Do paper stickers protect an EPROM from getting erased?
![Adrian's Digital Basement ][](http://yt3.ggpht.com/ytc/AIdro_k7CrQRe71qN6c0Ua7A1P_I8KkxjqLa70u0zYeCvdvnKw=s48-c-k-c0x00ffffff-no-rj)
ฝัง
- เผยแพร่เมื่อ 5 ต.ค. 2024
- I have always wondered if a paper sticker on an EPROM is enough to keep it from getting erased. Let's test!
EPROMs are designed to be erased by 253.7 nm UV, which is in the "UVC" band. From the World Health Organization: "Short-wavelength UVC is the most damaging type of UV radiation. However, it is completely filtered by the atmosphere and does not reach the earth's surface."
-- Video Links
Using the sun to erase an EPROM: (and testing Xenon flash as well)
hackaday.com/2...
• EPROM vs Słońce - miga...
EPROM Wikipedia Article:
en.wikipedia.o...
UVC at Wikipedia:
en.wikipedia.o...
-- Tools
Deoxit D5:
amzn.to/2VvOKy1
store.caig.com/...
O-Ring Pick Set: (I use these to lift chips off boards)
amzn.to/3a9x54J
Elenco Electronics LP-560 Logic Probe:
amzn.to/2VrT5lW
Hakko FR301 Desoldering Iron:
amzn.to/2ye6xC0
Rigol DS1054Z Four Channel Oscilloscope:
www.rigolna.co...
Head Worn Magnifying Goggles / Dual Lens Flip-In Head Magnifier:
amzn.to/3adRbuy
TL866II Plus Chip Tester and EPROM programmer: (The MiniPro)
amzn.to/2wG4tlP
www.aliexpress...
TS100 Soldering Iron:
amzn.to/2K36dJ5
www.ebay.com/i...
EEVBlog 121GW Multimeter:
www.eevblog.co...
DSLogic Basic Logic Analyzer:
amzn.to/2RDSDQw
www.ebay.com/i...
Magnetic Screw Holder:
amzn.to/3b8LOhG
www.harborfrei...
Universal ZIP sockets: (clones, used on my ZIF-64 test machine)
www.ebay.com/i...
RetroTink 2X Upconverter: (to hook up something like a C64 to HDMI)
www.retrotink.com/
Plato (Clone) Side Cutters: (order five)
www.ebay.com/i...
Heat Sinks:
www.aliexpress...
Little squeezy bottles: (available elsewhere too)
amzn.to/3b8LOOI
--- Links
My GitHub repository:
github.com/mis...
Commodore Computer Club / Vancouver, WA - Portland, OR - PDX Commodore Users Group
www.commodorec...
--- Instructional videos
My video on damage-free chip removal:
• How to remove chips wi...
--- Music
Intro music and other tracks by:
Nathan Divino
@itsnathandivino
Back in the late '70s/early '80s a co-worker discovered a neat trick with the Motorola 68764 (oddball 8Kx8 EPROM in 24-pin package instead of 28-pin). If you plugged it in backwards and turned on the power, a little light came on in the window. For some reason it no longer worked if you put it back in correctly.
At least he discovered the LEEPROM (Light-Emitting EPROM).
The coolest thing I had ever happen concerning sunlight erasing EPROMs was when I fixed a Popeye bootleg.
It had some major sprite faults. After finding a short between a sprite ROM data and address pin, the sprites looked much better, but had ragged stripes through them. Injecting signals into the data lines showed the stripes coming out of the ROMs. I dumped the ROMs and one had these jagged lines through the sprites across the entire address range. When I looked at the ROM, I could see a scratch in the sticker on the window that had the exact shape as the stripes in the sprites. (well I just burnt the image that was supposed to be on the ROM over itself)
Pretty crazy!
tl;dr: Sticking a thin sheet of processed tree on your chips makes them remember your data
Trees make me remember things too
I. Mr. Carfenharghen was 5 mineral late too.
I put paper on my chips thanks to you. They tasted awful and didn't even reply when I asked them how much they remember about the downfall of British Leyland. Thanks for nothing Rene :(
Trees never forget.
Most brands list a ‘cooking’ time in the datasheet.
I looked up a couple of datasheets, and they even specified wavelength and intensity.
