The Kuartlotron - keantoken's simple error-correction superbuffer

That is what the DC offset trim is for, if you have a stepped attenuator you can turn it, listening for the zipper effect and trim the DC offset to eliminate it. If you have any DC offset remaining after doing this (should be very small), then you need to go through trimming the rest of the circuit.
 
Well mine is up and running. The only odd thing is that my overall volume was greatly reduced. I have zero DC output on one channel and around 5mv on another. That's good without twisting and screws! But as I turn up my volume I can't get nearly as loud as I use to before the last notch, which drives the amp into a full out distortion clipping.

Bel Canto CD1 player 100ohm (2.1v max output)
DACT 10kohm
Kuarltotron Stock
Amp 25kohm (1.1v max input)

Maybe it's related to my aging CD player. I don't know for sure.

On the bright side initial impressions are very good. But even in the hour that it's been playing the sound has changed a fair bit. It started out a little too soft, but the vibrancy is coming back. As you can see I didn't even give it a fancy PSU (12.5v +/-, 9v tranny). I'm going to add a pair of Kmultipliers to it later. And maybe at some point I'll get fancy with my AC filtration techniques but it's pretty good already, even with a generic IEC cord.

I know the switch is unused right now, I might wire it up again, but it's a lot of work since it's only 2 poles and I have to join all the ground wires... And I won't run them untwisted with signals.

20160721_153941.jpg


http://www.diyaudio.com/forums/gallery/showfull.php?photo=14096
 
Last edited:
Important info for Kuartlotron PCBs!

The connection between the 270R and 15pF capacitor must be made off the PCB. Those vias are mistakenly grounded, which generally causes reduced volume and low input impedance. If you have already soldered the boards, you can clip the leads where they go into the board, lift them up and solder them back together above the board.

Sorry for the mistake. I did not discover the error because it did not cause any obvious faults.
 
My voltage is over 10v, so I can't get my 5M trimmer on one board to turn far enough, and on the other board it's very close to to the limit. For every 1 volt over the target 10v I'd consider adding a 500ohm worth of resistance in series to the 5M trimmer just to be sure. In my case I'm going to stick something around 2M in to be sure. But I think 14v or so might be the absolute limit before degrading sound? 12.5v sounds fine to me.

Thanks for figuring out the 270r and 15pf. Oh the fun of DIY :)

*Tip, make sure to not have any volume device connected when you measure DC offset from the INPUT, while the buffers are unconnected to source&amp.
 
Last edited:
That is why 10V rails. With 10V rails the default values will get TP close enough that you can set offset with the trimmer. Otherwise you may not be able to get 0 offset.

Using a 10M resistor for R9 will help at higher voltages, but I had problems with harshness when trying to run it at more than 10V. plus the transistors heat more and thermals start to become problematic.
 
That is why 10V rails. With 10V rails the default values will get TP close enough that you can set offset with the trimmer. Otherwise you may not be able to get 0 offset.

Using a 10M resistor for R9 will help at higher voltages, but I had problems with harshness when trying to run it at more than 10V. plus the transistors heat more and thermals start to become problematic.

I'll let you know if I have any "harhness" but I wouldn't make that claim so far...

In retrospec I should have got a 266J14, but have a 266K18. But it's been so long since I ordered it, I'm not even 100% sure it was for this project. Won't adding some Kmultipliers drop my V a little? (bugger, I'll have to rim again)

Here's the "fix" and my volume level is correct now.

20160722_125119.jpg
 
I'll let you know if I have any "harhness" but I wouldn't make that claim so far...
This does not do anything disproportionate.
Harshness is not on this page.
An epic length list of what it won't do is entirely appropriate for a buffer; and, this circuit is a lovely and helpful buffer. It is rather good at refusing to do anything else, so you won't be getting any harshness here.
 
Last edited:
That looks like English, but I still need an interpretation to understand it... sorry.
I built it wanting a bass boost.
Wasn't entirely disappointed.
It boosted everything's resolution.
Well, everything was slightly clearer.
Why else have a buffer anyway?
Slight problem is that it didn't boost the bass disproportionately.
Oops!
I should have asked for a bass booster.
I didn't get it.
Also,
If you wanted more harshness, you're not going to get that.
This may be a great buffer but it absolutely sucks at signal modification capacity.

Can re-try with the southern version if you're sure. Wouldn't want to try that on the first go, so if that's the thing to do, then just mention it.
 
Um no...

I'm reporting I don't have any harshness because Keantoken noted higher voltage can come across with a harsher sound. I'm only at 12.5v.

And if you have a lack of bass maybe you're looking in the wrong place. It's an error correction buffer. So ya, I doubt it tolerates silly attempts to change the signal.