long studs clamping the thermostat housing. The only option was to drain the system to lower the level of coolant until it was below the offending leakage.

I’d been advised that it wasn’t necessary to use a gasket sealant and the clamping force alone would be sufficient to seal the joints. This advice was now looking rather dubious as I mopped up the pool of anti-freeze that hadn’t managed to soak into the carpet.

Fortunately the ethylene glycol breaks down within a couple of weeks so will not pose a on-going health hazard. The gaskets were dried out and some silicone based gasket sealant purchase for the reassembly.

However on closer inspection it was an incorrect gasket that caused the leak. The interface between the inlet manifold and the thermostat housing is an odd one. The inlet manifold has small chamber off the main water channel. The chamber is capped off by a flat face area on the thermostat housing.

I can’t work out what this chamber is for and had assumed it was non-functional as far as the cooling system was concerned. I suspected it was probably providing additional strength to the casting. What I hadn’t spotted was a thin slot joining the chamber to the main water channel.

The triangular gasket supplied did not extend to cover the perimeter of the chamber and so was the cause of the leak, not the lack of sealant. A correct one was soon on its way from SNG Barratt. Even so I decided to use the silicone sealant just to make sure!

Once it had cured for 24 hours, the system was refilled without incident. With hindsight, I should have initially filled the system with plain water to confirm there were no leaks. It could then be drained and refilled, adding the coolant.

Dec 112013

The heater housing was too far gone as almost all the panels had either rusted through or were paper thinUnfortunately the heater box was beyond economical repair. The entire bottom section was paper thin and had rusted through in places. The side joints hadn’t fared much better and had rusted from within causing the joints to swell.

Every time the heater housing was rotated to inspect it, showers of rust fell from every opening! The motor and fan cage were also missing.

The heater matrix within is surrounded by a thick felt material and I suspect that this had acted as a sponge. The absorbed water had sat against the base and sides causing them to corrode over a prolonged period of time.

The padding material for the inside of the heater are available as a 'complete' kit ... excluding the padding between the heater and the bulkheadFortunately new heater units are available (although not that cheaply), so a new one was ordered along with a heater matrix and a kit containing all the various padding materials.

Oddly the padding kit didn’t include the square rubber seal fitted between the heater body and the bulkhead.

If you were going to the effort of replacing all the internal padding, it’s likely you’d be working on the fan housing off the car and so would probably need to replace the bulkhead seal too!

On the positive side, the new heater has an improved design for the fan cage which hopefully might address the reported problems with the original, asthmatic unit. The blades on the original fan cage were flat and aligned radially which isn’t the most efficient in generating a throughput of air. The new cage has curved blades angled towards the direction of rotation.

The heater motor can be switched between two operating speeds and is achieved by introducing a resistor into the circuit to reduce the voltage across the motor. It’s riveted to the motor’s mounting flange and the loom wires soldered in place rather than using spade connectors. I’d stupidly expected a new unit would come with the resistor attached!

The padding around the heater matrix is a thick fibrous material and it was a really tight squeeze to fit it all in. At least the matrix won’t be able to move around!

I found it was necessary to glue the square foam seal to the heater box with contact adhesive, before fitting the heat to the bulkhead. Otherwise, with only one pair of hands, it tends to fall out of place when attempting the fiddly task of fitting the mounting bolts while supporting the heater unit.

Heater Matrix & padding Bulkhead seal glued in place Heater unit installed

Fortunately the rubber connectors and ducting behind the dash were all present and in good order. So they only needed cleaning in soapy water to remove the grime that have built up over the years.

Sep 202013

There wasn’t time to investigate why the bonnet wouldn’t shut the last few inches, so I had a chance to sleep on it. The problem was that it’s not possible to see inside the bonnet area at the point its travel begins to be obstructed.

Only the engine and radiator has been installed so it shouldn’t be too difficult to work out. I remember reading somewhere on the E-Type forum that the Series 2 had an extra spacer somewhere in the engine mountings. Although I thought this only affected the carburettor clearance but these were still to be fitted.

Some off-cuts of the foam rubber Dynaliner were place strategically on all the high points. I was aiming to lower the bonnet to the obstruction and then inspect the Dynaliner. The foam would take a little while to return to its previous state so the indentations of any impact would remain and be easily traced.

The radiator support struts had been fitted at a jaunty angle so this was almost certainly the cause of the problem .... or was it?A couple of attempts still didn’t reveal an area of compression in the rubber …. more head scratching. It was definitely not the engine as this was almost entirely encased in Dynaliner.