"UV light is not only damaging for EPROMs but also for humans"
I love this line :D
The shit can permanently blind you over time if you stare at it. It is not immediate but it is something that will happen when exposed to it over time.
It is the same thing with welders, as they have to wear safety goggles or face masks when welding, the UV from the welding arc will blind them over time without protection.
@@elfenmagix8173 Yes, I know, it is extremely dangerous. Some stupid people hang up UV lamps just because they look cool. crazy.
no rickets though!
yes, its called skin cancer, and diminishing eyesight.
Yup we did that indeed a long time ago. We had bad and good eproms. We had a test with 2 eproms in a window pane, and one was cleared in 3 weeks, and one (in the same place) was there for more than 2 years and was still programmed and verified correctly.
I tried to clear EPROM on sun, left it there for a day, no change, tried plasma cutter plasma, no luck, 1W uv-c led done this in 5 minutes.
Next time, try sunscreen on ones without stickers?
Nice little experiment here! I'd been exploring and researching how to erase some old eproms I salvaged a while ago, and you brought up a ton of good points that I learned along the way.
Being in Florida, I assumed the sun would be enough to erase my eproms. After letting them sit out between 11 AM and 3 PM, (the brightest and most intense output of UV from the sun during the day), I only had one out of four go corrupt, and even still that wasn't a complete erasure. Then I let them sit under some indoor florescent lamps for a similar length of time, no results. Then I borrowed an LED UV grow lamp from a friend, and they never successfully erased after 12 hours sitting under the lamp.
After I did some research, reading the datasheet for my particular eproms, I learned that they will only erase with a specific wavelength of UV light, and that all the other sources I'd been trying to use, up to that point, didn't output the correct wavelength. Even the sun itself only outputs a fractional percentage of the specific UV wavelength protons that are needed to erase the eproms, so I literally got lucky when that one eprom I mentioned earlier had some corruption.
That's when I bought a vintage Datarase eraser off of eBay, after learning of the reliability issues the Chinese erasers have, and that was the best $40 I spent.
quartz glass in the eraser and the chip window, regular glass blocks a lot of UV anyway.
In my EE classes we always just used masking tape which was seen as sufficient protection against the crappy mid-90s tube fluorescent lights at my university.
Incandescents and CFLs both emit some UV as well. So, maybe it's less relevant now, but "back in the day" there was indeed some concern about leaving them out uncovered. Maybe it's not the exact right wavelength of UV, but it may be close enough. And it's not like it needs to erase the entire thing, just messing up a few bits could be enough to bork it.
The glass envelopes on both types of bulbs are largely opaque to UV light.
In my experience, people like to repeat things over and over as fact, and rarely do they ever have a primary source where the person actually tested if the claim is true. There tends to be a degradation of knowledge the further you get from the source of it.
Given that many people have put EPROMs in direct sunlight, and it's taken weeks/months to erase the EPROM, color me skeptical that a florescent light will erase an EPROM in any amount of time....
I will say it's worth testing though.
I have read of a test a looong time ago .. the conclusion was that their eprom got corrupted from normal office CFLs in a little under 8 years. But eprom loose bits after 10s of years even if perfectly shielded
edit: meant EPROM not eeprom ..sorry
True that it only needs one bit to change for the whole eprom to essentially be corrupted. I guess what he can do is read 4 more chips, and place them in various locations around his home. One by a window, one just on the bench, one in a drawer and one inside a computer. Then read them in a year later. My assumption is that both the ones in the computer and in the drawer would be perfectly fine, the one by the window would likely be bad.. would be curious to see the one on the bench. I think most people's gripes on his channel are related to eproms inside computers where they'd have almost no UV light reaching them.
@@brianv2871 also a reminder that modern windows (at least double glazed ones) I think commonly all use Low-E glass. Vs the old single glazed plate glass ones which had no extra special formulations. So yeah, would be interesting to see it infront of a normal window, and also just, out in the open air or open window.
Now the intro logo is perfect! I love how it changes from orange to blue!
It's worth pointing out that the concern with uncovered EPROMs isn't them getting completely erased, it's bit rot accelerated by the occasional errant high energy photon.
I think I would be more worried about a power spike.
