fixing weird resonance behaviour?

wackelpeter · Posted: Wed Mar 05, 2014 1:21 pm Post subject: fixing weird resonance behaviour?

first of all hello to all and thanks to all those who left their schematics public so that someone with very limited knowledge in electronics may also can try to build one of these modules.

Well, i'm not working on a modular System at all i just put a few modules in a old metal briefcase to have some more effects or modifying sources for my analog gear...

so far i have built a +/-15 V power supply... regulating that down to 12/9/5 V +/- too
the VC LFO from Fonitronik, the Wasp clone, Steiner Parker filter, simple AR Env. VCA, coloured White noise, Bandpass Filter from Ken Stone, two clock dividers, etc.

at least i have finished the korg 770ish style VCF and the Moog Ladder filter from the 1900 series but both don't work to my satisfaction when i increase the resonance... for the first time there's not much audible Variation but then they are really fast self oscillating in high frequency range... the self oscillation starts a bit abruptly for my taste and in the range before self oscillation there's not really much of an audible harshness or more edgy behaviour of the filter to hear...
makes no difference if i use logarhythmic or linear pot's
the steiner parker VCf worked in a similar behaviour first and then i read that this can be fixed to some extend by using 2 diodes parallel and in opposite direction to each other...
is the part with that 6 diodes something similar? should i maybe increase or reduce the amount of them to get an better resonance tracking? or changing the cap size before the Output where one wiper of the resonance pots is?
btw. i would like to ask if the 770 VCF is supposed to be driven at 15 or 1
2 Volts... had run it at 15 Volts...

any help would be nice and when i'm ready with these i still have to fix an error in my EMS Diode VCF where always one of the Transistors blew away...

first had Problems with these 2 aforementioned filters too... on the Transistor ladder i forgot to cut one stripe between the base of two Transistors and at the 770 board there was a bad solderpoint at the Feedback resistor of the opamp... well these failures fixed i'm only unhappy with the resonance...

thanks in advance for your help.

p.s. having again a look at the schematics of the korg 770 filter my next assumptions would be changing the 2K2 resistor betwenn gnd an Center wiper of resonance pot or the 100k resistor in the Feedback Loop of the op amp, but this would rather change the gain of the op amp as getting a smoother Transition before self oscillation i guess...

wackelpeter · Posted: Thu Mar 06, 2014 3:25 pm Post subject:

okay korg 770ish style filter resonance fixed just reduced that one resistor going from resonance pot center wiper to gound from 2,2k to 1,5k now that's nearly starting with a lower pitched hum and then finally reaching full oscillation at high frequency...

also equaled output levels of all 3 filter modes as bandpass was much louder as the other 2 by increasing their Input resistor to ground from 47 ohms to 270 ohms... still tracks well and levels are fine when switching between modes...

hmmm... looking at the transistor ladder filter schematics (VCF2 - 3522) i don't really know where to start with... just changing value of R18 (10k) doesn't reflect the behaviour only delays the Point where it starts when turning resonance pot on... maybe experimenting with R13 to ground (2,2k) or any other useful advice?

wackelpeter · Posted: Mon Mar 31, 2014 10:26 am Post subject:

well, so far the resonance behaviour is almost fixed...

Now i'm a bit unhappy with the Output volume of the Transistor ladder filter (VCF2 - 3522) from fonik's/Matthias' site... it's only half as loud as the 3 other filters i've built so far (steiner, Wasp, that Korg 770ish Thing) pushed up the volume a bit with reducing the Input resistors for signal input to 47K...
but don't want to reduce them too much...
next tried to increase the feedback resistor (470K originally) in the Output stage... just a small increasing of it turns into selfoscillation of the filter with resonance pot fully down... next tried some different cap values paralleling the resistor...
with more than 1nf or so it worked and selfoscillation only occured with resonance turned on...
but that sounds not as good as before...
tried further to put instead of caps some pairs of parallel reversed diodes... best result i got with 2 diodes (4 and 6 sounded awful) but again it didn't left me that satisfied as with the original design...

does anyone have any hints how to tweak the output volume else without affecting the overall filter behaviour much?

thank you in advance helpful advice...

cheers bastian

fonik · Posted: Fri Apr 25, 2014 3:47 pm Post subject:

raise R16/R17? you may have to raise R18 as well to get the resonance in bounds...

wackelpeter · Posted: Mon Apr 28, 2014 9:12 am Post subject:

Hi Matthias,

thanks for your reply... had already fiddled with the mentioned resistors 16/17 but just played with just one of them... totally ignored that it could be Logical that both have the same value and could correspond with each other... Smile

have know because i'm lacking 680k used 820k and the Output is as expected nearly the double... luckily we have build it with the level pot...

well that worked and as you suggested the fixed resonance resistor should be raised too... so far i'm not sure if i should use 22, 20 or 18K as when too much resonance occurs the filter sounded not as good what's a matter of taste... think i'll go for 22k with just slight ringing with full resonance... sounds best to me...

Again thank you and be prepared for further noob questions... Razz

fonik · Posted: Mon Apr 28, 2014 10:02 am Post subject:

wackelpeter · Posted: Mon Apr 28, 2014 10:17 am Post subject:

have already downloaded the "op amps for everyone" pdf by TI...
pretty much sites to read (464pg) and i'm afraid to print them all...

will have a look at those golden rules too... hopefully just a few sites less to read... Very Happy

anyway thanks again and have a nice week.

cheers bastian

fonik · Posted: Mon Apr 28, 2014 12:43 pm Post subject:

much less:

Two "golden rules" of ideal OpAmp:
1.The output attempts to do whatever is necessary to make the voltage difference between the inputs zero.
(This rule, which applies only to closed-loop amplifier circuits, means that the feedback sent from the output to the input causes the two input voltages to become the same.)
2.The input draws no current.
(This rule means that the input terminals look at the voltage placed across them but don’t allow any current to flow into the op amp.)
(from dummies.com)

and especially the 1st one explains a lot. just apply it to the inverting and the non-inverting set-up.