Hi guys,
This is my first post to this forum and the motive for it is as follows:
A few months ago, my old and beloved T3 (upgraded to EX status by myself many years ago) decided to quit. I know a bit of electronics (I said "a bit"!) so I decided to open it up and investigate. My findings were not heartening: not only was the fuse as black as my soul, but there was a rather large charred area in the power supply board with a hole in the middle, under two burnt up resistors. I managed to get a service manual, but after some failed attempts at repairing the power supply (which is of the switched variety), a friend of mine suggested a different course of action, which was taken: install a new, non-Korg power supply. So, I now have a power supply with all the correct voltages, including the display's, and the T3 now comes to life when switched on (I sized the opportunity to replace the battery and also all the switches - all 44 of them... - something I had already done once, some fifteen years ago).
The next step was to test the T3, so I performed all the tests as described in the service manual. All but two passed with no errors. But those two are proving rather difficult to solve, hence my request for help.
Error 1:
All tests of the disk drive end with a message "drive not ready". In fact, the drive problem was the first thing I noticed after the "resurrection", because the disk was not accessed as it usually was at switch on. I tried another drive, which is working perfectly in a PC, but the result was the same (BTW, I could not get the T3 drive to work in a PC). My problem now is, I do not know how to determine whether this is a problem of the drive or the controller IC and associated circuitry. I also thought the disk might be bad, so I "T3-formatted" a blank disk with OmniFlop and successfully copied a "T3_factory_sounds.img" file into it. But the tests with this disk also failed miserably. Erm... HELP!
Error 2:
The RAM test fails with an error message pointing to address "DRAM 10000 - 3FFFF". Now, in the T3 there are three blocks of identical DRAM (a block of five ICs and two of eight - one of which is the EX expansion) and I have no idea which of these blocks does this address refer to, nor do I know which individual ICs within the block are faulty. Also, all the ICs in the block of five heat up quite a lot and I'd like to know whether this is normal or it is a sign that they are about to go up in flames soon! So, again, HELP!...
Would someone be able to please supply some relevant info that may help with these problems or know where else I should look for help?
Thanks in advance.
Cheers,
Slurpy
Massive T3 disaster - I need HELP!
Moderators: Sharp, X-Trade, Pepperpotty, karmathanever
Hi guys,
Given the overwhelming response of ZERO replies to my appeal for help, I couldn't but keep you all informed about the evolution of the situation, which is as follows:
The floppy drive has resurrected from the dead and is now working fine and passing all the tests. Don't ask me how it happened (I know you won't!), but the fact is, it now works just fine.
The RAM memory thing. I am now 99% sure that the five RAM IC's that serve the effects unit will have to be replaced and that will solve the problem. As soon as I get the new IC's I'll try this out and will, of course, share the results with all of you (how could I not?!!!).
Cheers,
Slurpy
Given the overwhelming response of ZERO replies to my appeal for help, I couldn't but keep you all informed about the evolution of the situation, which is as follows:
The floppy drive has resurrected from the dead and is now working fine and passing all the tests. Don't ask me how it happened (I know you won't!), but the fact is, it now works just fine.
The RAM memory thing. I am now 99% sure that the five RAM IC's that serve the effects unit will have to be replaced and that will solve the problem. As soon as I get the new IC's I'll try this out and will, of course, share the results with all of you (how could I not?!!!).
Cheers,
Slurpy
I'm not as lucky as you, slurpy - I have no service manual for the T. If you care to share that with me, I might be able to help some.
conundrum is right as far as I'm concerned, your questions were too specific to answer without proper documentation - I have no clue what specific address chips are located at, but might - or might not - be able to tell from a schematic, for example.
Only two (given the circumstances, mildly useful) educated guesses I can make based on your description:
1. the resistors and the PSU board was 'charred' likely because of the filter capacitor drying up and resulting in larger peaks on the rails, which in turn overheated the resistors (and the rest of the PSU went with it)
2. DRAM's of the era used to warm up quite a bit (hot - but not burning - to the touch) and therefore I'd assume it is rather normal. I'd be concerned about the others that aren't warm. I'm inclined to think it is the wave RAM area, as I doubt that the CPU has access to the DSP's RAM (hence no check routine for that), but I might be wrong here without seeing the schematic, again (basing myself on the similar M1's architecture).