I have an eraser very much like that one and the timer is in 10s of minutes, so turning it to 6 runs for an hour
So, maybe Adrian's eraser isn't broken? He thinks it will end after 6 minutes but it takes 60 minutes :)
Even on his eraser there's a little black marking that says x10
Every TH-cam expert knows that you have to recap the eeproms, because their memory cells are all little capacitors with a transistor!!!
I think you're thinking of DRAM, that uses a small capacitor to hold a state for a very short time, hence it requires constant refreshing.
(E)EPROMs basically use a MOSFET with an isolated gate and hence don't require and refreshing except maybe after 10's of years. Programming involves inducing a charge near to the gate to cause it to change to represent a 0 or 1.
MOSFET gates have capacitance, but they aren't capacitors.
@@TomStorey96 ...another spoiler and victim of meta-level irony. facepalm
@@TomStorey96 whoosh
@@TomStorey96 I bet you're fun at parties! 😁
@@RetroJack I bet he is really fun. :)
My first "exposure" to EPROMs was in the late 80s when I started a 30 year career in marine electronics. I was servicing/installing JRC GPS navigators. Each month JRC would send updated firmware for their JLR-4000/4100/4200 series of GPS receivers. I had a multi gang EPROM programmer and used a UV-B plant grow lamp to erase the old EPROMS. This was most certainly the wrong lamp but it did erase the EPROMs after a few hours or so :-)
I am glad that you adressed this topic, i wished i saw this video earlier, since I was one of the commenters complaining that you keep those uncovered. I assumed the EPROMS were far more sensitive to general lighting and get erased far more quickly. You brought a god argument olso saying that the chips stay in the computer itself where there is barely light penetration. Great video again Adrian! Keeep up the good work!
Back in the early-mid-'70s I worked in the Launch Control Center at Kennedy Space Center in a department that handled "special" non-launch related research instrumentation and data processing. (Great fun! Our small department got to do the really interesting stuff and not worry about schedules or launch preparedness. :-) We had an expensive piece of equipment arrive and anxiously put it our rack, wired it up, pushed the "on" button and nothing useful happened. :-(
We sent the device back to the manufacturer, but they reported back that all the EPROMS (they had no paper cover) had been erased! The EPROMs were replaced, returned to us, and the unit worked OK. It turns out that, at the time, the Space Center had a policy of photographing all new equipment as it arrived, before it was delivered to the department that ordered it. The photography department documentation procedures would remove the covers and photograph the inside of equipment. And of course they used strobe flash lighting just inches from the device. Once we knew that, the reason for blank EPROMS was obvious. Apparently EPROMS were still a relatively new invention but almost immediately after that instance, the policy of the photography department changed. Go Figure! Moral of story: Don't photograph your uncovered EPROMs.
I haven't touched an EPROM in 30 years but during that time, I don't remember having to "cook" EPROMs for long periods of time. I did a lot of EPROM programming in that job and others, but I don't remember long waits for an EPROM to be erased. Perhaps just 5 minutes? Is my memory that fragmented? Or did we have faster erasers back then?
Interesting! I really have to wonder actually if these EPROMs were just never programmed before the gear was sent. I linked a video in the description where a guy tested an EPROM outside in the sun but also under a Xenon flash tube -- and he was not able to erase an EPROM with a flash tube. Although perhaps those super early EPROMs were more susceptible :-) But I would personally assume the gear just had blank EPROMs installed in the first place.
@@adriansdigitalbasement2 Seems like a reasonable explanation. Did the gear have an "Inspected by #5" and golden oval QC sticker anywhere? ;-)
Cool intro animation and fixing the zoom-in. Don't think anybody else noticed but I did.
Straight from the National Semiconductor 27C256 datasheet.
"The erasure characteristics of the 27C256 are such that erasure begins to occur when exposed to light with wave-lengths shorter than approximately 4000 Angstroms. It should be noted that sunlight and certain types of fluorescent lamps have wavelengths in the 3000-4000-range.After programming, opaque labels should be placed over the 27C256’s window to prevent unintentional erasure. Covering the window will also prevent temporary functional failure due to the generation of photo currents"
You will find similar instructions in most EPROM datasheets.
