Commodore PET 2001 voltage regulators

[retromit]

Originally Posted by SiriusHardware

Tim also asked if he should continue using the UD5 (Basic Toolkit) PROM following on from its successful pin graft - I don't see why not, unless the running temperature alarms you.

I decided yesterday to install this in the PET, and it looks to be working correctly. The SYS command sets everyting up, and the commands are all available. So that's good news. In view of the repair, and with a couple of the other legs on the PROM being a bit wobbly (although it is now in its own socket), I might still create an EPROM for this. There was a lot of corrosion on the legs, so it might be a safer option as well.

[SiriusHardware]

Running the test EPROMs for long periods of time will give you a fair idea of how warm EPROMs will get if you clone the PROMs into them - probably not very warm.

Unfortunately the specific EPROMs which replace those PROMs (TMS2532) are quite expensive and most device programmers don't support them, although the first adaptor you made -may- allow a programmer which can program a 2732 to programme a TMS2532 by rearranging the pins appropriately. You'd need to be very sure you had the right adaptor for the right purpose otherwise you'd risk applying the high programming voltage (VPP) to the wrong pin of your expensive TMS2532s.

My old device programmer can natively program the TMS2532 so that's another option for you as and when you get to that point. I don't have any though, so you'd need to send some down to me.

[ajgriff]

Originally Posted by SiriusHardware

Unfortunately the specific EPROMs which replace those PROMs (TMS2532) are quite expensive .....

Would these be any good?

https://www.ebay.co.uk/itm/222249893...oAAOSw4shX2ljs

As everyone is aware I know little about these things.

Alan

[SiriusHardware]

They might be, although that description is unpleasantly mangled as it has '27C256' (a completely different device) right in the middle of the description line. Unfortunately while my programmer can definitely programme the TMS2532 it doesn't support the Hitachi version, someone else's programmer may do so though.

The pinout looks OK, with A11 in the right place (on pin 18).

https://www.jrok.com/datasheet/HN2532.pdf

[ajgriff]

Originally Posted by SiriusHardware

They might be, although that description is unpleasantly mangled as it has '27C256' (a completely different device) right in the middle of the description line.

Yes I can see how that's a concern. I suppose the fact that the advertised devices have 24 pins is of some comfort. I think the DIP 27C256 has 28 pins. Could always query the discrepancy with the seller before ordering.

Alan

[SiriusHardware]

In the interests of science, I have ordered some. Although my programmer has no specific support for the Hitachi device I'm wondering it it will 'just work' if I just select the Texas TMS2532 and try programming one of the Hitachis that way, as long as VPP is the same.

They'll be handy to have if it works, there are quite a few old 80s computers which use 24 pin PROMs with this pinout.

[ajgriff]

Brave!

[SiriusHardware]

Or stupid, possibly. Time will tell.

On a brief read through the respective data sheets both use VPP of +25V and both use a 'standard' programming pulse (low on pin 20 for 50mS) so it looks quite hopeful.

[retromit]

Originally Posted by SiriusHardware

Or stupid, possibly. Time will tell.

On a brief read through the respective data sheets both use VPP of +25V and both use a 'standard' programming pulse (low on pin 20 for 50mS) so it looks quite hopeful.

It will be interesting to see how you get on with these. When you mentioned the TMS2532 last night I did find these on eBay and was tempted to try them with the first socket I built, but was then a bit unsure (although the programmer I have can apparently handle 2732s).

If they work, then it could be a good solution going forward.

[SiriusHardware]

It's up to you of course but the Hitachi devices are due to arrive at the start of next week, so maybe wait until I've had a chance to try programming one as they are quite a bit cheaper than TMS2532s.

Buying ICs of any sort online can be a bit of a minefield, the more desirable or rare a part is the greater the possibility that it has been faked, often by rebranding / rebadging a relatively common device as the rarer one. You really don't know what you're going to get these days.

I'll let you know how these ones work out.

[Timbucus]

Originally Posted by SiriusHardware

Or stupid, possibly. Time will tell.

On a brief read through the respective data sheets both use VPP of +25V and both use a 'standard' programming pulse (low on pin 20 for 50mS) so it looks quite hopeful.

Unusually I resisted the urge to add to my collection of PROMS and EPROMS... well done Sirius.

[SiriusHardware]

On more careful inspection I notice that my All07A Eprom programmer has a fixed algorithm for TMS2732 which applies 25V VPP but also 6V VCC during programming. The Hitachi datasheets don't make any suggestion of doing this on their chips, or not that I have found so far, so it might be unsafe to try to directly program the HN462532s as though they are TMS2532s.

If I look up how the programmer programs more conventional 2732s it uses +25V VPP but only 5V VCC, so it might be safer to make an adaptor to twist the TMS2532 / HN462532 pinout into the footprint of a 2732 and select 2732 (generic) in the programmer's menu to make sure it only uses the basic 50mS programming pulse, and not some manufacturer specific fast algorithm.

I'll probably try it both ways - the second way, if that works, would be more widely useful for the rest of you who probably can program 2732 / 2732A but not 2532 / 2532A due to its less common pinout.

I hear a lot of noise about the limitations of those popular TL886 programmers, mostly that they can't generate enough output voltage to properly support 25V VPP devices like the 2716, 2732, although they may be able to program the '...A' versions which are usually 21V.