Again, if you care to share the service documentation (please PM) I might be able to help a little more.
conundrum is right as far as I'm concerned, your questions were too specific to answer without proper documentation - I have no clue what specific address chips are located at, but might - or might not - be able to tell from a schematic, for example.
Only two (given the circumstances, mildly useful) educated guesses I can make based on your description:
1. the resistors and the PSU board was 'charred' likely because of the filter capacitor drying up and resulting in larger peaks on the rails, which in turn overheated the resistors (and the rest of the PSU went with it)
2. DRAM's of the era used to warm up quite a bit (hot - but not burning - to the touch) and therefore I'd assume it is rather normal. I'd be concerned about the others that aren't warm. I'm inclined to think it is the wave RAM area, as I doubt that the CPU has access to the DSP's RAM (hence no check routine for that), but I might be wrong here without seeing the schematic, again (basing myself on the similar M1's architecture).
Again, if you care to share the service documentation (please PM) I might be able to help a little more.
T3 down
Hi guys! I mean, HI GUYS!!!
Boy, this is like finding life in other planets - I'm not alone! Sincere thanks for your replies!
But before I get down to business, please allow me to set the record straight:
All I did in my previous post was state a fact (zero replies) - nothing unfair about that. OK, I admit I allowed a modicum of sarcasm to seep through, but that was absolutely intentional in trying to elicit some response, an objective that was, as you know, fully achieved... And again, thanks for replying.
Done. To business then:
The PSU and the floppy drive are no longer issues, as both problems are now solved.
As for the RAM, there is a small RAM space attached to the main processor which I've ruled out, as most of the T3 is actually working (MIDI, keyboard, joystick, pedal and external switches ADC, the auto-test, the display, etc). Without system RAM this would be impossible, me thinks.
Then there's the PCM RAM (the EX expansion ram). This is composed of 8 identical IC's which share a board with the Mask ROM, another 8 IC's which contain all the internal samples and waveforms (nothing else in this card, except for some filters and capacitors). I've ruled these out as well. Here's why:
When you enter the auto-test mode (by simultaneously pressing the lower zero + PROG A switches while turning power ON) there is a message saying something like
"PCM RAM Exist. Now writing check data" - after a while the battery check is executed and all LEDs light up indicating the tests were successful. Then it automatically moves on to the second phase, where you test everything else. When you quit the auto-test, the internal RAM is then checked. This test fails (all LEDs blink) and the screen looks like this:
RAM Check
PCMRAM
DRAM 00000-0FFFF DRAM 10000-3FFFF (inverted - dark background, clear characters)
SRAM
As per the manual, the inverted address points to the bad RAM range.
Now this ROM/PCM RAM board links to the main board via a single large connector. If you perform the same tests with this board disconnected, the results are mostly the same. The ONLY differences are the following:
1. When you enter the auto-test there is no PCMRAM check (no message) - it starts with the battery check.
2. When you quit the auto-test and the internal RAM is checked, the test fails and the screen looks like this:
RAM Check
DRAM 00000-0FFFF DRAM 10000-3FFFF (inverted - dark background, clear characters)
SRAM
followed by a message "Error: Internal RAM"
What is the RAM this address points to? From the above I can be fairly sure that it's not the PCMRAM.
There are only two remaining RAM blocks: the eight IC SRAM block (the sequencer Work RAM which apparently passes the test) and the five IC block, which are the ones that get burning hot and are connected to the MDE (the effects unit). I'm inclined to believe the latter RAM block is the culprit (or the MDE itself - perish the thought!).
In the block diagram one can see that the MDE is the last stage before the output, and that every sound must pass this stage before being sent out. It seems that this RAM area is used for the intermediate results of the MDE calculations. As such, if this RAM is bad, we can expect to get some horrible noises in the output, which is exactly what I'm getting.
