Commodore 8026 printer / typewriter

[SiriusHardware]

Colin did say this earlier:

I believe that the legs go CBE from left to right.

When making a statement like this though, it's always best to be explicit about how and from which angle the device is being viewed - sometimes people depict them with the flat face facing towards them and the legs pointing downwards towards the floor, other times they depict them with the legs pointing towards the viewer's face and the flat face of the device pointing towards the floor.

A labelled sketch, even a photo of a quick pencil sketch, is always better than a description.

[ortek_service]

Originally Posted by SiriusHardware

Colin did say this earlier:

When making a statement like this though, it's always best to be explicit about how and from which angle the device is being viewed - sometimes people depict them with the flat face facing towards them and the legs pointing downwards towards the floor, other times they depict them with the legs pointing towards the viewer's face and the flat face of the device pointing towards the floor.

A labelled sketch, even a photo of a quick pencil sketch, is always better than a description.

Thanks, I'd missed that - I presume it was before Colin got the 'Universal' LCR-T7 tester, which should be able to confirm this (as long as you know which of the transistors leads was connected to each of the tester's '1','2' & '3' labelled pins on its ZIF test socket)

And if it does confirm CBE from left to right with transistor orientated as I'd specified (looking at front part-numbered flat face, with legs pointing downwards, as fitted to PCB) when I quoted it for the BC327 etc, then that is standard for most BCnnn transistors (without an 'L' alternative-pinout suffix) / the same as the other BC-series PNP transistors Colin has.

[ScottishColin]

I am attempting to learn KiCad and draw a schematic of the relevant area armed with my continuity tester. It's taking a little time but I hope to have something (that will undoubtedly be wrong) tomorrow.

Colin.

[SiriusHardware]

I did see that you had posted the requested images of the top and bottom sides of the PCB, apologies but I have not had much time to myself this weekend. Maybe I'll get a chance to have a look at work tomorrow.

[ScottishColin]

OK - I've tried to produce a drawing of the location of each component and mark it as best as I can. There are some I cannot identify so apologies for them. I'll try to get on with KiCad tomorrow.

Hope this helps - if it's of no use or needs corrections, please let me know.

Colin.

[SiriusHardware]

Nice looking layout diagram - eventually you will need to draw in the top side tracks and where they go to, and the underside tracks and where they go to, but for the time being just do it in the area local to those components.

[ortek_service]

Yes, and if you've built the symbols for the various IC's yourself, then you may also want to add the pin names to these to make them easier to follow.

There maybe also some KiCad libraries out there with some of these already done, but maybe as more of a schematic orientated symbol than the package-pinout view - using the later to start usually with makes it easier to trace the connections, but it can sometimes be better to re-draw later-on with ones typically used on other schematics by Commodore etc. This is especially the case for analogue IC's, where there are usually conventions of inputs on the left, outputs on the right and power at top / ground at the bottom of the symbol. With digital memory etc IC's, a lot of the pins may be bi-directional or on a common bus, so might want them all on one-side to pass through to next IC.

[ScottishColin]

Yes - that was a freehand drawing in Powerpoint, but I thought I had to start somewhere.

I had KiCad loaded and the 74LS chips and Commodore's 65xx series loaded into it ready.

Here's a question - when I mark resistors, do I mark them as the markings they have on them or should I measure them with my meter in situ and use that measurement? I guess that's a daft question as what was installed (ie the markings) are what they should be.

Colin.

Originally Posted by ortek_service

Yes, and if you've built the symbols for the various IC's yourself, then you may also want to add the pin names to these to make them easier to follow.

There maybe also some KiCad libraries out there with some of these already done, but maybe as more of a schematic orientated symbol than the package-pinout view - using the later to start usually with makes it easier to trace the connections, but it can sometimes be better to re-draw later-on with ones typically used on other schematics by Commodore etc. This is especially the case for analogue IC's, where there are usually conventions of inputs on the left, outputs on the right and power at top / ground at the bottom of the symbol. With digital memory etc IC's, a lot of the pins may be bi-directional or on a common bus, so might want them all on one-side to pass through to next IC.

[SiriusHardware]

-As you rightly guessed, use the markings they have on them, and if the markings are unreadable, desolder / disconnect one end of the unreadable component and measure its resistance by itself, not in-circuit. I did notice you have one resistor which appears to have totally lost its paint coating, therefore also its markings.

If you happen to find any components whose measured values are significantly different from their marked value, then flag that up.

[SiriusHardware]

I don't know if this will help but here are a couple of crops from Colin's main photos. I've flipped the 'underside' view so that it shows the pads and traces as they would appear if the PCB was transparent and you could see the lower side traces through it from above. I've outlined the location of the pads for the missing transistor with a red triangle on both images.

[ScottishColin]

If you see the component I have circled in black, I am assuming that is a diode; is that correct?

If so how do I measure it please? Do I need to unsolder one end to be able to measure it?

Thanks.

Colin.

P.S. I am working on the schematic - honest.

[SiriusHardware]

I don't think that is a diode, I think it is a resistor which has completely lost its clothes. The spiral track you can see running around it is the actual resistance element of the resistor - something you don't normally see.

