29 Jan
2009
29 Jan
'09
6:22 a.m.
Working with EPROMs, one must have an eraser. I built my own a while back, using a
germicidal UV lamp tube, a flourescent fixture, a sealed box and a programmable timer. It
works well, and I typically can erase an batch of eproms in like three minutes.
At least, when I was working primarily with 2732's and 2764's, I could. Now,
I've been needing to use higher capacity chips - and these seem to take longer to
erase. Five minutes isn't enough for some. AMD 27C010's seem to take closer to ten
minutes to erase.
My eraser has a drawer that can accept about a foot of eproms (I used a 12" tube). I
could probably cram more on there, but I try to keep them centered under the light. So, if
I stick a dozen chips on there, of varying sized and types, and 'nuke' them for
five minutes, I'll go and blank-check them, and some will be blank, others will be
nearly blank (lots of FF's, but some garbage left), and some don't seem affected
at all. I've been trying to come up with a correlation between manufacturer and erase
time, but it seems to vary. Last time, I nuked four AMD 27C010's that I pulled off of
an old LaserJet font cartridge. Three were unaffected and the other was blanked.
I worry about leaving the chips in there too long, since I know too much UV can damage
them. But at the same time, I wonder about the lifespan of the UV tube - and how many
hours it's good for. (I guess I should have checked the package when I bought it...).
So, how much UV exposure does it take to damage an EPROM? Is UV exposure cumulative? For
example, if I have a chip that's unaffected by five minutes, and I put it back in for
another three - is that the same as as running it for 8 minutes total, or do the charges
that the UV is supposed to dissapate simply linger? Are CMOS chips harder to erase, taking
a longer time than the older parts?
-Ian
29 Jan
29 Jan
7:20 a.m.
the chip docs should give an erase time, iirc the docs would say to
get to the clear stage and add x% time to guarantee the chip is set to
1/0
Dave Caroline
7:42 a.m.
--- On Thu, 1/29/09, Jules Richardson <jules.richardson99 at gmail.com> wrote:
Presumably more expensive erasers do things in-place -
i.e.
periodically do a blank-check during the erase cycle?
Hmm, I've never even heard of such a device. Even the professional EPROM erasers are
little more than what I've built, a box with a UV light and a timer. I've never
heard of an eraser in which the chip was actually connected to anything while being
erased.
If a machine like this exists, however, it would be quite cool! But probably very
expensive, and a pain in the neck to use, as you'd have to tell it what kind of chip
you were erasing, and it would need ZIF sockets and all kinds of electronics to read the
chips.
-Ian
8:29 a.m.
Mr Ian Primus wrote:
--- On Thu, 1/29/09, Jules Richardson
<jules.richardson99 at gmail.com> wrote:
True. I suppose my primary use is in recycling old EPROMs, so I grab one from
the junk box, erase it, then immediately program it. In that scenario, I'd
just be setting up the EPROM prior to the erase step rather than after it, so
it'd be no more work. In a commercial environment, the usage patterns are
probably different.
Like Dave I've normally found they need somewhere between 15-30 minutes to
erase - it just bugs me sometimes that I might be 'overcooking' the chip by
erasing for too long, so the concept of an intelligent eraser popped into my
head... :-)
cheers
J.
Presumably more expensive erasers do things
in-place - i.e. periodically
do a blank-check during the erase cycle?
...
If a machine like this exists, however, it would be quite cool! But
probably very expensive, and a pain in the neck to use, as you'd have to
tell it what kind of chip you were erasing, and it would need ZIF sockets
and all kinds of electronics to read the chips. 
10:04 a.m.
Mr Ian Primus wrote:
Even the professional EPROM erasers are little more
than what
I've built, a box with a UV light and a timer.
I have such a device, and the timer on it goes up to 60 minutes.
I normally run it for about half that, with 2716/2732/27128 EPROMs.
--
John Honniball
coredump at gifford.co.uk
10:21 a.m.
On Jan 29, 2009, at 10:42 AM, Mr Ian Primus wrote:
It would be neat, but kinda pointless. Leaving them in the eraser
for too long will definitely fry them, but it's not a matter of "five
minutes too long and it's toast".
I leave chips in the eraser for 20 minutes, and it's good for
everything I use, which is usually higher-capacity EPROMs, but
sometimes EPROM-based MCUs and UV PALs.
-Dave
--
Dave McGuire
Port Charlotte, FL
Presumably
more expensive erasers do things in-place - i.e.
periodically do a blank-check during the erase cycle?
Hmm, I've never even heard of such a device. Even the professional
EPROM erasers are little more than what I've built, a box with a UV
light and a timer. I've never heard of an eraser in which the chip
was actually connected to anything while being erased.
If a machine like this exists, however, it would be quite cool! But
probably very expensive, and a pain in the neck to use, as you'd
have to tell it what kind of chip you were erasing, and it would
need ZIF sockets and all kinds of electronics to read the chips. 
11:34 a.m.
--- On Thu, 1/29/09, Jules Richardson <jules.richardson99 at gmail.com> wrote:
I've never seen a commerical one (but then I've never seriously looked),
ut I am pretty sure there was a design for ome in one of the magazines
over here <mumble> years ago. Probably Elektor. Given the time, it
probaly only worked with 2764/27128/27256 devices which have essentially
the same pinout, so telling it what sort of device it was erasing is not
a problem.
Presumably more expensive erasers do things
in-place - i.e.
periodically do a blank-check during the erase cycle?
Hmm, I've never even heard of such a device. Even the professional
EPROM erasers are little more than what I've built, a box with a UV
light and a timer. I've never heard of an eraser in which the chip was
actually connected to anything while being erased.
If a machine like this exists, however, it would be quite cool! But
probably very expensive, and a pain in the neck to use, as you'd have to
tell it what kind of chip you were erasing, and it would need ZIF
sockets and all kinds of electronics to read the chips. From what I rememebr, the cirucit just read through all
the addresses of
the EPROM until all of them read as FF. And then geve <n>
times more
erasure time. In other words what you expect. I think it was built from
TTL (no microcotnroller, no PLDs of any type), so it should still be
possible to make one.
-tony
1:49 p.m.
Tony Duell wrote:
From the article:
"The contents of the EPROM are continuously read, until all cells read FFh.
[...] Once all cells read FFh, the lamp is left on for about one minute. This
is done to make sure all cells are properly cleared, including those which are
at an undefined state between erased and still programmed."
