thoughts on Digitech Space Station eprom copying?

[Gui113m]

Marmaliser hooked me up with a set of each chip and an extra ram chip just recently, if anyone else needs one or a 27C1001 he's your best bet if you're in europe. Thanks again, man! My 400's a 300 for now.

+1

Thanks marmaliser!!

[caleboaks]

@caleb

First of all, if you are still getting some sort of "proper program" signal under the white noise, then this tells me that your EEPROM is programmed correctly AND the U8 chip was soldered correctly.

My #1 suspect in this situation would be the RAM chips. Specifically, U16, U17, or U18. Try swapping these 3 RAM chips (one at a time) into U19 while in Whammy/Wah mode. See if one of them will cause your Whammy/Wah to act the same way. Even if they don't, they could still be causing the issue though.

As for the busted traces on the U8 chip, you might be in luck. From U8, those three traces each go to the EEPROM chip U9.

U8 Pin 14 ---> U9 Pin 5
U8 Pin 15 ---> U9 Pin 6
U8 Pin 16 ---> U9 Pin 7
U8 Pin 17 ---> U9 Pin 8

Good Luck with the fixes.

Thanks for the advice! Unfortunately, after making sure that all the connections on the SMD chip were good, I still don't get anything on the display or any output, with or without white noise... Could I have ruined the SMD chip?

[Govmnt_Lacky]

Thanks for the advice! Unfortunately, after making sure that all the connections on the SMD chip were good, I still don't get anything on the display or any output, with or without white noise... Could I have ruined the SMD chip?

My advice would be to upload some good pictures of your mod. Specifically, try to get good shots of the RAM chips, the R30 area, U8, U9 area, your switch and wiring..... anything that could help.

[Govmnt_Lacky]

Throwing this out there in case someone has any fresh ideas....

I built an XP ALL with stacked 27C256 EEPROMs. This was built on an XP100 PCB so I had to remove R30, swap R38 over to R47, Install U8, and new IC sockets for the RAM chips.

Everything looked great..... then the problem! Randomly, the pedal will blank out and/or show hieroglyphs in the display. No sound output on any of the jacks. There is no particular thing that I do which induces this. It appears to be totally random. The timing is random as well. I can power the pedal for 2-3 hours with no problems. I can unplug the pedal power and then plug it back in and 5 minutes later it will hang up 

I leave it overnight and then plug it in and it will be fine for 30-40 minutes and then hang up again. It is frustrating because there is no particular trigger and it is totally random.

I thought I found the problem being some questionable solder points on U8 however, the problem has reappeared. I am starting to think this might be a problem with the onboard CPU or sound processor.

If anyone has a fresh idea... I'm all ears. 

[slacker775]

swap R38 over to R47

Whats this part? Haven't seen anything other than remove R38 before.

As for your issue, I wonder if it could be potentially bad memory - though odd that it could go awhile before flaking out.  I would be more inclined to guess there is some kind of instability either with the CPU or the memory that over time due to heat or the like sends the unit into lala land.  Maybe even a weak solder joint?

[Govmnt_Lacky]

Whats this part? Haven't seen anything other than remove R38 before.

R38 is an SMD 10 ohm resistor. On XP100s where they use the CPU internal memory for the program, this resistor is in the R38 position. To have the CPU look to the EEPROM for the programs, the resistor must be removed from the R38 position and soldered into the R47 position.

I would be more inclined to guess there is some kind of instability either with the CPU or the memory that over time due to heat or the like sends the unit into lala land.  Maybe even a weak solder joint?

Right now, I am thinking it is inherent to the CPU or the Sound processor. Weak solder joint is what led me to checking U8 and resoldering it's joints which appeared to work however, this was a false indication as the pedal started hanging up after 2 more power cycles.

The fact that I get total loss of sound AND the display goes blank/funky tells me that it is most likely in the CPU 

[slacker775]

Whats this part? Haven't seen anything other than remove R38 before.

R38 is an SMD 10 ohm resistor. On XP100s where they use the CPU internal memory for the program, this resistor is in the R38 position. To have the CPU look to the EEPROM for the programs, the resistor must be removed from the R38 position and soldered into the R47 position.

Well crap, thats not in the doc!  I'm working on my second conversion and I'm actually getting somewhat similar results to what you have - except it happens right out of the gate.  I haven't tried audio yet, but when I power it on, most of the time I just get gobbledygook on the display.  On rare occasions, it has actually seemed to function normally for a few seconds before flaking out.  I haven't done anything with R47 though.  I'll have to look at that when I get home.

I would be more inclined to guess there is some kind of instability either with the CPU or the memory that over time due to heat or the like sends the unit into lala land.  Maybe even a weak solder joint?

Right now, I am thinking it is inherent to the CPU or the Sound processor. Weak solder joint is what led me to checking U8 and resoldering it's joints which appeared to work however, this was a false indication as the pedal started hanging up after 2 more power cycles.

