Troubleshooting haraldswerk VC LFO

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
unvrsnyrgi
Learning to Wiggle
Posts: 20
Joined: Tue Mar 05, 2019 10:36 am

Troubleshooting haraldswerk VC LFO

Post by unvrsnyrgi » Mon Dec 02, 2019 6:26 am

Hi all!

So I ordered PCBs using the provided gerber files.
https://www.haraldswerk.de/LFO/VC_LFO/VC_LFO.html

I've build one LFO but the problem is that it's not oscillating. There's no triangle wave in IC4 pin 1 but constat negative voltage. I'm running +/-12V system. I've build the module using 1/8W resistors (since I had them already). Nothing hasn't fried so I'm thinking that's not the problem, or is it?

I couldn't source MPF102 so they are substitued to Tayda's 2n5458 (They have the same pinout). I've checked power pins from every IC and they're ok. Also no shorts between GND, -12V, +12V. I changed all the ICs and it didn't help. Resoldered every joint didn't help. I changed BC560C to new ones - nope.

I'm clueless how to continue troubleshooting and all help is very much appreciated. Thank you!

Here's more info about same circuit but the flat version. It's the same but some labels are different:
"C1 and associated components comprise a linear voltage to log current converter. IC1A sums the control voltages. IC1B provides the temp compensation realized with KTY81-110. TR3 adjusts the V/Oct characteristic. Q1 and Q2 forms the log converter with IC1D as constant current source. IC1C scales the control voltage for the linear FM input. The transconductance of IC1OTA1 controls the frequency of the oscillator. IC2C, C1 and associated components comprise an integrator. When current flows into IC1OTA1 output the integrator ramps up, when current flows out of IC1OTA1 the integrator ramps down. When the integrators output goes above the threshold of comparator IC2D its output goes high. The output of IC2D is fed to the non-inverting input of IC1OTA1 OTA through D1, D2, R1, R2 and TR1. TR1 balances the current flowing during the high and low periods of IC2D. With TR1 you can adjust the symmetry of the triangle. While IC2Ds output is high current flows out of IC1OTA1 OTA and the integrator ramps down until the voltage at the input of IC2D goes low enough to overcome the hysteresis provided by R13 and its output goes low. When this happens the comparator starts to ramp up again and thus we have a triangle wave at the output of IC2C. The bias of the comparator IC2D is controlled by the current generated by the linear voltage to log current convertor. This controls the current that flows in and out of IC1OTA1 and thus the frequency of the oscillator.

The sawtooth is created by mixing portions of the original triangle wave and an inverted version of the triangle wave. N-FETs Q1 and Q1 are used as analog switches."

User avatar
SphericalSound
Wiggling with Experience
Posts: 366
Joined: Thu Feb 01, 2018 9:23 am
Location: Spain

Post by SphericalSound » Mon Dec 02, 2019 11:01 pm

That´s a hell of schematic for a LFO...

If IC4 is not producing a triangle it may be for the 100nf cap is you are super lucky or something related to the 13700 or the rest of the circuit.

I would recheck that c3 is 100nf, is not shorted and doing good conection.

Then, I would try to feed some volts at the FM_LIN and recheck if theres triangle in IC4.

Then I would try to feed some volts at the V/OCT and CV1 and recheck if theres triangle in IC4.

Then, post results. Maybe another user can give you more hints

unvrsnyrgi
Learning to Wiggle
Posts: 20
Joined: Tue Mar 05, 2019 10:36 am

Post by unvrsnyrgi » Tue Dec 03, 2019 6:04 am

Thanks for the hints! I've checked C3 and tried to feed CV inputs but no luck.

I'll put this project on the side for a while. It's too frustrating to put more hours into this :) Earlier I've succesfully build Bastl Kastle so I'm thinking of making multiple Kastle LFOs instead of these.

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 315
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Wed Dec 04, 2019 2:14 pm

unvrsnyrgi wrote:Thanks for the hints! I've checked C3 and tried to feed CV inputs but no luck.