It finally dawned on me – it must be the radiator support struts impacting the vertical sides of the air intake channels. When they were first installed, it was mentioned whether they should be fitted on the inner or outer side of the damping rubber grommet.

With hindsight, they did look angled when mounted on the outer sides. The brackets were removed and the bonnet was lowered to prove the theory. Spot on …. the bonnet reached the landing rubber without being obstructed.

The struts were reinstalled but this time on the inboard side of their grommet. The bonnet was lowered, having finished the job, only the problem was still happening!

I thought I'd found the issue had been caused by the radiator struts. Re-fitting them differently resulted in a gouge in the paint workHowever this time, when the bonnet was raised to investigate, it revealed a 4-5 inch scrape in the paint on the air intake duct. Gutted.

The strut positioning was only part of the problem. They were causing the ‘springiness’ of the impact because the bonnet edge was hitting the strut and just pushing it aside.

Once it had be repositioned and was out of the way, it allowed the bolt head, securing the strut to the radiator, to take a gouge out of the paint work.

The bolt head was sitting too proud because I’d used a washer between the radiator and the strut and one between the strut and the bolt head. Without them, it would probably have been ok. However I’m going to remove all the washers, fit a thin-headed bolt and reposition the radiator on the bottom mountings to try to centralise it with the bonnet.

 Posted by at 8:59 pm
Sep 162013

The heater valve was another part that was difficult to remove, as the bulkhead heater pipe had seized solid into the valve body. I didn’t want to apply heat in case it damaged any internal rubber seals and so I tried to break the joint by rotating the valve body. All this achieved was to deform the pipe, which eventually had to be cut to remove the valve.

There were signs of weeping from the valve so I suspected an internal seal had started to deteriorate and it would need replacing. The valve consists of a pot metal valve body and a plated end cap. The body has protrusions around its circumference which interlock with corresponding hook shaped protrusions on the end cap and then a single rivet stops the end cap rotating relative to the valve body.

The rivet was drilled out and then it was fairly easy to split the valve in two by rotating the end cap. This revealed the cause of the weeping – a sprung rubber diaphragm, that is used to control the passage of water, had become furred up.

The deposits had compressed the rubber seal in several places so it no longer made a complete seal. The rubber had also hardened over time so wouldn’t spring back fully once the deposits had been removed.

Even after extensive internet searches, I haven’t been able to find a supplier that just supplies the internal rubber diaphragm. Unfortunately the options are very limited.

Either purchase a complete repro valve or a repair kit from an American site, which includes everything but the valve body. However the kit was considerably more expensive than the repro valve, so I went for the latter.

Overall the quality of the new valve was fine, only let down by the finish of the valve body casting. It wouldn’t make any difference to the operation of the valve but I would have preferred to keep the original body.

Sep 122013

Quite early on the radiator had been sent off to Northampton Autorads to be ‘re-cored’. It was then stored for a number of years as progress with the restoration ground to a halt. Partly due to lack of time but also a lack of enthusiasm once it became clear how much work was involved in a full restoration.

It was only once the rebuild had restarted in earnest that the cooling fans and shroud were refurbished. Both, in different ways, had been a trial in perseverance to get the desired finish. So it was a real disappointment to find out all the mounting holes down each side of the radiator were too far from the edge to enable the shroud to fit.

There was no way I was going to hack the shroud to fit after the palaver to get the correct crinkle finish. Northampton Autorads usually have a stand at the Stoneleigh spares day, so I took it up on the off chance that they might be able to have a look at it and suggest the best way to modify it.

To my amazement they agreed it was wrong and simply replaced it on the spot, despite the considerable time since it was manufactured. I was quite willing to pay for the modifications as I should have reported any issues when it was first returned. I was impressed with their customer service!

Generally you learn from your mistakes, however I wrapped up the radiator on my return and set about tackling the ever growing list of rebuild tasks. It was some nine months later that I started to prepare for the big installation weekend, transforming it from a bodyshell to a rolling chassis, when I again tried to reunite the radiator and shroud.

I couldn’t believe it – the mounting holes were again out of alignment. Neither the shroud nor the radiator mounting brackets could be fitted. I sheepishly emailed them with the pictures below and they immediately arranged to collect the radiator by courier. This time the shroud and brackets were included so they could ensure it all fitted.

Only two shroud mounting holes aligned Radiator mounting bracket holes were also wrong
The lower mounting hole for the shroud didn't align. The problem is that this hole is also shared with the radiator mounting bracket The radiator mounting bracket holes were only marginally out but enough to stop the brackets being attached

Less than a week later, everything was returned all made up. I really can’t fault their customer service as they addressed the issues without question. I’m not sure why the manufacturing process is such that positioning of the mounting holes is so prone to error. I think nowadays, when a radiator is re-cored, all but the top and bottom sections are replaced so the original sides are thrown away.

At least the unit is now ready to be fitted when the engine goes in ….

The radiator, shroud and cooling fans all ready to go on as a single unit

Feb 142013

The dash heater controls operate plastic vent outlets on the underside of the dash, one in each footwell. When the vent is open, the air follows the passage of least resistance into the footwells. By closing the vent, this path is blocked and therefore the air is forced to exit via the dashtop windscreen vents.

The vents themselves consist of five interconnected vanes with the central vane connected to the dash control. Operating the dash heater control rotates the central vane, and with it the other vanes, between the fully open and fully closed positions.

Somehow the central vane of one of the vents has either been misplaced or lost during the constant sifting through the boxes of parts. Unfortunately the vents seem to be unique to the Series 2 and, as far as I’m aware, are not available any more.

After fruitless searches of the parts boxes and keeping an eye out at Stoneleigh spares day, I had to bite the bullet and start researching if and how I could fabricate a new vane. The problem is that without the central vane the vent is useless.

I think most plastic parts are generally injection moulded which isn’t really a DIY option. However there are some very low viscosity polyurethanes available that are suitable for moulding which may produce a good replacement. At least having two vents meant I still had a central vane to make a mould from!

A order was placed with MB Fibreglass Supplies who were very helpful in explaining the moulding process and several days later some RTV Silicone Mould Making Rubber (Polycraft GP-3481), Fast Cast Polyurethane Liquid Plastic Casting Resin (Polycraft FC-6720) and black polyurethane pigment arrived. Some white modelling clay (water clay) was also required, which can be obtained from most craft suppliers.

The first step was to produce a two piece silicone mould of the vane. Four ‘L’ shaped pieces of plywood were fabricated with a depth of around 3″ to make a mould housing. Using ‘L’ shaped pieces has several benefits; they can easily be moved relative to each other to obtain the desired mould footprint, clamping together is straightforward and they can easily be removed at the end without damaging the mould.

The mould housing is then half filled with the modelling clay and clay rubbed along the each of the corner joints to seal them. An off-cut of wood and some coach bolts was used as a mini tamping device. The vane was then pressed into the clay until the long lengths of the vane were flush with the clay (ie half above and half below the clay). Finally a number of indentations were made in the clay which will act as key for both sides of the mould.

It was now time to make the first half of the mould with the two-part silicone system, mixed by weight – 10 parts rubber to 1 part catalyst, ably assisted (hindered) by my two nieces who were on mixing and pouring duties. Being a red colour, it was easy to see when the catalyst had been fully mixed into the white rubber part. The mixture was then slowly poured into the mould housing, covering the clay and vane. The technique is to pour slowly and in the same place so that the silicone pushes out the air as it flows over the part being moulded.

The Room Temperature Vulcanizing (RTV) silicone normally cures in around 4 hours although I left it overnight as a precaution as it still felt tacky after 4 hours, probably due to the cold weather. The mould housing can then be turned over so the clay can be removed, to reveal the first half of the silicone mould with the clay indentations now appearing as small peaks.

Traces of residual clay were removed by wiping with a damped cloth to prepare for the making of the second half of the mould. Once dry, the first half of the mould was lightly brushed with Vaseline, diluted in white spirit.

This should act as a releasing agent stopping the second half of the mould sticking to the first. Some more two-part silicone was then poured into the mould housing as before and again left overnight to cure. Now for the moment of truth …. will two halves separate?

They actually separated very easily and the original vane came out without damaging the mould. The quality looked very good although the proof will only come once the new vane had been cast. The final preparation of the mould was to cut a conical channel for pouring in the polyurethane casting resin and an air vent to help prevent trapped air bubbles in the cast.

The polyurethane resin used was a two part product which naturally cures to an ivory white colour so a small amount of black pigment is required to get the desired finish. The mixture ratio by weight of resin part A, part B and pigment was 10:10:1 so the main difficulty was weighing the three parts accurately as the part only weighs 4 grams.

The resin cures in approximately 60 minutes so it wasn’t long before the first cast was ready. The initial impression was very good – even the original casting marks were faithfully reproduced. However the part was far too flexible so the nieces rudely declared it a ‘FAIL’.

MB Fibreglass Supplies were again helpful and thought the cure process had probably been compromised, most likely caused by having insufficient temperature in the component liquids when they were mixed.

A second casting was made after first heating the liquids on a radiator. This produced a much stiffer vane which seemed to stiffen even further once it had been removed from the mould and left on the radiator overnight. I now had two operational heater vents!!