electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
go to the radio page Live at electro-music.com radio 1 Please visit the chat
poster
 Forum index » DIY Hardware and Software » MusicFromOuterSpace.com designs by Ray Wilson
What else? Soundlab Troubleshooting
Post new topic   Reply to topic Moderators: dnny, Uncle Krunkus, v-un-v
Page 1 of 1 [15 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Author Message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sat Oct 27, 2012 8:05 pm    Post subject: What else? Soundlab Troubleshooting Reply with quote  Mark this post and the followings unread

I have two problems I'm struggling with on a soundlab (original, not plus or ultimate) build.

1) I can calibrate Osc 1 for 1V/Oct pretty easily, following Ray's instructions. I'm actually quite happy to have his advice about "if it's flat, make it flatter and try again, if it's sharp, make it sharper and try again" that worked excellently compared to my previous attempts to calibrate VCOs...) BUT Osc 2 will not calibrate for anything. The trimmer was all the way to max resistance (103 ohms on this one, actually), and still flat, so I tried adding more resistance by replacing the 470 with a 560. Not much improvement at all. I'm honestly not sure whether I should be going up or down with resistance on this leg either.

So the questions for this one:

Should I be increasing or decreasing the 475R for O2?

2) I get a ton of bleed from Osc 2 and the LFO into the white noise. This is regardless of the level output of the LFO or the mixer setting for the Osc 2, and it *sounds* like most of the O2 bleed is from the PWM section, or at least that part has effect as well--changing the pulse width is audible in the noise section. I've checked at pin 8 in the middle of that subcircuit as well, same effect.

With O2 being on a different, not directly adjacent chip, I'm not sure what's the likely cause here. Any suggestions?

Thanks....
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Oct 28, 2012 12:21 pm    Post subject: Reply with quote  Mark this post and the followings unread

Some more analysis and reading and re-reading stuff I should have read the first time....the part of Osc 2 that's giving me trouble is the expo converter. And with the help of electronotes (I *really* need to spring for that complete set sometime soon, but I am thankful for what's available online) I've realized that the issue is probably that my op amp is not putting out a sufficient voltage. It didn't get built with a tempco, and the 2K resistor is a standard 10% carbon resistor. I'm guessing that the resistance there is throwing off the scaling from 1V/Oct to ~18mV (in this case 20mV if perfect components were used)/Oct, and if that's too low, I'll never get it to adjust to the value my transistor needs.

I'm going to put the 475 back down to a 475 or 470, and I'm going to try increasing the 2K resistor, and cross my fingers.
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Oct 28, 2012 6:45 pm    Post subject: Reply with quote  Mark this post and the followings unread

Progress, but ARG!

Put in a 2.2K 1% resistor. Returned what I increased to 510 back to 470.

I can now calibrate per Ray's method....

100Hz, 207Hz (?), 400Hz. Yay!

Then, 689Hz. WHAT? Mad Then 1.02KHz. ????

Do I need to replace a transistor? That pattern is just insane. I suppose I should measure my base voltages to be sure that it's not coming in wrong....but that's crazy.
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Oct 28, 2012 8:37 pm    Post subject: Reply with quote  Mark this post and the followings unread

Not the transistors. Replaced the two (presumed randomly selected) with a pair hand picked to have Vbe within 1.45mV (tested the previous pair and they were over 3mv) but I get the exact same result.

So now I'm stumped as to the next most likely cause for trouble. The MPF-102? Some issue with the bit of the circuit meant to sync the two oscillators?

I really wish the chips were socketed, I'd swap IC5 and IC6 and see if the problem followed the op amps, but I don't really want to desolder all these just as a test. (this is not originally my build, hence non socketed).

Edit: oh yeah, with the previous two transistors in place I measured base voltages, and they were a very consistent 20.7mV per step in voltage. So the first stage of the expo converter is just fine. With the matched transistors I'm fairly confident in the second stage.
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Thu Nov 01, 2012 11:22 pm    Post subject: Reply with quote  Mark this post and the followings unread

Leaving aside the O2 issues (haven't gotten better even after recalibrating my calibration voltage source), I'm now focused on the Modulation behavior with O1.