4000 Angstroms = 400 nm.
I used to use the write protect labels from 5 1/4 floppy disks as EPROM labels. Worked nicely as those were typically some kind of metallic tape. Have the same datarase II eraser, too! Cool
We discussed this in college. UV has a lot less penetration power than visible light and cannot even penetrate glass. The window on an EPROM is made quartz because it is transparent to UV. I checked on the web to be sure I was remembering correctly and, apparently, my brain cells are working today 🤓
Back in the 80s I had a job of testing electronic circuit boards used for building controls. Each board had 3 EEPROMS on the front. We had over 15 test/burn-in stations all running the same building control board. Initially the boards had blue vinyl labels covering the EEPROM window. The blue labels seem to work fine until someone came in one day and took a photo using a camera with a flash attachment. Every controller started to reset right after the flash. Engineering immediately changed the blue labels to foil labels which wasn't affected by a flash camera.
Loved this video. The question you answered here is one of those that we have on the back of our minds and never find the time to do a quick test, as you always have "better things" to do. Thank you for this Adrian !!!. Very comprehensive and, as you showed us, an analogic excel suffices.
Many stickit posters has a nasty habit of losing adhesive strength, so it is somewhat risky:)
12 hours under UV? The chips don’t have any germs on them, I’m sure!
Thanks for answering some questions which were disturbing me since decades. 😅👍🏻👍🏻
Ohhh! You should have done a part 2, and peel off the stickers, and see how little it took to erase them! :D
Why? What would that prove that we don't already know? He even proved that 30 minutes was more then enough to erase the chip that didn't have a sticker.
Forget erasing for a sec - what the heck is on the Grass Valley Group chip right there you show in the video? For those who don't know, Grass Valley was Atari's little R&D group tucked away in Grass Valley, CA where the original VCS/2600 was designed, and much of the work on the 400/800 happened there too, I think.
Apparently leaving them too long under UV light can cause the bits to 'stick'. I had some stubbon EPROMs that wouldn't quite fully erase in the 10-15 minutes expected so left them in for ages (2hours +).
The did erase after that but I then noticed they couldn't be written to anymore. Even previously 'good' EPROMs I'd written to 6 months ago seem to be knackered after to longer exposer under an erase light...
"knackered" - fellow Brit hello
I've cooked my share forgetting about them
Thanks for doing this. It'll save my neck from all the head-shaking I do when I watch some random TH-camr scramble to cover an exposed EPROM when they open the case of some piece of equipment.
Interesting experiment! I always wondered how effective the paper labels were.
Most glass will block UV so even by a window it wouldn't erase it, unless the window was opened. Same reason you can't get a sunburn inside your car.
If you do need UV to pass through glass, like in aquarium UV sterilizers for instance, it has to be quartz glass.
I absolutely loved this video. I was somewhat aware that paper interferes with UV light, but as a complete test, removing the paper and then trying to erase the EPROM would prove that those CAN be erased still or using the EPROM that WAS successfully erased and adding paper to it to prove it is protected. Nonetheless, I am certain that paper blocks enough UV rays so that it isn't strong enough to erase those EPROMs.
I have accidently damaged the rom data on an eprom due to the xenon flash on my slr camera, when taking pictures of PCBs. Since that day, I always keep a sticker on windowed eproms.
I was wondering about xenon flash effects .. but more about flashing it while in operation. (like the raspberry pi 3 that crashes when fotographed with a flash)
@@VVerVVurm Same issue. There is an exposed silicon chip on the pi, I don't remember what it does. UV can induce voltage like a solar panel. It's this same effect that causes eproms to be erased by exposure to UV.
So yes, it's exactly the same 'issue'.
The LED flash on many modern cameras and phones will not cause an issue. If you have a higher end camera with a proper xenon flash though, careful.
@@rich1051414 ikr .. every transistor is a photo transistor if you leave the cover up
Wow, what a synchronicity! Just yesterday night I was browsing the web looking for info on EPROM stickers. Where to get them etc. I found very nice suggestions on several forums. Tonight, I open youtube, and THIS video was first youtube suggestion to watch!. Today, I have learned from you a couple of things I did not know. Adrian, I love your videos, you're the best. Keep up your excellent work!
What I remember back in the day was foil stickers and mostly used as safety precaution.
