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

[CitroJim]

Oh yes, I can definitely say that's a massive improvement Zel Are you going to take the old 'box apart and see what's worn in it?

[Zelandeth]

Oh yes, I can definitely say that's a massive improvement Zel Are you going to take the old 'box apart and see what's worn in it?

Definitely my intention. Guessing the answer is going to most likely be "everything."

One thing that is hard to convey in the video is how much less backlash there is in the drivetrain now. It's actually possible to smoothly put the power back on when coming off the freewheel, before there was always something of a jolt that was impossible to avoid.

The bearing set to rebuild it isn't expensive so if the gears themselves and the case aren't damaged I may well look to rebuild the box that came out. Either as a good spare or to pass on to someone else in need - though they don't fail particularly regularly so demand isn't huge!

[CitroJim]

Attending the Redborn Classic Car Show on Saturday...

[Zelandeth]

Attending the Redborn Classic Car Show on Saturday...

Not one I've seen before. I shudder to think what one must be like when lowered... they're bouncy enough at the best of times! Not as though you're going to gain anything in terms of handling either, they don't roll meaningfully anyway. Biggest improvement from stock is just a spacer above the inner semi trailing arm mounts which removes the significant amount of negative camber they had from the factory for some unknown reason. That can make them a bit twitchy, but otherwise they're very chuckable - as you'd expect for a tiny car that weighs 600kg.

Rather less scruffy than mine though. Not that I honestly care, it's a car which earns it's keep after all.

[CitroJim]

It wasn't sounding very healthy either Zel; It was smoking like a chimney when it drove out of the show...

[Zelandeth]

It wasn't sounding very healthy either Zel; It was smoking like a chimney when it drove out of the show...

Way too much oil then you'd tend to think...or it's never used properly so never gets to clear itself and the exhaust out.

You know what two strokes are like and how grumpy they get if not given a good pasting on a regular basis!

Or it's that classic case of a car that's been made to look very pretty but the underlying mechanical bits being largely overlooked. Which is a phenomenon I've never really understood... surely you'd want to make sure a car was reliably and comfortably usable before worrying about what it looked like. From what I've seen at shows it seems to be a very common situation though!

[moizeau]

Or it's that classic case of a car that's been made to look very pretty but the underlying mechanical bits being largely overlooked. Which is a phenomenon I've never really understood... surely you'd want to make sure a car was reliably and comfortably usable before worrying about what it looked like. From what I've seen at shows it seems to be a very common situation though!

I agree completely
Here is my 'Ugly Beast'.

Starts on the button, everything works, just looks a little rough.
Having said that, that this is her sister,

a little prettier but more importanty, everything works.

[Rp0thejester]

But still, the Trabant is a good looking lawnmower. I do agree that mechanical work is more important than bodywork otherwise you end up with a rather large paper weight.

[CitroJim]

Great to see you yesterday Zel and hear the new, quiet Trabant with my own ears

[Zelandeth]

Great to see you yesterday Zel and hear the new, quiet Trabant with my own ears

Well the gearbox is quiet(ish), and mechanically the engine is (as you'd expect of a two stroke really given the very limited number of moving parts), the exhaust on the other hand is far from quiet...especially once a few revs are involved. Subtle it is not.

Nothing dramatic to report so far. The exhaust manifold to expansion box nuts managed to work themselves loose and started buzzing - though that surprises precisely nobody, myself included. It took me several attempts to get the buggers to stay put last time.

Finally got around to fixing this light on the dash today. I nicked the bulb out of this for the speedometer a fair while ago.



It's kind of redundant given the visibility of the switch clearly being in the "on" position and that the indicator tell tale on the dash also flashes, but it's now working as intended again. Yes, the red "button" on the dash in a Trabant isn't a button. It's the tell tale for the hazards.

[CitroJim]

the exhaust on the other hand is far from quiet...especially once a few revs are involved. Subtle it is not.

It sounds lovely Zel

[Zelandeth]

Well this wasn't something I expected to be picking up this weekend.





Not a machine I'd ever really gone looking for as they tend to change hands for daft money. However if you're in the right place at the right time things can sometimes just work out right.

[CitroJim]

Oh wow! That's great Zel Is it functional?

[Zelandeth]

Oh wow! That's great Zel Is it functional?

Functional, yes. Working, no.

Though I don't reckon it's far away from it.



This tells me a lot, and it's all good news.

The power supply rails are all present, machine is correctly coming out of reset, the clock is working, CPU is running and executing code, there's something in the ROMs and the CRTC is working. Despite a bit of burn in, the CRT is bright and sharp too which is always good.

Most likely we've just got some bad memory or a dirty socket that's causing our issues. Assuming the power supply is actually good anyway - I've not actually got a meter on it yet (the previous owners have had it powered up already or I'd have checked that before applying power to the board).

Shouldn't be too difficult to debug - get a scope on each of the data and address lines and look for one that doesn't look right will probably let me home in on it pretty quickly.

I could actually probably narrow it down more by looking at what is on the screen as well as those characters will exist in a specific space in memory. Though physically checking data lines sounds an easier place to start than trying to decode memory maps.

I had hoped this would just be using 4116s for the memory as I have an Apple II spares machine I could rob some out of if needed, however it looks to be based around 2114s instead (for a whole 8K of memory!).

Based on the screen burn it looks like this was originally used as part of an inventory management or sales system - for plumbing things I think given I can read mentions of copper, PVC and gland nuts.



For all screen burn isn't desirable, it does sometimes offer an interesting glimpse into the history of a computer.

Edit:
Hmm. Methinks before I delve any further some checking of the power supply is indeed going to be needed.

Pretty sure two out of the four diodes in the bridge rectifier for the -5V rail shouldn't be getting hot...



That's the two leftmost ones below.



As it's only two diodes rather than the whole bridge getting hot I'm suspecting shorted diodes rather than the rail being short to ground itself. If the -5V rail is MIA though, the system definitely won't work properly.

I mean step one of any diagnostic work is always checking voltages anyway, so that's the first port of call before I dig any further.

Those diodes are 1N5402s, which sadly are more substantial (200V/3A) than anything I think I have in stock at the moment I think.

[Zelandeth]

Had ten minutes to look at this in the evening today.

First progress is that not being half asleep, I actually found the memory on the board. It's a bunch of TMS4108s. Which makes far more sense given the general architecture of the machine.



This is good. They are far less unusual chips, and can actually be replaced with 4116s to double my memory to 16K (and in theory if I populated the adjacent vacant sockets I could go all the way to 32K).

There's a helpful donor machine with two chips short of 48K worth of 4116s up here.



I have done the first round of troubleshooting though in checking our voltages.

+12V: 11.49V. That smells on the low side to me...
+9V Unregulated: 10.41V.
-9V Unregulated: -10.58V.
+5V: 5.04V.
-5V: -5.02V.
+16V Unregulated: 20.1V.

I need to get the scope involved to ensure the rails are clean. Especially given those two diodes are getting warm which makes me suspicious. Once that's all nailed down we can actually start the proper debugging process. No point in digging too deeply until we've confirmed that we have good and clean power.