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
I've never seen a commerical one (but then
I've never seriously looked),
ut I am pretty sure there was a design for ome in one of the magazines
over here <mumble> years ago. Probably Elektor. Given the time, it
probaly only worked with 2764/27128/27256 devices which have essentially
the same pinout, so telling it what sort of device it was erasing is not
a problem.
Rietjens, A.: Intelligent EPROM Eraser. Elektor Electronics, June 1994, p.44-46.
It handles the more common 28-pin EPROMs -- 2764, *128, *256, *512. Can't say
I'd actually bother building and using one, though. 15 minutes is usually more
than enough for most EPROMs and MCUs.
ISTR someone asked why higher-capacity chips take longer to erase. Newer chips
use smaller transistors -- the smaller the transistors, the smaller the
"floating gate" is, and the less SiO2 there is to be hit by the UV. Less area
= less energy absorbed = less charge leakage = longer erase time.
From what I
rememebr, the cirucit just read through all the addresses of
the EPROM until all
of them read as FF. And then geve <n> times more
erasure time. In other words what you expect. I think it was built from
TTL (no microcotnroller, no PLDs of any type), so it should still be
possible to make one. 
2:34 p.m.
ISTR someone asked why higher-capacity chips take
longer to erase.
Newer chips use smaller transistors -- the smaller the transistors,
the smaller the "floating gate" is, and the less SiO2 there is to be
hit by the UV. Less area = less energy absorbed
True, but...
= less charge leakage = longer erase time.
I'm not convinced. Less area = smaller floating gate = less charge to
leak, too; it seems to me that UV flux per nm^2 and the thickness of
the oxide would be the relevant variables, not the area of the floating
gate per se.
Or do you have some hard reference for this? It's not clear whether
you're repeating something you found in an authoritative source or an
"it sounds reasonable to me" or "somebody told me" or what.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
30 Jan
30 Jan
11:10 a.m.
ISTR someone asked why higher-capacity chips take longer to erase. Newer chips
use smaller transistors -- the smaller the transistors, the smaller the
"floating gate" is, and the less SiO2 there is to be hit by the UV. Less area
= less energy absorbed = less charge leakage = longer erase time.
Yes, but I would have thought that smaller devices had less charge to
discharge (if you see what I mean). Assuming the voltage on the floating
gate is much the same in both cases (and if anything it'd be smaller on
later/higher capacity devices), the charge is going to be roughly
proportional to the area of the floating gate (since the capacitance
between that gate and whatever you reference it to is proportional to
that area).
I believe that if you have a number of charged bodies in a mediaum of
resistivity rho and permitivity epsilon (as normal), the time constant of
that system (to equalise the charges) is independnat of the shapes and
positions of the charged bodies, it is simply rho * epsilion.
th return ot the 'intellegent' EPROM ereaser, my worried is that do we
_know_ that the rest of the chip (addresss decoder, sense amplifiers,
etc) will work correcting when irradiated with short-wavelength UV? In
other words is the readout of the chip even meaningful under such conditions?
-tony
12:29 p.m.
Tony Duell wrote:
th return ot the 'intellegent' EPROM ereaser,
my worried is that do we
_know_ that the rest of the chip (addresss decoder, sense amplifiers,
etc) will work correcting when irradiated with short-wavelength UV? In
other words is the readout of the chip even meaningful under such conditions?
I'd go with "maybe, but it isn't guaranteed".
To my knowledge, no EPROM manufacturer (if there are any left, I seem to
recall STMicroelectronics were one of the last) guarantee that their chips
will work under UV light -- I suspect the expectation is that you hit them
with UV light for one reason only -- to erase them. There may be a little bit
of shielding over the more sensitive circuitry, but I wouldn't bet any money
on it.
One thing that's sometimes worth a quick demo is the propensity of EPROMs to
temporarily lose their programming when nailed with high-intensity light from,
say, a camera flash. That caught me out a few times -- I've got a couple of
old HTEC "Kitty Card" 8052 microcontroller boards (formerly sold by
Greenweld). Basically, it's a board with an 8032, an EPROM, a bit of RAM, an
RS232 line-driver and a few I/Os (notably a HD44780 LCD driver). The original
code in the EPROMs contained text referring to the old Argos "Premier Points"
scheme.
Anyway, I finished hooking the board up, and figured it'd be nice to have a
photo of the whole arrangement. The camera flash fired and the software
promptly crashed.
A nice explanation as to why semiconductors tend to come in light-proof
packages, and why you really should put reflective labels on windowed ICs when
you're done playing with the software...
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
1:09 p.m.
--- On Fri, 1/30/09, Philip Pemberton <classiccmp at philpem.me.uk> wrote:
One thing that's sometimes worth a quick demo is
the
propensity of EPROMs to temporarily lose their programming
when nailed with high-intensity light from, say, a camera
flash.
<snip>
Anyway, I finished hooking the board up, and figured
it'd be nice to have a photo of the whole arrangement.
The camera flash fired and the software promptly crashed.
That's interesting! Did the thing work after being restarted, or was the data
permanently altered? It doesn't take much to disrupt running software, after all.
I remember one article that described an incredibly creative, low cost digital camera-like
device. Instead of a CCD, the thing used an ordinary DRAM chip with the cover pried off.
The light would affect the cells of the DRAM, and then the 'image' could be read
by the computer. Of course, the thing was incredibly primitive - the pictures weren't
great, but it was an ingenous hack, and a clever utilization of the light-sensitive nature
of these things.
-Ian
1:36 p.m.
Mr Ian Primus wrote:
That's interesting! Did the thing work after being
restarted, or was the
data permanently altered? It doesn't take much to disrupt running software,
after all.
It was a temporary glitch -- I power-cycled the board and it worked again.
I remember one article that described an incredibly
creative, low cost
digital camera-like device. Instead of a CCD, the thing used an ordinary
DRAM chip with the cover pried off. The light would affect the cells of the
DRAM, and then the 'image' could be read by the computer. Of course, the
thing was incredibly primitive - the pictures weren't great, but it was an
ingenous hack, and a clever utilization of the light-sensitive nature of
these things.
"Kuckuck", by Martin Kurz --
<http://www.vampyr.msk.ru/electroniks/kuckuck.zip>
Interesting, even if you don't read German. The source code, schematics and
pretty diagrams explain most of it :)
Basically you fill the DRAM with "1" bits, then read it back out after a
predetermined time. That gives you an image. Take more than one image and you
can even get a greyscale image :)
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
3:20 p.m.