So, what I'll have to do is what I said I would: replace the five IC RAM block and hope for the best.
Stay tuned - do not miss the next chapters!
Cheers,
Slurpy
Boy, this is like finding life in other planets - I'm not alone! Sincere thanks for your replies!
But before I get down to business, please allow me to set the record straight:
All I did in my previous post was state a fact (zero replies) - nothing unfair about that. OK, I admit I allowed a modicum of sarcasm to seep through, but that was absolutely intentional in trying to elicit some response, an objective that was, as you know, fully achieved... And again, thanks for replying.
Done. To business then:
The PSU and the floppy drive are no longer issues, as both problems are now solved.
As for the RAM, there is a small RAM space attached to the main processor which I've ruled out, as most of the T3 is actually working (MIDI, keyboard, joystick, pedal and external switches ADC, the auto-test, the display, etc). Without system RAM this would be impossible, me thinks.
Then there's the PCM RAM (the EX expansion ram). This is composed of 8 identical IC's which share a board with the Mask ROM, another 8 IC's which contain all the internal samples and waveforms (nothing else in this card, except for some filters and capacitors). I've ruled these out as well. Here's why:
When you enter the auto-test mode (by simultaneously pressing the lower zero + PROG A switches while turning power ON) there is a message saying something like
"PCM RAM Exist. Now writing check data" - after a while the battery check is executed and all LEDs light up indicating the tests were successful. Then it automatically moves on to the second phase, where you test everything else. When you quit the auto-test, the internal RAM is then checked. This test fails (all LEDs blink) and the screen looks like this:
RAM Check
PCMRAM
DRAM 00000-0FFFF DRAM 10000-3FFFF (inverted - dark background, clear characters)
SRAM
As per the manual, the inverted address points to the bad RAM range.
Now this ROM/PCM RAM board links to the main board via a single large connector. If you perform the same tests with this board disconnected, the results are mostly the same. The ONLY differences are the following:
1. When you enter the auto-test there is no PCMRAM check (no message) - it starts with the battery check.
2. When you quit the auto-test and the internal RAM is checked, the test fails and the screen looks like this:
RAM Check
DRAM 00000-0FFFF DRAM 10000-3FFFF (inverted - dark background, clear characters)
SRAM
followed by a message "Error: Internal RAM"
What is the RAM this address points to? From the above I can be fairly sure that it's not the PCMRAM.
There are only two remaining RAM blocks: the eight IC SRAM block (the sequencer Work RAM which apparently passes the test) and the five IC block, which are the ones that get burning hot and are connected to the MDE (the effects unit). I'm inclined to believe the latter RAM block is the culprit (or the MDE itself - perish the thought!).
In the block diagram one can see that the MDE is the last stage before the output, and that every sound must pass this stage before being sent out. It seems that this RAM area is used for the intermediate results of the MDE calculations. As such, if this RAM is bad, we can expect to get some horrible noises in the output, which is exactly what I'm getting.
So, what I'll have to do is what I said I would: replace the five IC RAM block and hope for the best.
Stay tuned - do not miss the next chapters!
Cheers,
Slurpy
T3 down
Hi Synthjoe
Disabling the effects does nothing - just tried it. Curiously enough, even with the effects disabled, fiddling with the effects parameters changes the type of noise I get - this and changing the effect are the only actions that affect the output sound. Oh, and this only on effects unit 2.
Come to think of it, even if you're not using the effects unit, the sound still has to be written to and read from RAM, although, as you rightly suggest, unchanged (unless the RAM is bad, as I suspect, which would totally garble the sound as it does).
If you have any further suggestions, please post them.
Cheers,
Slurpy
Disabling the effects does nothing - just tried it. Curiously enough, even with the effects disabled, fiddling with the effects parameters changes the type of noise I get - this and changing the effect are the only actions that affect the output sound. Oh, and this only on effects unit 2.
Come to think of it, even if you're not using the effects unit, the sound still has to be written to and read from RAM, although, as you rightly suggest, unchanged (unless the RAM is bad, as I suspect, which would totally garble the sound as it does).
