The Quest for >2000 Vppss

[kevin gilmore]

q12 is the current limit. so if the amp board shorts, at least the power supply will not explode.

what do you want me to replace the pzta part with?  remember 20+ma

Edited October 28, 2025 by kevin gilmore

[simmconn]

I was just pointing out, as how Q12 is connected today, it’s not doing what it is supposed to do. The output should be connected to gate instead.

I remember last time you recommended TTC004B as a replacement for the PZTA42 (different package). We just need something with a flat hfe-Ic curve to, say >100ma at VCE=10V.

[JoaMat]

Seems simmconn reads shematics. Regarding Q12, schematics and gerbers don't match, I think.

[kevin gilmore]

power supply schematic updated. amp board updated.

circlotronpsupdated.pdf

 

kgsshvcarbonv6.1c3cs - CADCAM.ZIP

Edited October 29, 2025 by kevin gilmore

[Michelag]

Wow!

But just for my understanding, this is the limit after which damage to earspeakers can occur, right? 

I mean, so much power is really needed to drive the most demanding e-stats? 

Or is it thought as an exercise of pushing it to the boundaries ?

 

[kevin gilmore]

max voltage swing is 4 x bias.  so 580 x 4.  the amp does loose about 20 volts of voltage swing, so 600 is a nice number. same voltage the koss uses. also the rating on the pnp transistor.  could cut the voltage to 550v for a bit more margin.

 

[psyduck]

Though running it close to max bias will probably damage your hearing.

[kevin gilmore]

surprisingly on something like a 727, the peaks are already clipped at moderate volumes.  so this will let all the wonderful dynamics thru.

and damaging you hearing, definitely.

[15lyl]

Such high voltage is too close to the head, will it affect the brain? say cancer.

[spritzer]

No.  Given the simple physics that the force of an electrostatic field falls with the square of the distance, it quickly vanishes into background noise before even reaching the skin.  An amp like this would never be run at full tilt, not into Stax compatible drivers at least.  Well past 120dB so permanent hearing damage until they simply burn up.  That means the field is even weaker than that in reality.  

Ideally you'd never want to run the headphones above say 1200-1400Vppss (which is already crazy loud) as depending on the air parameters around you, is could destroy the headphone drivers.  Now there are a lot of variables at play here but a rule of thumb says insulation of air is 3kV/mm and we are running at 0.5mm spacing so 1500V.  I go a bit lower as humidity is a huge factor here and the new Stax drivers appear to be anything but stable so I'd err on the side of caution with them.  I've noticed some very alarming behavior from the X9000 and the 007S so yeah... Stax will never side with you if you have driver damage.  

[15lyl]

Thank you for your explanation, Spritzer.

[psyduck]

Merry Christmas 

I made the output board for D&G and tested it with +/-400V and 845 tubes. Seems to be working as there's 44v across the cathode resistor.

I was planning on using +/-1250V but the C3M mosfet blew up on the circlotronps because 1n4007 diodes aren't rated for that voltage.

[MLA]

Test build of Super-Carbon up and running to see what is what: 

PSU angle brackets are recycled from a previous, very tight, build, where those cutouts made perfect sense . 

Two small issues worth mentioning on amp boards. First, TTC004b and PZTA42 have opposite pinouts (ECB vs BCE) so if using TTC004b, transistor front needs to face heat sink. Second, you need to add some current limiting on the CS LEDs if you intend to use them.

Running at +/-550V. 

No listening yet; need to build a more permanent CS supply solution first and also tidy up in general.      

  

[kevin gilmore]

new board in the stax mafia google drive has the orientation of the ttc004b fixed.

edit:  current limiting on CS LED???

 

Edited December 27, 2025 by kevin gilmore

[Omega_ELS]

2 hours ago, MLA said:

Test build of Super-Carbon up and running to see what is what: 

PSU angle brackets are recycled from a previous, very tight, build, where those cutouts made perfect sense . 

Two small issues worth mentioning on amp boards. First, TTC004b and PZTA42 have opposite pinouts (ECB vs BCE) so if using TTC004b, transistor front needs to face heat sink. Second, you need to add some current limiting on the CS LEDs if you intend to use them.

Running at +/-550V. 

No listening yet; need to build a more permanent CS supply solution first and also tidy up in general.      

  

its a new carbon?

 

"kgsshv-carbon board with the 3rd generation sicfets as both current source and gain stages.
+/-400 to +/-600 - probably limit to +/-550" (c)

[MLA]

7 hours ago, kevin gilmore said:

edit:  current limiting on CS LED???

Low precision writing on my part. No LEDs directly in CS circuit itself, but on the boards I have, two LEDs sit across the incoming CS power rail, so they see 15V.

Edited December 28, 2025 by MLA
I realized it wasn't complicated.

[psyduck]

I think there's no resistor to limit the current so it blows up.

1 minute ago, MLA said:

For some reason, the ones I just put in place blow up when I power on. Standard LTL-4213 ones. When powering up the CS directly to 15V, they make one blink and then they are gone. Slow power on with variac makes them literally blow up, as in into actual pieces. I really don't get it. Maybe the converter is a slow starter? 

 

[MLA]

27 minutes ago, Omega_ELS said:

its a new carbon?

 

Maybe more like a power up-Carbon ; same design but with higher PSU voltage and new current sources, using current production sicfets. 

[Shawn]

On 10/29/2025 at 2:41 AM, kevin gilmore said:

power supply schematic updated. amp board updated.

circlotronpsupdated.pdf 29.72 kB · 40 downloads

I carefully cross-checked the circlotron schematic against the PCB layout and noticed a few discrepancies between the two. I’m posting them here so others can help double-check the circuit.

All the locations where the schematic and PCB differ are marked with red lines in the file.

CirclotronV.11.pdf

[kevin gilmore]

 i know how the one error got there.  the other errors may have been there for a very long time.

will go and look at the original circlotron file later.

updated schematic

 

circlotronpsupdated.pdf

[Shawn]

On 12/27/2025 at 11:06 AM, MLA said:

Running at +/-550V. 

Quick question: how did you connect the floating PSU outputs to B+, B−, and GND in your build?

[psyduck]

1 hour ago, Shawn said:

Quick question: how did you connect the floating PSU outputs to B+, B−, and GND in your build?

Take one + from a PSU and - from another, connect them together to form ground. Then connect the other +/- to the B inputs.

[Shawn]

36 minutes ago, psyduck said:

Take one + from a PSU and - from another, connect them together to form ground. Then connect the other +/- to the B inputs.

So, like connect the PSU1 +  and PSU2 -. Then the remaining - and + from both PSU to the B+ and B-, leave the GND unconnected.

Do I understand it correctly?

Edited December 30, 2025 by Shawn

[MLA]

Almost, but you also need to connect to GND on the amp board, so: 

First, connect GND on the amp board to PSU1+ and PSU2 -. This is now your relative ground point. 

For the PSU1 board, since it has + connected to GND, it's negative terminal will sit at -550V relative to GND, so connect that to B-.

For the PSU2 board, since it has - connected to GND, its positive terminal will sit at +550V relative to GND, so connect that to B+.   

 

[Shawn]

Can we replace the Q7 KSC5026 with STN0214 ? Since someone mentioned it can be replaced in GRHV build.

Also, an extra 1N4007 may be good here(photo 2) for further protection i guess.

Try to shrink the size of the PCB and see how far we can go.