The fact that I get total loss of sound AND the display goes blank/funky tells me that it is most likely in the CPU 

Since I've started dealing with these things, I've been looking for a good way to test the RAM chips.  Just some way I could drop them into a breadboard with an Arduino or RaspPi and run it thru standard memory tests to verify if the chips are good.  So far I haven't come up with anything (ARGH!).  At least the EEPROM can be read back and verified against the BIN file.  Outside of that, it's a bit of a guessing game.  Even the CPU on these things is proprietary as far as I can tell so no datasheets or the like to give us some diagnostic info to help with debugging.

[Govmnt_Lacky]

Since I've started dealing with these things, I've been looking for a good way to test the RAM chips.  Just some way I could drop them into a breadboard with an Arduino or RaspPi and run it thru standard memory tests to verify if the chips are good.  So far I haven't come up with anything (ARGH!).  At least the EEPROM can be read back and verified against the BIN file.  Outside of that, it's a bit of a guessing game.  Even the CPU on these things is proprietary as far as I can tell so no datasheets or the like to give us some diagnostic info to help with debugging.

The CPU is not proprietary. Only the Digitech version with the masked internal memory is. The actual CPU chip info can be found. Look up "Philips 80C31 Microcontroller"

The sound processor (the biggest chip on the PCB) unfortunately IS proprietary. It is also not made anymore 

As for the RAM chips... yeah, it's a guess. All you can do is verify all solder joints and connections to the rest of the circuit 

How many EEPROM chips are in your build? Are you using the CPU masked memory to retrieve the XP100 program and EEPROMs for the other 3?

[slacker775]

I'm doing the XP-All style with a 27C1001.  My first conversion worked flawlessly, but I was also helped a bit in that my XP-100 base already used an EEPROM and already had U8 in place.

[slacker]

I've got a quick question for anyone who's done an XPall+ conversion, could you confirm that the switch wiring diagram on page 56 of the build doc is correct http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=51931
The reason I'm asking is me and Greg have been working on electronic switching system, basically replaces the rotary with a stomp switch to scroll through the modes, we've got the XPAll sorted but are having problems with the XPAll+
Looking at the build doc (page 40) it says the different modes start at hex addresses 0000, 8000, 10000 and 18000 which makes sense because that's splitting the EEPROM up into 4 equal chunks. However, the values given for A14, A15 and A16 in the table on page 39 for 10000 and 18000 are incorrect and this seems to have carried through to the switch diagram on page 56.

10000 hex is  A14: 0 A15: 0 A16: 1
18000 hex is A14: 0 A15: 1 A16: 1

Cheers in advance

[slacker]

You can ignore my question, just heard back from Greg and we've got it working. The switching diagram makes sense now, in the  xp300 and xp400 settings the switch leaves A14 floating and it's controlled by the 83C154 so this must pull it low when it starts reading the EPPROM and the start addresses will then be either 10000 or 18000, the A15 and A16 settings are correct, so it's only the table that's wrong.
Actually this leads on to something else, you don't actually need the rotary to control A14, it's driven by the 83C154 and will always be 0 when it starts reading the eeprom because it always starts at what it thinks is address 0. You only need to switch A15 and A16.

[slacker775]

I like the sound of an electronic switching setup!   Especially if that could be put onto a pcb with the leds to greatly reduce the amount of wiring!

[Govmnt_Lacky]

Thanks Ian! 

So far, the XP ALL+ build that I have with the electronic switching is electrically functioning properly. No sound check yet but I will get to that at a later time and report back.

To sum up. The A14 connection (in my build) is only connected to its place on the XP PCB. No pull up resistor to 5V and nothing to pull it down to 0V. Essentially, the 83C154 is controlling the switching. The only lines that I am currently controlling is A15 and A16.

I am using the 27C1001 EEPROM built by Dino and one of marmaliser's adapter PCBs. No relay installed for Dry signal in XP100 mode.

So, according to this.... here is the coding for the A15 and A16 lines per pedal:

XP100 = A15: 0V  A16: 0V
XP200 = A15: 5V  A16: 0V
XP300 = A15: 0V  A16: 5V
XP400 = A15: 5V  A16: 5V

[MüThing]

I'm liking the idea of a footswitch over a rotary, if a PCB does end up being made I think I'd consider an XP-ALL+ mod a lot more.

[slacker775]

Ultimately the wiring isn't all that bad, but I'm always a big fan of less wiring.

[mitchschaft]

Cool, so we won't need the relay anymore?

[MüThing]

Cool, so we won't need the relay anymore?

I think the relay would still be needed for the stereo dry out jumper.

[mitchschaft]

Oh yeah, I forgot. Sometimes mine doesn't switch, but I haven't gone through troubleshooting yet.

[Govmnt_Lacky]

Cool, so we won't need the relay anymore?

You still need the relay for the Dry Out while in XP100 mode. I don't use it in my build(s) because I just do not think it is worth it for the Dry output 

[marmaliser]

The thing that slows me down to a standstill on doing the XP all+ is the switch wiring , Electronic switching on a PCB sounds great to me .  Happy to do a PCB if someone can come up with a scheme