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

Set SAM to show
VC-LFO issue
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author VC-LFO issue
ppkstat
Hello.

I am having an issue with my lfo, specifically I can not get it to oscillate below 0.5Hz. The first part of the coarse freq pot is unusable. As it approaches the low frequencies the waveform gets distorted and the bicolor led does not follow a 50% duty cycle. Below you will find some scope photos that illustrate this well.

Saw at 100Hz
http://imgur.com/NxwQf6B

1Hz
http://imgur.com/zEqgnhn

0.5Hz
http://imgur.com/xoByFmD

No oscillation below that, just steady +5V. The led is constantly green.
http://imgur.com/CEsyySy

Moreover, I cannot get rid of the glitch completely. It is very short but still there.
Synthbuilder
In your VC-LFO the timing capacitor is discharging at a different rate to the charging. That is to be expected at very low frequencies but not at the frequencies you are getting. A few things that would be worth checking:

1. The timing cap, C12, may be too big. It should be a 4n7 or 4.7nF.

2. U9 is leaky. Op-amps can go leaky if they are damaged by static electricity. It should be a TL072 or other FET input dual op-amp. Try another TL072 in there first.

3. The LM13700, U8, may be be damaged. Again it may be worth swapping out your LM13700 for a fresh one.

4. The board surface around U4, U9, U8 or C12 may be dirty or damp. If you've water washed your board make sure it is really dry. That part of the circuit is very sensitive to unwanted current flow.

5. It could be U4 itself. But I'd look at the above first.

The narrow glitch when the waveform passes through 0V is probably what I would expect. That's part of the triangle to sawtooth conversion. It may not look nice on a scope but in use it's not normally a problem.

Tony
ppkstat
Hello Tony thank you for your time.

1. It was 4.7nF but I soldered a new one just to make sure. Nothing. It is an acrotronics polyester film capacitor.

2. I replaced it, nothing.

3. I replaced it, nothing.

4. I use rosin-based solder and I clean the flux with an isopropyl alchohol/ acetone mixture (75%-25%). I checked the areas and appear very clean.

5. I replaced it, again nothing.
Synthbuilder
I'm not really sure what to suggest next. Clearly something is causing the timing capacitor to charge and discharge at different rates. And that normally means current is going where it shouldn't. If it isn't the devices I strongly suspect it could be still be a contaminated board surface.

I think I would avoid using acetone to clean PCBs. I use acetone to dissolve certain plastics and it may well dissolve component cases and might - I've not tested it - remove the solder mask or PCB legending. Pure Isopropyl Alcohol is recommended.

Tony
ppkstat
Thanks. Would a high-res picture help?
Synthbuilder
ppkstat wrote:
Would a high-res picture help?


It may do. You can upload it somewhere and then put a link to it in this thread.

Tony
ppkstat
Thanks Tony I will.

Just to let you know I tried some troubleshooting yesterday. With the fine freq pot at min I placed my oscilloscope probe on the leg 2 of R42, varying the coarse freq pot. At maximum it was arourd -12V but as I moved it towards the minimum the voltage stopped changing mid-way at around -7V. This is the point oscillation stops as well. Of course when I touched the specific point the frequency of the oscilator increased, I am guessing due to the sensitivity of the LM13700.

At the summer stage (U2, U3) the response of the pot was normal.

I am planning to desolder R42 and feed some negative voltage through varying resistors to see if I can get it to oscillate at lower frequencies. Let me know what you think.
ppkstat
Pictures:

http://imgur.com/AiCpfSU

http://imgur.com/di82A88
Synthbuilder
Can you try a different TL072 in U9? I mean from a different batch. All the ones you have in that board have a identification 18MDSHY. Can you see if one from another batch behaves differently?

The reason I say this is that all my Texas TL072s come in a different package to that one. Mine have pin 1 marked with a dot on the package and not a indent in between pin 1 and pin 8. For example:



I'm wondering whether you have genuine op-amps. There's a lot of fake chips out there at the moment and poor input leakage currents could be a possibility with these fake ones.

Tony
ppkstat
My god, you're right very frustrating

The TL072s did not come from a reputable source but I ve used many of them in your projects without any problems.

Because I didn't have any from another batch I replaced U9 with an LF412CN. The lfo works perfectly. I should have some legit TL072 on monday I will let you know.

I just can't beliave this very frustrating
Synthbuilder
OK that's good news. The LF412 is a good substitute for the TL072 and will work fine in the VC-LFO.

Often these fakes are either some other op-amp, like the bipolar 4558 for example, or the correct device that failed one test. This could the case here - it is a TL072 but with really bad input bias currents. I guess what could be happening is that the devices fail and instead of getting trashed the wafers sneak out the back door to be packaged and labelled by someone else.