If you have any further suggestions, please post them.
Cheers,
Slurpy
I assume you have checked effects off/on, 'no effect' and bypass using a pedal switch.
However, I think output to C/D for the very least (in combi or sequencer mode), should eliminate all processing by the MDE (DSP in my vocabulary), surely so if effects processing is set to 'serial'. Same for an appropriately panned drumkit in program mode.
Have you tried these, as well?
However, I think output to C/D for the very least (in combi or sequencer mode), should eliminate all processing by the MDE (DSP in my vocabulary), surely so if effects processing is set to 'serial'. Same for an appropriately panned drumkit in program mode.
Have you tried these, as well?
OK, having looked at the schematic I'm almost sure the DRAM area 10000-3FFFF refers to IC37-44. I agree that PCM RAMs are likely intact.
IC32-36 are the 'fivepack' which I believe you're referring to as 'too hot', but these are only accessible by the MDE (MB87405 - IC31), so I don't think the CPU is able to directly address, hence to verify these. However, the noise you've described supports the likelihood of malfunction. Check No. 5 in diagnostics mode. it is performing tests on that DRAM, I believe.
Which brings me to the conclusion, that the PSU failure might well have entailed a general failure of DRAM chips, those being the most sensitive components of the era. Unfortunately, the diagnostics is not very verbose about the deficit in 10000-3FFFF, so it might be that only one or a couple out of the 8 chips are affected among IC37-44.
However, I think that the lower 64k (00000 to 0FFFF) is covered by the battery backed CMOS SRAM (IC28?-29), which is a completely different animal. These might have survived the PSU failure, which would explain why programs are retained and most check routines and operations seem to be OK. I don't understand, however, why there's a check for 00000-0FFFF and then separately for SRAM? Maybe I've overlooked something on the schematic, will check again later.
I'd suggest loading a sequence and trying to play that back, visually verifying whether the sequencer operates, possibly sending sequencer data to another module via MIDI to verify results. Sequencer uses the DRAM chips referred to above, some SYX (MIDI data filer) operations might be another bunch, but not much more I'm aware of, hence it is a bit difficult to see how much that RAM is affected. Programs, combis (and hence their edit modes) are using the lower 64k SRAM area, which is reported intact accordingly by the diagnostics.
IC32-36 are the 'fivepack' which I believe you're referring to as 'too hot', but these are only accessible by the MDE (MB87405 - IC31), so I don't think the CPU is able to directly address, hence to verify these. However, the noise you've described supports the likelihood of malfunction. Check No. 5 in diagnostics mode. it is performing tests on that DRAM, I believe.
Which brings me to the conclusion, that the PSU failure might well have entailed a general failure of DRAM chips, those being the most sensitive components of the era. Unfortunately, the diagnostics is not very verbose about the deficit in 10000-3FFFF, so it might be that only one or a couple out of the 8 chips are affected among IC37-44.
However, I think that the lower 64k (00000 to 0FFFF) is covered by the battery backed CMOS SRAM (IC28?-29), which is a completely different animal. These might have survived the PSU failure, which would explain why programs are retained and most check routines and operations seem to be OK. I don't understand, however, why there's a check for 00000-0FFFF and then separately for SRAM? Maybe I've overlooked something on the schematic, will check again later.
I'd suggest loading a sequence and trying to play that back, visually verifying whether the sequencer operates, possibly sending sequencer data to another module via MIDI to verify results. Sequencer uses the DRAM chips referred to above, some SYX (MIDI data filer) operations might be another bunch, but not much more I'm aware of, hence it is a bit difficult to see how much that RAM is affected. Programs, combis (and hence their edit modes) are using the lower 64k SRAM area, which is reported intact accordingly by the diagnostics.
T3 down
Hi Synthjoe,
See? You should never underrate yourself, man! Your latest post was extremely useful and has already allowed this tech adventure to evolve to a higher level and reach some interesting conclusions!
As per your suggestion, I tried outputs CD/34 and there they were, all the ROM sounds in all their dry glory! Dry, because all the effects are missing... (BTW, disabling the effects by any other means does not cure the garbled sounds problem).
Conclusions:
Every sound sent directly to the output is ok. Every sound that goes through one or both effects units - even if OFF - gets silenced or garbled. I still think and hope this will be cured by replacing the fivepack RAM.
Then I progressed to the sequencer. Loaded the original demos from the floppy and pressed Play in song 00 (Hop something...). It did start playing but soon stopped and crashed the T3. Power cycled it, loaded the sequences again, moved to song 01 (Interlude?) and pressed Play again. The song played to the end without a glitch. More, it then automatically went and played song 02 (The One). This one went well almost to the end but it did stop a few notes before and crashed the T3. The last two songs (Duke and Marshall Man) didn't even start - the T3 crashed immediately after pressing Play. I repeated everything a couple of times with the exact same results. Because I was loading the sequences from a very old original floppy (which Omniflop refuses to read!), I loaded the sequences by sysex (which I had downloaded from KORG.UK) and repeated the tests. No change. The same crashes at the very same points in the sequences.
Conclusions:
The sequencer proper is fine, the sequencer RAM eightpack (SRAM?) is bad.
There's one very important test that I was not yet able to perform - the PCM RAM. The fact is, I do not have any T3 PCM samples. I tried to create a DSM-1 disk using Omniflop with some .dsk images (first formatting the floppy to DSM-1, then writing the file to it). It goes fine up to 91%, then aborts with an error message just saying it was unable to conclude the operation. Am I doing something wrong here?
So, this is the current situation. Any more suggestions?...
Before I go (to bed - it's 5:10 am!), I want to really thank you for your help, man. I was quite stuck and now there's a light at the end of the tunnel!
Cheers,
Slurpy
See? You should never underrate yourself, man! Your latest post was extremely useful and has already allowed this tech adventure to evolve to a higher level and reach some interesting conclusions!
As per your suggestion, I tried outputs CD/34 and there they were, all the ROM sounds in all their dry glory! Dry, because all the effects are missing... (BTW, disabling the effects by any other means does not cure the garbled sounds problem).
Conclusions:
Every sound sent directly to the output is ok. Every sound that goes through one or both effects units - even if OFF - gets silenced or garbled. I still think and hope this will be cured by replacing the fivepack RAM.
Then I progressed to the sequencer. Loaded the original demos from the floppy and pressed Play in song 00 (Hop something...). It did start playing but soon stopped and crashed the T3. Power cycled it, loaded the sequences again, moved to song 01 (Interlude?) and pressed Play again. The song played to the end without a glitch. More, it then automatically went and played song 02 (The One). This one went well almost to the end but it did stop a few notes before and crashed the T3. The last two songs (Duke and Marshall Man) didn't even start - the T3 crashed immediately after pressing Play. I repeated everything a couple of times with the exact same results. Because I was loading the sequences from a very old original floppy (which Omniflop refuses to read!), I loaded the sequences by sysex (which I had downloaded from KORG.UK) and repeated the tests. No change. The same crashes at the very same points in the sequences.
Conclusions:
The sequencer proper is fine, the sequencer RAM eightpack (SRAM?) is bad.
There's one very important test that I was not yet able to perform - the PCM RAM. The fact is, I do not have any T3 PCM samples. I tried to create a DSM-1 disk using Omniflop with some .dsk images (first formatting the floppy to DSM-1, then writing the file to it). It goes fine up to 91%, then aborts with an error message just saying it was unable to conclude the operation. Am I doing something wrong here?
So, this is the current situation. Any more suggestions?...
Before I go (to bed - it's 5:10 am!), I want to really thank you for your help, man. I was quite stuck and now there's a light at the end of the tunnel!
Cheers,
Slurpy
You're welcome, glad I could contribute to some progress - would not have been possible without the proper documentation... 
So, I'd suggest to replace the full 'fivepack' - from the slight changes you mention when changing effects I'd assume the MDE chip is probably fine. Do test the chips if you can, maybe you'll find one or two amongst them which are still OK - this/these can be handy when fixing the work RAM (even though I wouldn't trust too much a chip that survived others nearby - it might have been affacted by the PSU failure too, and operating only marginally, ready to fail at any moment).
From the symptoms you describe, it seems that seqencer RAM is affected only to some extent, not all 8 chips. Furthermore, not all failures will crash the sequencer. For one, only command byte corruptions are likely to cause trouble, as opposed to data byte corruption. Plus, some memory failures can go undetected. If memory fails to '1', for example, and the original data contained also '1' on that position, then it will go unnoticed. The crashing of the sequencer in a given tune per se is not an accurate indication of where the problem lies.
So, the challenge will be to figure out which chips need replacement, without unsoldering all eight of them. This memory is organised in two banks of 64k x 16 bits each (2x128 kB). It is difficult to read from the schematic, but I believe that ICs 38 (bits 13-15), 40 (b9-12), 42 (b4-7) and 44 (b0-3) are the lower address bank, while the others are the upper bank. If you were able to determine which half of the memory is affected, then you'd get closer to the ones needing careful inspection. Temperature differences among these chips can be a tell-tale. Obviuosly, testing data pins with an oscillioscope triggered from pin2 (/CAS) of each IC while running memory diagnostics would be a much better way. Also, dumping a large sequence from PC via sysex and then loading it back and comparing the result in the binary domain can be revealing.
As to the DSM images, if it comes from Larry's page then they are 'packed' image files created by Copy QM (another tool like Omniflop, but only works under Win9x), so Omniflop will not be able to create a disk from that.
Lemme see what I can do for you in the PCM disk department (even though I'm sure those chips are fine, diagnostics would have already reported, if otherwise)...
So, I'd suggest to replace the full 'fivepack' - from the slight changes you mention when changing effects I'd assume the MDE chip is probably fine. Do test the chips if you can, maybe you'll find one or two amongst them which are still OK - this/these can be handy when fixing the work RAM (even though I wouldn't trust too much a chip that survived others nearby - it might have been affacted by the PSU failure too, and operating only marginally, ready to fail at any moment).
From the symptoms you describe, it seems that seqencer RAM is affected only to some extent, not all 8 chips. Furthermore, not all failures will crash the sequencer. For one, only command byte corruptions are likely to cause trouble, as opposed to data byte corruption. Plus, some memory failures can go undetected. If memory fails to '1', for example, and the original data contained also '1' on that position, then it will go unnoticed. The crashing of the sequencer in a given tune per se is not an accurate indication of where the problem lies.
So, the challenge will be to figure out which chips need replacement, without unsoldering all eight of them. This memory is organised in two banks of 64k x 16 bits each (2x128 kB). It is difficult to read from the schematic, but I believe that ICs 38 (bits 13-15), 40 (b9-12), 42 (b4-7) and 44 (b0-3) are the lower address bank, while the others are the upper bank. If you were able to determine which half of the memory is affected, then you'd get closer to the ones needing careful inspection. Temperature differences among these chips can be a tell-tale. Obviuosly, testing data pins with an oscillioscope triggered from pin2 (/CAS) of each IC while running memory diagnostics would be a much better way. Also, dumping a large sequence from PC via sysex and then loading it back and comparing the result in the binary domain can be revealing.
As to the DSM images, if it comes from Larry's page then they are 'packed' image files created by Copy QM (another tool like Omniflop, but only works under Win9x), so Omniflop will not be able to create a disk from that.
Lemme see what I can do for you in the PCM disk department (even though I'm sure those chips are fine, diagnostics would have already reported, if otherwise)...
Hi Synthjoe,
Here is today's SITREP (situation report...):
I managed to load some PCM sounds from floppy disks. The new multisounds show up as Dxx, as they should, and play perfectly. This quietens my paranoia about the PCM RAM. You know, part of this paranoia stems from the fact that I found two electrolytic caps with stuff oozing out of them in the PCM ROM/RAM board (I replaced them, of course).
As for the sequencer, I loaded the same demo songs one by one. They all played back with no glitches, provided they were loaded into internal song positions 01 or 02 (I did not test the higher positions). This gives strength to the conclusion that some RAM chips are good, some are bad. But, you know, I'm not inclined to trust any of those IC's - after all they were all subjected to the same supply surge that killed some of them. So I think I'll simply scrap the lot and replace all thirteen. But let me tell you that I liked your "cunning plan" of comparing the sequences before and after being loaded to RAM!
Just out of curiosity, I'd like to know what messages you get when you perform the RAM tests in your T2/3.
The next step is replacing the RAM chips. That will take some time, but I will certainly let you know when it's done.
Thanks again for everything.
Cheers,
Slurpy
Here is today's SITREP (situation report...):
I managed to load some PCM sounds from floppy disks. The new multisounds show up as Dxx, as they should, and play perfectly. This quietens my paranoia about the PCM RAM. You know, part of this paranoia stems from the fact that I found two electrolytic caps with stuff oozing out of them in the PCM ROM/RAM board (I replaced them, of course).
As for the sequencer, I loaded the same demo songs one by one. They all played back with no glitches, provided they were loaded into internal song positions 01 or 02 (I did not test the higher positions). This gives strength to the conclusion that some RAM chips are good, some are bad. But, you know, I'm not inclined to trust any of those IC's - after all they were all subjected to the same supply surge that killed some of them. So I think I'll simply scrap the lot and replace all thirteen. But let me tell you that I liked your "cunning plan" of comparing the sequences before and after being loaded to RAM!
Just out of curiosity, I'd like to know what messages you get when you perform the RAM tests in your T2/3.
The next step is replacing the RAM chips. That will take some time, but I will certainly let you know when it's done.
Thanks again for everything.
Cheers,
Slurpy
T3 down
Hi all,
Would anyone know (and please tell me) where to get 13 x Mitsubishi M5M4464AL-12 (ZIP case) RAM memory chips in Europe?
If you've been following this thread you'll know I need them to bring my T3 back into full life quality (it must be really nasty for a synth to live without the juice of DSP effects...).
I'd be eternally thankful for that information.
Cheers,
Slurpy
Would anyone know (and please tell me) where to get 13 x Mitsubishi M5M4464AL-12 (ZIP case) RAM memory chips in Europe?
If you've been following this thread you'll know I need them to bring my T3 back into full life quality (it must be really nasty for a synth to live without the juice of DSP effects...).
I'd be eternally thankful for that information.
Cheers,
Slurpy
T3 UP!
HI guys,
Just a short note to let you all know that after replacing 5 RAM chips (no mean feat, I tell you...) my T3 EX is now playing perfectly!
The sequencer is still only half-operational - I was unable to get all the 13 chips I needed (I got 7), but if I can get them I'll replace those as well.
The RAM auto check still fails (and those 5 chips do get hot!), but I was unable to detect anything wrong with the sound or anything else apart from the sequencer, so I guess the check involves all 13 chips and will only pass successfully after all of them have been replaced.
Thanks again to Synthjoe (and Conundrum, who also bothered to reply) whose help was indeed fundamental in solving the problem.
In fact, Synthjoe, I still have a question for you: what does your floppy drive do when you switch on? Does the LED come on or does it simply make that faint spinning noise for a brief while? Please let me know.
Cheers,
Slurpy
Just a short note to let you all know that after replacing 5 RAM chips (no mean feat, I tell you...) my T3 EX is now playing perfectly!
The sequencer is still only half-operational - I was unable to get all the 13 chips I needed (I got 7), but if I can get them I'll replace those as well.
The RAM auto check still fails (and those 5 chips do get hot!), but I was unable to detect anything wrong with the sound or anything else apart from the sequencer, so I guess the check involves all 13 chips and will only pass successfully after all of them have been replaced.
Thanks again to Synthjoe (and Conundrum, who also bothered to reply) whose help was indeed fundamental in solving the problem.
In fact, Synthjoe, I still have a question for you: what does your floppy drive do when you switch on? Does the LED come on or does it simply make that faint spinning noise for a brief while? Please let me know.
Cheers,
Slurpy