KOD NG-01 Compressor (Koda)

himister · Post by **himister** » 05 Oct 2009, 13:00

This one is the old Polish production pedal from the 80's.
Disassembled some time ago but functional so I made gut shots that could be useful for redrawing the pcb and I had hope traced schematic. These production pedals had a "funny" fet switching and status led was in inverse function when it's engaged and it hums on any adapter. Very noisy pedals, I've seen them a lot in the 90's and kinda miss them 'cause they were cheap.

My idea was to change old passive components with less noisy ones and mod it to true bypass since it used BF245A for a fet switching. 4 caps are wihout markings but I intend to desolder them and check the values. It may be interesting to trace it and find on which design this one was based.

Guts:

: kod_ng-01_001.jpg (47.7 KiB) Viewed 3290 times

himister · Post by **himister** » 15 Oct 2009, 14:49

Layout done. Need some help with schematic tracing if anyone would like to help.
Flip-flop switching board needs to be finished but I would go for true bypass and rid of the fets.

pedalgrinder · Post by **pedalgrinder** » 18 Sep 2011, 06:23

Hi everyone has anyone built this and found out what its like seems like an interesting unit. Has a schematic been done for this???? I have half drawn one but my penciled out sketch looks like a bowl of spaghetti. Cheers

jonasx26 · Post by **jonasx26** » 18 Sep 2011, 13:30

Just threw this together from the posted layouts. It's likely to contain errors.
Seems like an interesting design, maybe a bit similar to Rockman comps.

Maybe you could check this out and add the pot wiring, himister?
Is there a second circuit board in the unit?

/ Jonas

jonasx26 · Post by **jonasx26** » 18 Sep 2011, 15:52

First schem just didn't look right. I retraced it from the photos and made some corrections.
There is some errors in your layout, himister. Either that or I've traced incorrectly.

Corrected schematic attached. Still not sure I got everything right though.

jonasx26 · Post by **jonasx26** » 18 Sep 2011, 20:39

Updated. Suggestions welcome.

mictester · Post by **mictester** » 18 Sep 2011, 22:22

himister wrote: My idea was to change old passive components with less noisy ones and mod it to true bypass since it used BF245A for a fet switching.

Changing any passive components won't make the slightest change to the noise inherent in this circuit. The FET switching is MUCH quieter than the mechanical option and isn't really worth messing with.

The circuit is simply an input buffer followed by an FET shunt to ground that's being used as a "voltage variable resistor". This is followed by an amplifier that introduces some gain to the audio path to feed to the output socket (through the switching and a further buffer). The output of the audio path amplifier also feed a rectifier circuit to derive a voltage proportional to the level through the audio amplifier op-amp stage. Some basic passive filtering is used to clean up the DC control voltage for the "voltage variable resistor" FET.

There are several weaknesses in this design - the shunt FET will introduce some distortion, the amplifier is always running flat out so will also introduce some noise and distortion, and the control loop filtering is pretty crude. The best part of the circuit is (probably) the FET switching!

pedalgrinder · Post by **pedalgrinder** » 19 Sep 2011, 03:10

Wow fantastic work. You guys are absolute legends. That shits on my scrible i had on paper but i wasn't wrong you just have to be able to decipher it. which i think i was the only one who could understand my own scribble. But thats great

pedalgrinder · Post by **pedalgrinder** » 19 Sep 2011, 03:18

Hi I also had this that i done crudely with paint to get a design from the original to etch. I was hoping to find a easy way of doing this from photos from pcb's but this was the only way i could figure on how to do it. If anyone has some better suggestions please let me know.

pedalgrinder · Post by **pedalgrinder** » 19 Sep 2011, 04:03

I was having a good think about this. The rockman of what i could find uses half wave dc control voltage this one uses full wave dc control voltage but also incoporates an attack function now how effective that is in the particular position its at iam not sure. But the famous 1176 compressor does a similar type of attenuation. It just uses more averaging of the signal to achieve the dc control voltage so your attack and release is manipulated in a more controlled way. Iam a big fan of the fet for compression but the main point to what i can see is how the control voltage is derived for the fet be it either n channel or p channel. It's finding also weather the fet is used as a gain control or attenuation also. i feel gain control in some ways can be more transparent where straight attenuation can tend to distort the signal more. I think from a bass players perspective ( don't hold that against me) the gain control will give an overall more cleaner and transparent tone. thats my thoughts but iam still learning also

jonasx26 · Post by **jonasx26** » 19 Sep 2011, 11:38

I was way off base with the rockman-similarities. I had drawn the schematic wrong and there seemed to be extra, global feedbackpaths. But I was wrong, this is a plain standard fet-shunt-comp.

himister · Post by **himister** » 19 Sep 2011, 12:33

Greetings fellas...sorry I haven't reply...didn't watch this thread and for some time no one was interested in this one.
As fr the layout, I just redrew the original layout but didn't done the FET switch circuit board and since this pedal was disassembled it would be messy to find out all connections. I still have the main board and I wanted to make a true bypass but didn't have the time to play with it.
As mictester said it looks like a poor design. From memory this pedal was really noisy but I thought it's cause the very noisy BF245 FET's and some old 4558 IC.
Also I thought that poor quality of the resistors and capacitors would add to the noise so 2nd of my ideas was to change all the components with better quality ones but in the end I give up...loads of work to get small changes in sound.
Still, if you guys are interested I'd be glad to help...for starters, bypass fets are on the main board, and signal switching is made through the diodes in the middle. Lemme check it first and give you the rest of info I have...don't hesitate to ask what ever you need!