I'll put this project on the side for a while. It's too frustrating to put more hours into this :) Earlier I've succesfully build Bastl Kastle so I'm thinking of making multiple Kastle LFOs instead of these.
Remove the connector to SW1. It is possible to shut off the LFO with wrong adjustments of TR1 or TR2. Measure the voltage drop over R42. It should change when you turn the coarse pot. If there is no change your problem is in the CV source. Otherwise it is in the core.
www.haraldswerk.de
Home of Next Generation Formant

unvrsnyrgi
Learning to Wiggle
Posts: 20
Joined: Tue Mar 05, 2019 10:36 am

Post by unvrsnyrgi » Fri Dec 06, 2019 4:51 am

Thank you Harald! I removed the SW1 connector and now I get all the waveforms. :party:

unvrsnyrgi
Learning to Wiggle
Posts: 20
Joined: Tue Mar 05, 2019 10:36 am

Post by unvrsnyrgi » Sun Dec 08, 2019 12:44 pm

I got some motivation to troubleshoot more. So far I have all the waveforms and I've been able to trim the ramp and sine wave properly. There's still problem with the frequency range of the oscillator. Maximum frquency is now about 40Hz and it should be around 700Hz. TR1 and TR2 need to be dividing almost all the voltage to GND or the oscillation will stop. I've checked that resistors in the CV section and next to LM13700 should be correct.

Any ideas what to try next? Thanks!

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 315
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Post by Haralds:Werk » Mon Dec 09, 2019 2:39 am

Nothing wrong here. 40Hz max is perfect for a LFO. The emphasis is on Low. The fault was on my side. I forgot to correct the figures from 700 to 40Hz on my website when I changed the value of the capacitor. I wanted to have a LFO which can go very slow on the low end. Doing so you'll lose on the higher end. If you need to go up to 700Hz you have do decrease the 100n capacitor to 10n or smaller.

I have corrected the figures on my site. Thank you for finding this one.
www.haraldswerk.de
Home of Next Generation Formant

unvrsnyrgi
Learning to Wiggle
Posts: 20
Joined: Tue Mar 05, 2019 10:36 am

Post by unvrsnyrgi » Mon Dec 09, 2019 5:02 am

Thanks! This information made me very happy :tu: :banana:

User avatar
frequenzteiler
Learning to Wiggle
Posts: 33
Joined: Wed Nov 14, 2018 12:46 am
Location: Berlin

Re: Troubleshooting haraldswerk VC LFO

Post by frequenzteiler » Thu Jan 16, 2020 5:41 am

hey.
here everything is working fine except the sine . it looks almost the same as the triangle . i exchanged the 13700 , double checked the values of R18, R20 ,R22, R23 and R19. the sine volume trimmer works fine . but the shape trimmer for the sine is basically (just) shifting the wave in voltage and phase - up or down . i thought the shape trimmer is going to round off the edges of the triangle-- :hmm:

nigel
Veteran Wiggler
Posts: 743
Joined: Wed Jan 16, 2013 6:49 am
Location: Melbourne, Australia

Re: Troubleshooting haraldswerk VC LFO

Post by nigel » Thu Jan 16, 2020 8:33 am

frequenzteiler wrote:
Thu Jan 16, 2020 5:41 am
[...] the sine volume trimmer works fine . but the shape trimmer for the sine is basically (just) shifting the wave in voltage and phase - up or down . i thought the shape trimmer is going to round off the edges of the triangle-- :hmm:
No, the volume trimmer rounds off the edges, when you turn it up far enough. If the top and bottom don't match (one is less rounded than the other), adjusting the shape trimmer will even them out.

User avatar
frequenzteiler
Learning to Wiggle
Posts: 33
Joined: Wed Nov 14, 2018 12:46 am
Location: Berlin

Re: Troubleshooting haraldswerk VC LFO

Post by frequenzteiler » Thu Jan 16, 2020 12:29 pm

NICE . :hail: you saved me !! . ... i was just not turning the sine volume trimmer up enough .
the sine is perfect now .. :yay:
Thank you, nigel .. !!!
I am very thankful . learning so much .

and thank you harald for your work and sharing all of this .

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 315
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Re: Troubleshooting haraldswerk VC LFO

Post by Haralds:Werk » Tue Jan 21, 2020 9:04 am

I am always happy to hear from other builders. Every feedback is welcome. This keeps me going and helps me to improve my designs and my website.

The edges from the triangle are rounded off by pushing the OTA over his limits. So, as said above, turn up the volume.
www.haraldswerk.de
Home of Next Generation Formant

Post Reply

Return to “Music Tech DIY”