I remember one article that described an incredibly
creative, low cost digital camera-like device.
Instead of a CCD, the thing used an ordinary DRAM chip with the cover pried off. The
light would
affect the cells of the DRAM, and then the 'image' could be read by the computer.
Of course, the
thing was incredibly primitive - the pictures weren't great, but it was an ingenous
hack, and a
clever utilization of the light-sensitive nature of these things.
This topic seems to come up here every few years...
In case anyone wants it, I have a scan of a two-part 1983 Circuit Cellar
artical called "Build the Micro D-CAM" which is essentally this, except
that he uses the IS32 "Optic RAM" from Micron.
I have the datasheet for the IS32 scanned as well, which says "Improved cost
standards for binary image sensing (less than $.0003 / element) are delivered
through Micron's MT4264 64K dynamic RAM memory chip as the sensing device.
The OpticRAM(tm) image sensor is differentiated by packaging with a glass lid
(vs the opaque lid used in memory chips) and connection of the normally unused
pin 1 for optical threshold reference adjustment".
So they actually state that they are using their standard 64k DRAM as the
sensor, but it's not clear how modified it is ... ie: what are they doing with
pin1 at the die?).
I've also got a scan of a 1975 Popular Electronics artical titled "Build CYCLOPS
first all solid-state TV camera for experimenters" - it's not clear what he used
for the sensor, he just referrs to it as a "1024 element MOS sensor" and
mentions
that it's packed in "a conventional 16 pin DIP". It's only a 32x32
display, and
he uses an oscilloscope for the display device...
Dave
--
dave09 (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
31 Jan
31 Jan
5:34 p.m.
New subject: DRAM Imager ( was: EPROM erase times and lifespan )
Dave,
The Cyclops you mentioned below ( from 1975 ) used a MOSTEK 4008 with an
Optical window. There is a guy still selling them for a " mere " $35 if you
are so inclined.
You cannot just use a standard 64K DRAM chip for an imager because the cells
are not necessarily contiguous ( because of yield problems there is
redundancy and sections are added or subtracted to yield a usable device ).
They're also usually built into at least two 32K sections.
Best regards, Steven
cost
> I remember one article that described an
incredibly creative, low cost
digital camera-like device.
> Instead of a CCD, the thing used an ordinary DRAM
chip with the cover
pried off. The light would
> affect the cells of the DRAM, and then the
'image' could be read by the
computer. Of course, the
> thing was incredibly primitive - the pictures
weren't great, but it was
an ingenous hack, and a
clever
utilization of the light-sensitive nature of these things.
This topic seems to come up here every few years...
In case anyone wants it, I have a scan of a two-part 1983 Circuit Cellar
artical called "Build the Micro D-CAM" which is essentally this, except
that he uses the IS32 "Optic RAM" from Micron.
I have the datasheet for the IS32 scanned as well, which says "Improved standards for binary image sensing (less than $.0003 /
element) are
delivered
through Micron's MT4264 64K dynamic RAM memory
chip as the sensing device.
The OpticRAM(tm) image sensor is differentiated by packaging with a glass
lid
(vs the opaque lid used in memory chips) and
connection of the normally
unused
pin 1 for optical threshold reference
adjustment".
So they actually state that they are using their standard 64k DRAM as the
sensor, but it's not clear how modified it is ... ie: what are they doing
with
pin1 at the die?).
I've also got a scan of a 1975 Popular Electronics artical titled "Build
CYCLOPS
first all solid-state TV camera for
experimenters" - it's not clear what
he used
for the sensor, he just referrs to it as a "1024
element MOS sensor" and
mentions
that it's packed in "a conventional 16 pin
DIP". It's only a 32x32
display, and
he uses an oscilloscope for the display device...
Dave
2 Feb
2 Feb
7:55 a.m.
New subject: DRAM Imager ( was: EPROM erase times and lifespan )
Dave,
The Cyclops you mentioned below ( from 1975 ) used a MOSTEK 4008 with an
Optical window. There is a guy still selling them for a " mere " $35 if you
are so inclined.
You cannot just use a standard 64K DRAM chip for an imager because the cells
are not necessarily contiguous ( because of yield problems there is
redundancy and sections are added or subtracted to yield a usable device ).
As noted in the quote from the IS32 data sheet, it states that the sensor is
based on the Micron's standard DRAM - however addition of a treshold adjustment
suggests it's not exactly the same.
> I have the datasheet for the IS32 scanned as well,
which says "Improved cost
> standards for binary image sensing (less than $.0003 / element) are delivered
> through Micron's MT4264 64K dynamic RAM memory chip as the sensing device.
> The OpticRAM(tm) image sensor is differentiated by packaging with a glass lid
> (vs the opaque lid used in memory chips) and connection of the normally unused
> pin 1 for optical threshold reference adjustment".
They're also usually built into at least two 32K
sections.
The data sheet also says:
"Dual 128 x 256 element arrays"
"32768 light sensing elements per array"
"Each of the two array on the chip contains 32,768 sensors arranged in 128 rows
by 256 columns of sensors. Thpical applications will utilize only one of the
arrays since the arrays are separated by an optical "dead zone" 120 microns
wide."
Dave
--
dave09 (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
31 Jan
31 Jan
12:53 p.m.
One thing that's sometimes worth a quick demo is
the propensity of EPROMs to
temporarily lose their programming when nailed with high-intensity light from,
say, a camera flash. That caught me out a few times -- I've got a couple of
That also caught IBM when introducing the 701,
though it wasn't EPROMs. If I remember the story
correctly, they used one of the forms of CRT memory
systems in the machine. Originally, it seemed like
a good idea to make the tubes visible both because
it looked cool and because you could see what the
machine was doing. But at the press introduction,
they fired off flashbulbs and the light interacting
with the memory devies crashed the machine.
Watson dictated that the plexiglass be replaced
by a steel plate.
(Caution, this is the story as my fuzzy memory
recalls it. I may have gotten some details wrong
or have lost some due to wetware bitrot. YMMV)
BLS
10:38 p.m.
On Jan 31, 2009, at 3:53 PM, blstuart at bellsouth.net wrote:
I build a lot of stuff with 8751 chips. They will usually crash
when a camera flash is aimed at them.