I noticed that when DC coupled, the modulation would have an effect on O1's frequency even with no input. Looking at it in more detail, i now see why. I'm using the banana modulation setup from the Optional Mods section, and basically when the DC coupling switch is closed, I'm injecting a -9V through the pot and a 100k resistor into the main voltage control point, "AA" on the schematic. Changing the pot changes how much of that voltage can effect the frequency. If I flip it to AC coupled, then this basically DC voltage is cut out of the circuit and there's no effect from it when there's no input at a frequency that would get through the cap. So the pot has no effect. (well, I didn't notice one, but later simulation has shown there ought to be a spike).

But then the O1 frequency is incredibly high.


After a while more analyzing things, I realized that part of the problem is that the diagram in the mods section has the mod amount pot low end connected to -9V rather than to ground. Using ground there helps--the mod amount no longer has any effect unless there is an actual modulation signal input, regardless of the AC/DC switch. But now the Oscillator 1 frequency is "too high" all the time, and I only get about half of its range as anything useful, and the lowest frequency I get is somewhere well north of 400Hz.

Interestingly, O2 has no such problems, it has useful frequencies for its entire range. I've verified that all the components that are in the oscillator loop (caps and resistors anyway) have the same values for both oscillators. I don't see any obvious solder problems.

Not sure where to look, though I'm suspicious of the MPF-102s because of the weirdness in O2's voltage response.

Anyone have any suggestions there?
Back to top
View user's profile Send private message
Uncle Krunkus
Moderator


Joined: Jul 11, 2005
Posts: 4761
Location: Sydney, Australia
Audio files: 52
G2 patch files: 1

PostPosted: Fri Nov 02, 2012 5:47 am    Post subject: Reply with quote  Mark this post and the followings unread

The only things I would suggest trying before replacing parts, is to maybe add a couple more bypas caps, and make sure your batteries are top notch. I have seen many people report odd behaviour from a SoundLab which ended up being dodgy batteries. Wink
_________________
What makes a space ours, is what we put there, and what we do there.
Back to top
View user's profile Send private message Visit poster's website
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Fri Nov 02, 2012 8:05 am    Post subject: Reply with quote  Mark this post and the followings unread

Regarding batteries: would one oscillator seem to be behaving "reasonably" (aside from the range) and the other not if that were the issue?

That's a good point though, I should swap the ones I have out, I've played enough with this thing outside of troubleshooting that I'm definitely below 9V.

Thanks for the help Smile
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Nov 04, 2012 10:40 am    Post subject: Reply with quote  Mark this post and the followings unread

Replaced the batteries, but no luck with O2. I have to think it's either something non obvious (buried in the spaghetti) in the panel wiring, or something else I won't find without completely replacing the components. The panel will get partly rewired when I redo it on the actual aluminum panel--right now it's on an MDF mock up; I don't have to rewire the pots and switches, but I will have to redo the bananas (and planned to). The client is satisfied that I've given it a good shot, and neither of us is interested in spending the amount of time it's likely to take to find why it tracks so strangely.

I increased the integrating cap on O1, because even with nothing but it's base frequency pot it was still ridiculously high frequency--this despite the values appearing the same as O2 and O2 at least having a reasonable frequency range despite not tracking. I couldn't find any extra source of voltage/current going into O1 that would otherwise explain its frequency shift.

I'm still getting bleed into the noise from the LFO and O2. But as far as I can tell not from any other sources. O2 is on IC6 and the LFO and noise generator are on IC7, I put .01uF bypasses on both of them (the base PCB has no bypasses) but no help. I can see the noise on the 9V above the noise transistor, so I tried putting a cap across the 1M pull up there, to no avail (tried .01uF, .1uF, 1uF, none of which had any discernable impact). I've tried to see if the noise is in the ground instead (measuring "local" ground potential back to the battery ground), but not seeing it there.

Anyone have advice for ferreting out this kind of bleed through?

