The Kuartlotron - keantoken's simple error-correction superbuffer

I won't predict what people will hear but there is no technical reason for the transistors to be matched except for DC offset. The Kuartlotron doesn't work against the transistors' nonlinear nature so the effect of a mismatch is minimal. I don't know if it makes an audible difference (probably insignificant if so) but it won't break the circuit.
 
Last edited:

RCruz

Member
Paid Member
2006-09-08 5:30 pm
BIEL
A breath of fresh air came into my lab....

Two days ago I received two Kuartlotron boards (new version) from Kean. I started populating those asap and it was a breeze to do so... all parts are perfectly identified and the pads have good sizes. The transistors holes are far apart so soldering is quite easy to do.

I did not match the transistors better than 10% and got output offset lower than 10mv.

Second day (today) I prepared a case, connected a pot and two 10v V12r (modified) regs from Salas.

The scope showed wonderfull squares up to 500MHz and absolutely no phase shift whatever voltage or freq I dialed in.

Then I connected the rig to my lab system and guess what.... a true breath of fresh air... The Kuartlotron is an exquisite piece of elegant engineering and it delivers.

Effortless dynamics, punchy tuneful bass blending to the wonderful flowing mids and topped by some of the best highs I ever experienced.

After I built my p2p DCB1 with modified V12R I believed there could not be anything better but now I have a serious contender.

Must thank Kean for sharing this special buffer and for letting me beta test it's new board.
 

RCruz

Member
Paid Member
2006-09-08 5:30 pm
BIEL
Some pics
 

Attachments

  • IMG_0959.jpg
    IMG_0959.jpg
    1,004.3 KB · Views: 751
Kuartlotron R2 PCB info:

The Kuartlotron R2 is the second PCB after learning from the first one. Improvements were made to solderability, component selection, pads, etc. Simple shunt regulators were added to the board so it could accommodate a variety of input voltages. Side-screw trimmers are now used so that the trimmers can still be accessed when the boards are mounted in a stack.

Short-form instructions:

1: The square pads in the upper left need to be shorted with a solder bridge to use the PCB inductor, unless you want to use a discrete 2uH inductor. No one has felt the need so far. The +10V and -10V vias are not to be used to power other circuits.
2: The resistors marked with an asterisk on the PCB need to be determined based on your input voltage. Included in the Mouser cart are resistors for a range of input voltages:

12V - 68R
13V - 120R
14V-15V - 150R
15V-16V - 196R 1/2W
17V-19V - 332R 1/2W
You can delete from the cart the resistors you don't need.​
3: Preferably before installing the trimmers, use a DMM and set them to the values given on the PCB ("set 56R"). You can just omit the trimmers and use resistors of those values between the outer trimmer pads if you don't plan on trimming.
4: You can start trimming if you so choose.

BOMs:
Standard pair:
http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=0fb260334f
Topscrew trimmer pair:
http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=1fb10c8580
Single board:
https://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=736d60e7c9
Single topscrew trimmer:
https://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=7f09cb94f7


Details and FAQ:

1: Each board is one channel, so you need a pair for stereo.

2: Some quick notes on construction:

a: Component values are shown on the PCB.
b: It accepts screw terminal inputs and outputs and M3 or 6-32 standoffs.​


3: Customizations:

a: Input resistance is normally 100k. You can use 10k if you want.
b: Output impedance is set to 47R for driving cables to reduce reflections. If you want lower output impedance you can use 4.7R in place of the 47R resistor.
c: Those who want to use superregulators can use the +10V and -10V vias on the PCB as sense points.​


Trimming procedure:

1: Each trimmer should be set to the value indicated on the PCB. This can be done before soldering them.
2: Short the input and adjust R5 so that TP reads half of V+ (5V).
3: Remove short from the input and then adjust the DC trimmer so that input voltage reads 0.
Output offset should now be within 5mV. This generally results from small variances in components and won't reduce performance.
4: Refer to the Kuartlotron webpage for distortion trimming if you have distortion measuring equipment and the inclination.
The Kuartlotron
 
Last edited:

lgabco

Member
2006-11-07 10:27 pm
A breath of fresh air came into my lab....

Two days ago I received two Kuartlotron boards (new version) from Kean. I started populating those asap and it was a breeze to do so... all parts are perfectly identified and the pads have good sizes. The transistors holes are far apart so soldering is quite easy to do.

I did not match the transistors better than 10% and got output offset lower than 10mv.

Second day (today) I prepared a case, connected a pot and two 10v V12r (modified) regs from Salas.

The scope showed wonderfull squares up to 500MHz and absolutely no phase shift whatever voltage or freq I dialed in.

Then I connected the rig to my lab system and guess what.... a true breath of fresh air... The Kuartlotron is an exquisite piece of elegant engineering and it delivers.

Effortless dynamics, punchy tuneful bass blending to the wonderful flowing mids and topped by some of the best highs I ever experienced.

After I built my p2p DCB1 with modified V12R I believed there could not be anything better but now I have a serious contender.

Must thank Kean for sharing this special buffer and for letting me beta test it's new board.

Thanks Ricardo for your comments. I have very similar experience. I was almost sure that nothing can beat Erno Borbely EB325 preamp. However Kuartlotron changed my opinion in two ways.
The first - There can be even better pre than EB325, especially in low level details
The second - every preamp with bipolar input stage I have ever built/heard did not sound to my taste. The hights were almost allways harsh and some kind of wildeness was present. This is absolutely not issue with Kuartlotron.
There is no sign of anything I am used to with bipolar input stages.

Note to my construction:I am feeding Kuartlotron with +/-12V and I have no capacitor at output since there is already one at the input of my power amp. Another mod on my side is absence of trimmer R9 for minimizing the distortion. I do not have distortion meter so I just tried to set it with my ears. I finished with 1Meg resistor on this position.

Congratulations to Kean one more time.
Ladislav
 
R9 is the DC trim. You may be sending ~1uA of DC through your cart (if I assume correctly that is what you're doing). Either that or your BC560 has ~250Hfe. Or possibly it is running at double it's intended current (could happen if you don't use Fairchild transistors). To adjust R9 all you need is a DMM.