Is it possible to damage an EPROM by exposing it to UV light for too long? What if a portion of the EPROM stays unprogrammed ('FF') and the chip goes through some program-erase cycles? The unprogrammed area should fail sooner or later, shouldn't it?
Oh no, that`s an ultra secret: burn before reading!
SCNR
Saludos (an alle Vernünftigen, Rest sh. sig) Wolfgang
--
Wolfgang Allinger, anerkannter Trollallergiker :) reply Adresse gesetzt!
Ich diskutiere zukünftig weniger mit Idioten, denn sie ziehen mich auf
ihr Niveau herunter und schlagen mich dort mit ihrer Erfahrung! :p
(lt. alter usenet Weisheit)
No, and certainly not with a typical UV EPROM eraser with a typically
4watt UV lamp. Typical of most erasers is about 15mW/cm2 at 254nm for
15 minutes. Crank up the output much more and you'll smell ozone but still won't burn out the EPROM. If the EPROM was manufactured using a UV exposure masking process, typical UV levels during manufacture are
12watts/cm2.
It's a bad EPROM. If the blank check on your unspecified EPROM burner says it's not blank, it's not blank. The mostly likely culprit is wrong settings on your unspecified model EPROM burner. It's also possible for something in the circuit, or ESD to blow up the EPROM. It's also possible that your unspecified EPROM burner has a weak UV lamp and did not erase the entire EPROM. My favorite was having the label peel off and stick to the lamp inside my Data-Erase cheapo eraser. That cast a rather large shadow and caused numerous bad erasures. I didn't find the label until I took it apart to replace the lamp.
It might be helpful if you disclosed the manufacturer and model number.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
I used only sunlight (Phoenix, AZ, 32N latitute, sky probably 90% clear), a= nd for EPROMS written with all 0s, over 90% of the bits would erase in a da= y, 100% in 2 days, so I'd erase for 3-5 days. One summer I left a couple o= f EPROMs outside for a couple of months straight, and they all worked when = I wrote all 0s to them, and the programming didn't take longer than for my = other EPROMs, according to the programming method described in a TI data bo= ok, where a short programming pulse was used while the normal 5V supply wen= t to 6V and the byte was tested after each write. =20
I left some 25C64's In my manually controlled (that is, no timer) UV eraser for a couple of weeks, when I forgot to turn it off. They all still programmed and worked fine.
-- Lee K. Gleason N5ZMR Control-G Consultants snipped-for-privacy@comcast.net
In the late 1960's, I was working on a nifty new product that included some of the first EPROMs. The prototype was done and the boss declared that it was time to take a proper photograph for the product release, manuals, promos, literature, etc. After the photographers with their bright lights were done, I couldn't make it work. I eventually determined that all the EPROMs had at least been partially erased. My guess is about an hour under the very bright lights. I didn't see the lights but suspect they could have been arc lamps as incandescent lamps doesn't have enough UV content to erase EPROMs. Installing a spare set of EPROMS got it working again. After that experience, I made it a habit of putting labels over the EPROM window.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
_Foil_ labels. Paper sometimes wasn't opaque enough.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Agreed. I still have a lifetime supply of 5 1/4" floppy disk write protect tabs. Most are black vinyl(?) or some opaque black plastic over paper. They go on easily, and remove without leaving a mess.
Incidentally, I just remmembered that I once won an arguement whether a stick arc welder will erase an EPROM. An Intel 2764 was used because I had plenty of them. My guess is that it takes about 2 minutes of exposure to trash the EPROM and about 10 minutes to totally erase it. Keeping an arc going for 10 minutes was impossible, but 15 seconds at a time was easy enough. I also managed to get slag on every EPROM, but that was expected.
When I tried to use some of the test EPROMs later, I found that I had zapped some cells. I don't know how I did that since I shoved the leads into a crumpled block of aluminum foil.
Assuming 200A at about 10V across the arc, that's 2000 watts. My wild guess is about 1/10th of that is UV. Assuming a spherical radiation pattern, a distance of about 10 cm, and an old EPROM chip size of 0.25 cm^2 chip area, the UV power density is very roughly: Surface area of sphere = 4 Pi r^2 = 4 * 3.14 * 10cm^2 = 1256 cm^2 Power density = 2000 watts / 1256 cm^2 / 0.25 * 1/10 = 0.64 watts/cm^2 That's 130 times the UV that's delivered by the typical 15mw EPROM eraser. I would have expected it to erase the EPROM 130 times faster, but that didn't happen.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Somewhere in the 80s I salvaged a proprietary controller in a steel sheet cutting machine using this. We were rushing a repair late at night in the middle of nowhere in France. About three sticks managed to wipe 8 devices. It needed some convincing the customer though, he believed I was joking. The nicest part was the French cuisine afterwards ;-)
I meant another thing: what will happen if a chip undergoes several write/erase cycles where an area is intentionally left unprogrammed every time? It's like erasing it over and over. Probably nothing, I think. It must have happened quite often that some byte (or bit, why not?) would always be unprogrammed at every cycle in the prototyping process, so it shouln't be a problem.
Nothing will happen. Erasing sets all the bits to a "1". If the bits are already set to "1", then erasing won't change or wreck the bit.
EPROMs do have a limited number of erase cycles. It really depends on the age and technology of your unspecified EPROM. Usually it's about
1000 erase cycles before the silicon dioxide wrapped around the gate falls apart.
Well, there's an easy test. Erase your unspecified EPROM. Verify that every byte is set to FF with the blank check on your unspecified model EPROM eraser. Create a bin file where all the bytes are programmed to "OO". Write the file to the EPROM and compare the bin file with a read from the EPROM. If they're identical, then the programmer was able to change every cell and there are no bits stuck at "1". If there are any bytes NOT set to "00", you have at least one stuck bit.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Joining this late, I want to mention that the 27C series EPROMS can still be bought. You can buy them from China for about $2 each including postage for erased and tested "pulls" and new ones can be had for around $45.
They even come in the small flat packages that are common these days. (sorry, it's late at night, and I don't remember their name)
I specifically needed 27C256's and 27C512's, last summer, so I bought an eraser for around $15, a USB programmer for under $50 (including the adaptor for flat packaged chips) and a bunch of $2 chips.
The eraser was an odd combination of a 230 volt unit with a US 120 volt plug, and a mechanincal timer with a separate on off switch. I replaced the plug with one more appropriate for my needs and just leave the timer in the always on position, using an electronic kitchen timer and the on/off switch.
Geoff.
--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
Yes, sometimes. The reason is, EPROM designs sometimes use difference amplifiers with dummy cells that are factory-programmed and covered with a light-tight lid. So, if enough light leaks around the lid, the bias of the sense amplifiers changes (and the chip can become incapable of proper readout).
It's a matter of bleaching out the bias current sources...
Yep. I just stocked up on 27C512 EPROMS (for Kantronics TNC's) at about $1.50/ea it took about 2 weeks to arrive. All 20 were good. However, I had to peel the original labels and UV erase them.
Atmel or Winbond AT27C512
formatting link
I have a Willem version 5.0 programmer.
Highly recommended. There's also the 6.0 version, which allegedly has more features and requires fewer adapters:
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Cydrome Leader wrote in news:k1c45e$jsi$ snipped-for-privacy@reader1.panix.com:
flash memory(NOT a "ROM" aka Read-Only Memory) is not going to have the same pinouts,because one pin has to have the write enable command that a PROM lacks.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.