sh101 troubles..

[asciwhite]

morning all,

I was going to bend my 101 but hit a snag. While moving it out of one room and in to another, it suddenly stopped working. All it does is a very quiet noise which will only play, if you can call it that, in one key. I've read loads online which all point to a dead slider (strangely white noise still works) thanks to roland using them for earthing.

Suppose I'm looking for a little reassurance that by getting some new sliders installed will at least get me back to a working machine.

cheers,

c

[Circuitbenders]

Does the filter and ADSR still work? With the resonance up to maximum you should be able to hear it working even if theres no osciallator signal.
When you say that the noise works do you mean that it works perfectly at normal volumes but for some reason the oscillator is very quiet in comparison?

A lot of faults on 101's can be traced to dead sliders but the first thng you should probably check is that the (very expensive) CEM3340 VCO is working. If you have a scope take a look at pins 4 an 8 on IC13 where you should be seeing a square and sawtooth wave. Another option is to probe with an audio jack when the waveforms are mixed before going into the filter. Set all the mixer faders at max, connect the sleeve to ground and then find R109. R110. R111 and R112 on the synth board and play a note while connecting the tip to the right hand sides (nearer to IC13).
These are the raw waveforms just after the mixer stage so if you're not getting anything here it might be time to worry.

I take it you've eliminated knackered jack sockets as the problem?

[asciwhite]

Sorry, I didn't explain that well. Filter works as does the adsr. There is some sound coming from the oscilator but it's minuscule compared with what it had previously at normal volumes). Moving the resonance up to max and sliding the cut off still work too.

I literally moved it from one room to the next; But this particular  one has had a hard life so that seemed to help point at a slider..

Alas no scope.

Yeah I checked the jacks and all seemed ok. No bad joints and I cleaned out the connectors.

c

Does the filter and ADSR still work? With the resonance up to maximum you should be able to hear it working even if theres no osciallator signal.
When you say that the noise works do you mean that it works perfectly at normal volumes but for some reason the oscillator is very quiet in comparison?

A lot of faults on 101's can be traced to dead sliders but the first thng you should probably check is that the (very expensive) CEM3340 VCO is working. If you have a scope take a look at pins 4 an 8 on IC13 where you should be seeing a square and sawtooth wave. Another option is to probe with an audio jack when the waveforms are mixed before going into the filter. Set all the mixer faders at max, connect the sleeve to ground and then find R109. R110. R111 and R112 on the synth board and play a note while connecting the tip to the right hand sides (nearer to IC13).
These are the raw waveforms just after the mixer stage so if you're not getting anything here it might be time to worry.

I take it you've eliminated knackered jack sockets as the problem?

[Circuitbenders]

Hmmm, it doesn't sound like a fader issue to me. I could be wrong but i can't really see which fader it would be that would cause you to lose all of the VCO waves but leave the noise intact. I'm not sure if you're saying that the noise still works perfectly and can be filtered normally at normal volumes or not.

If the filter is definately working then you need to trace the VCO signals back to where they are being lost. It might just be something like a loose cable or dry joint, especially if it died just from moving it around. You should be able to to this with the schematic and an audio jack connected to some PC speakers or something.

Having said that, if tit will only play one note, albeit very quietly, that might indicate some kind of CV problem. Does the note actually change when you press different keys and have you tried running it from an external CV source like a midi-cv convertor? Its going to be very difficult to track down a CV problem without a scope.

[asciwhite]

White noise goes through the filter and adsr without any problems or any change to how it played pre failure of the unit. I had a good look around for a loose connection and came back blank. Not stripped it down completely so a dry joint could be the issue.

The note stays the same no matter which key is pressed. Had a feeling this was going to be nye on impossible without the correct equipment.

I didn't try to play it via the cv connector (was in a blind panic as it's not mine) but I'll give that a go next. Typical really, finally get my hands on a 101 and it lasts al of three months...