-Dave
--
Dave McGuire
Port Charlotte, FL
One thing
that's sometimes worth a quick demo is the propensity of
EPROMs to
temporarily lose their programming when nailed with high-intensity
light from,
say, a camera flash. That caught me out a few times -- I've got a
couple of
That also caught IBM when introducing the 701,
though it wasn't EPROMs. If I remember the story
correctly, they used one of the forms of CRT memory
systems in the machine. Originally, it seemed like
a good idea to make the tubes visible both because
it looked cool and because you could see what the
machine was doing. But at the press introduction,
they fired off flashbulbs and the light interacting
with the memory devies crashed the machine.
Watson dictated that the plexiglass be replaced
by a steel plate.
11:14 p.m.
New subject: KAYPRO 10 - Need Help Please
I rescued two ( more like 1 3/4 ) KAYPRO 10 machines headed for euthanasia.
I repaired the LVPS in the first unit and it comes up (?) but I think it is
waiting for something ( like a BOOT disk ? ). They both ( machines ) have
one half-height 5 1/4 floppy drive and one half-height 10 MByte hard drive.
So far both of the CPU boards do exactly the same thing, after RESET it puts
up the following prompt on the CRT;
* KAYPRO 10 v1.9e *
Anyone know what if anything it is waiting for ( I can't believe it wouldn't
BOOT off of the hard drive, but I don't know any better ) ?? The 10 MB Ready
light for the hard disk does come on ( for what it is worth...).
Better yet, anyone want a couple ( or one ) KAYPRO 10s to screw around with
and add to your collection ? They are built pretty tough ! Make an offer
that will cover my Super Bowl beer fund and you're in business ! Because of
their weight, probably limited to domestic USA shipping ( from Southern
CA ).
Best regards, Steven
2 Feb
2 Feb
10:53 a.m.
New subject: KAYPRO 10 - Need Help Please
On Sunday 01 February 2009 02:14:15 am Scanning wrote:
I rescued two ( more like 1 3/4 ) KAYPRO 10 machines
headed for euthanasia.
I repaired the LVPS in the first unit and it comes up (?)
In other words you get some stuff on the screen? That was a real common
problem with those, and it was typically the soldering at the single
connector on those boards.
FWIW, I have schematics of at least three variants on the supplies used on
those machines, if somebody needs one.
but I think it is waiting for something ( like a BOOT
disk ? ).
Probably, yeah.
They both ( machines ) have one half-height 5 1/4
floppy drive and one
half-height 10 MByte hard drive.
None of the ones I have were equipped with a HD.
So far both of the CPU boards do exactly the same
thing, after RESET it
puts up the following prompt on the CRT;
* KAYPRO 10 v1.9e *
Anyone know what if anything it is waiting for ( I can't believe it
wouldn't BOOT off of the hard drive, but I don't know any better ) ?? The
10 MB Ready light for the hard disk does come on ( for what it is worth...).
Does the floppy drive light come on too? They were a little weird in that
respect, on my Osborne for example, the light on the floppy drive would
only come on when there was actually some activity going on there. On Kaypro
machines, the light would stay on for as long as the drive was _selected_.
This was almost disastrous on more than one occasion when Kaypro users were
introduced to Osborne hardware and had a tendency not to wait until disk
operations were finished. :-)
Better yet, anyone want a couple ( or one ) KAYPRO 10s
to screw around with
and add to your collection ? They are built pretty tough ! Make an offer
that will cover my Super Bowl beer fund and you're in business ! Because of
their weight, probably limited to domestic USA shipping ( from Southern
CA ).
I wouldn't turn 'em down, anyway, though I'm not sure whether or not
it'd be
worth it to me, perhaps we better take that aspect of it up offlist...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
3 Feb
3 Feb
4:18 p.m.
New subject: KAYPRO 10 - Need Help Please
List ( et al );
The KAYPRO twins have been adopted and have a new home. Thanks to all for
the inquires, comments and feedback.
Best regards, Steven
1 Feb
1 Feb
8:19 a.m.
New subject: Optically switched control panel?
A member of another list thinks he remembers a large computer with a
control/maintenance panel that used optical switches to set and clear
register bits. The operator would place a finger in a hole/depression,
interrupting a light beam to a photosensor, toggling a flip-flop.
Did such a system exist, or is this just wishful thinking?
Mike Loewen mloewen at cpumagic.scol.pa.us
Old Technology http://sturgeon.css.psu.edu/~mloewen/Oldtech/
19 Feb
19 Feb
3:57 p.m.
New subject: Optically switched control panel?
I'm aware of something close...
The HP2114 computers use a capacitive proximity switch.
There is a small variable capacitor behind the front panel to adjust the
sensitivity of the switch.
If its over-adjusted, a wave of a finger will toggle the switch register
bits, which then light up
from behind.
----- Original Message -----
From: "Mike Loewen" <mloewen at cpumagic.scol.pa.us>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Sunday, February 01, 2009 11:19 AM
Subject: Optically switched control panel?
A member of another list thinks he remembers a large computer with a
control/maintenance panel that used optical switches to set and clear
register bits. The operator would place a finger in a hole/depression,
interrupting a light beam to a photosensor, toggling a flip-flop.
Did such a system exist, or is this just wishful thinking?
Mike Loewen mloewen at cpumagic.scol.pa.us
Old Technology http://sturgeon.css.psu.edu/~mloewen/Oldtech/
20 Feb
20 Feb
1:01 a.m.
New subject: Optically switched control panel?
Part of the US NAVY UYQ-21 AEGIS program used an input / output computer
device called a Computer Controlled Action Entry Panel ( CCAEP ) that had
"switches" that were really just holes that you poke your finger into to
interrupt a light beam ( IR LED ) that was shining on a phototransistor.
Couldn't use capacitive switches because the guys had to wear gloves in the
event of a chemical threat.
Best regards, Steven
> A member of another list thinks he remembers a
large computer with a
> control/maintenance panel that used optical switches to set and clear
> register bits. The operator would place a finger in a hole/depression,
> interrupting a light beam to a photosensor, toggling a flip-flop.
>
> Did such a system exist, or is this just wishful thinking?
>
>
> Mike Loewen mloewen at cpumagic.scol.pa.us
30 Jan
30 Jan
1:36 p.m.
Tony Duell wrote:
th return ot the 'intellegent' EPROM ereaser,
my worried is that do we
_know_ that the rest of the chip (addresss decoder, sense amplifiers,
etc) will work correcting when irradiated with short-wavelength UV? In
other words is the readout of the chip even meaningful under such conditions?
