UVPROM

[veedub565]

Hope this is the right part of the forum for this. I have a HP 8341B Signal Generator with four UVPROM's containing the firmware on the processor board. The firmware is an old version and I want to update it. I want to retain the old UVPROM's so I still have a good copy of the old firmware.

The UVPROMS's are AMD AM27256-1DC and long since obsolete now.

Couple of questions,

1. Is there a suitable modern alternative ?
2. Is it worth buying used UVPROM's  ?

[Andy Watson]

I've purchased a few batches of EPROMS from ebay without any problems. All of them used. Note, you'll probably want a UV eraser to make sure that theyr're blank.

Alternatively, take a copy of the erpoms that you have and then re-program them.

[veedub565]

Thanks Andy, got a Stag prom eraser available to me to wipe them with.

[Peabody]

They appear to still be available.  Jameco has a few hundred in stock, which I assume are new.

But I wonder if an AT28C256 EEPROM would work.  But the old UV version may be cheaper and easier to find.

[srb1954]

Hope this is the right part of the forum for this. I have a HP 8341B Signal Generator with four UVPROM's containing the firmware on the processor board. The firmware is an old version and I want to update it. I want to retain the old UVPROM's so I still have a good copy of the old firmware.

The UVPROMS's are AMD AM27256-1DC and long since obsolete now.

Couple of questions,

1. Is there a suitable modern alternative ?
2. Is it worth buying used UVPROM's  ?

If you are doing a simple S/W update and don't need to repeatedly erase and reprogram parts you should also consider OTP (one-time-programmable) parts e.g. Microchip AT27C256R-70PU. These are still available new from reputable distributors and are somewhat cheaper as they are packaged in a standard plastic DIP package rather than the expensive windowed ceramic package.

[westfw]

You can still buy 27c256 EPROMs (modulo current chip shortages.)
the OTP versions are pretty cheap.


a 5V parallel EEPROM or FLASH chip should work, too.  And probably 27c512s or equiv, which might be easier to find.

[veedub565]

Thanks for the suggestions, I've ordered some used ones with the same part number. I'll erase those and program with the new firmware, see how it goes.

[mon2]

https://www.jameco.com/z/AM27256-1DC-Advanced-Micro-Devices-EPROM-UV-256K-32Kx8-8-Bit-200ns-5V-DIP-28_2303055.html

[veedub565]

problem with Jameco is shipping to the UK, $30 shipping plus import duty.

[veedub565]

Well the UVPROMS from EBAY finally arrived after a fortnight. I've been waiting on those so I can crack on with updating my 8341B with the latest firmware. So you can imagine my annoyance at finding that the clucking tw*t of a seller has stuck the IC's into nothing more than plastic packing foam done up with sellotape. Couldn't have done any more to blow the sh1t out of them with ESD if they'd tried 

So that was a total waste of time and money. Going to explore some of the other options mentioned in this thread.

[Peabody]

I remember the UV parts as being pretty tolerant of abuse like that.  Perhaps you should at least try them to see if they will program.

[veedub565]

Yeah the older stuff like that is more tolerant then the more modern stuff. I will give them a try, It's not ideal though, and a little frustrating. Such is the vagaries of EBAY I suppose.

[m k]

It's very irritating when people are dealing with these things seemingly without any idea of ESD.
Legendary is a situation where seller has opened the package and placed components on what ever surface, beside the package.

One other thing, I don't understand this ROM access time uncertainty, it's just a memory.
Maybe it's an understanding of not available, it's not complete, something is always available.

If you upgrade a computer memory you're checking how fast replacement is still accessed faster.

Last time I bought 256k 28-series EEPROMs.
They are also drop in replacements for 64k and 128k of 27-series.
Of course for non authentic cases only.

[veedub565]

Well I got around to erasing these chips, 1hr with a Stag prom eraser. My friend was going to program them for me, but it turns out his Dataman 48UXP doesn't list that device. I think he said he'd have to spend £150 to get a license or something, not worth it.

Any suggestions ?

Dataman 48Pro2 does list that device, but really expensive to buy.

I'm looking to see if there are any businesses nearby that would program them for me.

Or I could use one of the other IC's previously mentioned, and maybe a cheaper prommer would work.

[PCB.Wiz]

Well I got around to erasing these chips, 1hr with a Stag prom eraser. My friend was going to program them for me, but it turns out his Dataman 48UXP doesn't list that device. I think he said he'd have to spend £150 to get a license or something, not worth it.

My EEtools programmer says this about those chips
Manufacturer:AMD, Device name:Am27256
Package Info.:DIP28, Organization:8000h x 8 bit, Adapter name:None, Manufacturer ID.:0001, Device ID.:0004, Algorithm:Quick, Vcc pin(s):28, Gnd pin(s):14, Vpp pin:1, Vccp:6.25V, Vpp:12.75V, Tpwp:100us


So you may be able to select a same-algorithm device ?

[jonsikul]

Did you ever get around to finish the task? If yes, how did it work out?

[veedub565]

Yeah I cheated and paid a professional to program them for me. Worked out fine

[josip]

Just note that I erased mine (original owner) by sun lite this spring in few weeks. Was not in hurry.

[coppice]

Just note that I erased mine (original owner) by sun lite this spring in few weeks. Was not in hurry.

Interesting. Many years ago we put a couple of UV EPROMs out in the sun to see how long they would take to fail a CRC check. After a few months we gave up. This was in London, where rust is a bigger issue than sunshine, but this was our experience.

[m k]

I bought an UV-C LED, can't remember the wavelength but close to optimal it was.
Erasing took some minutes.

Mini-Pro was checking on the fly how the thing advanced and same stubborn bits were the final resistance every time.

[Postal2]

I don't know about uvc-led, must be hi-power led for erase (with passive cooler). I use uv-lamp. Need to bind with wire uv-chip directly to lamp tube. After 2 hours no change. Need 8..10 hours to erase. My lamp is exactly as photo, but wavelength is 120 nm and much ozone produced. Power of my lamp is 20W.

[coppice]

I don't know about uvc-led, must be hi-power led for erase (with passive cooler). I use uv-lamp. Need to bind with wire uv-chip directly to lamp tube. After 2 hours no change. Need 8..10 hours to erase. My lamp is exactly as photo, but wavelength is 120 nm and much ozone produced. Power of my lamp is 20W.

A LED seems ideal for this job, when you only need to erase a single chip. The target area is small, like a LED. The traditional UV tubes used in EPROM erasers wasted 99% of the light.

[Postal2]

A LED seems ideal for this job, ....

I agree, if you have such - bring your experience with exact data about led and other.

[retiredfeline]

You can buy 3W 10V UV disinfection E17 base bulbs from eBay for about $5 that put out the right wavelength.

[Postal2]

You can buy 3W 10V UV disinfection E17 base bulbs from eBay for about $5 that put out the right wavelength.

Signed as no ozone with 250 nm wavelength. Bind with wire to chip and wait. Success will be, but I don't know how long wait.

I erased M27C512 from ST 4 pcs with my lamp at 8..10 hours.

The recommended erasure procedure for the M27C512 is exposure to short wave ultraviolet light which has wavelength 2537 Å. The integrated dose (i.e. UV intensity x exposure time) for erasure should be a minimum of 15 W-sec/cm The erasure time with this dosage is approximately 15 to 20 minutes using an ultraviolet lamp with 12000 μW/cm power rating. The M27C512 should be placed within 2.5 cm (1 inch) of the lamp tubes during the erasure. Some lamps have a filter on their tubes which should be removed before erasure.

Now, of course, I bought a lot of W27C512 and W27E512 for use instead.