Hmmm, it doesn't sound like a fader issue to me. I could be wrong but i can't really see which fader it would be that would cause you to lose all of the VCO waves but leave the noise intact. I'm not sure if you're saying that the noise still works perfectly and can be filtered normally at normal volumes or not.

If the filter is definately working then you need to trace the VCO signals back to where they are being lost. It might just be something like a loose cable or dry joint, especially if it died just from moving it around. You should be able to to this with the schematic and an audio jack connected to some PC speakers or something.

Having said that, if tit will only play one note, albeit very quietly, that might indicate some kind of CV problem. Does the note actually change when you press different keys and have you tried running it from an external CV source like a midi-cv convertor? Its going to be very difficult to track down a CV problem without a scope.

[Bogus Noise]

Does sound like something to do with the oscillator chip to me - the note you're hearing may actually be the 101's own gate signal. The fact that it's not changing and very quiet makes me think this - I've repaired an SH101 with a damaged CEM chip in the past and it was behaving very similarly.

However, as Paul says, because it happened while moving it, it's much more likely to be that something came loose rather than a dead chip. Hope you can get it sorted!

[asciwhite]

Guys, cheers for all the input.

As you can believe I'm pretty desperate to get this back in working order.

I'll have a proper look around it in the morning.

Thanks again!

a

[Polysixx]

Hi 101 experts I wonder if you can help me?

I have 2 101s with different problems and I'm  hoping it's just new sliders needed in both cases but not sure.

#1 this works fully but fault 1 is the PWM adjust (not vol) slider does NOTHING at all to the wave, it's 'dead'. The volume of the PWM (compared to the saw/sub) is about half as loud too, again it could be a slider with too much resistance? The volume does change with the slider but it doesn't go as loud as my other 101 on that wave.

(edit) have used some slider lube/cleaner.. the PWM now appears to be doing 'something' it kinda works but sounds very weak - not like my other one, however the fact it's doing something now where it wasnt before (after some switch cleaning/lube) would suggest it's a slider prob? it had no effect on my other slider probs though

Also on this unit the FREQ does nothing at all, it appears (comparing to my working one) that the cut off is permanently set around the 3 mark, the slider does nothing. Does this sound just like a dead slider or could it be the filter chip/ir3109?  The resonance slider fully works as expected, so if res works but freq doesn't then slider? The p/o looked after it, it's good cosmetically, all original packaging no missing screws but has sat unused for years so I'm hoping it's just sliders?

Aside from that all is well and works as expected

#2 on the other unit everything works except VCO modulation (from either the joystick OR the slider in VCO section) it just remains 'un modulated'. I did managed to test the slider in the bend section for this and it didn't change at all through the range and the resistance was way high compared to the other 2 so I'm sure that is one (i'm gonna swap bend boards between units to see if it rules it out - unless the modulation up top also needs to route through that for it to work?) . until I can check the resistance on the mod slider in the VCO section I won't know if it appears to work ok

so.. all things considered does this sound like the typical 'new sliders needed' problems or can any faults in the CEM 3340 or IR3109 cause these problems? The reason I doubt it's those is as I said everything else works great and sounds great but I suppose whatever pins are responsible for certain things could lead to dead 'stuff' inside the chips?

ANd lastly - does anyone know of other sources (pref UK) for replacement sliders, esp if not full kits, other than tech transplant?

Thanks

[Circuitbenders]

At an educated guess i'd say all those issues you describe there could be down to dodgy faders.

Theres nowhere else on the planet that will sell you 101 faders apart from technology transplant, although if you pay for membership to the TT site you can get the bender board faders individually as well as all the pots. Unfortunately they only sell the faders in full sets. Having said that, you would probably find that one full set might fix the issues on both your machines.

I've found that its often possible to actually open up a 101 fader and either clean out the crap thats stopping  it working or to use some conductive paint to bridge the worn out part of the track, which is inevitably at the top or bottom. If you're careful you can probably reassemble it as good as new.

[Polysixx]

Thanks, well i'm gonna have to buy a full set anyway and at least have some spares and fix both (hopefully). I wonder how much I'll be hit for import duty on these (anyone in the UK bought them and know?)