It's a quality control thing. The reason why manufactured developed eprom have a sticker is a quality step to prevent it from happening accidently(especially because manufacturing floor have have a variety of stray UV source). It is very hard to erase an eprom accidently. But if you flood it with UV the changes are almost 100% no matter the covering of the window. The one's in your shop are pretty safe from being erased.
Actually, the purpose of the sticker is to block light because light can also make faulty or partially programmed cells work!
I remember a time in the 80's when were designing and testing a computer mainboard with a programmable EEPROM with no label on the window. It was all running great, so we put it all back together and replaced the plastic lid, switched it on and it didn't work (no boot - POST fail)! So we took off the lid to see what we did wrong but it seemed to work OK again. So we replaced the lid and it didn't work! We found that just by holding the lid over the top of the system then it would stop working.
So the light was actually making the EPPROM work, if we stuck a label over the clear window it stopped working (lid or no lid).
After that, we always stuck a label on the window BEFORE we programmed/verified them!
Timer goes to 6 minutes? What about the x10 marking? Great vid as always btw!
Haha, he probably forgot... the timers on these things barely work. his writing should say 'x10 to ∞' :)
3:46 - lol, today every gamer PC is brighter than a regular room lamp. Yeah, not UV, but full of light :)
Please tell me I'm not the only one that read "Dataarse II"
I guess they're telling you where to put your data...
Where the sun don’t shine? 😆
i've seen 'aluminium' stickers used on many eproms commercially, although sometimes just paper stickers,
Stay cool Adrian. I’m seeing it’s gonna get mighty toasty in your area. No pun intended
That's a pretty snazzy EPROM eraser! Mine's not fancy like yours, although it does have 5 slots. It's actually a cheap toothbrush steriliser that runs off 5V. The one good thing about mine is that it erases the chips in like a minute or less, since it uses a UV-C tube. And it's lucky that it automagically switches off quickly, because you can definitely smell the ozone that it creates while it's running!
good tip with the toothbrush device.
I believe that the numbers on the EPROM eraser represents 10s of minutes, meaning that the time goes up to 60 minutes, not 6...
I've seen some EPROMS that had what looked like kapton tape over the windows.
I wasn't able to hear your A/C either, but I was also hoping to see you try and verify an eprom with a piece of post-it note on it. I'll take you word for it, but seeing is believing :)
I could not erase old EPROMS using the sun 🌞, even after weeks outside... Thank you for the good content!
Yeah, I think we've all done this hoping we could avoid buying an eprom eraser and it doesn't really work. It will likely corrupt them, but the time it will take to fully erase them could take months so it depends on how soon you need to reuse that eprom :)
The label is a convenient place to put the date and version number of the code contained within. Then again, I always worked with florescent lighting.
UV eraser tubes are made of quartz, and not just glass. I suppose that someone was afraid of erasing eproms from florescent tube lighting but the glass there is pretty good at filtering out much of the UV so I would imagine that it would take a very long time to have a measurable effect.
The White EPROM labels I used to use at work had a black layer inside the label to stop all chance of light getting through.
I left an EPROM in the window for a month or so to test how quickly it would erase and it still had all the data on it
Is the glass in your window UV resistant?
Sometimes even the adhesive residue on the EPROM window prevents it from being erased in expected time 🙂
Also you could have programmed all 00h in those EPROMs to see if at least one bit in any cell gets erased, that would be a more sensitive test.
Cheers from Poland 👍
Interesting experiment, thanks for sharing! 👍
I don't have an EPROM eraser, but the steriliser of my Braun toothbrush works great to erase two at a time!
I never bought myself an EPROM eraser and use mostly non-UV reprogrammable EPROM chips like the W27E257. But when the need comes up to program an UV EPROM I do use sunlight and I need to leave a chip several days outside in the sun before it tests empty. Therefore no need to worry about a little bit of light indoors, you really need ultra-violet violence to erase them.
So the conclusion is: use electrically erasable EEPROMs if possible (or even Flash) and avoid all of these concerns ;-)
that's a bit redundant... the first 2 E's stand for Electrically Erasable. UVPROMs are the ones that can only be erased via UV.
I switched long ago since they're not expensive and easy to flash.
