2732 and TL866II Plus

[Wawavoun]

Hello there,

I use the well know TL866II Plus programmer for my old sbc computers... I use the up to date Xgecu software.

I have programmed a lot of 2716 memory with the help of an external 25VDC power supply connected to pin 21 (pin of course disconnected from the programmer).
I also set programming pulse duration to the maximum value...
I have to disable "pin detect" and "verify" before programming but this is normal... After programming I disconnect the 25 VDC supply, reconnect the pin 21 to the programmer and made a separate verification which is always successful.

Now today I have try to do the same with 3 different 2732 eproms... No way to got a result. The programming progress bar even not appear !!!
Of course on 2732 eprom Vpp is applied to pin 20.

I have check both datasheet of 2716 and 2732 and don't understand why this work for one and not for the other.

I have noticed that some Xgecu software set a voltage level to 6.5 V instead of 5 V for some 2732 model eprom but both settings give the same result.

I have not found any log file or something like that for founding the problem...
The programmer read 2732 eprom without error.

Some other users report success in the same situation so any advice will be helpful.

Regards.
Philippe

[dcac]

I think 2716 has a dedicated VPP pin - but 2732 has VPP and OE is on the same pin.

And VPP on 2732 needs a specific timing sequence for programming. But on 2716 you can probably get away with keeping it at a fixed level.

[Wawavoun]

You are perfectly right !

I dont see this evidence reading the datasheet.

Thanks a lot.
Philippe

[Wawavoun]

Hello there !

Okay some works about 2732 programming with TL866II Plus.

I have quickly made a hack based on the joined schematics...

Now VPP on the eprom follow the timing give by VPP pin of the programmer but at a level adapted to the eprom (usually 25 VCC).

This hack works for 2716. Pin detect works, reading eprom works, programming and verifying eprom works !

Scope show that 25 V timing on the eprom pin follow the VPP programmer pin at the correct level !

But no success for 2732. Pin detect pass, reading works but programming fail !
At programming start scope show that eprom VPP pin start to switch between 0 V and ~25V then everything stop and I got the fail message...

I have try with a very short usb cable to avoid voltage drop and 3 eprom gave the same result so I dont think that the problem is the eprom themselves...

Any help will be appreciated !

Regards.
Philippe

 

[Wawavoun]

A try with minipro and a pulse duration of 50000 µs give the same result...
Switch to usb 3 port also didn't change anything...

[g4tuz]

I have just ran into the problem with 2732 eproms. Looking at the 2732 data sheet gives pin 20, VPP high 25V and pin 18 pgm low for programming. However looking on a dual trace scope I have pin 18 high at the same time as VPP high. This is with the V11.90 windows software, have not tried this with minipro under Linux yet.

[PKTKS]

It may  sound obvious but these ROMs are really old.  They can be replaced with modern ones  that won't require 25v

This programmer can read the old ones and easily burn new versions...

Sounds hassle free.. 

Paul

[Phil_G]

I'd suggest the Atmel AT28C32 if you can source any.   EEPROMS are so much more convenient 
 
I recall from the bad old days that 2732s were always troublesome. Either neighbour, the 2716 and 2764s were all generic, but for some reason 2732s from different manufacturers behaved differently. IIRC Texas ones were to be avoided, though my recollection is poor.
Then there was the 2732A and 2532 to further confuse matters... 

G4TUZ I think I have a few varieties of 2732 in my stash, if so I could send some to try?

Edit: so far I've found an ST, two different AM's, a Texas and an NEC if you'd like them

[m k]

I have a rule of thumb recollection that non flats are for the bin.
Can't remember why.

[g4tuz]

Noted about the AT28C32, but some vintage restorers want the 'genuine' article. I never had a problem with 2732/32A with a homebrew programmer based on an ISA card with 8255 I/O that I built in the mid 80's. Ironically I did make the same error with the pgm pulse when writing the driver software that the TL866II software suffers from. I have yet to try minipro under Linux to see if the error has been copied there. If it has it should be an ease fix. Thanks for the EPROM offer, I have a good supply, just need to get the 866 working correctly.

[g4tuz]

OK, after a bit more testing and digging. The open source minipro software also manages the oe line (pin 18) incorrectly. It would appear that the original windows device data base was dumped to provide the starting point for minipro, the windows issue has been carried over to linux etc.

[Dilbert66]

I know this is an old thread but for others having the issue with 2732's , here is a solution with the tl866ii plus.

See my post here for more details: https://www.eevblog.com/forum/testgear/modification-to-add-the-missing-21-and-25-volt-vpp-ranges-to-your-tl866ii-plus/msg3959678/#msg3959678