The exact sound is like when the oscillators are low, it suppresses the noise; if I turn the frequencies up on both the LFO and O2 to their maximums, the noise level is more consistent (though then I have a very clear frequency component as well). This is taking the sound from the first op amp after the noise transistor as well, and I even get it (albeit at a very low level, of course) at the emitter of the transistor.
Back to top
View user's profile Send private message
mrmrshoes



Joined: Feb 19, 2011
Posts: 73
Location: Newcastle Upon Tyne
Audio files: 4

PostPosted: Sun Nov 04, 2012 3:53 pm    Post subject: Reply with quote  Mark this post and the followings unread

Alright man


Well I have built a Soundlab mini synth. I etched a double sided PCB from ray's site a couple of years ago now. Populated it and did a prolonged and lame wiring job. fired it up and it worked. Well I would say it was 97% working. All the modules worked fine but like you I had bleed problems with the white noise.

Quote:
2) I get a ton of bleed from Osc 2 and the LFO into the white noise. This is regardless of the level output of the LFO or the mixer setting for the Osc 2



The Noise sounded crazy with the bleed from OSC/LFO, you could hear it modulating. Really weird like. When looking for a solution, people suggested isolating the LFO wires from the rest of the cabling. I've got a feeling that the noise circuitry itself could be part of the problem, Its sensitive to external noise/signals

From rays troubleshooting page.

Quote:
When you finally find the noisiest one cut its collector off and use it. The reason you do this is because the collector acts like an antenna and unless you WANT radio signals in your white noise you probably want to reduce the size of this impromptu antenna to the shortest possible length.
 

I would try using shielded cable for all signals (EG,OSC,LFO).

If this fails I'm not sure what else you could try, I ended up giving my SL away to a good pal of mine and I never managed to fix the problem. I couldn't face rewiring it after the amount of time I spent on it.

Have you read casper electronics soundlab build notes I find them reassuring.

Quote:
This thing is quirky all over the place. The VCAs are noisy, the filter warbles when the resonance is all the way up and the noise output only worked for a day. But overall it’s a real beast and makes some amazing sounds.
 

http://casperelectronics.com/finished-pieces/modular-synth/



Quote:
After a while more analyzing things, I realized that part of the problem is that the diagram in the mods section has the mod amount pot low end connected to -9V rather than to ground. Using ground there helps--the mod amount no longer has any effect unless there is an actual modulation signal input, regardless of the AC/DC switch. But now the Oscillator 1 frequency is "too high" all the time, and I only get about half of its range as anything useful, and the lowest frequency I get is somewhere well north of 400Hz.
 

Is this part of the FM modification? What version are you building (SLMS,SLMS-extras, SLMS-Plus) Also did you implement the circuit improvements? This is strange, you should get full range (freq Pot) when all modualtion sources are switch out of the osc's CV input.

I hope you get these problems ironed out man.

shoes
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Nov 04, 2012 6:27 pm    Post subject: Reply with quote  Mark this post and the followings unread

I'm doing SLMS (the original), and yeah, that's the modulation mod for Oscillator 1.

I did not do the circuit improvements, but I think I may try that, particularly for the Noise circuit.

Also of note, I found a ground loop in the foil diagrams. There's a via on top of the board that closes a loop in the upper left corner of the circuit board. I cut the trace (actually cut it on the far side near the top of the leftmost "flow" area, but it breaks the loop); sadly, that didn't resolve any of these issues.

Unfortunately I don't think I have any handy shielded cable, and since this is also modded to be patched out, that seems like it would be an AWFUL lot of shielded cable. Shocked

In other news, I hate when all my breadboards are tied up, I'm going to have to see if I can find some space to build that noise circuit and choose another transistor candidate. I expect a noisier one could help things as well (higher levels output?)

Edit: also thank you for the Casper link, I will be checking that out as well. What you described about the noise being modulated by the other sections is exactly what I am hearing too.
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Sun Nov 04, 2012 9:03 pm    Post subject: Reply with quote  Mark this post and the followings unread

The improved noise circuit seems to be the ticket, I can't hear the same modulation now Very Happy
Back to top
View user's profile Send private message
mrmrshoes



Joined: Feb 19, 2011
Posts: 73
Location: Newcastle Upon Tyne
Audio files: 4

PostPosted: Mon Nov 05, 2012 6:59 am    Post subject: Reply with quote  Mark this post and the followings unread

awesome man. one down two to go. Wink

I've built that noise circuit and it's defos an improvement.
I've done separate PCB layouts for each SL module (VCO,VCF,VCA,EG,LFO) as a test run to building a full fledged modular. I based each layout on the SLMS plus schematics. All of them work fine and dandy. I much prefer it this way coz its far easier to troubleshoot and wire up.