Can I nip this one in the bud and state that I was thinking of some for of
intelligent programmer which turned the lamp off before doing a blank check?
But having said that, the comment made about heat from the lamp operation is a
good one - I suspect that would mean that there's a potential for damage if
the device was read without letting it cool properly first.
cheers
Jules
29 Jan
29 Jan
10:16 a.m.
On Jan 29, 2009, at 10:27 AM, Jules Richardson wrote:
I've never seen nor heard of an EPROM eraser that does this, in
nearly 30 years of using EPROMs.
-Dave
--
Dave McGuire
Port Charlotte, FL
the chip docs
should give an erase time, iirc the docs would say to
get to the clear stage and add x% time to guarantee the chip is
set to
1/0
Presumably more expensive erasers do things in-place - i.e.
periodically do a blank-check during the erase cycle? 
10:39 a.m.
On Thu, Jan 29, 2009 at 1:16 PM, Dave McGuire <mcguire at neurotica.com> wrote:
On Jan 29, 2009, at 10:27 AM, Jules Richardson wrote:
I have never seen that either... the ones I've worked with
professionally have had a variety of features including timers,
conductive plastic rails at 0.6" spacing for ESD reasons and easy
loading, but I've never seen any sort of eraser that applies power to
the chips in any way.
I myself did some experiments a couple of years ago at the South Pole
when I didn't have a proper UV-A bulb (I just had a UV inspection
lamp)... I could get the chips to superficially erase, but when I
built a rig out of sockets and a trim pot to turn down Vcc to the
EPROM for testing, I could reliably read the former contents at
voltages below 5V even when the chips appeared empty at exactly 5V.
In the end, I gave up on trying to use the lamp I had. Fortunately I
had a couple of 1MBit FLASH chips and was able to muddle through, but
in the future, I'd probably pack some 27F256s or pack a real EPROM
eraser and use ordinary EPROMs.
-ethan
Presumably more expensive erasers do things
in-place - i.e. periodically
do a blank-check during the erase cycle?
I've never seen nor heard of an EPROM eraser that does this, in nearly 30
years of using EPROMs.
10:44 a.m.
On Jan 29, 2009, at 1:39 PM, Ethan Dicks wrote:
I myself did some experiments a couple of years ago at
the South Pole
when I didn't have a proper UV-A bulb (I just had a UV inspection
lamp)... I could get the chips to superficially erase, but when I
built a rig out of sockets and a trim pot to turn down Vcc to the
EPROM for testing, I could reliably read the former contents at
voltages below 5V even when the chips appeared empty at exactly 5V.
Yup, because the bit comparator threshold is determined by Vcc.
Lower Vcc, and you lower the comparator's threshold.
In the end, I gave up on trying to use the lamp I had.
Fortunately I
had a couple of 1MBit FLASH chips and was able to muddle through, but
in the future, I'd probably pack some 27F256s or pack a real EPROM
eraser and use ordinary EPROMs.
Speaking of which, I've recently become enamored with Hitachi
58C256. I bought a tube of them cheap a couple of years ago and
haven't really messed with them until recently. They're very nice
chips; I hope to find some more of them at some point.
-Dave
--
Dave McGuire
Port Charlotte, FL
11:21 a.m.
On Thu, Jan 29, 2009 at 1:44 PM, Dave McGuire <mcguire at neurotica.com> wrote:
On Jan 29, 2009, at 1:39 PM, Ethan Dicks wrote:
Uh, yeah... I meant 28F256s.
-ethan
in the future, I'd probably pack some 27F256s
or pack a real EPROM
eraser and use ordinary EPROMs.
Speaking of which, I've recently become enamored with Hitachi 58C256. I
bought a tube of them cheap a couple of years ago and haven't really messed
with them until recently. They're very nice chips; I hope to find some more
of them at some point.
11:46 a.m.
On Jan 29, 2009, at 2:21 PM, Ethan Dicks wrote:
Uh, yeah... I meant 28F256s.
Yeah I figured as much. :)
-Dave
--
Dave McGuire
Port Charlotte, FL
in the future, I'd probably pack some 27F256s or
pack a real EPROM
eraser and use ordinary EPROMs.
Speaking of which, I've recently become enamored with Hitachi
58C256. I
bought a tube of them cheap a couple of years ago and haven't
really messed
with them until recently. They're very nice chips; I hope to find
some more
of them at some point.
3:07 p.m.
Ethan ( et al );
Never ever power up an EPROM fresh out of the eraser. If the eraser is
putting out enough UV to actually do it's job, the EPROM gets really frikkin
hot, and the die is way the hell past the safe operating temperature for the
device. Also, just as we had a " metal migration " problem with fuse PROMs
that would heal when enough metal re-grew the fuse, EPROMs have a " charge
migration " problem. Insufficient UV erasure will not bleed all of the
charge out of the cell and enough charge will sneak back in to make the cell
a "zero" again ( if that was the state before the erasure ) or it will be
fuzzy ( not quite one or zero, depending on Vcc ) as some of the guys here
have observed.
We had hundreds ( literally ) of industrial strength UV erasures and still
had problems. We finally had to have our CAL Lab test and post stickers on
each device because the UV output was all over the place. Plus the position
and distance to the lamp made a big difference. The range seemed to be
between 30 minutes to an hour to assure the device really was erased.
Devices that would not totally erase in that time were sh!t-canned and
erasers that would need more than an hour would have the UV tube replaced.
All of our erasers had conductive foam on the bottom of the trays.
Best regards, Steven
> On Jan 29, 2009, at 10:27 AM, Jules Richardson
wrote:
>> Presumably more expensive erasers do things in-place - i.e.
periodically
>> do a blank-check during the erase cycle?
>
> I've never seen nor heard of an EPROM eraser that does this, in nearly
30
years of using
EPROMs.
I have never seen that either... the ones I've worked with
professionally have had a variety of features including timers,
conductive plastic rails at 0.6" spacing for ESD reasons and easy
loading, but I've never seen any sort of eraser that applies power to
the chips in any way.
I myself did some experiments a couple of years ago at the South Pole
when I didn't have a proper UV-A bulb (I just had a UV inspection
lamp)... I could get the chips to superficially erase, but when I
built a rig out of sockets and a trim pot to turn down Vcc to the
EPROM for testing, I could reliably read the former contents at
voltages below 5V even when the chips appeared empty at exactly 5V.
