Zel's Fleet Blog - BX, Jag XJ-S, Sinclair C5, Mercedes 208D & 230TE, AC Model 70.

[mickthemaverick]

If you go down that route Zel, I'd suggest you couple it to a decent generator and then feed the output to 4 hub mounted motors via a quadruple speed controller. Then you dispense with the steering system and install twin power walking poles connected to two horizontally opposed rheostats each mounted in a front to back configuration to allow tank style directional control

[MattBLancs]

6V92TA transversely mounted behind the front seats?

A with a nod to originality (i.e. keeping to two cylinders) you need Mr Milliard to create a 2V92TA surely!
(If the 9 litre, V6 makes 270 bhp, says Wikipedia) Then about a third of that: 90 horsepower (from 3 litres!) would be a fair upgrade without getting too silly

[CitroJim]

And brakes

[Zelandeth]

To be fair, a 2-71 would probably still be a giggle given the relative torque values on hand. Also probably still weighs more than the whole Trabant!

As far as I'm aware the only two cylinder Detroits were inline, think V6 was as small as the V configuration engines went.

[mickthemaverick]

Just for inspiration Zel!!

[Gibbo2286]

Back in the day when Austin and Morris came together we had a customer who worked in the experimental department at Longbridge, he was involved with the design of the Morris Minor when the first fitted the Austin OHV engine from the Austin A30.
He told us they tried the 1200 cc Austin A40 engine and even the A70 engine (for fun), the A40 engine would work to some extent but the A70 engine ripped the axle shafts to bits and the teeth off the crownwheel and pinion.

[Hell Razor5543]

This looks like a lot of fun;

[Zelandeth]

Fallen behind a bit.

Somewhat idiotically, I didn't make sure that the Rover was drowned in wax before the winter. This was a mistake and made for a lot of scrubbing.



Likewise the Renault (which I do confess to having been kind of ignoring and hoping it will miraculously fix itself for a while - it hasn't) was even worse.





Couple of hours with the pressure washer and some scrubbing had things looking a bit more presentable.









Of course something fell off the Rover. Because Rover.





So had to glue that back on.



Which naturally meant I managed to get glue everywhere when the badge slipped out of my hands, because I'm a nightmare where anything involving adhesives is involved.


Driving the P6 through built up areas at low speeds with the windows down, the really deep burble just off idle definitely puts a grin on my face when it gets bounced back by surrounding buildings. It was quite pleasant to be driving it again.

Of course because it's this car we couldn't have a journey without something new cropping up. This time it was the temperature gauge which decided to go on strike.



Perfect on a hot day! That didn't add any anxiety to the drive home at all...no, I definitely wasn't at all worried...

Looks like we have an issue with the power supply to it as I'm only seeing 1.3V at the sensor. Which I promptly broke the terminal off.



New one added - which of course didn't fix the gauge.



Not that I really expected it to given the voltage reading. My guess is that the 5V regulator for the instrument panel (which I rebuilt using a solid state regulator) has died. Usually this would be made obvious by the fuel gauge also dying, but as that doesn't work anyway it's harder to be certain. Not going to have time to look into this today anyway.

Very exciting news to me though was that after having the enging idling for a couple of minutes while I fiddled with that was that I had no oil drips!



Coolant drips instead! Which appears to actually be coming from around this top radiator hose.



I have moved the hose clip a bit closer to the end of the hose, but given that end tank has obviously seen "some repair work" in the past, to be making sure there's not a crack or other damage to the hose mating surface which would cause us issues.

There is a tiny weep from a porous spot you can see in the photo, I think this drip is totally separate to that though.

Did get a nice photo while out though. Right up to the point I realised the bonnet wasn't fully latched.



Guess I'll need to snap that again then!

Which brings us to yesterday.

The instrument panel was once again pulled out of the car.



Which is actually really easy once you know how it comes apart. Less than ten minutes had the panel on the operating table. Any resemblance to our dining table is entirely coincidental.



One very important step was ensuring that the speedometer drive was securely sealed to ensure it didn't get grease everywhere.



A little more strip down then followed. I have to give the designers credit, this is by far the easiest instrument panel I've ever dismantled. Though there are some slightly questionable design decisions hidden within. These panels have a bit of a reputation for having connection issues. One source of these is some riveted vias through the PCBs, though I think I've found another obvious potential trouble spot. I give you the connection interface between what I have termed the backplane and the instrument sub assemblies.



Yes, that is just a hole in the solder mask where the PCB traces on the two boards are squeezed together by a nearby machine screw. While I'm sure this was a real boon in terms of production streamlining (the P6 I believe I read was actually the first mass produced car to use a PCB based instrument cluster, though I believe that was the original version, by the time this P6B was made plenty of others would have followed suit), 49 years and quite a bit of damp ingress later it sounds like a potential continuity nightmare.

The first specific bit of the panel I investigated was actually the ammeter. I had seen this twitch a few times but never really saw any significant deflection. More recently it had appeared to be completely dead. This turned out to be an entirely physical problem.

Here's a close up of the meter movement as it came out of the panel.



The issue is actually that the little circuit board on which the coil is mounted has over time and goodness knows how many heat cycles, worked itself loose. This has allowed it to drop downwards by about 1/8". This in turn meant that the armature of the meter had shifted too far away from the coil to be influenced by the magnetic field. This is actually where things should sit.



The board was originally held in place by the grommets around the spade terminals on the back, these however have lost their grip over the years. I eventually opted for just putting a couple of blobs of epoxy in strategic locations to hold it together.

I've always found the difference in designs of instrument panels of cars through the years always to be fascinating. So I was quite interested in a discovery I made while pulling the movement for the ammeter out. Warning...instrument panel design nerd rabbit hole ahead.

