PROMs and EPROMs, specifically 2332/2532/2732 - test-drb@ccmp.vtda.org

jrr＠flippers.com

24 Oct 24 Oct

7:18 p.m.

On 10/24/2015 5:43 AM, Adrian Graham wrote:

Hi folks, PET4032 repair continues with all ROMs, video RAM and dodgy sockets removed thanks to a hot air gun. Holes cleaned and I have new turned pin sockets for everything I've removed which I'll be fitting this afternoon. Since the ROMs came out OK I'm trying to dump them using my Pinmaster48 programmer, being from the 90s it doesn't read 2332/2532 PROMs but it WILL read a lot of variants of 2732 so I've made an adapter as found thanks to google and USENET: 2332 pin 18 to 2732 pin 21 (A11) 2332 pin 20 to 2732 pin 18 (Chip enable/Power Down) 2332 pin 21 to 2732 pins 20 and 24 via diodes with banding at the 2332 end (2332 Vpp) Wiring checks out and the diodes are aligned correctly so pins 20 and 24 don't interfere with each other, however the pinmaster continually gives me "continuity error on pin 20" Have I goofed somewhere? Cheers,

You just tie pin 21 of the 2332 to Vcc (+5) - Pin 24 of the 2732 - to read. There are only two modifications needed normally for reading a 2532/2332 in a 2732 socket. 2332 Pin 21 - Vcc (2732 Pin 24) Pin 18 - A11 (2732 Pin 21) If your reader coughs up a /OE error then use diodes: 2332 Pin 20 - 2 Diodes, one to 2732 pin 20, the other to 2732 pin 18. Banded end to these pins, and you will also need a pullup resistor on the 2532/2332 socket pin 20 so /CE (2332/2532) goes high when not selected. 2K2 would do fine. Anything from your junk box between 1K and 4K7 should work though. I don't think you will need the diodes though. John :-#)# -- John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9 Call (604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames) www.flippers.com "Old pinballers never die, they just flip out"

Reply

mhs.stein＠gmail.com

1:19 a.m.

Probably a silly question, but I assume that if you're using a 28-pin socket you are inserting it aligned at the end *opposite* pin 1, with pin numbers offset by 2 (ie. 2x32 pin1 is 2764/256 pin 3, etc.)? m ----- Original Message ----- From: "Adrian Graham" <witchy at binarydinosaurs.co.uk> To: <jrr at flippers.com>; "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Saturday, October 24, 2015 8:56 PM Subject: Re: PROMs and EPROMs, specifically 2332/2532/2732

...

On 24/10/2015 19:18, "John Robertson" <jrr at flippers.com> wrote:

On 10/24/2015 5:43 AM, Adrian Graham wrote:

Hi folks, PET4032 repair continues with all ROMs, video RAM and dodgy sockets removed thanks to a hot air gun. Holes cleaned and I have new turned pin sockets for everything I've removed which I'll be fitting this afternoon. Since the ROMs came out OK I'm trying to dump them using my Pinmaster48 programmer, being from the 90s it doesn't read 2332/2532 PROMs but it WILL read a lot of variants of 2732 so I've made an adapter as found thanks to google and USENET: 2332 pin 18 to 2732 pin 21 (A11) 2332 pin 20 to 2732 pin 18 (Chip enable/Power Down) 2332 pin 21 to 2732 pins 20 and 24 via diodes with banding at the 2332 end (2332 Vpp) Wiring checks out and the diodes are aligned correctly so pins 20 and 24 don't interfere with each other, however the pinmaster continually gives me "continuity error on pin 20" Have I goofed somewhere? Cheers,

You just tie pin 21 of the 2332 to Vcc (+5) - Pin 24 of the 2732 - to read. There are only two modifications needed normally for reading a 2532/2332 in a 2732 socket. 2332 Pin 21 - Vcc (2732 Pin 24) Pin 18 - A11 (2732 Pin 21) If your reader coughs up a /OE error then use diodes: 2332 Pin 20 - 2 Diodes, one to 2732 pin 20, the other to 2732 pin 18. Banded end to these pins, and you will also need a pullup resistor on the 2532/2332 socket pin 20 so /CE (2332/2532) goes high when not selected. 2K2 would do fine. Anything from your junk box between 1K and 4K7 should work though. I don't think you will need the diodes though. John :-#)#

