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

[xantia_v6]

Re the drop links, if you find some that have the same ends, but are longer, then a cut and weld operation may be the least hassle, if you know someone with a welder.

[Zelandeth]

Re the drop links, if you find some that have the same ends, but are longer, then a cut and weld operation may be the least hassle, if you know someone with a welder.

Assuming you don't then fall foul of welded repairs not being allowed to suspension components when it comes to MOT time.

I played that game a good few years ago when the originally (very poorly) welded on lower shock absorber mount sheared off the wishbone on one of my Saabs. Not an unknown issue on them. It was a clean break in the original weld rather than rust damage to the arm or anything.

This was repaired to a far higher standard than it ever left the factory with - yet the MOT tester refused to pass it. Ended up having to buy a whole new lower wishbone at no small expense. On which said mount was held in place by what looked like a couple of half hearted tack welds done by someone who had about 20 seconds welding experience.

Plus given how short these are I think I'd most likely end up damaging the bushes welding the ends together to a decent standard. The distance between the centres is only 131mm, so probably only 10cm or so of metal between the ends, and they are pretty chunky. So ideally would want to take them apart before welding anyway...so rebushing the original ones probably makes as much sense.

[Zelandeth]

Back to the technolgical necromancy.

After having rebuilt yet another iteration of the T1600 power supply only to have it explode more violently than any I've come across before, I've run out of patience with that approach. Without a schematic, even a block diagram or a working example to compare to I've not a hope in hell of getting one of these things working.

So this happened to one of the parts boards.





This has now been relegated to a cable termination platform and something to which I will secure the replacement power supply to (assuming it will fit, otherwise it will go in the expansion slot). Helpfully it also has most of the doubled up pins tied together right at the connector which saves me some hassle. With all of the components removed (I suspect I'll be finding SMD resistors around my desk for months) it's basically just an inert slab of board now.

A not insignificant amount of peering at the manual and very awkward pin tracing occupied a fair chunk of my afternoon.



I've said it before and I will again. The whole flex-PCB backplane the T1600 is built around is really good for keeping to overall volume of the system down, but it really does make working on the thing awkward.

The manual helpfully gives a nice detailed pinout of every connector in the system. Except for the one on the power supply board! Thanks guys.

We need the following.

+5V. Which basically the whole actual computer runs on, unsurprisingly given the era it's from.

+12V. Used for the display backlight and hard drive motor.

Ground. Obviously.

+9V. Only used by the modem, which isn't fitted. So we're ignoring that.

-22V. Bias voltage for the display panel.

That one has proven slightly tricky to trace. The manual clearly shows it as being on pin 14 of the display connector.



However there's no continuity from there through to any of the pins on the power supply. I suspect this may well end up needing some more time spent on it and possibly a dedicated line routed for it.

The absence of that rail though shouldn't stop the rest of the system from working...so time for some really messy prototyping to happen.



This mess is obviously only for testing to see if the idea works. The end result will be rather less messy and more robust.

Given the history with this machine I was kind of expecting smoke despite having been careful to ensure things were going where they should be.



No smoke. No fireworks. No actual display as expected, but the backlight is working and responding correctly to the brightness control. Even better, we get a POST beep, and the system is responding to the keyboard. However I obviously can't see what I'm doing.

Next step will obviously be to try to figure out something for the -22V rail or to have a dig around to see if I can find a CGA to VGA adapter in the giant box of cables.

I'm curious to see if hooking up -12V to the -22V rail would get us any sort of display at least for testing.

Longer term, I do have one of those little Pico ATX supplies floating around somewhere, which may well actually work well for most of what we need.

Still work to be done, but it feels like we're a lot closer to a working system than we were yesterday.

[CitroJim]

Interesting stuff Zel Keep is posted

[MattBLancs]

Ended up having to buy a whole new lower wishbone at no small expense. On which said mount was held in place by what looked like a couple of half hearted tack welds done by someone who had about 20 seconds welding experience.

So did you reweld up the new one (and slap on some black paint to hide the "modifications") ?

[Zelandeth]

Ended up having to buy a whole new lower wishbone at no small expense. On which said mount was held in place by what looked like a couple of half hearted tack welds done by someone who had about 20 seconds welding experience.

