The Kuartlotron - keantoken's simple error-correction superbuffer

I did not find the time to research why my build needs such a high value cancelation resistor but i solved another problem that bugged me since i build the first version :
Sometimes the buffer works and sometimes not !
No matter what i did : silence !
That really drove me nuts. Two days ago again i could not get any sound out of the buffer although it had worked fine the other day.
On the bench it tapped the TKD pot and here we go : the pot has a mechanical deffect on both channels. I bend the pot a little bit and got it working. Maybe that was the problem that Self found. The TKD seems to be quite delicate and may not like strong misuse like i do sometimes when i test worst case.
 
Sounds like oscillation

I build one channel with point to point. in fact this is the secong build point to point and both have very similiar problems.
the circuit is from the keantokean blog page as stated in the beginning of the thread.
There is a constant hump that increases with volume and changes when the input leads and volume control are touched.
Power supply is with 2 LM 317. output V is +/- 10V.
Transistors are BC 550 c and 560c. Different brand . R1 , R2 are both 56 R.
Vbe at Q3 is 61Mv. Vbe at Q4 is 68 Mv. Tp measure 5 v after trimming.
Output offset is 6 mv .

Hearing through the humming, the sound is very good and no obvious distortion.

Any suggestions to solve the humming sound.??
Anyway to troubleshoot without analayser. ??
Do i need a gate stopper at Q1?
could be the Inductor --wrong spec ?? . i salvage this from somewhere and is this value critical ?? the previous version is self wound ---15 turns over a AA battery .


I am building this to power a class D amp with an input impedance of about 10 K. This amp has too much gain and hence i have a voltage divider before the volume control. The remote for the volume is bought from ebay .

please excuse for an untidy build.

cheers

kp93300
 

Attachments

  • IMG_1910.JPG
    IMG_1910.JPG
    163.5 KB · Views: 909
  • IMG_1912.JPG
    IMG_1912.JPG
    128.9 KB · Views: 911
  • IMG_1911.JPG
    IMG_1911.JPG
    159.4 KB · Views: 897
Last edited:
I would not increase the voltage, because self-heating effects are proportional to Vce. All the resistor values would also need changed. The transistors here have exceptional low-Vce behavior, so the input voltage should not be more than necessary.

A resistor and TL431 shunt regulator would be the simplest way to reduce the rails and fully meet the Kuartlotron's needs.
 
I would not increase the voltage, because self-heating effects are proportional to Vce. All the resistor values would also need changed. The transistors here have exceptional low-Vce behavior, so the input voltage should not be more than necessary.

A resistor and TL431 shunt regulator would be the simplest way to reduce the rails and fully meet the Kuartlotron's needs.

Okay, shall look at doing it that way - thanks all.
 
Followed the thread and your website fascinating circuit so simple and low component count great stuff!.
You make mention the max output current is -10mA but the "positive has no definite maximum" not sure what you mean by this.
Another point in the thread you mention this circuit is fast and suitable possibly for oscilloscope inputs any idea of the slew rate?
Going to be building this next!