Hi John and others, Thanks for that. I removed the diodes and wired 2332 (21) to 2372 (24) leaving the A11 swap in place, the programmer complained about pin 18 missing. The adapter in front of me is now wired like this: 2332 (18) to 2372 (21) 2332 (20) 2x diodes to 2372 18 (band) and 20 (band) 2332 (20) 2k2 resistor to 2332 (24) 2332 (21) to 2372 (24) Now I get "reverse insertion" hinting I've wired something upside-down which isn't the case. I can read 27256 and 2764's no problem so I'm mounting things the correct way. Further digging into available eproms that the programmer can read reveals it CAN read 2532s, specifically the MCM2532 which the datasheet tells me has the same pinout as the 2332. I still get "reverse insertion error" so I'm guessing my PROMs are toast. Cheers, -- Adrian/Witchy Binary Dinosaurs creator/curator Www.binarydinosaurs.co.uk - the UK's biggest private home computer collection?

Reply

ard＠p850ug1.demon.co.uk

7:27 a.m.

...

Further digging into available eproms that the programmer can read reveals it CAN read 2532s, specifically the MCM2532 which the datasheet tells me has the same pinout as the 2332. I still get "reverse insertion error" so I'm guessing my PROMs are toast.

As others have said, normally when putting a smaller chip into a programmer with a bigger socket you keep the 'bottom row' of pins -- including ground -- in the same position for all chips. A look at the pinouts will show why. But as you can read 2716s that's not the problem. Often 'reverse insertion error' means the chip drew the wrong supply current -- often far too much. Which does suggest your ROMs are dead :-( -tony

Reply

jrr＠flippers.com

5:19 p.m.

On 10/24/2015 5:56 PM, Adrian Graham wrote:

...

On 24/10/2015 19:18, "John Robertson" <jrr at flippers.com> wrote:

On 10/24/2015 5:43 AM, Adrian Graham wrote:

Hi folks, PET4032 repair continues with all ROMs, video RAM and dodgy sockets removed thanks to a hot air gun. Holes cleaned and I have new turned pin sockets for everything I've removed which I'll be fitting this afternoon. Since the ROMs came out OK I'm trying to dump them using my Pinmaster48 programmer, being from the 90s it doesn't read 2332/2532 PROMs but it WILL read a lot of variants of 2732 so I've made an adapter as found thanks to google and USENET: 2332 pin 18 to 2732 pin 21 (A11) 2332 pin 20 to 2732 pin 18 (Chip enable/Power Down) 2332 pin 21 to 2732 pins 20 and 24 via diodes with banding at the 2332 end (2332 Vpp) Wiring checks out and the diodes are aligned correctly so pins 20 and 24 don't interfere with each other, however the pinmaster continually gives me "continuity error on pin 20" Have I goofed somewhere? Cheers,

You just tie pin 21 of the 2332 to Vcc (+5) - Pin 24 of the 2732 - to read. There are only two modifications needed normally for reading a 2532/2332 in a 2732 socket. 2332 Pin 21 - Vcc (2732 Pin 24) Pin 18 - A11 (2732 Pin 21) If your reader coughs up a /OE error then use diodes: 2332 Pin 20 - 2 Diodes, one to 2732 pin 20, the other to 2732 pin 18. Banded end to these pins, and you will also need a pullup resistor on the 2532/2332 socket pin 20 so /CE (2332/2532) goes high when not selected. 2K2 would do fine. Anything from your junk box between 1K and 4K7 should work though. I don't think you will need the diodes though. John :-#)#

