Erasing Eproms, Can you cook them?

amsterdamny

New member
Joined
Aug 23, 2007
Messages
619
Reaction score
1
Location
pomona park, Florida
I am trying to burn some eproms but the burner keeps saying there not empty, at first I was not cleaning the window spotless, Now i am and they still say not empty, can you cook them leaving them in the uv to long? I used the burner before with no problem
 
Yes. You most definitely can cook EPROMs if you leave them in the eraser too long. Similarly, EPROMs can be damaged by plugging them in backwards, or powering them improperly, or programming them on the wrong setting. I've had a couple EPROMs that were "stubborn", and refused to erase all the way the first time. A blank EPROM will be all 1's (FF). Read it in and see how 'clean' it looks. I've had luck in such cases by programming the EPROM with all 0's, and then erasing it again. If you still can't clean it up, it's bad.

-Ian
 
I believe that's a newer version of the Willem that will, in fact, produce the necessary 25v programming voltage. But, at the moment, your problem is that the chip is not getting blank. That is a very old style EPROM, sometimes those do take longer than the newer ones to erase. According to the datasheet, it needs to be one inch away from a suitable UV light for 15 to 20 minutes.

How long was it in for?

Read it in, and see if it's "getting blanker" - most bytes should be FF, or something thereabouts.

Try programming it with all 0's, and see if it will take. You can program all 0's over anything, it does not need to be erased first - since you're just going through and programming all the bits that aren't programmed yet, writing to every cell.

-Ian
 
Ok as for the chip being old, I have tried 4 or 5 other newer chips and get the same thing, chip not blank, I started out with all the chips in the uv for about 10 min then 15 then 20 up to a hour, checking every 5 min
 
Ok it says Error at 0x000001 Chip = 0x01 buffer = 0x00
Size & check sum 0x7ff 0x0000
I have no idea what the hell any of that means

What operation were you trying to perform when you got the error? I take it you were trying to write 00's over the chip?

Basically what this means is that it hit an error at the very first cell of the EPROM, it's comparing the data on the chip to the data in the buffer, and the first byte came back as 0x01, but it's supposed to be 0x00, since that's what you're trying to write to the chip. Size came back as 0x7ff, (2 kilobytes), which is the size of the EPROM, and the checksum of the buffer would be 0x0000, since the checksum on all zeroes should be zero.

I also am assuming that your programmer software allows you to "patch" - i.e. write over a chip that is not blank. It should...

It sounds like you either have a bunch of bad chips, or your programmer is having problems with the 2716's. The latter actually seems more likely, espeicailly since I know you've been able to program 27128's before.

Bring 'em over next time you come to Ray's, I can check the chips on my system and see if they're bad or not.

-Ian
 
Not to rain on your parade, but you have that chip plugged in wrong. Pin 1 should be at the top left corner of the ZIF socket. You have the notch on the EPROM the correct way but it is plugged in at the bottom of the ZIF socket.

No, he's got it right. Unless his programmer operates differently from mine. If you compare the differences in pinouts between the 2716 and longer chips (like the 271001), it'll make more sense. The extra address lines of the longer chips are "above" the existing pins on the smaller chips. Since the 2716 doesn't have those upper address lines, it goes at the bottom, not the top.

-Ian
 
You can reverse polarize the floating gates if you leave the eprom under UV too long... many will still burn with enough patience, but not on the first attempt... you'll end up with stuck 1's that take a bunch of programming attempts before they'll clear.
 
No, he's got it right. Unless his programmer operates differently from mine. If you compare the differences in pinouts between the 2716 and longer chips (like the 271001), it'll make more sense. The extra address lines of the longer chips are "above" the existing pins on the smaller chips. Since the 2716 doesn't have those upper address lines, it goes at the bottom, not the top.

-Ian

Then his Willem is different than mine was. On mine Pin 1 was always in the upper left corner.

It also required an external power supply in order to be able to burn NMOS EPROMS reliably. I used a 12V HP laptop computer ps and then it worked. It never burned the older chips right with USB power.

ken
 
Ok, from what I read on there forum, This Willem thing is a useless piece of SH*T it will not work with 2716 or 2732 like it said when I purchased it, I would like to stick this piece of SH*T UP THERE AS*!
That said, Can someone recommend a burner, I would like to be able to do 25 & 27 series but at least 2716 2732 something that is easy to use.
 
I have had good luck with a BP Microsystems EP-1140, but all the BP's on EPay right now are rather expensive.

I was also able to burn 2716s and 2732s with the older Willem I had, but it took a 12v external power supply to be able to do it. The problem is that the older EPROMS burn at 21 or 25 volts and the Willems use a DC to DC power converted and the USB port just can't deliver the amperage necessary to keep the voltage up high enough.

Even with a 12V 1.2A power supply it wouldn't reliably burn 2532's so I sold it and bought the BP. I am sure some of the others here can let you know what their favorites are.

ken
 
I find it odd that the thing has indicator LED's for the various programming voltages, and yet doesn't have any sort of power supply connection. It seems to have been designed to generate the necessary programming voltage from the USB connection.

What happens if you press that button right next to the programming voltage indicators? Can you just change the voltage that way?

-Ian
 
Back
Top Bottom