I'll desolder the 'bad' sliders from both + 1 working one which I can test in place to double check before ordering. They are all 100k on the upper control area but some are marked A some are B so they are different spec even though the same length and apparent resistance yes?

thanks

[Circuitbenders]

A & B refer to the response curve of the fader. A is probably logarithmic and B is linear, but it might be the other way around. They would certainly work in place of each other but you might find that about 80% of the effect is in 20% of the travel distance if you get them mixed up.

. I wonder how much I'll be hit for import duty on these (anyone in the UK bought them and know?)

If you even get charged i can more or less guarantee that the parcelforce £13.50 'clearance fee' will be more than the customs VAT fee. This clearance fee is illegal and you don't have to pay it despite what parcelforce will tell you. You never entered into any contract with them to do anything with your parcel and its illegal for them to ransom items of post for payments other than VAT.
If they send you a letter saying you have to pay an import VAT fee and a clearance fee just phone them up and say you want to pay the VAT now but you're not going to pay the clearance fee. Pay the VAT over the phone and you'll get your package the next day, and probably an invoice for the clearance fee in a week or so that you can ignore. They won't do anything about it as they can't afford to go to court and get a judgement against them as it would set a precedent.

Thats my parcelforce rant for the day. 

[Polysixx]

many thanks!

Someone should really order a job lot of these kits into the UK in one go (or smuggle them in in a suitcase ) and sell them to us 101-desperados!

[Polysixx]

Hmm so I've had #1 apart. It all looked original, you could tell by the screw sound when I undid the boards and th general tightness of the wiring compared to my other one (which has had slider repairs in the past). And yet, for some reason I've found 'Jumper 5' on the synth board has been 'cut' and bent over slightly so it doesn't connect!

I'm wondering if this could ever happen by itself (heating up?) and if not why would someone cut it, unless they had previously modded it? but I see zero signs of any mod work and it was that dusty inside behind the controls etc that it certainly hadn't been cleaned out before either.

On my other 101 J5 is connected of course. gonna look at the schematics to see what J5 does/where it leads before I solder it back up but any thoughts in the meantime would be welcome.

Also what is the best way to test the sliders with a multimeter, I seem to be getting erratic readings (like the covering they use after assembly on the PCB is stopping a good reading, should I scrape this clear first) and which pins are best to test on the slider and what resistance am I looking for on FREQ, PWM and PWM VOL? at it's lowest and highest positions - Thanks!


edit > also on #2 there is a white jumper wire on the top of the synth board (I've seen this in other pics of other 101s but on #1 there is none but there are 2 resistors on the back?) #2 is an earlier serial number so could this explain these differences? DId they change from resitors (on the trace side) to the white wire on the component side? If not then perhaps those resistors were also part of a mod.. must google for 101 innards to check this

Oh sorry for all the qs but other than a mild soapy water solution and a soft sponge / tissue is there any better way to get a grey 101 back to grey instead of 'off grey/smoke tinged'? I've cleaned them both up and they look better but still not as grey as some I've seen, not sure you can get it completely new looking but any tips?



---------- edit / slider findings ----------

So I tested the lot using the 'obvious' testing points of the sliders using 200k scale. I tested them fully open and closed: from left to right (starting with lfo ending on release)

open: 0.3 closed: 55
open: 0.3 closed: 50
open: 1 closed: 1 (PULSE WIDTH: Dead/slider contacts not touching)

open: 0.3 closed: 49
open: 0.3 closed: 20
open: 0.0 closed: 86
open: 0.3 closed: 59

open: 1 closed: 1  (FREQ: Dead/slider contacts not touching)
open: 0.3 closed: 40

open: 0.3 closed: 40
open: 0.3 closed: 37
open: 0.0 closed: 80

open: 1 closed: 1
open: 1 closed: 1
open: 0.3 closed: 50
open: 1 closed: 1


So from this test it would seem those 2 sliders (on unit #1) are defintely acting up. It's either the broken tracks inside or the contacts not touching (can try bending them down again), as PW slider was doing 'something' when cleaned it may just be the contacts need a bend.