Hi John and others, Thanks for that. I removed the diodes and wired 2332 (21) to 2372 (24) leaving the A11 swap in place, the programmer complained about pin 18 missing. The adapter in front of me is now wired like this: 2332 (18) to 2372 (21) 2332 (20) 2x diodes to 2372 18 (band) and 20 (band) 2332 (20) 2k2 resistor to 2332 (24) 2332 (21) to 2372 (24) Now I get "reverse insertion" hinting I've wired something upside-down which isn't the case. I can read 27256 and 2764's no problem so I'm mounting things the correct way. Further digging into available eproms that the programmer can read reveals it CAN read 2532s, specifically the MCM2532 which the datasheet tells me has the same pinout as the 2332. I still get "reverse insertion error" so I'm guessing my PROMs are toast. Cheers,

Reverse insertion may just mean you are drawing more current on the /OE and /CS than expected. How about this - go back to my original suggestion (no diodes) and this time add a small resistor to the 2732 pin 18 and Vcc to act as a load. Try larger resistors if the reader still complains - and try reading with NO 2332 in the reader (all FFs). Once you can trick the reader into reading air as highs then try your 2332 again with the working resistor values for the unused select. Oh, and what reader are you using? Did you check with the manufacturer (or archives somewhere - archive.org?) to see if they have a trick for reading 2532/2332s? John :-#)# -- John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9 Call (604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames) www.flippers.com "Old pinballers never die, they just flip out"

Reply

jrr＠flippers.com

26 Oct 26 Oct

5:09 p.m.

On 10/25/2015 4:37 PM, Adrian Graham wrote:

...

On 25/10/2015 17:19, "John Robertson" <jrr at flippers.com> wrote:

Hi John and others, Thanks for that. I removed the diodes and wired 2332 (21) to 2372 (24) leaving the A11 swap in place, the programmer complained about pin 18 missing. The adapter in front of me is now wired like this: 2332 (18) to 2372 (21) 2332 (20) 2x diodes to 2372 18 (band) and 20 (band) 2332 (20) 2k2 resistor to 2332 (24) 2332 (21) to 2372 (24) Now I get "reverse insertion" hinting I've wired something upside-down which isn't the case. I can read 27256 and 2764's no problem so I'm mounting things the correct way. Further digging into available eproms that the programmer can read reveals it CAN read 2532s, specifically the MCM2532 which the datasheet tells me has the same pinout as the 2332. I still get "reverse insertion error" so I'm guessing my PROMs are toast. Cheers,

Reverse insertion may just mean you are drawing more current on the /OE and /CS than expected. How about this - go back to my original suggestion (no diodes) and this time add a small resistor to the 2732 pin 18 and Vcc to act as a load. Try larger resistors if the reader still complains - and try reading with NO 2332 in the reader (all FFs). Once you can trick the reader into reading air as highs then try your 2332 again with the working resistor values for the unused select. Oh, and what reader are you using? Did you check with the manufacturer (or archives somewhere - archive.org?) to see if they have a trick for reading 2532/2332s?

It's an MQP Pinmaster48, a 90s-era programmer. As it happens tonight I got round to dumping some other EPROMs I had for someone else and one of them was an SGS2532 which read fine as an MCM2532 so I know the machine works with that age of chip. All my CBM ones give the same results so I'm assuming they're dead. Thinking about it there may be some 2532s at work so I can try burning a PET tester. I saw the madrigaldesign adapter on Friday but after re-re-remaking the one I was working on yesterday it was beginning to look a bit rough around the edges :) Cheers,

Hi Adrian, Perhaps one final test for your 2332s is to do a Diode Test on the pins relative to the ground pin (pin 12) and Vcc pin (24). These should sow either open or something like 0.6 or higher voltage drop across the pins - exchange the probes to check both directions. If your gates all read OK (check between Vcc and GND as well!), then it might be that the brand of 2332 you have simply draws more current than your programmer likes. What brand is the PROM? Perhaps it is in one of our reference book libraries... If your 2332s are bad, then have you put out a call to see if anyone else has archived them already? John :-#)# -- John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9 Call (604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames) www.flippers.com "Old pinballers never die, they just flip out"

Reply

witchy＠binarydinosaurs.co.uk

11:12 p.m.