@@misterkite The rare and elusive spicy version of RAS Syndrome
Redundant RAS
its telling that that even in 2020 CNC manufacturers still use battery backed CMOS for machine settings not flash memory.
hardware people don't trust flash memory for mission critical stuff. PC industry are corporate and thus charlatans.
EPROM and EEPROM require UV C spectrum , not produced by room lighting (flourescent does produce some UV but that is AV A) . If you have room lighting that produces UV C, you have bigger problems such as cellular damage to your body , rather than erasing Eproms
1. EEPROM doesn't need light.
2. EPROM requires UV light of less than 400nm, UV-A is between 315nm and 400nm.
@@Okurka. Yeah, if there's enough "light" to erase an EEPROM, your house is probably on fire. :)
EEPROM is an *Electrically* Erasable Programable Read Only Memory chip, they're not erased by light. Flash memory is a form of EEPROM.
some specialists say that you should periodically read, backup, erase and rewrite the important EPROMs in your devices. the recommended time interval is around 5 years. so take out the eprom, dust it off, read it, UV erase it, and write the same data back in it.
One of my friends would use black write-protect stickers to protect his eproms 🙂.
On a side note, I read that UV-C can't penetrate clothing (why we don't get sunburns under our clothes), so that may give a hint as to why paper is an effective UV-C block (the paper fibers are denser than clothing?).
We would repurpose the foiled-coated stickers that were used to write-protect (or write-enable--I don't remember) 8" floppy disks!
Thank you again for your videos. Had a refrigerator fault caused by two electolytic caps. Had to source, remove and solder the new ones on. No big deal although
Got the same eraser with of course a broken timer. Actually replaced the filmsy mains cable with an AU plug and bypassed the timer while I was in there. Haven't changed the bulb but from what I've found it only takes around 2 or 3 minutes to fully erase the EPROM with the stock bulb
Back in the day when I was doing EPROMS I bought the cheapest UV EPROM eraser I could find out of The Computer Shopper. It was cobbled together out of a plastic cassette tape case which had a UV bulb in the top half and a piece of foam glued into the bottom. No timer, no safety interlock, nothing. It worked fine though. Never had a need to get anything else.
The paper is thicker than the wavelength of the ultraviolet light so it's not too surprising that the paper covered eprom's weren't erased, but still a good test regardless :)
I feel like I half prompted this mentioning the unprotected EPROM in the Laser.
To note: I save some Photoresist PCB protective covering, the PCB is exposed with UV light to transfer the track layout, then you develop it in a chemical, so the cover is designed to protect against UV light (and minor scratches).
When I put a sticker on an EPROM, I tend to put the UV protective cover on the EPROM, then put the paper sticker on top of that.
I put one on that dash of my car for a full workday and didn't have any data loss. I theorized that the glass of the windshield was probably opaque to the band of UV light that is needed to erase the data, and that the window on the chip is probably a special quartz glass that passes that band.
Australian sun light can do it in summer!
I used to do it 20 years ago before getting a proper UV eraser.
Sitting on a workbench with no natural sunlight is unlikely to affect the contents.
The EPROMS datasheet will advise the correct erase time, it varies between manufactures.
I assume you mean eprom and not eeprom, and yeah, I guess if you have the entire summer to wait to erase an eprom. :)
Shouldn't be a problem! We used normal stickers in our test rig bioses for years and no issues.
Hell even no stickers without direct sun light has never caused me any issues. We had dedicated NIC BOOT EPROMS on our test bench in my first job. They laid there for 2 years, no stickers and were still fine. We just had a very small run roof strip for natural light the rest was fluorescent. And where those EPROMS lay was often under that roof window. Windows do keep out a lot of UV-B.
I built my own EPROM programmer with a 24" (clear) germicidal lamp, ballast from the hardware store, and a discarded IBM keyboard box. Worked great. I still have it, but the lamp sockets flop around now as between the UV and decades of age, the tape is in bad shape.
I remember a story about the photographer taking pictures of a PC for an upcoming ad. With the case off, every time he took a flash picture the PC crashed. Narrowed down to uncovered windows on the BIOS EPROMs.