Cheers!

pedalgrinder · Post by **pedalgrinder** » 20 Sep 2011, 03:17

Hi joanasx26 and everyone else,

I found some errors in draft 3 of your schematic and sense i didn't know how to modify your schematic i printed it liquid paper and pencil the scanned it back in so here is the mistakes i found please correct me if iam wrong. the 200k is part of the bypass and the emitter of the first transistor actually travels through the attack into the 100n. I personally don't see how this could act as a attack feature other than changing the impedance the op amp sees. But thats what they call it. Also the two doides coming out of the op amp go the pnp emitter. This would be and interesting place to maybe put a release control with the ref2 voltage just a thought maybe a 1 meg pot. other than that the rest of the schematic seems fine you've done a great job. I personally would use a 2n5457 and some jfet op amps such as TL072 or a LF353 much cleaner and warmer. The other thing is if its hissing could be a thought of putting a low pass filter on it but i would have to give that some more thought. The thing i like about this one is that it had a bit of modding pottential to play with. But i still prefer the gain to adjusted rather than attenuated as i believe this gives more distortion. Any ways here is the schematic with the changes i hope you don't mind me posting it changed. cheers

himister · Post by **himister** » 21 Sep 2011, 00:36

Here you go...more photos taken earlier. Hope it helps a bit resolve tracing.
Also I include a Sprint Layout pcb file which I drew so if anyone uses this software can edit and make proper changes!
Cheers

KODCompLay.rar: UnRar it to C:\ otherwise the bitmap photo file needs to be inserted again in the program!; (140.4 KiB) Downloaded 96 times

pedalgrinder · Post by **pedalgrinder** » 21 Sep 2011, 03:55

Hi himister,
Thanks so much for the photos there great. Any chance you could read what the potentiometer or (pot) values are as i think most of the circuit has been traced but the pot values are still missing. The pictures will also help verify everything to i think from memory there was a few resistor values that also needed to be addressed but over all i really do think with some better quality parts this unit will sound half reasonable. I have also always been a fan of fet compressor's. Cheers

jonasx26 · Post by **jonasx26** » 21 Sep 2011, 17:20

I found some errors in draft 3 of your schematic and sense i didn't know how to modify your schematic i printed it liquid paper and pencil the scanned it back

Dedication!

Actually, I lost all interest in this effect when it turned out to be a somewhat poorly designed, standard topology circuit.
So I won't be looking into this any more.
I'll update the schematic with your corrections though!

I personally don't see how this could act as a attack feature other than changing the impedance the op amp sees.

The "attack"-pot is probably wired as a variable series resistance, forming a voltage divider with the fet. My guess is this will actually control the ratio and sensitivity, not attack..
It will also control the frequency respone as it forms a high/band-pass filter with C3, Q3, C10, R19 and C11.

I personally would use a 2n5457 and some jfet op amps such as TL072 or a LF353 much cleaner and warmer.

Altering resistor/capacitor VALUES would make a way more notable difference than replacing the opamps. The opamps/fets aren't to blame.
There is a lot of attenuation going on before IC1 and then a lot of gain in IC1. Crappy signal/noise-ratio..
(which is difficult to avoid with these kinds of circuits, I know)

Will post updated schematic shortly.
/ Jonas

himister · Post by **himister** » 22 Sep 2011, 02:36

pedalgrinder wrote:Hi himister,
Thanks so much for the photos there great. Any chance you could read what the potentiometer or (pot) values are as i think most of the circuit has been traced but the pot values are still missing. The pictures will also help verify everything to i think from memory there was a few resistor values that also needed to be addressed but over all i really do think with some better quality parts this unit will sound half reasonable. I have also always been a fan of fet compressor's. Cheers

No problem mate

...for the pot values I don't know how much I can help you there..pedal...disassembled...long time ago...

I just remember that there was a 100kLIN and a 220KLOG pots which of these was for what, sustain or attack I really don't know...
for the level pot 10k might be used, but don't get me for sure! That's all I can say! Sorry!
Hope the rest of it would be helpful!
Cheers

jonasx26 · Post by **jonasx26** » 01 Oct 2011, 01:27

Updated schematic. I rechecked the photos and the suggested corrections does not seem to be right.
So this will be my final schematic. Feel free to continue adding on to it. I can post the eagle-files if anyone is interested.

/ Jonas

pedalgrinder · Post by **pedalgrinder** » 05 Oct 2011, 23:25

I disagree if you follow the emitter from that first transistor on where that 200k resisitor goes and the 3n3 you will see that that schematic is wrong i traced it closely myself. the strange part as i said earlier is the attack seems to act as just an attenuator not even going to ground but just essentially putting more resistance in the signal path. As far as my eyes can see that is incorrect. But hey great schematic better than i can do.