In the end, I gave up on trying to use the lamp I had. Fortunately I
had a couple of 1MBit FLASH chips and was able to muddle through, but
in the future, I'd probably pack some 27F256s or pack a real EPROM
eraser and use ordinary EPROMs.
-ethan
7:31 a.m.
Working with EPROMs, one must have an eraser. I built
my own a while back, using a germicidal UV
lamp tube, a flourescent fixture, a sealed box and a programmable timer. It works well,
and I
typically can erase an batch of eproms in like three minutes.
...
Three minites seems awfully fast for complete erasure. For example, an Intel 1982
databook
lists erase time for the 2764 as 15-20 mins at a distance of 1 inch from the lamp (2537
Angstroms, 12000 uW/CM2).
This is pretty consistant with what I've observed - I normally bake em for about 15
mins.
I've seen some take as long as 30 mins (usually older "fixed program time"
devices).
Keep in mind that although the output is digital, the actually process of an EPROM is
analog in nature - the output is decided based on the amount of charge on the gate.
This brings a few important points to understand:
1) Some EPROMS are "more programmed" than others.
Early devices with a fixed proramming time per cell normally are saturated, and well
programmed. Later "intelligent" algorithms would watch the device, note how much
charge
was needed to make a 1->0 transition occur in the cell, add a margin and "your
done".
Much faster - but depending on the equipment, sometimes "less programmed".
2) The comparitor is voltage dependant on most devices.
An EPROM bit near the 0/1 threshold may change state with a difference in power supply
voltage - Intel spec'd a supply of 6V when using their "intelligent
algorithm" because
that would raise the comparitor voltage and cause it to require more charge to get a
zero bit. This insured that you were programmed "well past" the normal boundary
at 5V.
You can sometimes read an EPROM that is suffering from "bit rot" by powering it
at a
lower voltage.
A side effect of this is that:
3) Some EPROMs are "less erased" than others.
An "erased" EPROM read at 5.2v might not be erased when read at 4.8v ... so just
because
it all reads FF's doesn't mean it's really erased. The fact that you see some
"mostly
erased" suggests you are a little close to the boundary. Just like programming, you
need
to add a margin to insure adaquate erasure.
My guess is that "lightly" programmed devices are showing just under the 1/0
threshold
after 3 mins, but probably aren't fully erased. You might also want to try and find
out the specs for your lamp and see if it's close to a normal eraser, and adjust your
distance/time accordingly.
Dave
--
dave09 (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
7:55 a.m.
--- On Thu, 1/29/09, Dave Dunfield <dave09 at dunfield.com> wrote:
Three minites seems awfully fast for complete erasure. For
example, an Intel 1982 databook
lists erase time for the 2764 as 15-20 mins at a distance
of 1 inch from the lamp (2537
Angstroms, 12000 uW/CM2).
Hmm. Well, either I got lucky with the previous EPROMs, or my lamp is a little more
powerful. I don't know the specs on it, I bought it a couple years ago at a pet supply
store. It's a 12" germicidal UV tube, the thin kind, what's that, T6?.
It's intended for use in pond filters.
The EPROMs sit on a specially designed non-conductive UV absorbant insulating surface
(it's a piece of wood), right at about an inch from the lamp.
This is pretty consistant with what I've observed
- I
normally bake em for about 15 mins.
I've seen some take as long as 30 mins (usually older
"fixed program time" devices).
Hmm, that might explain why I have some that just won't blank. I probably need to put
them in longer.
Keep in mind that although the output is digital, the
actually process of an EPROM is
analog in nature - the output is decided based on the
amount of charge on the gate.
This brings a few important points to understand:
1) Some EPROMS are "more programmed" than others.
<snip>
2) The comparitor is voltage dependant on most
devices.
An EPROM bit near the 0/1 threshold may change state with a
difference in power supply
voltage -
<snip>
3) Some EPROMs are "less erased" than
others.
Wow. Good to know. I had no idea. I always tried to give them the least amount of UV that
they needed to come up blank on my programmer. It's a PC/clone based ISA programmer,
so I'd assume that the 5v it's giving the chip is just the 5v from the PC's
supply, so I would hope that it's close.
My guess is that "lightly" programmed
devices are
showing just under the 1/0 threshold
after 3 mins, but probably aren't fully erased.
Very good to know - so far, I haven't had any problems, but I should probably go back
and do some more erasing. Hmm, perhaps a programmer adapter with the ability to lower the
supply voltage is in order. I've got some experimenting to do.
But, yeah, I guess I'll try setting the timer for longer now, to ensure they're
good and blank. I was always told that too much UV would damage the chips - I've never
left one in for more than say, 15 minutes before. Next time I need to erase some, I'll
put 'em in for longer than three or five minutes...
-Ian
8:11 a.m.
But, yeah, I guess I'll try setting the timer for
longer now, to ensure they're good and blank. I
was always told that too much UV would damage the chips - I've never left one in for
more than say,
15 minutes before. Next time I need to erase some, I'll put 'em in for longer
than three or five
minutes...
Which is why a timer is a good thing ... I can tell you from personal experience
that forgetting a tray of EPROMs in an untimed eraser overnight leads to non-
functional devices. The ones I left in for an hour seemed to be OK, but I wouldn't
do that on a regular basis - after about the third time I left the tray in
overnight, I ripped the timer out of a dryer we were disposing of and made it an
integral part of my eraser.
Dave
PS: Since you don't know the charactistics of your lamp, I'd err on the cautious
side ...
--
dave09 (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
8:28 a.m.
--- On Thu, 1/29/09, Dave Dunfield <dave09 at dunfield.com> wrote:
From: Dave Dunfield <dave09 at dunfield.com>
Subject: Re: EPROM erase times and lifespan
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
Date: Thursday, January 29, 2009, 11:11 AM
Hehe, yeah, I really didn't want to risk ruining any chips, so when I built the
eraser, the timer was top priority. I used the timer out of a junked microwave oven,
replacing the horrible membrane keypad with the numeric keypad sawed off the end of a
damaged Apple IIe platinum keyboard. So mine even has a nice little VFD telling me just
how much time is remaining, a nice little *feep* *feep* *feeeeep* to tell me that it's
done, and best of all, the little VFD flashes "COOK" at the bottom while
it's running :)
Oh, and it doubles as a clock...
