• Please review our updated Terms and Rules here

2716 eprom

Mike_Z

Mike_Z

Veteran Member
Joined
Dec 1, 2013
Messages
1,713
Location
Near Milwaukee Wisconsin
My 8080A machine has 64K of memory. The lower 32K is a 32k x 8 SRAM. The upper 30K is 16 2k x 8 SRAM. The highest 2k is a 2716 EPROM, that holds a small monitor program that I wrote long ago. The last time I reprogrammed this EPROM was in 1989. The chip works fine, but I'm looking to make some additions and improvements to the program. The programming is complete, so I dragged out my eraser and a hand full of old 2716 chips. I placed a half a dozen 2716's into the eraser over night. Only one chip was completely erased, most had one to six errors, each error byte had all zeros in it. These old chips were purchased new in 1986. I thought that maybe if I cycled the chips maybe it would help, but no soap. Most chips would program but there were a few errors, one chip would not program at all. I read on this forum about bit rot, but that seems to be about individual bits of a programmed EPROM changing.
I looked around for some new 2716's, but seems they are not made any more. There are some 'REFURB' units, but that must be some of what I already have. So....I'm thinking about a EEPROM. I don't know much about them. Is there some that is pin compatible to a 2716? In a nut shell how are they programed?

Thanks
Mike
 
Maybe a 28C16 (example AT28C16, CAT28C16) would work, although those might also be obsolete in DIP packages now too.
 
Mike,

2732 chips are pin compatible with 2716's. The only trick is to take the 2716 image and 'double it', that is copy it twice into the 2732. The easiest way I know to do this is : copy/b 2716.bin+2716.bin 2732.bin in a DOS command. Then burn the 2732.bin image to the 2732. This guarantees the image is in the right place (upper or lower bank) regardless.
 
Mike_Z said:
, so I dragged out my eraser and a hand full of old 2716 chips. I placed a half a dozen 2716's into the eraser over night.
Click to expand...
So, six to ten hours under UV. Normally, depending on circumstances, five to twenty minutes is sufficient. The Internet contains lots of references to EPROM damage caused by UV overexposure.
 
kyeakel said:
I'll second the 28c16 or the DS1220AD, I've used both. I prefer the 28c16 and have bought them by the tube.

Kipp
Click to expand...
FWIW, they're both rated for 10 years data retention, and depending on the programmer one may be easier to program than the other.

I'd just get another 2716 or 2732, but definitely don't leave it under the UV overnight!
 
Chuck(G) said:
An adapter for a readily available 27256 (or 27C256) would be very easy to construct.
Click to expand...
...or cheap to buy. And you could use '128s as well, of which I just happen to have a few hundred... ;-)

A bonus when using a larger EPROM is that you can add a jumper or a switch and select among alternate images.
 
Last edited:
Chuck(G) said:
An adapter for a readily available 27256 (or 27C256) would be very easy to construct.
Click to expand...

As long as were talking about making adapters, why not ditch the EPROM and the UV Eraser thing and go with an EEPROM. I believe the Atmel AT29C256 is mostly pin compatible with the 27256. Here is where I got mine. 40 bucks seems like a lot for one EEPROM, but if you scroll down and read the listing carefully, you'l see that it's a 40 bucks for 10 pieces. I sent the seller a message stating that I only wanted 2 pieces and he sold me two pieces for 8 bucks plus 6 bucks shipping.
 
Some great ideas. Which I will be looking into. But..... last night the machine developed an intermittent problem. It seems it wants to reboot all by itself. So, I'll be spending my time trouble shooting. I like the switch idea of a large EPROM and I still want to investigate the EEPROM and probably will purchase a few, either to use or to just play with. Let you know what I find, when I find it.
Thanks
Mike
 
Well, I pretty sure that my problem was self inflicted. I added a programmable interrupt controller 8259. Wired it in and programmed it. The programming included an unmask command. I had read somewhere that the 8259 had internal pull down resistors on the IR lines. Well, apparently they are not big enough. I had left all the IR line open and I figure that they were floating enough that interrupts were being generated. I added some 1k pull downs on the IR lines and been watching the machine for the last few days and seems we are back to normal.
Mike
 
It's best not to assume the operation of any pullups or -downs in a device, even if it's mentioned in the datasheet. Most of the time, these are very high resistance (~~100K) and not to be trusted.
 
Personally, I get my EPROMs (UV) for about $1 a piece when I pick them up bulk. If you already own the eraser, it's more economical to use the old stuff sometimes. Just troll eBay for working pulls. :)

ibmapc said:
As long as were talking about making adapters, why not ditch the EPROM and the UV Eraser thing and go with an EEPROM. I believe the Atmel AT29C256 is mostly pin compatible with the 27256. Here is where I got mine. 40 bucks seems like a lot for one EEPROM, but if you scroll down and read the listing carefully, you'l see that it's a 40 bucks for 10 pieces. I sent the seller a message stating that I only wanted 2 pieces and he sold me two pieces for 8 bucks plus 6 bucks shipping.
Click to expand...
 
The best lessons are gotten through experience. Now I know. I have a home brew eraser. It uses a GE Germicidal lamp G8T5. It's 30 years old and I wonder if it is still good. I erase for maybe 20 minutes and most of the bytes are erased. I've been looking at some spec's for these lamps and the only last 3-5 thousand hours. Mine has not operated that long, but maybe it is weak. I'll get a new one. I have found a 2716 that does erase and I have programmed and erased it a couple times, successfully, so maybe I'm OK. I want to do more testing.
Mike
 
Well..... I picked up 15 2716 eproms on ebay. They arrived today. The seller said they were new, but I have my doubts. One chip you can see corrosion through the window, another all the internal leads are gone, a number of the chips have adhesive on the window. I erased the bunch for 15 minutes. I purchased a new UV lamp, a G8T5. I know this one is working, you can smell the ozone. Of the 13 that could have worked, 10 still had data in them. The data pattern of each was all the same 370 octal in all cells. The other three were erased. I then pulled a ram chip from my memory card and tried each 2716 in the computer. 11 of the chips must have internal shorts, because the machine would not boot. The other two allowed the machine to boot. I tried to program these last two chips, one worked and the other would not do anything. Rather poor showing.

So.... I started to look at EEPROMS. I found on ebay some 8x2k chips an Exel XL2816A, Intel D2817A and a MicroChip 28C17A. They are all reasonable in price. Not having any experience with EEPROMS, is there a difference between these, one better than the other? Should I go for a larger chip and switch out unused parts. Are the larger (and newer) chips easier to use?

Thanks for the help
Mike
 
You must log in or register to reply here.
Back
Top