On 26/10/2015 17:09, "John Robertson" <jrr at flippers.com> wrote:

...

Hi Adrian, Perhaps one final test for your 2332s is to do a Diode Test on the pins relative to the ground pin (pin 12) and Vcc pin (24). These should sow either open or something like 0.6 or higher voltage drop across the pins - exchange the probes to check both directions.

I'll make up a small wiring harness that'll let me do that, it'll give me something to do while I'm waiting for my lady to ring :)

...

If your gates all read OK (check between Vcc and GND as well!), then it might be that the brand of 2332 you have simply draws more current than your programmer likes. What brand is the PROM? Perhaps it is in one of our reference book libraries...

They're all stock MOS 2332 PROMs used in every CBM PET. Mike's already linked to the archive and tonight I've successfully burnt the 8032-flavour PETTEST.BIN so my programmer is happy with Ti 2532JL EEPROMS that I liberated from work. Cheers, -- Adrian/Witchy Binary Dinosaurs creator/curator Www.binarydinosaurs.co.uk - the UK's biggest private home computer collection?

Reply

jrr＠flippers.com

25 Oct 25 Oct

5:21 p.m.

On 10/24/2015 5:56 PM, Adrian Graham wrote:

...

On 24/10/2015 19:18, "John Robertson" <jrr at flippers.com> wrote:

On 10/24/2015 5:43 AM, Adrian Graham wrote:

Hi folks, PET4032 repair continues with all ROMs, video RAM and dodgy sockets removed thanks to a hot air gun. Holes cleaned and I have new turned pin sockets for everything I've removed which I'll be fitting this afternoon. Since the ROMs came out OK I'm trying to dump them using my Pinmaster48 programmer, being from the 90s it doesn't read 2332/2532 PROMs but it WILL read a lot of variants of 2732 so I've made an adapter as found thanks to google and USENET: 2332 pin 18 to 2732 pin 21 (A11) 2332 pin 20 to 2732 pin 18 (Chip enable/Power Down) 2332 pin 21 to 2732 pins 20 and 24 via diodes with banding at the 2332 end (2332 Vpp) Wiring checks out and the diodes are aligned correctly so pins 20 and 24 don't interfere with each other, however the pinmaster continually gives me "continuity error on pin 20" Have I goofed somewhere? Cheers,

You just tie pin 21 of the 2332 to Vcc (+5) - Pin 24 of the 2732 - to read. There are only two modifications needed normally for reading a 2532/2332 in a 2732 socket. 2332 Pin 21 - Vcc (2732 Pin 24) Pin 18 - A11 (2732 Pin 21) If your reader coughs up a /OE error then use diodes: 2332 Pin 20 - 2 Diodes, one to 2732 pin 20, the other to 2732 pin 18. Banded end to these pins, and you will also need a pullup resistor on the 2532/2332 socket pin 20 so /CE (2332/2532) goes high when not selected. 2K2 would do fine. Anything from your junk box between 1K and 4K7 should work though. I don't think you will need the diodes though. John :-#)#

Hi John and others, Thanks for that. I removed the diodes and wired 2332 (21) to 2372 (24) leaving the A11 swap in place, the programmer complained about pin 18 missing. The adapter in front of me is now wired like this: 2332 (18) to 2372 (21) 2332 (20) 2x diodes to 2372 18 (band) and 20 (band) 2332 (20) 2k2 resistor to 2332 (24) 2332 (21) to 2372 (24) Now I get "reverse insertion" hinting I've wired something upside-down which isn't the case. I can read 27256 and 2764's no problem so I'm mounting things the correct way. Further digging into available eproms that the programmer can read reveals it CAN read 2532s, specifically the MCM2532 which the datasheet tells me has the same pinout as the 2332. I still get "reverse insertion error" so I'm guessing my PROMs are toast. Cheers,

Oh, and here is what someone did in 2011 to read 2332s: http://www.madrigaldesign.it/forum/viewtopic.php?t=104 John ;-#)#

Reply