Tokamac build and programming XC1736 chips

[DrGonzo]

So, in tandem with building a couple of Performer 68030 boards for my SEs, I also put together a couple of the Tokamac-LC boards for my LCIIs. Everything is in and together, but I'm really struggling to get the XC1736 programmed. After determining that they weren't supported by my TL866Plus, I went out and snagged a Xeltek 280U which states it supports them. The SuperPro software does list them and will even do a blank check and read from the chips. However, whenever I try to write them it just throws an error. Anyone run into this and, if so, have any idea what I could be missing here?

Thanks in advance!
~Dr. G

 

[Forrest]

In your picture, the pin on the top left corner does NOT look like it’s in the socket. I wonder if the same pin is not in your programmer.

 

[DrGonzo]

Hi Forrest,

I'll take a closer look when I get back in front of it in a bit here. I actually just threw one of the 8 or so chips I'd been trying to flash on the board for the picture for completeness, but I'm quite sure they've been firmly seated in the ZIF socket (or most of the others were anyway).

I did look at the datasheet for the pins and the voltage required for programming. It looks like I should see 6v on pin 7 (Vpp) to enter programming mode and then like 15v during write. Perhaps it's not delivering that. Might take my multimeter to it and see what's what.

Thanks for the extra set or eyes and suggestion!

 

[SuperSVGA]

You're using the XC1736A setting on an XC1736D? I think those have different programming voltages.

 

[DrGonzo]

According to the data sheets, the XC1736/XC1736A want upwards to 15v while the XC1736D wants only 12v (bad if they are D I'd imagine). If memory serves, I did try selecting the XC1736D in the device list, but it not only wouldn't program, it wouldn't even read/blank check them. It also said something about selecting a different algorithm for an XC1765x or some such thing? I am now 2nd guessing myself so I'll give it another go tonight and report back to validate that my recollection isn't a complete fabrication. heheh Thanks!

 

[SuperSVGA]

I know my programmer does warn about the reset polarity when switching to the 1736D, as it's a feature that the 1736A does not have. You may want to double check you have this the right way around, because I think the default on the 1736D is active high rather than active low, while it looks like the 1736A might only be active low.

 

[DrGonzo]

So this is what I get when I select an XC1736D from the list:



As you can see. It not only fails to program as it does with XC1736/XC1736A, but also won't blank check or read the chip at all. Error! across the board. I did also do a continuity check on the ZIF socket to make sure that each pin is making good contact. All good there. I'll try to see if I can get a reading on the voltage on Vpp during program, but guessing that I won't be able to get a good reading as it fails so fast and likely drops back down very quickly. Will report back if I get anything though.

Thanks again for all the help/ideas,
~Dr. G

 

[SuperSVGA]

Just out of curiosity, if you switch the reset polarity under Dev. Config, does it read? I can't find any clear documentation on whether that setting matters for reading.

 

[DrGonzo]

No, unfortunately not. Tried it both ways and get the same error whenever selecting XC1736D for read and blank check. With XC1736/XC1736A the option isn't there to switch it, but for D it says high is the default so perhaps the default for all of them?

 

[SuperSVGA]

Yeah, I'm guessing high is the default. The data sheet for the 1736A is a bit confusing because it shows the reset net label as active low, but it looks like according to other parts of the data sheet it's active high.

It is strange that the 1736D setting isn't working, as it shouldn't be that different from the 1736A. Since it's reading with the 1736A setting that makes me think that the chip is working fine, but you can't program it because the programming voltage is too high.
It's possible that the programmer software is configured incorrectly for the 1736D. I'm not sure why else a read would fail when it works with the other.
I wonder if you could cheat with a voltage divider, put like a 2.2k/10k divider on the VPP to bring it closer to 12V.

 

[DrGonzo]

That might be an option. I think there are also files that control the behavior for each type of chip buried in the SP280U software, so perhaps if they're editable I can create my own custom config as a hybrid between the two. Will keep digging.

 

[Jockelill]

Galep supports that chip, but it’s an expensive programmer.

 

[mg.man]

Apparently (@max1zzz can confirm?) the Dataman 48LV does as well. I just bagged one for about $150 - not for this, but for something old it also supports.

 

[Phipli]

Apparently (@max1zzz can confirm?) the Dataman 48LV does as well. I just bagged one for about $150 - not for this, but for something old it also supports.

I'm not max1zzz, but can confirm he has successfully programmed AM1736D chips.

 

[JC8080]

@Bolle, do you have any suggestions for programming the XC1736 chips? I assume you were able to successfully program one for your clone project.

 

[Bolle]

None of the cheap modern programmers will work with them. You either want something really old or a really expensive modern programmer… Elnec, Xeltek, Dataman - something along those lines.

 

[Phipli]

I wrote an Arduino program for verifying the PROMs if needed.

Lets you check them independently of the programmer.

 

[max1zzz]

I have used my Dataman 48LV to program these successfully but as Bolle says these are not cheap. Cost me a shade over £200 iirc but you can sometimes find them much cheaper if you look around (Though keep in mind this specific model is a parallel programmer and the software only officially supports Win2K at the latest)

If anyone is struggling to get these chips programmed I am happy to program the chips for you, all I would ask is you to cover the postage both ways

 

[DrGonzo]

I'd picked up an old Xeltek SuperPro 280U that said it would program them. I guess technically it does but they don't seem to work. It may be because the github says to set the reset polarity high, but the software doesn't give me the option on the XC1736?

Max1zzz, I'll likely take you up on your offer if you wouldn't mind zapping a handful for me. Just have to find them again as it's been a while since I last worked on these.

Thanks!
~Dr. G

 

[trag]

Probably not useful, but the Xilinx XC1736D is on the list of supported devices for the Needham EMP-30 programmer.

That's a very old programmer. Needham went out of business in the early 2000s. It was a $1500 programmer in its day, IIRC.