UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 9th Oct 2012, 8:45 pm   #21
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Eprom data life?

Quote:
Originally Posted by Dave Moll View Post
It may also depend on how well the window has been covered - thus keeping out the UV that gets as far as the exterior of the chip.
For a while in the nineties I worked for a company which did some work on behalf of BT in their telephone exchanges, and I remember noticing that there were usually signs in there saying "No flash photography" - not '"No photography", bur specifically "No flash photography".

I never did find out the actual reason why, but at the time I did wonder whether they were worried about corruption of EPROM data due to the intense UV light content.

With hindsight, maybe it was something else entirely like a risk of explosion due to environmental factors in certain areas - I don't know.
SiriusHardware is online now  
Old 9th Oct 2012, 11:44 pm   #22
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,526
Default Re: EPROM data life?

There would have been a large lead acid battery standby supply somewhere.......
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 10th Oct 2012, 2:08 am   #23
Refugee
Dekatron
 
Refugee's Avatar
 
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
Default Re: EPROM data life?

New standby batteries have to have tin flues with fans to keep the risk low.
There should have been plenty of no smoking signs to go with that sign.
Refugee is offline  
Old 6th Nov 2012, 11:20 pm   #24
woodchips
Octode
 
Join Date: May 2010
Location: Grantham, Lincolnshire, UK.
Posts: 1,172
Default Re: EPROM data life?

Some real data on EPROM life. Just trying to resurrect my Motorola Exorciser that uses 2716 EPROMs, should be 2708 but couldn't program them so did a mod to use 2716s. These were programmed just over 30 years ago, and can still be read, and copied. I had most trouble with the tarnish on the programmer (Stag PP39) ZIF socket that wasn't making connection. The board with the EPROMs has been stored in the loft for the last 15 years which possibly explains that.

One puzzle was that the EPROMs were Intel 2616s, not 2716s, what is the difference? I can't at the moment find any useful data on them.

Bob
woodchips is offline  
Old 1st Dec 2012, 9:49 pm   #25
medge799
Tetrode
 
medge799's Avatar
 
Join Date: Feb 2009
Location: Saffron Walden, Essex, UK.
Posts: 98
Default Re: EPROM data life?

Hi Bob,

Quote:
Originally Posted by woodchips View Post
One puzzle was that the EPROMs were Intel 2616s, not 2716s, what is the difference? I can't at the moment find any useful data on them.
Looks like the 2716 (EPROM) was the erasable version of the 2616 (PROM). See http://www.intel-vintage.info/intelmemory.htm for more info.

Hope this helps,
Mark
__________________
Mark Edgeworth
____________________________________________
BVWS Member
medge799 is online now  
Old 2nd Dec 2012, 11:42 am   #26
barryknapp
Retired Dormant Member
 
Join Date: Apr 2012
Location: Gloucestershire
Posts: 3
Default Re: EPROM data life?

Reading these posts reminded me of an experiment I did with a 2716 EPROM in the early 80's. There were dire warnings at the time concerning the necessity to cover up windowed EPROMS to prevent data loss. Being a sceptical ****** I placed a programmed EPROM on top of a wall at the back of the building in full summer sunlight.

Every few days I would test the chip to see if it had lost any data. As nothing appeared to be happening I increased the times between tests untill eventually I forgot about it. About three months later I remembered the chip was still on top of the wall and tested it again expecting it to be well stewed. Much to mine and everyone else's surprise it still tested ok!

I conclude from this exercise that is more likely to be chemical degradation inside the chip over a long period of time than exposure to natural light which can cause failures. Ceramic cased chips are going to be better in theory than plastic OTP ones.

I definitely makes a difference which make of programmer is used. The traditional slow and steady ones are more reliable than some of the more modern fast algorithm ones. I have never encountered a problem using Data IO, STAG or SMS programmers, although a device programmed on one machine will sometimes show as faulty on another, due to more rigourous testing methods.

Happy days.
Barry
barryknapp is offline  
Old 2nd Dec 2012, 1:25 pm   #27
woodchips
Octode
 
Join Date: May 2010
Location: Grantham, Lincolnshire, UK.
Posts: 1,172
Default Re: EPROM data life?

Mark, yes, the 2616 is a OTP. Looking more closely they have paint on them, scraped off, and part numbers so I think they were firmware updates for the Intel MDS we had, and I rescued them from the bin.