What is puzzling me is the envelope readings, showing 3 'dead sliders' yet the envs were working fine before I dismantled it so I won't know if they are really dead until I assemble it again I guess.

Also some of those readings are way different (esp for closed) Is that just the various types (A/B and that other one I see in the schematics) or does this mean they are not operating in correct tolerance and should be replaced to get the 'proper sound' back?


And some good news (I think) is I have a roland 'made in japan' 100k B slider from a JX-3P that I replaced the slider on with a broken stem. I hope it's the same thing as i'll be able to test it at least. (after replacing stem from 101)


edit > sorry for this mega post but, I've just dismantled the 2 dead sliders and tested the tracks, looks like they are in good condition aside from somewhere the single connected side (the 0k signal) there is a lack of continuity in the track, so this may be a good candidate for the conductive paint trick mentioned above as the other side is testing fine (0 - 100k on the track). Also both spring connects had some greenish oxidisation which i cleaned off and bent the connects (and the pushy brassy colour thing) down a bit for better contact. I reckon if I can just add that paint  so the track has continuity it will be fixed, resolder them in and job done I hope. I can see why it would break there, it's where those two spiny/sharp contacts sit for 30 years in one position being pressed onto it.

I need to get some key contact paint for one of my polysix also so I should be ok to use that stuff on these?

So

[Circuitbenders]

J5 doesn't matter. It was sometimes used in the factory for the initial tuning setup. Some models have it and others don't but if your 101 oi staying in tune then i would leave it how it is.

Lots of 101's have the slight mods and changes you describe in them. I wouldn't worry about it as you can be pretty much certain they were done in the factory as different revisions came out.

From your readings there its pretty plain that the FREQ and Pulse width sliders are buggered. WIth regards to the different readings, it could be that some of the sliders aren't working quite right, but i think its more likely that you just aren't going to get a reliable reading of the faders resistance while its in circuit. For all you know the two ends of the fader are also connected via a different resistance via somewhere else in the circuit and thats whats giving different readings. The only way you'd know for sure is by pulling them out and testing them out of the circuit.
I'm not entirely sure why your ADSR faders should be reading like they aren't working but i can't imagine just opening the thing up would have killed them. Possibly you just weren't getting a decent contact on the solder with your probes.

Its very common on 101 faders to get that break in the resistance track, although its usually not actually in part of the track itself but more often in the thin metal section that joins the carbon track to the solder pin. A tiny amount of this should fix it:
http://www.rapidonline.com/productinfo.aspx?tier1URL=Tools-Fasteners-Production-Equipment&tier2URL=Service-Aids&tier3URL=PCB-Accessories&tier4URL=Conductive-silver-paint&moduleno=180423&kw=87-0781#

I've never tried to clean a brownish 101 to try and make it grey again but i doubt you'll get far. Thats probably decades of being in the sun or having people smoking around it that's changed the colour of the plastic. I could be wrong though. Let me know if you find anything that works.

[Polysixx]

Thanks again. Your advice sounds about right there, and very glad to know about the J5 thing. In which case it is all original inside, which my other certainly wasn't. Shame I had to go in and fix sliders and spoil that

yes I see the metal tab you mention where it joins the track, and of course I got  a perfect reading on the variable track with it dismantled (vs mounted to the PCB). On one of them at least there's a break on the track because I tested with the meter at different sections of the track up and down the length of it and around a slider sized space from the bottom there was no continuity (even though it looked 'ok' not obviously damaged (even with magnifying glass) so I think that paint on stuff should work there, i'll also put some around the metal tab where it joins to make sure there too as I think it's gone in both places.

And thanks for the link, I'll order some asap, I wonder if they do the paint on key repair stuff (which on my polysix contacts is worth a shot before I try and get some of those stick on replacements from the US).