Either way it is a pain that these devices are so common these days.

Tony
Paradigm X
Blimey, fake TL072s. eek!

where did they come from if i may ask ppkstat? reason i ask is i have had a number of failures of tl074s (replacing them with another fixes a problem) from a certain cheap supplier. i was starting to suspect something.

i think ill buy all ics from farnell now and leave the cheap supplies alone.

thanks.

and great support/diagnostics tony!
ppkstat
Well, Chinese eBay sellers, where else?

Farnell would be very expensive, I am going to get mine from banzai and local sources, I am sure they 'll be ok.

I 'll replace the tl072 and tl074s of the rest of my modules. Fortunately they're not many.

Rejection during quality control seems a much more likely scenario than counterfeiting. Who on earth would go to the extent to counterfeit something that cheap? The profit margin seems very small. Then again, I am not an expert in these things.
Paradigm X
ok, thanks, theres a popular thai 'guitar' effects place where ive been buying cheap ICs and alpha sockets.
ppkstat
Paradigm X wrote:
thai 'guitar' effects place


I buy a lot of stuff from there including all my pots, polyester capacitors and several ICs (not TL0 though). I never had any problems except with the metal film resistors.
ppkstat
Got some legit TL072s today, problem solved. They look like the ones in Tony's photo.

Tony thank you very much for your help. It was a tricky one.
Synthbuilder
thumbs up
ppkstat
It seems I was quick to judge. The problem of not oscilating at all at low frequencies was solved. However there are several other problems.

1) I cannot get it to oscilate under 0.1Hz. It is oscilating there, it never stops but it won't go under that. Ie something that seems like a range issue.

2) It is very unstable (drifting)

3) I cannot v/oct calibrate. Not that this is important right now but the v/oct trimmer seems to have very little impact on the actual frequency.

4)Again, the capacitor is charging and discharging in different rates. Its much better than before but still present.

Tony I am sorry for all this. I can take measurements would you like to indicate what to measure?
ppkstat
I made some time measurements to illustrate both the charging and drift issues a bit better. All measurements are with both the frequency pots on minimum. The tune trimmer is adjusted to give around 1 Hz when the frequency pot is halfway.

On startup I am getting 17 sec for the positive portion and 103 sec for the negative.

15 mins after I am getting 31 sec for the positive and 110 sec for the negative.
Synthbuilder
It would be well worth checking what pin 1 of U2 is doing. This voltage should move up and down as the tune pots and trimmer are varied. It should not move when the pots are not being moved. If you can see this voltage varying then the problem lies around the components attached to U2 pins 1, 2 and 3. Check the pots and trimmers are behaving themselves by looking at the voltage on their middle pins - it should be stable.

If all appears well check pin 1 of U3. Again the voltage here should only move when the tune pots and trimmer are adjusted. V/oct should also affect the voltage but only slightly and probably only noticeable if the voltage is not very small - ie above 1V or below -1V.

If all is well then measure pin 9 of the THAT 340. This should again move in tandem with the pin 1 output of U3 - but will be very much smaller, around 1/40th of the voltage from U3. If this is not stable I would guess the temp co resistor is faulty. Replace with a new device or even an ordinary 1K resistor just to see.

If this is stable then the THAT 340 could be bad but U3 could be faulty, as could C9.

Tony
ppkstat
This is going to be hard.

Quote:
It would be well worth checking what pin 1 of U2 is doing. This voltage should move up and down as the tune pots and trimmer are varied. It should not move when the pots are not being moved. If you can see this voltage varying then the problem lies around the components attached to U2 pins 1, 2 and 3. Check the pots and trimmers are behaving themselves by looking at the voltage on their middle pins - it should be stable.


The voltage does move up and down when the pots are varied and only then. The voltage on the pots and trimmer wiper is steady as well.
On pin 1 U2 I am getting:

11.9V when both pots are on min
-3.6V when pots are on max

On pin 1 U3:
-13.3V on min
4.24V on max

On pin 9 U7 (THAT340)
245mV on min
690mV on max

Quote:
V/oct should also affect the voltage but only slightly and probably only noticeable if the voltage is not very small - ie above 1V or below -1V.


Yes, this does happen.

I also replaced the THAT340, U3 and C9. Nothing. The THAT array was the only ic I haven't tried to replace because I didn't have any.

If I blow some air to the transistor array the frequency of the oscillator increases greatly. This is normal to an extent I guess but it does seem to be very sensitive.

Next week I am bringing to a friend who is a technician and he is much more experienced than me and has a better scope.

Tony if nothing works I will mail you to send the module to you to have a look at it. However I would like that to be a last resort so if you have any more ideas please let me know.

Thanks a lot for your time, I really appreciate it.
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group