Programming a 2716 does bring back just how slow they were to program, minutes. Doing a 271024 would be quite some job, hours. Presumably the time was because the charge storage wells were large, another reason why they last so well? The 1702 needs something like 60V to program, that is a pretty high voltage for a semiconductor like a memory.

Bob
woodchips is offline  
Old 3rd Dec 2012, 1:25 pm   #28
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
Default Re: EPROM data life?

Quote:
Originally Posted by woodchips View Post
Programming a 2716 does bring back just how slow they were to program, minutes. Doing a 271024 would be quite some job, hours. Presumably the time was because the charge storage wells were large, another reason why they last so well? The 1702 needs something like 60V to program, that is a pretty high voltage for a semiconductor like a memory.
Bob
Early EPROMs also became quite hot when being programmed: I wonder if the manufacturers mandated slow programming in order to prevent the semiconductor slice overheating?

--G6Tanuki
G6Tanuki is offline  
Old 4th Dec 2012, 2:16 pm   #29
PeterRoberts
Pentode
 
PeterRoberts's Avatar
 
Join Date: Dec 2011
Location: Camberley, Surrey, UK.
Posts: 145
Default Re: EPROM data life?

Hello All,

I have a home-built memory board in my Icom IC-R71 receiver that uses a 2716 EPROM. I programmed it in the late 1980's and its still working today. Similarly, I have another radio using a 2732, last programmed about 15 years ago and that is still working fine too.

Cheers,

Peter
PeterRoberts is offline  
Old 4th Dec 2012, 6:12 pm   #30
Guest
Guest
 
Posts: n/a
Default Re: EPROM data life?

It's odd thinking that the electrons stored in a weeny bit of silicon can stay there for so long. With modern, i.e smaller, chips this must amount to but a few thousand of the beasties. You have to admire modern technology.
 
Old 25th Dec 2012, 12:29 am   #31
VintageHarry
Retired Dormant Member
 
Join Date: Dec 2012
Location: Melbourne, Australia.
Posts: 2
Default Re: EPROM data life?

Here's one to watch out for:

Windowed UVEPROMS are erased by the UV photons knocking trapped charge out of wells in the chip. Now, any photons will contribute a bit to this process so,

1) It's very sound advice above to place opaque stickers over the chip windows

2) DONT PHOTOGRAPH AN OLD UVEPROM BOARD WITH A FLASH!

Sorry to shout but old UVEPROMS have been blown away like this and it just takes a moment to forget that your clever digital camera will often default to flash without asking.


PS The 1702s were programmed with 25V, far higher than later chips and I have wondered if this has helped data longevity by driving that much more charge into the cells. I have some similar vintage 25V chips that also seem to have done well.

H
VintageHarry is offline  
Old 25th Dec 2012, 10:52 am   #32
lesmw0sec
Octode
 
Join Date: Jul 2009
Location: Carmel, Llannerchymedd, Anglesey, UK.
Posts: 1,498
Default Re: EPROM data life?

The odd thing is that there seemed to be some "recovery" effect. I recall working on a prototype using a 2716 by a sunny window. When the sun shone directly on to the 2716 the programme went daft. Shielding the window with my hand restored normality.
One hopes that the life of the PIC types will be a bit longer. I want my bread machine to keep working. It has already done 10 years service!
lesmw0sec is online now  
Old 6th Feb 2013, 1:43 am   #33
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: EPROM data life?

Quote:
Originally Posted by lesmw0sec View Post
The odd thing is that there seemed to be some "recovery" effect. I recall working on a prototype using a 2716 by a sunny window. When the sun shone directly on to the 2716 the programme went daft. Shielding the window with my hand restored normality.
One hopes that the life of the PIC types will be a bit longer. I want my bread machine to keep working. It has already done 10 years service!
Semiconductor junctions may possibly all have inherent properties like a phototransistor does - ie, you shine a light on them and they change state for as long as the exposure to light lasts. This effect is not normally observable in conventional semiconductors (normal transistors, etc) because they are intentionally built in lightproof packages - but I remember way back when, converting OC71 (Germanium) transistors into usable phototransistors simply by scraping the black paint off the glass envelope.

Re: Your bread machine, I think typical FLASH devices are supposed to hold their data for many decades, and if (as I presume) the control PCB was commercially made it may even have used a dedicated mask-programmed version of the processor.
SiriusHardware is online now  
Old 6th Feb 2013, 11:58 am   #34
GMB
Dekatron
 
GMB's Avatar
 
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
Default Re: EPROM data life?

