Cookie Information, please respect my cookie handling while usage this website, or close this session. click here
Skip to end of metadata
Go to start of metadata

schematics can fins on Jürgen Haible FS1a Frequency Shifter


Issue 1.

after finishing the psu and connecting all potis,jacks.

i got in calibration process 1 a fine TrIangle waveform but without 90degrees phasedifferent.

(dualoscilloskop and separate frequency counter)

by usage the OSC BAL trimpot - i was able to get 2 identical triangle waveforms - but no 90degrees from eachother.


calibration process 1 picture  click to enlarge:



Issue - calibrating process 2:  failure - no 90degrees different (at 1khz)


after this i tried

racing the signal along the VCO structure:(U9, pin 7 is listed twice)
U8, pin 7: triangle. (You have that.)----yes was a triangle
U9, pin 7: triangle - yes is a triangle (U9 pin7)
U9, pin 1: square wave - yes  - but its very unsharp

(at u9 a  other new TL072 same result)


so i tried to exchange some resistors described at em forum


I've taken the board out, and have gone over it. No wrong resistors. Bear in mind that this is a copy of Dave Brown's build (and using his BOM), so there are some resistor changes - specifically R2 changed from 51K to 36K, and R51 is changed from 430 to 51K. There is also a 36K resistor added at pin 6 of U1. Would any of these make this effect? (Feel free to chime in, Dave, if you had a similar problem - probably not though)

  • R51 muss 51K sein statt 430ohm   - here its better to use a trimpot 100k and test value above 10K-100K
  • R2 51k zu 36K
  • 36K  U1 pin6


after this i get this:

(by usage of the sin sharp trimpot - i cannot trim the " waveform like square" to a sin wave.. and its not 90degrees different.


Troubleshooting: 04.09.13


issue: same triangle at step 1.

test oscilloscope setting for trigger - maybe a triggersetting issue


at U8 pin 1 - must have a triangle COS

at U8 pin 7 - must have a triangle SIN

both must have 90degree !



after checking my oscilloscope settings i found out, its my fault - the channeltrigger settings was set to "dual", needed is here single 1channel..

later i changed back the resistor R51 (board1)  ( origin 430ohm, )  to 330ohm so i have a same amplitude..

here correct Oscilloscope settings (click to enlarge)

correct result step1: fine

Step 2 measurement -  before R51 changed to 330ohm  i used 51K (electro-music-thread -info, was wrong in my case)


Step2  after changing to 330Ohm is like this:  fine

Issue 2:

if you got this or this with 90degrees and you use a external PSU (MOTM,dotcom header)

AND you have the internal psu parts soldered, remove the internal powersupply parts (its only need to extract both LM317, D1,D2,D3,D4,D5, R1, R6, on board 2)

or you get this on Step 2.. attention to bottom

  • No labels