[ScottishColin]

As a fallback if necessary, a reminder that I got a TMS2532 from 1980s_john, but they were a little more expensive then (£6 ea).

Colin.

Originally Posted by SiriusHardware

On more careful inspection I notice that my All07A Eprom programmer has a fixed algorithm for TMS2732 which applies 25V VPP but also 6V VCC during programming. The Hitachi datasheets don't make any suggestion of doing this on their chips, or not that I have found so far, so it might be unsafe to try to directly program the HN462532s as though they are TMS2532s.

If I look up how the programmer programs more conventional 2732s it uses +25V VPP but only 5V VCC, so it might be safer to make an adaptor to twist the TMS2532 / HN462532 pinout into the footprint of a 2732 and select 2732 (generic) in the programmer's menu to make sure it only uses the basic 50mS programming pulse, and not some manufacturer specific fast algorithm.

I'll probably try it both ways - the second way, if that works, would be more widely useful for the rest of you who probably can program 2732 / 2732A but not 2532 / 2532A due to its less common pinout.

I hear a lot of noise about the limitations of those popular TL886 programmers, mostly that they can't generate enough output voltage to properly support 25V VPP devices like the 2716, 2732, although they may be able to program the '...A' versions which are usually 21V.

[SiriusHardware]

I think that included programming with the code of your choice didn't it, so that's a pretty fair price if they are still available from 1980s_john (who is a forum member here with that username) for around that price.

A blank from somewhere else, plus postage to me to be programmed and then postage from me to Tim would probably cost more.

[retromit]

Originally Posted by SiriusHardware

I think that included programming with the code of your choice didn't it, so that's a pretty fair price if they are still available from 1980s_john (who is a forum member here with that username) for around that price.

A blank from somewhere else, plus postage to me to be programmed and then postage from me to Tim would probably cost more.

Thanks for the info. I might send 1980s_john a message to see if he still supplies these.

As a alternative, has anyone useda ROM adapter board in the PET? I was having a look over GitHub and found the following:

https://github.com/SukkoPera/Open2327RomAdapter

I was drawn to the section on replacing 2332 ROMs. Just thinking that if the TMS2532 (or equivalents) are hard to come by, then could this be an option?

Tim.

[SiriusHardware]

That board looks like it is intended to make 2764 / 27128 / 27256 / 27512 etc EPROMs 'look' like 24 pin PROMs as far as the host system is concerned. It won't look very pretty as there will be a 28 pin EPROM on a daughterboard overhanging a slot which should be occupied by a directly fitted 24 pin device, but you could take the view that looks won't matter when the unit's lid is on.

It does have the advantage that the 2764 and upwards are still widely available, whereas even 'standard' small EPROMs which don't have the abnormal pinout of the 2532 are getting slightly harder to come by, and as suggested earlier some of the popular cheap programmers include them in their supported device list but can't actually generate the correct VPP voltage.

I think if I was going to go the adaptor route I would use adaptors which convert 24-pin 2732 / 2732A - still relatively easy to get - to 24-pin 2532 pinout. Such adaptors are available if you look around, and then look for the 2732A to put in them because the cheap programmers are more likely to be able to get near their VPP of 21V than the 25V of the non-A versions.

[SiriusHardware]

Anyway, we seem to have become obsessed about PROMs and EPROMs when you have already established that the four main ones in your PET are OK and even that wounded optional one has been massaged back to life.

You'll be using the CPU you borrowed from the VIC for now, but how are you getting on with finding a replacement 6520 / 6521?

[retromit]

Originally Posted by SiriusHardware

Anyway, we seems to have become obsessed about PROMs and EPROMs when you have already established that the four main ones in your PET are OK and even that wounded optional one has been massaged back to life.

You'll be using the CPU you borrowed from the VIC for now, but how are you getting on with finding a replacement 6520 / 6521?

I'd agree! I think the EPROM discussion took off from a chat about heat on the original PROMs. I think it may also have been down to wanting to preserve the original PROMs in case of an issue. As it stands, the PET has been running fine for the last few days.

I have ordered a replacement 6520 (from Alan's eBay recommendation) and am hoping it may be here this week. If that works, then we should be back to the full complement.

In terms of other outstanding work, the cassette belt has been despatched; again, hoping it will arrive this week so that this can then be changed in the cassette player with the hopes of loading something.

Whilst waiting for these I've just been watching the video Colin provided a link for (with regards to the keyboard), so it will be one of the next jobs to see if I can get the shift key working.

All in all, and considering where we were a couple of weeks ago, things are looking quite good. I'd just like to send my thanks once again to yourself and everyone on this thread for all the help, advice, and support given. It really has been appreciated.

Tim.

[ScottishColin]

If you're bored, you can type in the program on page 184 of this book (actually p199 of the PDF) which tests the IEEE port. It found a fault in mine which meant another chip to be replaced, but I was in the mood to try to find every fault while I was working on it.

http://www.1000bit.it/support/manual...EEE488_Bus.pdf

[SiriusHardware]

A very quick diversion off onto EPROMs again. I have discovered that our ancient device programmer at work (A Needhams EMP20) has native support for the Hitachi 2532 with VPP=25V, VCC=5V and standard 50mS PGM pulse, so I'll try those inbound ones on that first.