But, yeah, I guess I'll try setting the timer
for
longer now, to ensure they're good and blank. I
was always told that too much UV would damage the
chips - I've never left one in for more than say,
15 minutes before. Next time I need to erase
some,
I'll put 'em in for longer than three or five
minutes...
Which is why a timer is a good thing ... I can tell you
from personal experience
that forgetting a tray of EPROMs in an untimed eraser
overnight leads to non-
functional devices. The ones I left in for an hour seemed
to be OK, but I wouldn't
do that on a regular basis - after about the third time I
left the tray in
overnight, I ripped the timer out of a dryer we were
disposing of and made it an
integral part of my eraser. PS: Since you don't know the charactistics of
your
lamp, I'd err on the cautious
side ...
Yeah, hence why I initially experimented with the three and five minutes. That gets most
of 'em 'blank', but I think I'm going to try aroud eight or ten minutes
next time, to get them 'blanker'. Some chips required two doses of three minutes,
others still weren't blank after that. I haven't tried those again yet, but
perhaps I need to do more experimenting.
-Ian
10:36 a.m.
On Jan 29, 2009, at 11:28 AM, Mr Ian Primus wrote:
Hehe, yeah, I really didn't want to risk ruining
any chips, so when
I built the eraser, the timer was top priority. I used the timer
out of a junked microwave oven, replacing the horrible membrane
keypad with the numeric keypad sawed off the end of a damaged Apple
IIe platinum keyboard. So mine even has a nice little VFD telling
me just how much time is remaining, a nice little *feep* *feep*
*feeeeep* to tell me that it's done, and best of all, the little
VFD flashes "COOK" at the bottom while it's running :)
Oh, and it doubles as a clock...
Ok, I take back what I said about building that. That's just
cool. :)
-Dave
--
Dave McGuire
Port Charlotte, FL
10:58 a.m.
On Thu, 29 Jan 2009, Dave McGuire wrote:
On Jan 29, 2009, at 11:28 AM, Mr Ian Primus wrote:
I'll not only second that, but beg for pictures of it too! :)
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
Hehe, yeah, I really didn't want to risk
ruining any chips, so when I built
the eraser, the timer was top priority. I used the timer out of a junked
microwave oven, replacing the horrible membrane keypad with the numeric
keypad sawed off the end of a damaged Apple IIe platinum keyboard. So mine
even has a nice little VFD telling me just how much time is remaining, a
nice little *feep* *feep* *feeeeep* to tell me that it's done, and best of
all, the little VFD flashes "COOK" at the bottom while it's running :)
Oh, and it doubles as a clock...
Ok, I take back what I said about building that. That's just cool. :)
10:51 a.m.
On 29 Jan 2009 at 11:11, Dave Dunfield wrote:
... I ripped the timer out of a dryer we were
disposing of and made it
an integral part of my eraser.
I just use one of those clockwork lightswitch-timers that are used to
turn off the lights when you're just getting to a critical piece of
code. :)
Good idea with salvaged appliance timers, though.
Your mention of the clothes dryer jogged an old neuron. My first
EPROM eraser was made from a coffee can and an "ozone lamp" from a
clothes dryer. I haven't seen one of those is decades, however.
Cheers,
Chuck
1:18 p.m.
On 29/01/2009 16:11, Dave Dunfield wrote:
Yup, my home-made eraser uses a timer and my bought (by someone else, I
merely inherited it) also has a timer. And despite having different
tubes, they both take about the same time to erase typical EPROMs -
around 20 minutes is what I usually give.
I remember in the dim and distant past being told that you can sometimes
recover over-erased EPROMs by baking them in an oven. Maybe that was
just 1702s or some similarly ancient device, I've never needed to find out.
And to answer Ian's other question, yes the UV exposure is cumulative,
and giving 5 minutes followed by three minutes is exactly the same as
giving one exposure of 8 minutes. What you're doing is exciting the
atoms in a floating insulated gate so the electrons leak away, and the
longer the exposure, the more leak away.
--
Pete Peter Turnbull
Network Manager
University of York
But, yeah, I
guess I'll try setting the timer for longer now, to ensure they're good and blank.
I
was always told that too much UV would damage the chips - I've never left one in for
more than say,
15 minutes before. Next time I need to erase some, I'll put 'em in for longer
than three or five
minutes...
Which is why a timer is a good thing ... I can tell you from personal experience
that forgetting a tray of EPROMs in an untimed eraser overnight leads to non-
functional devices. The ones I left in for an hour seemed to be OK, but I wouldn't
do that on a regular basis - after about the third time I left the tray in
overnight, I ripped the timer out of a dryer we were disposing of and made it an
integral part of my eraser.
10:35 a.m.
On Jan 29, 2009, at 10:55 AM, Mr Ian Primus wrote:
I have to say I'm curious as to why you'd do this. Nicely-
packaged commercial EPROM erasers are available all over the place,
and they're cheap. I just bought a really, really nice one (I mean
REALLY nice) for ten bucks to replace my beat-up one, and then I sold
my old beat-up one for...ten bucks. I have two more in the closet
that I bought for similar amounts, at least ten years ago.
Keep in mind that all electronic devices are analog...and some of
them are specified in digital terms. EPROMs store a bit as a charge
on a tiny piece of conductive material floating in an insulator
("floating gate"), and when that charge is above a certain threshold,
it's a 0, and if it's below the threshold, it's a 1...so there's a
varying amount of charge on that floating gate.
Erasing an EPROM drains that charge away by turning that
surrounding insulator into a partial conductor. So, you see what
happens when you erase them to just below the bit transition threshold.
Three minites
seems awfully fast for complete erasure. For
example, an Intel 1982 databook
lists erase time for the 2764 as 15-20 mins at a distance
of 1 inch from the lamp (2537
Angstroms, 12000 uW/CM2).
Hmm. Well, either I got lucky with the previous EPROMs, or my lamp
is a little more powerful. I don't know the specs on it, I bought
it a couple years ago at a pet supply store. It's a 12" germicidal
UV tube, the thin kind, what's that, T6?. It's intended for use in
pond filters.
The EPROMs sit on a specially designed non-conductive UV absorbant
insulating surface (it's a piece of wood), right at about an inch
from the lamp. 3) Some EPROMs
are "less erased" than others.