So did you reweld up the new one (and slap on some black paint to hide the "modifications") ?

Given I'd been completely without a car for a week by that point I didn't want anything left to chance so it was presented with it exactly as the arm had arrived, bare metal and bar code label still attached.

Suffice to say I was rather hacked off with the whole song and dance by that point.

[MattBLancs]

Ended up having to buy a whole new lower wishbone at no small expense. On which said mount was held in place by what looked like a couple of half hearted tack welds done by someone who had about 20 seconds welding experience.

So did you reweld up the new one (and slap on some black paint to hide the "modifications") ?

Given I'd been completely without a car for a week by that point I didn't want anything left to chance so it was presented with it exactly as the arm had arrived, bare metal and bar code label still attached.

Suffice to say I was rather hacked off with the whole song and dance by that point.

Fair enough, can understand the frustration there!

[Zelandeth]

Toshiba are now on My List for crimes against service technicians.

Those tables of connector pinouts, the sneaky sods haven't used consistent labelling! On the display I/O connector the -22V pin is labelled "M22V(-22V)."



That label however it turns out is suspiciously absent from anywhere else. Hence my wondering for a while if the voltage might actually be generated on the motherboard somewhere - with it being such a low power (0.22W) rail that wouldn't necessarily have been unreasonable - but I couldn't see anything which looked like a likely candidate, and the display obviously wasn't working.

It turns out that on the 100-pin connector to the "backplane" there is one pin however which is labelled "LCDV" which piqued my curiosity. Helpfully again, no further mention of that anywhere else in the manual.



Time to add another wire to my prototype. Which I also added a bit of hot glue to stabilise things having already pulled one wire off already.



Then followed a bit of experimentation. I don't have an easy way to generate -22V right now, but I did have a -12V rail from the ATX supply I'm currently hooked up to on hand. Figured it would maybe do something different (with a resistor in series initially to hopefully reduce the odds of me potentially blowing anything up) to give us some idea of whether I was on the right track.



It looks like we are! See that white border around the display? That's a clear sign of the panel is actually starting to try to work now. Essentially it's still an overloaded "white" screen as we've only got half the intended bias voltage, but it's a pretty clear sign I'm on the right track I think.

This is the power supply connector pin out as far as I'm currently aware. Which is the one table missing in the manual.



Key to acronyms:

ELV - Display EL Backlight supply.
RAMV - Memory board power supply.
LCDV - LCD Panel bias voltage.
VCC - Main motherboard 5V supply.
HDLV - Hard drive low voltage supply.
HDHV - Hard drive high voltage supply.
HDMV - Hard drive spindle motor supply.
GND - Ground, as I'm sure you figured out.
P12V - Expansion card slot 12V supply.

The currently greyed out pins are what appear to be data lines. Reverse engineering what those are up to is way, way, way, waaaaaay beyond my ken. As such they are essentially being ignored for now. There is a definite element of crossing my fingers that those lines are only used for communication regarding power management and then potentially being floating (I've no idea if the termination would be at the motherboard or PSU end) isn't potentially going to cause other issues - but at this stage I'm crossing my fingers. If we wind up having to add a bunch of pull up/down resistors later so be it.

[Zelandeth]

Had a dig around this afternoon and found the little Pico ATX supply I was thinking of.



If it works as hoped I'll figure out a way to directly secure this to the underside of the original supply PCB. The idea here is to keep everything I'm modifying attached to that PCB, so if I ever come across a proper original board again it can just be plugged in.

Having a bit of a ponder about the -22V rail, I was thinking about buying in a little DC-DC converter for the job, but given it's such a low current application I might actually see if I can cobble together a simple voltage doubler circuit to do the job. Seems worth a try. All I need is a 555, handful of caps and a few diodes if I remember the designs I learned back at college right...big if!

The Rover got a small tweak made to the windscreen wipers yesterday to address the issue with the driver's side wiper hitting the top of the screen surround. This was a very high tech solution.



Yes, I just attacked the end of the blade with a pair of scissors. Problem solved.



This has improved the performance of the wipers from "it's actually worse with them turned on than off" to "pretty poor."



