Digitech WH-1 - Anyone know the eprom chip models? + ROMs?

Started by MüThing, March 29, 2017, 07:42:36 PM

Previous topic - Next topic

MüThing

Hey there everyone,

I purchased a used Whammy mk.1 (Rev J board) from eBay, the seller was selling it for a crazy low price because they cannibalised it for the EPROM chips; namely U10, U11 and U22 or 23 (I can't find the actual chip number until I actually receive the whammy, it is the large labelled EPROM just below the rotary switch).

I was hoping someone may know what types of chips I would need to replace them as the chips are covered with labels in every photo I can find, as well as possibly knowing where I can find a copy of the ROMs for each of them (similar to how the old XP300 ROM was made available). I have a feeling those may not be around for people to download however, I hope it doesn't come down to it but maybe some intrepid user has a WH-1 of their own they're willing to take a look at.

If anyone has any info they can spare about any of this I'll greatly appreciate it, I'm not sure what I'll do with the pedal once I get it in working order but I'll be sure to keep anyone interested up to speed!

MüThing

Seems like I've found an example for U11, and more than likely for U10. A peeling label shows the letters "N825147AN" but looking that string up on Google only brings up datasheet websites that lead to 404s, and two single Alibaba listings for what seem to be the wrong chip. It might be worth finding whatever I can and seeing whether some kind of equivalent can be found.

Thanks to Govmnt_Lacky for finding the photo, now to find out what U22 is:

Quote from: Govmnt_Lacky on October 18, 2014, 07:25:56 PM
Could it be that the connections from the encoder to the PCB are wrong? Do you have any pre and post pictures of the encoder in the circuit?

Couple pictures I found of the encoder.



It looks like part of the encoder on one of the versions was "wired" and the wires were twisted.

MüThing

Derp! That 5 was actually an S (making it N82S147AN), I was sort of close though.


Looks like it's some kind of TTL PROM, the old kind that would be programmed by blowing out nichrome fuses to represent binary data. Doesn't look like the kind of thing I can do myself with one of those cheap TL866 units.

http://pdf1.alldatasheet.com/datasheet-pdf/view/115308/PHILIPS/N82S147AN.html

EBK

Quote from: MüThing on March 29, 2017, 07:42:36 PM
they cannibalised it for the EPROM chips
What an incredibly odd thing to do!
Did they take the EPROMs out to reverse engineer the pedal for Chinese knockoffs?
  • SUPPORTER
Technical difficulties.  Please stand by.

MüThing

Quote from: EBK on March 30, 2017, 09:23:50 AM
What an incredibly odd thing to do!
Did they take the EPROMs out to reverse engineer the pedal for Chinese knockoffs?

Doesn't seem like it, they said they picked it up to raid it for the chips but for what purpose I have no idea. Plus if they were going to reverse engineer anything you would imagine they'd take the L4A0256 from it (I personally wouldn't mind a knockoff of that thing).

digi2t

Send a request to Digitech service, and ask for the schematic. That's how I got all the XP schematics, so they might be kind enough to send you that one. The chip numbers will be on the schematic.
  • SUPPORTER
Dead End FX
http://www.deadendfx.com/

Asian Icemen rise again...
http://www.soundclick.com/bands/default.cfm?bandID=903467

"My ears don't distinguish good from great.  It's a blessing, really." EBK

MüThing

Quote from: digi2t on March 30, 2017, 12:47:14 PM
Send a request to Digitech service, and ask for the schematic. That's how I got all the XP schematics, so they might be kind enough to send you that one. The chip numbers will be on the schematic.

I sent them one this morning, checked my emails just a while ago and they came through for me.

According to the schematic, U10 and 11 are 82S147 chips (the name seems to differ by brand. And U22 is a regular 27C64, which is good news for ROM backup/flashing.

My only concern right now is how expensive the 82s chips are and how the flashing process isn't the same for them, I'm wondering whether someone else has figured out a more compatible alternative that can be flashed in the same way as the 27C would be. Only issue with that would be how quick they would need to be in order to work like the 82S chips do, they have a latency around 45ns iirc.

EBK

Quote from: MüThing on March 30, 2017, 06:46:36 PM
My only concern right now is how expensive the 82s chips are and how the flashing process isn't the same for them
You are facing a mountain of other unknowns, my friend.  You have no idea whether this pedal worked before the mysterious chip harvesting, and you haven't yet sourced the ROM code....

My advice?  Buy an identical but working version of this pedal.  Swap the ROMs from that into your original to test.  Then, if things go well, proceed to fix your original and sell one of the working units when you are done.
  • SUPPORTER
Technical difficulties.  Please stand by.

MüThing

Quote from: EBK on March 30, 2017, 07:19:31 PM
You are facing a mountain of other unknowns, my friend.  You have no idea whether this pedal worked before the mysterious chip harvesting, and you haven't yet sourced the ROM code....

My advice?  Buy an identical but working version of this pedal.  Swap the ROMs from that into your original to test.  Then, if things go well, proceed to fix your original and sell one of the working units when you are done.

Sound advice, my only issue is how expensive/rare they are these days. It could take months before I find someone else with a working WH-1, more if I'm hoping they'll let me experiment like that.

digi2t

  • SUPPORTER
Dead End FX
http://www.deadendfx.com/

Asian Icemen rise again...
http://www.soundclick.com/bands/default.cfm?bandID=903467

"My ears don't distinguish good from great.  It's a blessing, really." EBK

MüThing

You're right about this one Digi, but I'm determined to get an end result from it.

