goldenreference high voltage power supply (GRHV)

[sorenb]

11 minutes ago, Pars said:

OK, finally getting around to testing the GRHVs I'm building for the Carbon. Using Soren's post on pre-testing, I had some variances.

I marked up the schematic from James above.

1. For the first test, using ~17Vdc thru an ammeter, I'm getting 0.5mA (not 2mA) (points A1-B1)
2. For the second test, I'm getting ~1.3-1.4mA (points A2-B2), which matches Soren's figures.
3. For the third test, I'm getting 10.5-10.9mA (points A3-B3). This may be referring to a Megatron HV900 PSU in Soren's notes below, and I'm not sure I'm looking at the correct points anyhow.

I have a 100K resistor paralleling the two 390Ks. Bringing the positive supply up on a variac until it regulates, it seems to be regulating at 55Vdc, which is about what I calculated. Boards are 1.8 version, single fat sw variants. The voltage appeared to be stable; raising the variac a bit didn't change the output.

Just checking because the first CCS test was considerably lower than what Soren saw. I did unsolder one end of the 100K resistor and rechecked, but no change.

Any thoughts, or just go for full voltage?

"you should get ~2mA" probably a typo, ~1mA is what I usually get, depending on the DN2540 ...0.5mA might be just fine

[Kerry]

For the first test .5 mA is good for points A1 - B1. 

The third test seems off to me. I’m not sure if A3 - B3 makes sense.  The current through R10 is very small. How are you measuring that?
 

 

[Pars]

Yeah, when Soren said to measure the current limiter, I was just injecting a + voltage at point A3 with the negative lead at B3. Ammeter in series with the power supply. I know the 5.1R is the current limiter. Maybe I should inject the voltage at A3-B3 and measure the voltage drop across R11 instead?

I'll probably remove the parallel 100K and test at full voltage. Since it was regulating at 55V, it should be fine? Seems like everything is working correctly.

Edited November 20, 2022 by Pars

[Kerry]

If it was regulating at 55v, you should be fine at full voltage.

Good luck and let us know how it goes. 

[Pars]

OK, powered up and works great. 406Vdc, stable. Bias is 576-580 at bias TP (drifting up a bit). What is the max for pro bias?

[Pars]

Tested the negative GRHV today, same performance. Was regulating at -55Vdc with the 100K resistor in. Removed it and tested at full voltage. -406.5Vdc, stable. Now for the Carbon boards.

[demonkuro]

kgsshvpssicfetsinglenewleftSWS.zip    doesn't seem to exist in Google Drive, who has this gerber? Version 1.8, thin version and the one that contains the CPC1117N, many thanks.

[kevin gilmore]

kgsshvpssicfetsinglenewleftfatsws - cadcam.zip

12/25/2015

definitely on the google drive.

[demonkuro]

13 hours ago, kevin gilmore said:

kgsshvpssicfetsinglenewleftfatsws - cadcam.zip

12/25/2015

definitely on the google drive.

kgsshvpssicfetsinglenewleftfatsws - cadcam.zip 

It is a fat one.

 

The thin one with CPC1117N has only the kgsshvpssicfetsinglenewrightsws3.ZIP.It is only right, no left.

 

[kevin gilmore]

kgsshvpssicfetsinglenewleftsws - CADCAM.ZIP

[demonkuro]

3 hours ago, kevin gilmore said:

kgsshvpssicfetsinglenewleftsws - CADCAM.ZIP 52.06 kB · 6 downloads

Thanks a lot.

[railcannon]

Hey I'm newbie into Stax Amp.

Why does scheamtic configure the bridge rectifier this way?

I think Gerber has no problem. Am I wrong?

[Blueman2]

Hello all!  Long time since I have posted here but I continue to lurk. 

I just had the negative power board go out on my Blue Hawaii. The c2m1000 literally blew its top off.  3 other components also bit the dust in the process.  Parts are ordered and will replace in next couple days.  All other parts appear to test OK.  

Looking at the parts that failed, any ideas on the cause or which part failed first that led to the others failing?  And more importantly, are there any parts you recommend I replace regardless of how they test, given the mode of failure? 

The Amp had been working great over the past few years, though with the Carbon and my trusty old SRX Plus, it sometimes went several months between uses.  

 

 

EDIT: 

I just recalled that I did the build when the C2M1000 was out of stock, so used the SCT2H12NZGC11 from Mouser listed in my Feb 2021 post in this thread (right hand column below).  Maybe that was the issue?  

 

Edited Tuesday at 09:14 PM by Blueman2

[Kerry]

Could there have been any kind of short?

Seems like it blew out the protection stuff along with the pass FET. 

[Blueman2]

Good thought, Kerry.  I checked but the resistance levels on the negative side are about the same as positive side for the amp boards.  PSU itself does not show a short.  I plan to rebuild the PSU, test it on the bench at low input voltage, then ramp it up.  Could just be one of the silicon devices was weak and took the others along with it.  

Edited yesterday at 04:42 AM by Blueman2

[mwl168]

Did the PSU blow up while in play or at power-on?

[Blueman2]

mwl168, great to hear from you!  It blew just as the HV power delay relay clicked on (I have a 45 second delay after LV and filaments turn on before HV turns on).  So probably an inrush current issue if there was no short.  

Edited 12 hours ago by Blueman2

[Blueman2]

The Blue Hawaii lives!  Got my shipment of Cree C2M1000's and replaced the 4 parts that appeared bad.  Turns out the BC557B was not bad.  I had just tested it incorrectly.  So just the c2m1000, the 24V diode, and the 2N3904.  My guess is that the alternate part I used for the C2m1000 was not quite up to the task and the inrush current of the startup finally killed it.  I will probably replace the positive PSU's chip as well since I got 5 of the C2M1000's for future repairs as needed.  

I had not adjusted the balance of the amp since I first built it, and decided to check the voltages.  After 30 minutes warm up, balance and offset are still +/- 1V.  Impressive given the number of hours I have put on that amp over the past 3 years.  

Edited 4 hours ago by Blueman2