[Coco] MPROM woes (cured)

[Gene Heskett]

On Saturday 15 November 2008, Roger Taylor wrote:
>At 06:45 PM 11/15/2008, you wrote:
>>At 06:24 PM 11/15/2008, you wrote:
>>>You DID notice that it came with a harness for (3) 9V batteries,
>>>right? As I said - I have NO idea on the condition of that unit. It
>>>was in a box I got from Art before he moved, hence why I wouldn't
>>>"sell" it to you. Wouldn't have been right... Tony
>>
>>I resoldered the broken battery harness and it works great.  I do
>>recall with my old MPROM board I owned years ago that if the 9v
>>batteries were drained even just a little, it had trouble
>>writing.  And it didn't take long at all to suck the power out of
>>the batteries.  That was the part I hated the most, and not having a
>>25-27v unregulated power supply to feed into the board instead of
>>the battery leads.  I have several 12v regulated supply units, but I
>>don't want to risk trashing the board by doing any experiments like
>>that.  MPROM boards are hard to come by.
>
>Ok, the problem was that the source voltage was just not high
>enough.  I put 3 brand new alkaline batteries in the harness and now
>the unit is writing a 27128.  :)
>This is an issue I dealt with many years ago, going through many
>batteries just because they fell below something like 8.5v and didn't
>supply over 25v to the regulator.

Which is one of the reasons that the only burner I ever built from scratch 
used an LM317 to switch the Vpp for the programming burn voltage.  One trick 
that would only work with a Z-80, and that's what I was using then :-(, is 
that there is a difference between the read and write pulse timings from the 
Z-80.

It was a case of, if memory access is true and the read pulse isn't then it is 
going to be a write even if the WE doesn't go true for another clock cycle.

I used that logic to trigger a one shot that set the burn time, and when it 
fired it froze the Z-80 state so all the addresses and data were good, while 
it triggered the LM317 to go up to the needed burn voltage, in that case 25 
or 17, 12 volt parts hadn't arrived then.

The burn program did the write, then a read and if it didn't compare it wrote 
again, keeping tabs on how many write cycles it took to get a good read, then 
divided the writes by 2 as was recommended at the time, and did that many 
more writes before going on to the next address.

The lm-317 can switch its output up and down in about a microsecond with very 
little overshoot, so it made an ideal switch for that, just short part of the 
setting r with a 7406 for the higher voltage.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
I'm not a lawyer. I don't even play one on TV.

	- Linus Torvalds on the gcc mailing list