Quote:
converting OC71 (Germanium) transistors into usable phototransistors simply by scraping the black paint off the glass envelope.
... until the manufacturers realised that people had worked out that this was cheaper than buying the photo-transistor version, and started filling them with opaque grease.
GMB is offline  
Old 8th May 2013, 3:30 pm   #35
ricard
Octode
 
ricard's Avatar
 
Join Date: Mar 2006
Location: Lund, Sweden
Posts: 1,631
Default Re: EPROM data life?

Quote:
Originally Posted by merlinmaxwell View Post
It's odd thinking that the electrons stored in a weeny bit of silicon can stay there for so long. With modern, i.e smaller, chips this must amount to but a few thousand of the beasties. You have to admire modern technology.
Then again modern flash memory types such as NAND flash do have fairly severe limits on the number of times they can be programmed, not to mention that the data in the bit cells decay over time to the point of being unreadable, unless measures are taken such as error correcting codes.
ricard is offline  
Old 8th May 2013, 3:58 pm   #36
GMB
Dekatron
 
GMB's Avatar
 
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
Default Re: EPROM data life?

Quote:
Ceramic cased chips are going to be better in theory than plastic OTP ones.
Unfortunately, ceramic packages contain more natural radioactivity than the plastic ones, so this is by no means certain.

By the way, you can actually measure the state of leakage on some types of EPROM. The ones with the high-voltage program-verify mode will do this as the idea of the program-verify mode was to compare the stored voltage with an unusually high threshold to make sure they are programmed OK. So if the verify fails you can wind the programming voltage down until the bits come right again and from that tell how well programmed they are.

The snag with the above suggestion is that if you get something wrong you will end up smashing the data.
GMB is offline  
Old 8th May 2013, 4:06 pm   #37
jjl
Octode
 
jjl's Avatar
 
Join Date: Jan 2003
Location: Ware, Herts. UK.
Posts: 1,082
Default Re: EPROM data life?

Quote:
Originally Posted by ricard View Post
Then again modern flash memory types such as NAND flash do have fairly severe limits on the number of times they can be programmed, not to mention that the data in the bit cells decay over time to the point of being unreadable, unless measures are taken such as error correcting codes.
This is in large part because most consumer NAND FLASH is of the multi-level cell (MLC) type where each memory cell effectively stores 2 bits of data by representing these as 4 distinct voltage levels. This system is therefore more "analogue" than single-level cell FLASH where each cell stores a single bit. As such, MLC is more prone to errors due to manufacturing variability, exposure to radiation, device aging etc but has the advantage of having twice the data capacity of SLC for a given number of memory cells and hence silicon die area.

John
jjl is online now  
Old 10th May 2013, 5:33 pm   #38
Valvepower
Octode
 
Valvepower's Avatar
 
Join Date: Aug 2008
Location: Rayleigh near Southend-On-Sea, Essex, UK.
Posts: 1,852
Default Re: EPROM data life?

Hi,

Interesting thread, as I’m currently experiencing some problems with a Microchip PIC18F43 micro controller, where the ‘last configuration’ data stored in the EEPROM memory is being lost. We have been using this particular PIC for some time with no problems, but out of the blue we are experiencing problems with a couple of batches of the chip. We have contacted Microchip and they have suggested the same as Ricard, where measures must be taken such as error correcting codes to stop the loss of data in the EEPROM.

This also ties with manufacturing variability as pointed out by JJL above.

Has anyone else had similar problems with ‘current’ microcontrollers?

Terry.
Valvepower is offline  
Old 10th May 2013, 8:58 pm   #39
Guest
Guest
 
Posts: n/a
Default Re: EPROM data life?

The newer Microchip offerings don't have EEPROM just the flash but you can only use it a few thousend times rather than the few million of EEPROM. I wonder what the difference is?
 
Old 11th May 2013, 1:37 pm   #40
jjl
Octode
 
jjl's Avatar
 
Join Date: Jan 2003
Location: Ware, Herts. UK.
Posts: 1,082
Default Re: EPROM data life?

This article has a pretty good explanation of Flash memory.

Possibly of interest is the fact that MLC NAND Flash requires quite sophisticated error correction such as BCH and wear-levelling algorithms in order to be at all useful. Error correction tends to be implemented in hardware in special purpose logic in the Flash controller device.

John
jjl is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:12 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.