Heh yeah there was another comment about this as well -- I would personally more suspect the electromagnetic pulse caused an issue. There are stories about flash tubes erasing EPROMs but have not been able to find any evidence of this actually being the case. (Like take working EPROM with data, expose it to flash then immediately read it back in again.) I put a link in the description of a video showing this kind of testing.
@@adriansdigitalbasement2 But he didn’t erase it with the flash. The chip output bad data apparently. How about read testing it while exposing it? (I almost said “flash” but that’s something entirely different!)
EEVblog #901 - Raspberry Pi 3 Photoflash Problem
th-cam.com/video/dDcsTnqVgWc/w-d-xo.html
Opaque labels definitely were a must... Just white light extends the access time of EPROMs which would cause problems in old devices where you were running up to the timing limits.
You could also erase an EPROM instantly with an old-style flash gun (with a xenon tube in). There was a company that sold an 'Instant Eraser' which was basically a flash gun in a box.
I have some disposable cameras, I should try that.
One summer a few years ago I set one out in the hot TN sun, and I left it for a week, it did have corruption but it was not blank. I always thought the paper wasnt enough so I use copper tape but I can save that
Nice updated logo.
For a little extra insurance, I like to soak my paper in sunscreen to make super duper sure.
I have had a eprom that went corrupt. It had no sticker and the cabinet had an open mesh at the top and rear. It stood near a window for many years. After it failed it stood there for some years before it came in fore repair. The eprom was as good as empty, tried to get the hex file from the manufacturer but it was to old (25 to 30 years). It took long because UV is (partly ? I blocked by klass. My eprom eraser does it most times in 10 minutes. I tried erasing them on UV tubes from a face tanner but after many hours they were still not erased.
Thanks for testing this, I wondered for a while if stickers were enough, I use alu tape after programming to bre sure
I always thought the stickers were to prevent accidental reprogramming/erasure, you have to explicitly peel the label off first before you can erase it.
i have seen some later eeproms with stickers that just have the Winbond or SST logos under them.
Now this is something I have suspected for a long time! It seemed to me that the labels that gets stuck over the window is less about preventing erasure and more about identifying what is programmed into the chip!
Evidently it does also help prevent erasure, but maybe that is somewhat of a secondary effect because usually these things are buried inside a computer or other device where light sources, and let alone the correct type of light sources, are unlikely to penetrate at the right intensity.
Somewhat plausible theory!
6:19 - "Close the drawer and the light STAYS ON!" - (with apologies to Firesign Theater :) )
Awesome collection.
Hey, in the 80's computer VDUs had eproms facing the neck of the CRT tube and eventually they would get erased. We said the VDUs were suicidal, wiping their own firmware. At the time we burnt new proms for firmware updates all the time so it was not much effort to fix them. Probably a mixture of UV light and x-rays going through the paper labels.
I’m enjoying this video but remain disappointed that the channel is not called “Adrian’s Digital Sub-Basement” :)
I think this is the most ranty Adrian can get.
You can't have seen his old video about bad Michelin windshield wipers
Nope, I know adrian, he can get much worse... just talk to him after an iPhone keynote when they announce one of their new "innovations". :)
Sarcasm doesn't translate well to the internet
wow so that paper stickier work so cool and so simple
Id be interested to see whats in that EPROM eraser. I have one that looks very similar, and it looks like it was made from a recycled CFL lamp ballast! The wiring was dreadful, nothing was insulated and it was only by a miracle that the tube tails were not shorted out. Oh, no lamp holder for the tube either.
sounds like the average American domestic central air system, they sure do love their rats nests and burcaps.
aluminium wiring was an economy thing post WW2, IR2 losses overheats and catches fire, previous to that was cotton covered wire, gets chewed overheats catches fire...
It's a good thing EPROM's have been superseded by other technologies. I'm pretty sure the visible-from-space, disco-nightmare of high-intensity ARGB LED lighting that I (and my enablers at various hardware vendors) splattered all-over the inside of my current rig, can not only emit enough high-spectrum UV light to erase those EPROMs, but could burn the paper labels right off the top of those ceramic packages. And yes, ARGB does make my 5950x faster, and I will hear no different.
Good to know! I just bought a bunch of EPROMs to program and have been wondering how susceptible they would be.