Wow. Good to know. I had no idea. I always tried to give them the
least amount of UV that they needed to come up blank on my
programmer. It's a PC/clone based ISA programmer, so I'd assume
that the 5v it's giving the chip is just the 5v from the PC's
supply, so I would hope that it's close. But, yeah, I guess I'll try setting the timer for
longer now, to
ensure they're good and blank. I was always told that too much UV
would damage the chips - I've never left one in for more than say,
15 minutes before. Next time I need to erase some, I'll put 'em in
for longer than three or five minutes...
45 minutes sticks in my head as the rule of thumb from the
2708/2716 days, but we have no idea of how much UV light your lamp
emits.
-Dave
--
Dave McGuire
Port Charlotte, FL
8:12 a.m.
On Thu, 2009-01-29 at 10:31 -0500, Dave Dunfield wrote:
Yup. I've found that some devices can take as much as half an hour or
more under the lamp in my eraser. Of course, the tube could be going
soft...
Gordon
Working with
EPROMs, one must have an eraser. I built my own a while back, using a germicidal UV
lamp tube, a flourescent fixture, a sealed box and a programmable timer. It works well,
and I
typically can erase an batch of eproms in like three minutes.
...
Three minites seems awfully fast for complete erasure. For example, an Intel 1982
databook
lists erase time for the 2764 as 15-20 mins at a distance of 1 inch from the lamp (2537
Angstroms, 12000 uW/CM2).
This is pretty consistant with what I've observed - I normally bake em for about 15
mins.
I've seen some take as long as 30 mins (usually older "fixed program time"
devices).
11:52 a.m.
Keep in mind that although the output is digital, the
actually process of
an EPROM is analog in nature - the output is decided based on the amount of
charge on the gate.
This brings a few important points to understand:
1) Some EPROMS are "more programmed" than others.
Early devices with a fixed proramming time per cell normally are saturated,
and well programmed. Later "intelligent" algorithms would watch the device,
note how much charge was needed to make a 1->0 transition occur in the
cell, add a margin and "your done". Much faster - but depending on the
equipment, sometimes "less programmed".
I wonder if this has any effect on the longevity of what's stored on them?
2) The comparitor is voltage dependant on most
devices.
An EPROM bit near the 0/1 threshold may change state with a difference in
power supply voltage - Intel spec'd a supply of 6V when using their
"intelligent algorithm" because that would raise the comparitor voltage and
cause it to require more charge to get a zero bit. This insured that you
were programmed "well past" the normal boundary at 5V.
You can sometimes read an EPROM that is suffering from "bit rot" by
powering it at a lower voltage.
Ah. Interesting to know, I'll have to make it a point to remember that.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
8:54 p.m.
On Jan 29, 2009, at 2:52 PM, Roy J. Tellason wrote:
It seems logical that it would.
-Dave
--
Dave McGuire
Port Charlotte, FL
1) Some EPROMS
are "more programmed" than others.
Early devices with a fixed proramming time per cell normally are
saturated,
and well programmed. Later "intelligent" algorithms would watch
the device,
note how much charge was needed to make a 1->0 transition occur in
the
cell, add a margin and "your done". Much faster - but depending on
the
equipment, sometimes "less programmed".
I wonder if this has any effect on the longevity of what's stored
on them?
11:47 a.m.
On Thursday 29 January 2009 09:22:05 am Mr Ian Primus wrote:
Working with EPROMs, one must have an eraser. I built
my own a while back,
using a germicidal UV lamp tube, a flourescent fixture, a sealed box and a
programmable timer. It works well, and I typically can erase an batch of
eproms in like three minutes.
At least, when I was working primarily with 2732's and 2764's, I could.
Now, I've been needing to use higher capacity chips - and these seem to
take longer to erase. Five minutes isn't enough for some. AMD 27C010's seem
to take closer to ten minutes to erase.
My eraser has a drawer that can accept about a foot of eproms (I used a 12"
tube). I could probably cram more on there, but I try to keep them centered
under the light. So, if I stick a dozen chips on there, of varying sized
and types, and 'nuke' them for five minutes, I'll go and blank-check them,
and some will be blank, others will be nearly blank (lots of FF's, but some
garbage left), and some don't seem affected at all. I've been trying to
come up with a correlation between manufacturer and erase time, but it
seems to vary. Last time, I nuked four AMD 27C010's that I pulled off of an
old LaserJet font cartridge. Three were unaffected and the other was
blanked.
I worry about leaving the chips in there too long, since I know too much UV
can damage them. But at the same time, I wonder about the lifespan of the
UV tube - and how many hours it's good for. (I guess I should have checked
the package when I bought it...).
So, how much UV exposure does it take to damage an EPROM? Is UV exposure
cumulative? For example, if I have a chip that's unaffected by five
minutes, and I put it back in for another three - is that the same as as
running it for 8 minutes total, or do the charges that the UV is supposed
to dissapate simply linger? Are CMOS chips harder to erase, taking a longer
time than the older parts?
-Ian
Interesting questions, too bad I don't have more in the way of answer to
offer... :-)
The eraser I have uses a small bulb, probably intended for use as
a "germicidal" bulb in some appliance application, only it's not in a
socket, it's soldered in. This came from Jameco or JDR or one of those
places. I've been typically leaving chips in there (there are slots for 2 of
them on opposite sides) for 15 minutes, and perhaps longer if erasure seemed
incomplete as you describe.
The only ones I've had trouble with when I try to verify them as blank are
ones that show me one or more of the pins stuck at zero for _all_ locations,
so I'm thinking they have a problem and that the use of the erasion probably
doesn't have too much to do with this.
So far I've got stuff from 1702 on up to 27C512 or maybe a step or two
further, and have not had any issues besides what I describe above.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
5551
days inactive
5573
days old
44 comments
19 participants
participants (19)
-
ard＠p850ug1.demon.co.uk -
blstuart＠bellsouth.net -
bshannon＠tiac.net -
cclist＠sydex.com -
classiccmp＠philpem.me.uk -
coredump＠gifford.co.uk -
dave.thearchivist＠gmail.com -
dave09＠dunfield.com -
ethan.dicks＠gmail.com -
geneb＠deltasoft.com -
gordonjcp＠gjcp.net -
ian_primus＠yahoo.com -
jules.richardson99＠gmail.com -
mcguire＠neurotica.com -
mloewen＠cpumagic.scol.pa.us -
mouse＠Rodents-Montreal.ORG -
pete＠dunnington.plus.com -
rtellason＠verizon.net -
steven.alan.canning＠verizon.net