I've done some digging online and found some people in the retro arcade communities who needed replacements for older 82S bipolar proms. A couple years ago someone was selling adapter PCBs for 82S123s, 147s and the like, I managed to find their account on eBay where they were still selling the boards and they told me they only made a couple of the 147 adapter boards and didn't have any left, but they were fine with sharing a diagram of the adapter with me.



This means that I can both read from the 82S147 proms as though they were a typical 27C type eeprom AND use 27Cs as replacements for the prom. Their choice of EEPROM replacements are the W27C512-45 by Winbond; the -45 suffix standing for their 45ns access time (the same as the 82s147 proms). Meanwhile the 27C64 for the main instructions can be read as is in a universal programmer.

I have a second WH-1 arriving soon that I hope to use for reading the ROMs from, I'll be sure to swap the chips into the other WH-1 before doing anything else to make sure that the lack of the ROMs is the only thing wrong with the unit.

MüThing

Some excellent news!

The "donor" Whammy arrived this morning (from Slovenia), I swapped over the U10, 11 and 22 from that one into the other unit and it worked perfectly! The only issue was the rotary encoder didn't seem to match up with what it was supposed to be but that appears to be because of the shoddy rewiring job the previous owner did.



It seems as if they had some issues with the encoder and tried to rewire it by running the 4 pins from the encoder down underneath the board and connecting them to pins P1.1 through 1.4 on U20, that's something I'll have to get sorted out as well.

MüThing

More progress:





Connected the PROMs up to my programmer based on the pinout of the adapter board; it read it perfectly fine. I don't have the replacement 2716 EPROMs just yet, nor do I have the 27C64 to replace U22. Once I have both of those however I'll get to work on making a prototype adapter board out of vero and then maybe make a proper board via OSHpark or do it myself if possible.

I've also uploaded the .bin files to MEGA in the minute case that anyone wanted or needed them, I guess if Digitech doesn't have them or wants to give them out it's up to us to have these kinds of things available.

https://mega.nz/#F!wIlgDK6R!SpVNXL4LhpAtmIz9BtFPHg

EBK

I'm amazed and extremely happy for you at the same time.  I had previously expressed some doubts about your project, but I admire your ambition and perseverance.  Consider me a hybrid cheerleader-mime, silently hoping for your victory to come.  :icon_wink:

I'm especially glad that there appear to be no (serious) mysterious side effects from the still-perplexing chip harvesting that took place.  My current theories are:
1) generic chips with a certain number of pins may be worth something to counterfeiters, but STILL, that doesn't explain why they bothered to keep/sell the thing afterward.   :icon_confused:
2) chips easily removed because they were socketed were useful as an arts and crafts supply.  (Yay, good guys!  :icon_razz:)
  • SUPPORTER
Technical difficulties.  Please stand by.

MüThing



The easy part's done, the 2764 EPROMS arrived this afternoon (TMS2764JL-30 to be exact). I burned the ROM I made and stuck it in and found that it worked perfectly. I was worried the added access time of these chips would cause some problems (the original is 200ns according to the schematic while the TMS ones are 300).

MüThing



Another update, this one took longer than I was hoping because the 2716 eeproms I receive weren't erased or blank checked and for some reason no amount of UV exposure would make them completely blank. I stumbled upon the pin-compatible 2816 eeprom a while later and gave that one a try.

Unfortunately the Whammy doesn't seem too happy to work with them; my guess is the access time of the 2816s are way too slow for what the rest of the system is expecting (the 2816s have 250ns access time, the original 82s147s have 45ns) causing the while thing to hiccup and start outputting random glitchy noise.

I'll see if I can get in contact with some small business that can burn proper bipolar proms (I have my eye on eprompro.com but if anyone knows anyone else please let me know), it seems this is about as far as I can go by myself but I'm still rather happy with the progress I've made so far.

MüThing

A very small update, I've been putting off doing anything new for this project but I'm certain that the ROM I extracted from the bipolar PROMs myself wouldn't work even if I got it burnt onto the proper medium. I'll be looking into getting them properly duplicated from the working WH-1 PROMs very soon.

As well as this, I've recently found a store page in Russia that is supposedly selling suitable replacements for the rotary encoder (they specifically say they're for the WH-1). I attempted to place and order but the sysadmin sent me an email saying they'd get new stock in about 2 weeks. Finding replacement encoders seems to have been a big problem over the years with loads of old topics for it, so I'm leaving a link to their store here: https://pedal-parts.com/en/home/potentiometers/encoders/whammy-wh-1-encoder

Prehistoricman

Quote from: MüThing on January 21, 2018, 11:13:49 AMI'm certain that the ROM I extracted from the bipolar PROMs myself wouldn't work even if I got it burnt onto the proper medium.

Why's that?

MüThing

Because I extracted the contents via a dubious looking 2716 to 82S147 adapter used in reverse, and looking at the contents of the .BIN files I extracted the code seems to loop several times rather than looking like a single block of code like the WH-1 27C64 ROM I dumped.

mitchschaft

Thanks for sharing. I'm pretty sure the rotary on mine is shot.