Turns out that the spring in the arm isn't applying enough pressure to keep the whole blade in contact with the screen. However I've noted that there's a lot of play in the bush where the spindle comes through the scuttle, so that needs to be investigated - when it's not actively raining.

I noted today that there is one visible point of water ingress while driving in the rain.



Not really surprising given that you can see daylight through gaps in the seal between the centre and left hand glass panel up towards the top. Sure a couple of blobs of sealant can sort that.

[Zelandeth]

Finally got around to actually extracting these today.



Fun fact: Things held in place by snap rings come apart a lot easier once you've found and removed said snap ring.

I definitely didn't spend about half an hour staring at this thing and trying to figure out how on earth it came apart.



Also don't punch anti-roll bars. They're harder than your knuckles.

I've been given a tip off that the links used on VW T4 Transporters are very similar. They're a centimetre or two shorter, but I don't think that's likely to be a huge issue - provided they match both sides I don't imagine a little extra pre-load on the ARB being a huge issue.

More importantly the diameters of the bushed parts look to be right - top one accepts an M10 bolt (identical), and the lower bush inner diameter is 23mm.



Hard to see in the photo, but that looks like it's a match.

The inner portion of that bush is actually fine - however it's a very loose fit in the body of the link, which ties in with where I was seeing movement with it on the car. The rubber of the lower bush looks distinctly newer than the top one, so I do wonder if someone in the past has attempted to replace the bush but picked one with slightly too small an outer diameter.

They're really cheap so I'll grab a pair and see if they fit. Watch this space.

Edit:

Also ticked off a to do list item on TPA which has been pending for uuh...about four years today.



Replaced the missing valve cap on the front wheel. Riveting stuff, I know!

[CitroJim]

That's a nice result Zel Fingers crossed the VW ones do the trick!

[Gibbo2286]

I wonder if this company has parts for your links.
https://www.polybush.co.uk/

[Zelandeth]

I wonder if this company has parts for your links.
https://www.polybush.co.uk/

Already have the catalogue linked somewhere above I think! If the VW links don't work, yes it will be their stock I end up digging through for something appropriate.

Attempted to go pick up (or at least order) a pair of links today but the traffic had other ideas. I even checked before I set out that it was clear, however it wasn't ten minutes later when I got to the back of the queue on the A5. After we'd moved about 3 car lengths over the space of 20 minutes I ran out of patience (and heat tolerance) and Aborted the mission.

...Then got stuck in another queue trying to avoid the one in the other direction I'd seen on the way to getting stuck in the first one.

*Bashes head against wall*

Apparently I'm also about to be having a chat with someone about a car I was kicking myself for having been too slow off the mark to look at a couple of weeks ago too. The guy who bought it has apparently got their eye on something else now so it might be available again...what was that thing I said about clearing the decks? Granted this is actually something sensible though rather than my usual nonsense!

[Zelandeth]

I need to take some actual temperature readings on the Rover I think following a high speed run. Yes it's been moderately warm here this afternoon, but it's not exactly roasting. However the car felt it necessary to do this about 30 seconds after the engine was switched off after a run out today.



Yes, I always let it idle for a bit first rather than just pulling into the drive an immediately turning it off.

The odd little dribble from the overflow I can deal with, but this was a bit ridiculous. I've also now added a small hose to the end of the overflow as the water from it before was running down the inside of the front valance and making a bloody mess.

Temperature was sitting where it normally does around town, about a needle's width above the bottom of the white section on the gauge - this is as far as it crept up to after stopping the engine. So hardly roasty toasty according to the gauge.



I think I'd like to confirm that the gauge is reading accurately.

I'll need to wait for the engine to have cooled further before I can really check the coolant level, but I've not topped it up since last time it decided to burp coolant out (checking after that showed the level to be about an inch or so above the top of the core). This was rather more than a burp though so I imagine it must have impacted the level a bit. I have never seen it vent when the engine is running though, it's always after stopping when there's the inevitable temperature spike then due to the circulation stopping plus heat soak etc.

[CitroJim]

Is the rad cap good Zel?

I'd dispute yesterday's temperature... It was bloody roasting where I was and what I was doing didn't help I guess... See my blog