For best results it is always better to unsolder one end of any unknown component which you are trying to measure the value of so that you know your measurement is of that component only and not others connected across it.

When you already have the circuit diagram, or when a resistor has a very low known (marked) value, you can measure components in-circuit allowing for what you know is connected around the component under investigation, but to be absolutely sure the best approach is always to disconnect the component from the surrounding circuit before measuring it.

Desolder one end of that component and measure its resistance first with the probes one way around, then the other way around. If it measures the same resistance with the probes either way around, it is a resistor. (The same resistance both ways can apply to other linear components, like inductors or fuses, but that component looks much more like a resistor).

[ScottishColin]

Understood - than you.

What about this one? This is what I have called D4 on the diagram in post 65 (pretty much the centre of the board).

Colin.

Originally Posted by SiriusHardware

I don't think that is a diode, I think it is a resistor which has completely lost its clothes. The spiral track you can see running around it is the actual resistance element of the resistor - something you don't normally see.

For best results it is always better to unsolder one end of any unknown component which you are trying to measure the value of so that you know your measurement is of that component only and not others connected across it.

When you already have the circuit diagram, or when a resistor has a very low known (marked) value, you can measure components in-circuit allowing for what you know is connected around the component under investigation, but to be absolutely sure the best approach is always to disconnect the component from the surrounding circuit before measuring it.

Desolder one end of that component and measure its resistance first with the probes one way around, then the other way around. If it measures the same resistance with the probes either way around, it is a resistor. (The same resistance both ways can apply to other linear components, like inductors or fuses, but that component looks much more like a resistor).

[Cobaltblue]

That's a diode most likely an IN914 or IN4148

Unless it something special.

Cheers

Mike T

[SiriusHardware]

Yes, most likely to be a general purpose silicon diode as CB says, although there are certain types of tubular capacitor which can look similar. It could potentially, possibly, be a zener diode.

Un-solder one end and use your meter on diode test mode with the probes first one way around, then the other way around.

[ortek_service]

Now that Colin's now got a ' T7 Multi-function Tester', then it should be possible to just connect any 2 different pin numbers of its socket to this component, and it will automatically work out if it is a capacitor or a diode (even a zener, upto 30V on these T7 models IIRC?).
- Just got to lift one and extend some wires to it / or take it completely out of the board and place in the ZIF-socket.
If it was a diode, then there'd normally be a thick black etc. band around one end to mark the cathode - unless it was a (bi-directional) diac, but unlikely to use one here. And coloured bands or v.small text to give the type number (or most of it). So maybe also worth using an optical magnifier on it, to see if anything can be read.

However, it rather looks like there been some spillage in parts of this board, that has made some component-legs go blueish-green and removed some component markings / the outer protective casing on that spiral-cut resistor (unless it ran warm and that cracked and fell away).

From a brief look at top-side / underside, it seems that the emitter of this PNP transistor goes to a power-rail (as thicker track), and there's a 470R resistor feeding the base.
But it's difficult to see much-more from the photo's due to lack of contrast / resolution a bit low / corrosion over legs is similiar colour to PCB - Will some IPA etc on a cotton-bud clean this up a bit? (Or maybe some PCB flux-cleaner etc).

[SiriusHardware]

However, it rather looks like there been some spillage in parts of this board, that has made some component=legs go blueish-green

That would be electrolyte or vapour from the two enormous NiCD or NiMH batteries (now removed) which were less than an inch away from this circuit area (see earlier posts).

[ortek_service]

Yes, I hadn't realised quite how close those had been to this area, as the area they'd been wasn't visible in this small-area close-up.
But that colour is rather indicative of being caused by a leaking (often NiCd) battery. Although I've also had the copper of the tracks turn black, and very-difficult to get solder to stick to.
Further cleaning may remove more component markings on those already attacked a bit, so useful to have some good photos of these, before doing too much intensive cleaning of that area.

[SiriusHardware]

The suggestion for Colin to let the new tester decide what the component is is a good one, although as you said any component so tested will need to have one of its two leads lifted from the PCB prior to connection of the tester.

[TonyDuell]

Colin has sent me the logic board assembly, along with the PCBs from the printer carriage, to investigate.

I've not done much yet, and this is all 'to be confirmed', but...

Pin 1 of the power connector (black wire) seems to be the ground for the motor drivers, etc. Pin 5 is the logic ground. I suspect these are joined in the power supply.

The corroded PNP transistor seems to switch the 5V supply to the motor drivers, etc. Probably to prevent the motors being energised before the processor is running. So a fairly high-current device.

I am not sure the batteries are really useful in this model. They seem to supply a pin on the processor chip which, according to the Mostek databook is the internal RAM supply on one version and just a port pin on the others. The 'development' processor with the piggyback EPROM used here is one of the latter. Possibly the former was used in the 'standard' Olympia model.

The wires from the GPIB board go (mostly?) to a 28 pin DIL space on the main logic board. Looks like it might have been for a 8212. Again, probably used as such in the Olympia model (Centronics interface?)