80-Bus News

EPROM Programmers

This issue I start with a few comments about the Bits & PCs EPROM programmer that appeared as the Gemini G808. I bought one shortly after they first appeared, and have used it quite a bit since then. Some time ago I started to have trouble programming some of my 2716s, (I never used it for 2708s), and even after overprogramming them several times some bytes were still not programmed. I put this down to old-age on the part of the EPROMs, after all they had been through the program/​erase cycle many times, and so I marked them as suspect and put them on one side. Some months later I happened to check the programming voltage on an EPROM as it was being programmed, and found it was only 21v rather than the required 25v. Off load the programmer produced the correct 25v. This made me dig out the 2716 data sheet, and a quick glance showed that a maximum current of 25ma could be drawn from the 25v supply during programming. I removed the 2716 from the programmer, and replaced it by a 1k resistor from pin 21 to ground. (25v across 1k gives a current of 25mA). I then started “programming” and measured the programming voltage that appeared across the 1k resistor – 18v. Bingo, there was my answer, the charge pump that generates the 25v supply for G808 was obviously rather short of breath, and unable to provide the necessary power. The 2716s I had had trouble with were obviously not programming because the programming voltage was too low. Perhaps with use the programming current actually taken by a 2716 slowly increases, and these had reached the stage where the on-board 25v supply was seriously overloaded.

The 25v is generated by a charge-pump circuit based round a 555 timer running as an oscillator. This runs off the +12v and −5v supplies, and theoretically adds another 17v onto the 12v line (ignoring all losses!), which is then regulated to 25v. I scrabbled in the proverbial junk box and tried changing the 555 to no effect. Altering the values of various components around the 555 didn’t help either. So I went to the drawer containing “projects started but never finished” and dug out the last of my EPROM programmer designs that had failed to see the light of day (until now). This particular one used a Texas Instruments TL497 switching regulator in a step-up mode to generate 25v from a 5v supply. This I connected in parallel with the existing voltage generator on G808, and there-after I had my 25v. The 2716s previously labelled “defunct” now programmed perfectly. As this fix worked I progressed no further.

My solution (of geting the 25V from elsewhere), is one way out, but if anyone with the same problem can come up with a suitable simple modification to the G808 circuit – send it in. Either as a few words which I can include in a an column, or as a short article in its own right (in which case you get paid!).

Generous offer That No-One Should Refuse

Talking of articles for the NEWS, if an article is submitted on a Gemini-format CP/M disk then the odds of it being published shorten considerably. So, in order to help the balance of articles in favour of non-disk (or non-CP/M) systems, I am offering to accept articles on cassette tape (Nascom 2 or Gemini format – not Nascom 1 tapes I’m afraid) or disk (5.25″ or 8″ – any format). These I will transcribe and pass on to the Editor. Note: this offer applies only to machine readable media – I’m not offering to type anything in from paper! All I ask is that you include two things: