MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Prophet 5 issue - DIY'ers please suggest?
MUFF WIGGLER Forum Index -> General Gear  
Author Prophet 5 issue - DIY'ers please suggest?
Ranxerox
My rev 2 Prophet 5 has developed a fault whereby osc B is way way quieter than oscillator A or the noise source in the mixer section. The knob for osc B only affects loudness in the first 10% or so of the throw (and seems a bit spotty at that). Otherwise the oscillator seems to behave normally...

Could it be the pot itself? Or a chip in the mixer section? Any suggestions greatly appreciated hmmm.....
Dr. Sketch-n-Etch
Without knowing anything about the circuit, my first guess would be a dodgy pot. Is it easily replaced? If you can gain access to the pot terminals with the synth running, then you can determine whether the pot is receiving a full-strength signal or not.
Ranxerox
Thanks, your advice made me finally look into the service manual! The pot is a 100k linear that's buried behind one of the other boards, but shouldn't be too hard to get to in the service position. If that's not the problem, then the mixer section uses three 4051 op-amps, hopefully won't be too hard to trace if one of them is faulty...
Graham Hinton
Ranxerox wrote:
My rev 2 Prophet 5 has developed a fault whereby osc B is way way quieter than oscillator A or the noise source in the mixer section. The knob for osc B only affects loudness in the first 10% or so of the throw (and seems a bit spotty at that). Otherwise the oscillator seems to behave normally...

Could it be the pot itself? Or a chip in the mixer section? Any suggestions greatly appreciated hmmm.....


If the fault is the same on all five voices then it's not the mixer section. What about patches from memory? If they are higher volume then it's probably the pot or the multiplexer that routes it to the ADC, if they have OscB low too then the fault will be in the CV path from the DAC to the OscB level control.
Ranxerox
A good point, thanks Graham. Osc B seems to be quiet for all voices, and irrespective of whether from a recalled patch or in edit mode. So it would be logical to assume the problem is on the DAC side of things, as you say...
oldgearguy
without knowing your level of experience, I have to caution that opening up and diagnosing and servicing a rev 2 Prophet 5 is not fun nor easy. Standing it up on end so the machine can be opened and have the boards connected and yet not shorting out while you reach inside with a scope or DMM can be a challenge.

Of course, if you've spent a lot of time teching your own gear and have dealt with the various manufacturers methods of assembly, then it shouldn't be too much of a problem.
Ranxerox
oldgearguy wrote:
without knowing your level of experience, I have to caution that opening up and diagnosing and servicing a rev 2 Prophet 5 is not fun nor easy. Standing it up on end so the machine can be opened and have the boards connected and yet not shorting out while you reach inside with a scope or DMM can be a challenge.

Of course, if you've spent a lot of time teching your own gear and have dealt with the various manufacturers methods of assembly, then it shouldn't be too much of a problem.


Thanks, yeah I've opened her up a couple of times before and serviced a few of my bits a pieces over the years, but I'm hardly an expert.

I'm hoping it's something cheap and relatively simple to diagnose, like a faulty op-amp. It could be either the relevant 4051 de-mux chip (U346) or the TL082 down-stream of pin 15 (U349), which controls the osc B mix VCA.

I don't think it's the latching s/h chip up-stream of the de-mux (U356), or there would be multiple issues - that 4051 controls quite a few parameters as well as osc level.

If the above turns out not to be the source of the problem then I'll probably have to admit defeat - and be in for a costly service cry
sneak-thief
This is the order of likelihood of failed components that I would replace:

OSC Mix pot - you can measure the output pin to see if it's putting out the full range of 0-5V as you rotate it. Also, check the pot for dry joints.

From page 41 of the PDF manual, aka "PCB2 POT MUX":

U202 - CD4051 - these blow all the time


And from page 46 of the PDF manual, aka "3-13 PCB 3 CV DMUX/LATCHES":

U356 - CD4051 - these blow all the time
U361 - TL082
Q302 - 2N4250
R322 - 33k - not very likely, but still possible!


The Rev. 2 is the only synth repair I walked away from - and I've repaired most of the big polys, CS70m, Ob-Xa, Synthex, JP8, JP6, JP4, P5 Rev. 3, etc. In this case, a decoupling cap had too much solder on it and it eventually moved and shorted out the whole thing. I eventually got tired of testing and replacing all the broken chips while walking a tightrope to not blow up any/all of the unobtanium SSM chips and gave it back to the owner.

It also looks like crap inside, with flying bodge-wires right out of the factory.
Ranxerox
sneak-thief wrote:
This is the order of likelihood of failed components that I would replace:

OSC Mix pot - you can measure the output pin to see if it's putting out the full range of 0-5V as you rotate it. Also, check the pot for dry joints.

From page 41 of the PDF manual, aka "PCB2 POT MUX":

U202 - CD4051 - these blow all the time


And from page 46 of the PDF manual, aka "3-13 PCB 3 CV DMUX/LATCHES":

U356 - CD4051 - these blow all the time
U361 - TL082
Q302 - 2N4250
R322 - 33k - not very likely, but still possible!


The Rev. 2 is the only synth repair I walked away from - and I've repaired most of the big polys, CS70m, Ob-Xa, Synthex, JP8, JP6, JP4, P5 Rev. 3, etc. In this case, a decoupling cap had too much solder on it and it eventually moved and shorted out the whole thing. I eventually got tired of testing and replacing all the broken chips while walking a tightrope to not blow up any/all of the unobtanium SSM chips and gave it back to the owner.

It also looks like crap inside, with flying bodge-wires right out of the factory.


Thanks, it's reassuring to know the 4051 is an obvious target - however I think you're citing the parts references from the Rev 3 technical manual? In any case, this is a helpful list of likely suspects. I think from process of elimination we've determined the problem is post-DAC, as it affects all patches whether in edit mode or not, eliminating the pot - but I'll check it anyway, just in case.

It sure does look pretty funky inside, and with so many ways to go into critical failure mode, I've heard it's a servicing nightmare... I think Oberheim's contemporaneous approach with the OB-X (separate cards for each voice, rear-panel case hinge - like opening the bonnet on a Ford F150!) was much nicer.
sneak-thief
Oh dang, I opened the wrong PDF.

Lemme try that again...

From page 61 of the PDF manual, aka "CONTROL PANEL POTS":

U202 - CD4051 - these blow all the time


And from page 78 of the PDF manual, aka "TUN MUX, S/H, AUD OUT":

U346 - CD4051 - these blow all the time
U349 - TL082
C322 - 0.01 poly
Ranxerox
Ok, I have finally gotten around to sourcing replacement parts and am going to fix this bad boy in the next week or two.

While I'm at it, I have decided to do something about the transformer- the present one buzzes annoyingly, so I have.a nice toroidal one to replace it with.

I'm also annoyed at how hot the PSU gets, due to the overworked 20 into 5volt regulator. So my question is, would a higher efficiency regulator make a difference here? I'm thinking of the Oki 78SR, which is rated up to 36volts at 1.5 amps...
Graham Hinton
Ranxerox wrote:
I'm also annoyed at how hot the PSU gets, due to the overworked 20 into 5volt regulator. So my question is, would a higher efficiency regulator make a difference here? I'm thinking of the Oki 78SR, which is rated up to 36volts at 1.5 amps...


No, all regulators dissipate the voltage across them multiplied by the current through them. The 5V regulator gets hot because it is using the same secondary voltage as the 15V regulators. If you are having a custom toroid wound then get a separate 9Vac secondary added and use that for the 5V. Or just add another transformer.

The other thing you can do is reduce the load on the 5V. Replace the NMOS Z80 with a CMOS one and as many of the TTL logic chips as you can. 74HCTxx is a pretty safe substitution for 74LSxx.
Ranxerox
Thanks for this Graham. My toroidal trafo is sadly not a custom wound one - it's just a Hammond Manufacturing unit with 2 x 18v secondaries that are designed to bridge to 36v.

That's interesting regarding the regulators, the Murata brochure-ware for the OKI-78SR states:
Quote:

"Based on fixed-frequency buck switching topology, the high efficiency means very low heat and little electrical noise, requiring no external components."


Link: http://power.murata.com/data/power/oki-78sr.pdf

But are you saying the other regulators in the PSU circuit would negate these gains due to the overall current draw?

As you say, there is another way to reduce load on the 5v regulator which involves swapping the EPROM chips for a newer type, but on the Rev 2 Prophet this involves a bit of surgery on the PCB and a lot of messing around flashing binaries on to EPROMs which I would like to avoid!
Graham Hinton
Ranxerox wrote:
My toroidal trafo is sadly not a custom wound one - it's just a Hammond Manufacturing unit with 2 x 18v secondaries that are designed to bridge to 36v.


Then add a standard 2x 9Vac one with the secondaries in parallel. Disconnect the LM340 input and add another bridge and reservoir capacitor.

Quote:

But are you saying the other regulators in the PSU circuit would negate these gains due to the overall current draw?


No, most of the heat is due to the large voltage dropped across the 5.6V regulator.

I didn't look closely and realise that you were going to use a switching regulator. It might work, but it may be at its limit and you can't switch amps without some effect.

Quote:

As you say, there is another way to reduce load on the 5v regulator which involves swapping the EPROM chips for a newer type, but on the Rev 2 Prophet this involves a bit of surgery on the PCB and a lot of messing around flashing binaries on to EPROMs which I would like to avoid!


Leave the EPROMs, replace the CPU and TTL. If you update the memory you may as well design a whole new pcb.
Ranxerox
OK, success - the first thing I looked at was U349 (TL082), which didn't seem responsive on my scope when I moved the OSC2 mix knob in edit mode. Since it was socketed I decided I may as well try swapping it first - which fixed it immediately!

I think I was very lucky, as all of the TL082s on my Prophet 5's computer board are socketed, and the ICs that I was interested in were exposed without having to take out any of the other boards. Phew! Thanks to everyone for their guidance, especially Graham and Sneakthief - your advice helped a lot.

Chickened out on swapping the transformers this time; once it ran I decided I'd rather spend time playing the thing than doing more surgery - that'll be a job for next time... Regarding the regulator, more study needed regarding the impact of current draw on the 5V supply, with the aim of figuring out if there's a switched mode component that can act as a drop-in replacement.

The synth also has a long-standing issue with the NV RAM - only about 10 of the preset slots actually work out of a total of 40, so no idea what is causing this as each of the RAM chips relates to a different data bus line!? Lower priority but still - annoying!
Graham Hinton
Ranxerox wrote:
The synth also has a long-standing issue with the NV RAM - only about 10 of the preset slots actually work out of a total of 40, so no idea what is causing this as each of the RAM chips relates to a different data bus line!?


When did you last replace the backup battery? They are only good for about ten years lifetime, which seemed forever in 1976. I prefer backup capacitors now, there are special 1F ones for this job. They don't need to be replaced and backup for around 3 months, depending on load, enough for short term storage. For long term dump to a recorder.
Ranxerox
This RAM issue goes back at least 5 or 6 years; although I have noticed it getting worse recently - banks 4 and 5 were still OK, but are now totally filled with garbage.

It could be the RAM battery; I believe it was newly installed by Lucid 7-8 years ago, so should be about time to replace it. I have managed to source a couple of replacement CMOS SRAM chips, in case that doesn't cure it, but figuring which ones to swap will be tedious!

I take your point about long-term storage; programming the Prophet 5 is simple and quite fun, so having limited patch storage is no great hardship!
MUFF WIGGLER Forum Index -> General Gear  
Page 1 of 1
Powered by phpBB © phpBB Group