Hi all,

I have a rather general question about the connections of my EPROM I use in my 6502 project.
I power a ST M27C128A, however, unfortunately in contradiction to the data sheet, I connected VPP
to GND instead of Vcc during normal read operations.
The system is rather stable... rather means, that if I start more complex programs, like 8K Basic,
it sometimes runs and mostly crashes. This is also dependent on the quality of the power supply of course.
Does anybody have an idea, if this can have to do with Vpp connected to GND instead of Vcc as well, such that
maybe access times are higher or the EPROM may produce unreliable outputs?
There are EPROMs where the data sheets say: Does not matter if Vpp or Vcc, just it must not switch.

Of course you could just say: Why don't you just connect it as staed in the data shee,t however I would like to know if anybody has experience with this topic. I didn't find anything on google.

Kind regards,
Andi

I would think you'll be OK - VPP is allowed to be low, according to the absolute limits in the datasheet (It might have been that it must never be low) - and you only need VPP when programming. So, as long as you have \P high, you should be OK. Although as you note, it's not supposed to be used this way.
Datasheet: http://www.seeit.fr/uploads/docu/M27C128A.pdf

The absolute limits table is about avoiding damage to the chip, not making it work. It doesn't tell you anything about normal operation. For that, you need the rest of the datasheet. In particular, table 3 has VPP at VCC for every non-programming mode. That's how it's supposed to be used.

If you have VPP at 12.75V during reads, that's actually a verify. You will get data, but it might be slower - tables 9 and 7 say maximum 100ns from falling output enable to valid data in verify mode, versus 80ns in read. The times from chip enable and address aren't given for verify, as they're supposed to be stable from before the programming pulse (which needs them stable for 2us!).

With VPP at 0V, anything could happen. It's not intended to be used like this, so the datasheet makes no guarantees. I've never tried running an EPROM like this, so I can't say whether that is the cause of your instability. But it's definitely wrong, and an easy fix.

> I've never tried running an EPROM like this, so I can't say whether that is the cause of your instability. But it's definitely wrong, and an easy fix.

Yes, on second thoughts, I'd go with this advice!

In my POC units I connected Vpp to ground without any ill-effects.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

the ST data sheet doesn't explicit say that VCC requires to be 5.0V +/-5% or +/-10% during read. It just "recommends" that.

This Microchip data sheet explicitly says VPP shall be VCC during read mode operations (Table 1-2 bottom column). Most likely Microchip don't try to make their devices incompatible to others.

Thank you all for your thoughts! I made sure its tied to Vcc. However, instability continued. But in the end I recognized a pattern in the "random" crashes, which continued to occur. First I did not understand them, since I simulated my software and there it did never crash. However, what I did not see first is that the simulator erased all RAM at reset and my hardware did not. This meant,... after power up, software ran and after reset it crashed. The reason was the init routine of my system detected that it had to clean up variables because of old memory content, and jumped to a wrong memory address... *gnah*. At least changing Vpp did not change system behavior, but of course one should stick to the data sheets...

Hello there !

I just experienced something very strange related to this topic.

The chip is a SGS M2716F1 eprom.

After 6 months of storage i get troubles with my Elektor Junior Computer and I check everything and mainly the eproms (I know it's a sensitive point).

On my eprom burner Vcc is 5 V.
During reading the vpp pin is connected to Vcc (in fact 4.56 V = 5 V less a shottky diode direct voltage). There is also a pull down 10 KR resistor from vpp pin to gnd.

After 20 minutes into the uv eraser if I read the eprom with Vpp pin high the eprom look empty ($FF everywhere).
But If I read it with Vpp pin to gnd via the resistor some datas are still there.

Then I put the eprom again in the uv eraser by two minutes steps and progressively (2 or 3 steps) all the datas were erased (always reading with Vpp pin to gnd)...
Of course with Vpp pin high the eprom also look empty...

Then I program the eprom and everything went fine.

The question is if this eprom has some defaults or not and if I can use it safely (in the destination assembly Vpp is always high) ?
Because the datasheet say Vpp high during read operations what can explain these observations ?

Thanks for help and advises.

Regards.
Philippe

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?