I had noted back when I first got this car that this instrument panel seemed to be extremely well and evenly illuminated for what I thought was an edge lit panel. Which based on being an early 70s design and having clearly raised digits it appeared to be.



Really is impressively well lit for something this old.



Turns out that it looks really evenly lit because it IS actually backlit. The edge lighting is purely for the pointers. Though not backlit in the way you're more used to seeing it executed. Turns out that the dial face itself is solid translucent turquoise tinted plastic with the black finish painted on. This is then clamped onto a solid 1/4" or so thick block of clear acrylic which is formed to behave as a light guide to provide an even illumination across the whole face. As demonstrated by me using a torch here.



This must have been really quite advanced panel design for the early 70s. Hats off to the designers - the night time illumination of this dash is easily as good as on any other car I've driven. Never mind one that was designed in the opening years of the 70s.

While the illumination might be quite high tech for the time, the actual gauge movements are pretty old school. These are all thermal gauge movements. If you've not seen one of these before, here's what's behind the face of one.



About as simple as it gets. It is literally a heater coil wound around a bimetallic strip. One end of this strip is rigidly mounted, the other is attached to the pivot of the meter needle. So as the current through the coil increases, the strip heats up more, it bends more and pushes the meter needle further. Very simple, and has the advantage of being inherently heavily damped. The downside is that they're obviously more power hungry than moving coil or moving iron meter movements.

Here ends the really nerdy bit, and we get back to fixing things.

The main fault I had dismantled things to address was the temperature gauge having decided to go on strike - I was pretty sure because the voltage regulator had died.

I had rebuilt the original thermal 10V regulator using a 7810 shortly after I got the car. However I really hadn't thought it through when I did this, specifically quite how electrically noisy an environment a car is. More specifically, it's also an environment where there are some really quite spicy transients floating around - especially when the starter motor solenoid is disconnected. As I learned much to my cost when I got an *almighty* belt off a Bedford YNT coach when fault finding an intermittent ignition switch. I don't know exactly what voltages or currents were involved, but that one was easily up there in the top five list for the most painful shocks I've had. The poor little 7810 really wasn't designed to deal with this sort of nonsense. There are proper transient protection diodes (I'm assuming they're basically very fast zener diodes) to install in settings like this. I don't have anything like that in stock, but I'll get some ordered for the future. For now I've added a 10uF bypass cap on the input side to hopefully help out - if I need to swap this again in future it's about 15-20 minutes work to get the regulator out and on the bench. You don't need to fully remove the dash from. The car, just pull it forward a bit so the regulator can be unplugged from the back.

The regulator shortly before it was crammed back together.



Before things went back together I also bridged the riveted connections to the regulator sockets with a bit of solder. It's not pretty, but it means the rivets aren't the sole things involved in making the electrical connection now.



Obviously I also gave all of the contact pads a really good clean. The scratches on some of these shows that I'm definitely not the first person to be in here.

The panel was thrown back together, lamp holders cleaned, speedometer drive greased, and inside of the front cover cleaned, then it was out back into the car. Result?



One working temperature gauge.

The ammeter is also now showing a lot more deflection than it was. Still sits pointing at a slight discharge as someone has clearly slightly bent the needles on both that and the oil pressure gauge slightly at some point. I'll need to fine tune that at some point but that's a job for future me.

Feeling accomplished here I decided to dig a bit further into the fuel gauge. I had tested the gauge itself in isolation and confirmed that it worked. I'm pretty sure that I had checked and found the sender itself to be open circuit from under the car. However I hadn't (as far as I could remember) really investigated the wiring to the back of the car. I had suspicions about it though - especially this chopped wire visible in the boot.



Sure enough testing this wire on the meter showed it to have 10V on it, and grounding it for a couple of seconds caused the fuel gauge to spring to life.



The wiring in the boot has suffered quite a lot of hackery in general from when the tow bar electrics were wired in (masking tape has good long term electrical insulation properties doesn't it?) and also from a rather shoddy speaker installation. Now I just need to figure out where that wire actually should disappear under the car. The sender did measure open when I originally checked I'm 90% sure but I'll obviously need to hook this back up properly either way or the gauge won't work even once the sender is changed! I did have a poke around underneath but couldn't immediately see where it went - though it's hard to see anything useful given there's a whole load of suspension and brakes in the way. Car really needs to be up on ramps to see anything much at all. That's a job for future me though

[CitroJim]

Great work Zel, a huge amount achieved! Well done

[myglaren]

Good work spotting the ammeter fault. I would never have detected that.

[MattBLancs]

Interesting stuff and others said good progress.

Delightfully shonky solution on the "blobs of solder generally held somewhere near the corresponding other blobs = PCB connection" can't help but wonder why the machine screw didn't form one of the connections, then only rely on one "blob to blob" connection. If any of the pairs of blobs were sufficiently different than one another, it could impact the other pair!
Suppose I'm being unkind, given the other much worse (masking tape!?) electrical work done by others later!

[Zelandeth]

Suppose I'm being unkind, given the other much worse (masking tape!?) electrical work done by others later!

I wish I was kidding!

[MattBLancs]

I think you did very well buying it before it became mysteriously on fire!

[Zelandeth]

I just don't know why it seems to be universally true that any tow bar electrical installation has to be a complete disaster. Almost every car I've had with a tow bar on was a complete disaster in the vicinity of it - Jim's Activa being the one exception I think. I'll probably just remove the lot in this case as it's never going to be used and the trailer socket is knackered anyway and obviously all the wiring needs to be redone from scratch anyway. Plus I've nearly died tripping over the tow ball about a dozen times already.

[CitroJim]

One of those facts of life Zel...