Ten to 15 minutes isn’t always enough with old EPROMs… the last ones I had to erase and reprogram took a half hour in the eraser. Also, there are certain chips the modern TL866 programmer will not properly program, you have to use an older unit to do those. Ran into that when trying to do some 2732s and 2764s. As for labels, I tend to take the labels off any that are intended to be repurposed as I don’t care about the content (or have it safely backed up somewhere) and only put on a label once the chip is programmed.
Hi Adrian, to manage my chips, the solution I opted for is tackle boxes, they keep the parts separate and come in many sizes and can be put on the bench in a pinch...
just got an old Heathkit ID-4803 EPROM eraser on ebay for the same price as the new chinese plastic ones, this one is an old metal one that they used to sell as DIY kits in the 80s and 90s. works great, erases chips in 10 minutes.
I was also curious about it once and I put a bunch of EPROMs without the stickers into the sun for 10 hours. We had 35°C hot summer day, so the best sun we can get here. And guess what?After 10 hours not 1 bit flipped. I couldn't erase the chips with just using sunlight. For the test, my EPROM eraser got them blank in about 10 minutes. You need a very specific wavelength to erase the EPROMs, less then 270nm if I remember right. Most of UV light reaching the earth is above 400nm, that's why it is only possible to damage some bits, but not to erase the EPROM completely....
You remember wrong, below 400nm works.
@@Okurka. Just looked into couple of datasheets of some EPROMs, they all say, that you need 253,7 nm. Anyhow, my point is, that it takes forever in the direct sunlight, which is mostly above 400nm.
@@necro_ware Those datasheets say the erasure begins at 4000 Angstrom (400nm) or lower and that 2537 Angstrom (253.7nm) is the ideal wavelength.
None of them say you need 253,7 nm.
@@Okurka. Well, not arguing on that. At least they suggest 253,7nm. But, again, my actual point was, that the sun is not enough to erase an EPROM. At least not in couple of days. I didn't try to leave it in the sun for months though :D
Look at the drawer, it's much darker than the rest of the eraser's plastic. The UV is killing your plastic drawer. I remember my brother bought and EPROM eraser back when we had a C64 but he was too curious and looked directly into the light, the next they he and my mom or dad went to the doctor because his eyes turned red or something. This was a long time ago I don't remember the exact details.
But officer, Adrian on TH-cam says that in order to protect myself from the harmful effects of the suns UV light, I simply need to cover myself from head to toe in Post-It(tm) Notes, that's why I'm in the supermarket dressed like this.
Thanks for this. I wrote the completed release of Star Fox 2 on to an EPROM chip and hand wired it into a doner SFX2 cart, so I could play it on original SNES hardware (this was before flash carts could emulate Super FX chips). I was just wondering if the chip would need to be rewritten eventually. Good to know!
That timer is up to 6 minutes? Why does it say minutes x 10?
my guess number x 10 = minutes so if you chose 6 it would be 60 minutes
@@Silver_Adventures That would be my guess but he said up to 6 minutes. Plus earlier he said something along the lines of the timer is broken, being a good thing, because that way he could do an hour if he wanted. Also, what lazy/cheap company would not simply put a 0 after each number?
The whole thing seemed weird so I was curious.
I know Portland weather might not cooperate, but I wonder if exposure to full sun would erase them. It'd be great to see if 1 hour, 1 day, 1 week does it. Maybe leave one uncovered in your suniest window and check in on it periodically. Sounds like good second channel video series.
Actually others have done this testing -- but also if you check the weather here now, it'll be 111F/44C this coming weekend. (And every day on both sides hot and full fun.) Here's an article on using the sun to erase chips: hackaday.com/2016/09/14/staring-at-the-sun-erasing-an-eprom
@@adriansdigitalbasement2 That's a great link, thanks!
judging by all beige plastic that has been yellowed by UV, the risk isn't zero even inside the case.
just depends on whether light enters through a vent or space and hits an EPROM.
it doesn't have to be totally erased, just one wrong bit can ruin it.
also, CRTs emit UV. strong enough or for how long... who knows?
i would say you're safe, esp since you can test and remake EPROMS. but if you give something away, be a good guy and cover it up.
Okay so this answered some questions. Can prolonged exposure to UV after blanking damage an EPROM to where it won't retain data at all or partially?