As for your VCO1 problem could you remove the FM Mod to see if the VCO works standalone? The phantom voltage might have something to do with that.

Also if you look at SLMS + modulation schematic, ray done away with the AC/DC switch and tied the amount pot to ground. (like you did)

http://www.musicfromouterspace.com/analogsynth_new/SOUNDLABMINISYNTH_PLUS/pdf/SLMS-Plus%20Page%204_schem.pdf

Ray also does this for each modulation source amount pot (LFO,EG, ect)

the only other mod for the VCO is a added ramp to saw output level shift. It centers the saw output to ground (oscillates +/-)

from rays site

Quote:
The value of R84 was changed to 220K to change the range of the pulsewidth control slightly.

Inverting Level Shifters (IC8-A and IC8-B) have been added in order to change the ramp oscillator outputs to sawtooth and to bias them to oscillate about ground. In the original the ramp oscillators were taken directly from the integrators in the VCOs. Those signals go from ground to about 5 volts. By making the ramps bipolar (oscillating about ground) the outputs of both the VCA and the VCF are more balanced.

This change requires a major kludge mounted in the PROTO AREA of the PC Board. It is documented as clearly as I know how below this section.

R209 "VCO-2 Mod VCO-1" 100K pot added which allows modulation of VCO-1 frequency by VCO-2's output. R210 is kludged onto the board as shown on the PCB info page.


Another cool addition would be to add rays saw to triangle circuit to the VCO. This would be added after the level shift circuit. I'm thinking of redesigning my SL VCO board to accommodate this (whats one extra OP-Amp in the greater scheme of things). Also using matched trannys and a tempco, i bet it would be a canny VCO.

Happy hunting man

shoes
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Mon Nov 05, 2012 7:24 am    Post subject: Reply with quote  Mark this post and the followings unread

Quote:
As for your VCO1 problem could you remove the FM Mod to see if the VCO works standalone? The phantom voltage might have something to do with that.


I had everything but the coarse adjust pot disconnected physically and still had too high frequency. I just increased the integrating cap and had done. I did it by soldering a second cap onto the legs of the first one, so if when I repanel it I find the problem, I can take it back off easily.


Quote:
Also if you look at SLMS + modulation schematic, ray done away with the AC/DC switch and tied the amount pot to ground. (like you did)


Yah, doing that (even with the ac/dc switch in place) works great.

I'm happy enough with the state of the circuit now, as is the client, so onward to finishing up the panelling and final assembly....[/quote]
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Tue Nov 27, 2012 9:53 pm    Post subject: Reply with quote  Mark this post and the followings unread

I can't remember now, but I thought I was able to get decent tracking out of O1 for more than a couple octaves, but with the mods I list here, that seems to be no longer. I can still track reasonably in the 0 -> 2V range, but outside of that it just falls apart. I'm done sinking time into this right now though, so I think this will chalk up to live and learn...and I'm going to use the SLMS+ circuits for my own stripboard version now that I have big enough boards, so maybe that's a bit better....
Back to top
View user's profile Send private message
elmegil



Joined: Mar 20, 2012
Posts: 2177
Location: Chicago
Audio files: 16

PostPosted: Wed Nov 28, 2012 10:20 pm    Post subject: Reply with quote  Mark this post and the followings unread

Aaand I decided to futz with the expo converters some more and realized that the resistors in the current mirror had been placed incorrectly, so I had two 10k's on O1 and two 1Ms on O2! Oops...

Working on matching some transistors (to improve that part as well) and hoping maybe I have this solved.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic Moderators: dnny, Uncle Krunkus, v-un-v
Page 1 of 1 [15 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
 Forum index » DIY Hardware and Software » MusicFromOuterSpace.com designs by Ray Wilson
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use