KGSSHV Carbon Build Thread

[mwl168]

Got it. Thanks.

I got work to so then. 

[mwl168]

So I did indeed install 150K and 200K chain for the bias resistors of the C2M1000170D when I first built my Carbon.

Today I replaced the 200K with a 150K, I had also dialed the bias back to 18mA and disengaged the servo. Then took a few measurements  after letting the amp warmed up. Now the PZTA42 Vce is around 13.7V, voltage drop of the 20K R is 25.4V and the Vgs of the C2M1000170D is around 3.3V. 

And I did not blow up the amp. 

[Pars]

Nice job Michael. I didn't do any of these changes as I know of to the Carbon I am building for another member. It will be off to another member to do the casework and finish it up.

I hadn't posted pics of the whole thing as it exists, so here is one:

 

[Pars]

I did the mods per simmconn/mwl168 prescribed for 400V supplies, using a pair of 150K resistors replacing the pair of 174K. Offset seems quite stable (no servo), better than many dynamic amps I've built in terms of not bouncing around. I didn't recheck the bias.

Channel 1:

PZTA42 Vce: 13.5/13.6

Drop across 20K: 24.7

C2M Vgs: 3.37/3.25

Channel 2:

PZTA42 Vce: 7.4/7.5

Drop across 20K: 24.7

C2M Vgs: 3.36/3.03

Amp was not warmed up, offset at ~10V per side and dropping. The channel 2 PZTA42 Vce was a little concerning. Let me know what you guys think.

[mwl168]

Did you take another measurement after the amp warmed up?

Channel 2 PZTA42  Vce don’t look right. It’s also odd that it happens to both PZTA42 and yet the voltage drop across the 20K is correct?

[Pars]

Yes, both the drop across the 20K and the Vgs look OK. No, I didn't let the amp warm up, though it was on about 10 minutes. I did go recheck the PZTA since that board is flipped 180 degrees from the other to make sure it was C and E (and not C and B ) I was measuring. The reading was correct. I'm normally pretty careful when building stuff, and don't rush things, but I'm not getting any younger

I guess I'll look the board over making sure no misplaced resistor(s), no solder bridges, questionable solder joints, etc.

[simmconn]

If the voltage measurements are correct, chances are the long tail resistor plus the offset pot on channel 2 is dropping too much voltage (it should be ~9V). It could be due to incorrect component values, or the idle current from the CCS is adjusted too high.

[Pars]

The long tail resistor is the pair of 175K I just replaced, I presume? I'll go and recheck the CCS currents as well.

[simmconn]

It’s the offset pot and the resistor in series, the latter being 180 ohm or 120 ohm.

[Pars]

OK, thanks. I'll check that for ~9V.

 

[Pars]

Checked the voltage across the zener diode (in parallel with the offset pot and tail resistor, which is 121R). Got ~6.8V on both boards. The CCS across the 50R was still at ~17.5mA (~0.875V). PZTA42 Vce on the good board was still at 13.2V or so, the bad board at 6.8V. I ohmed out the pair of 150K resistors. On both boards they behaved similarly, when ohming from the top of the string to ground both acted as a cap charging (which should have been 300K to ground).

I'll keep looking, but any more suggestions welcome. I uploaded pics of top and bottom of the board.

[simmconn]

On 3/4/2023 at 1:14 PM, simmconn said:

Vce(PZTA42)+V(offset res)+Vgs(SiC)=Vg(SiC)-V(B-) //10+8.6+3.2=21.8

The equation can be made more accurate by including the voltage drop on the 100 Ohm resistor on the emitter of the PZTA42s:

Vce(PZTA42) + 0.1*I(CCS) + V(offset res) + Vgs(SiC) = Vg(SiC)-V(B-) //13.5 + 0.1*17.5 + 6.8 + 3.3 = 25.35

Since you've measured all voltages except the 100 Ohm resistor, it may be the one that's off and you didn't notice.

In rare occasions, the DN2540 may be 'walking wounded' with a gate leakage, in which case the voltage drop on the 50 Ohm resistor no longer reflects the actual current from the CCS. Otherwise the voltage on the 100 Ohm resistors should corroborate with the reading from the 50 Ohm resistor in the same arm, assuming the output offset voltage is near zero so the current drain through the feedback resistors is negligible. 

[Pars]

I'll try to make sense of your equation(s) later, but thanks!

I went and ohmed out all resistors on board between the two and everything matches. I then violated my own rule of no HV when drinking, and measured the V drop across the 100R resistors in series with the PZTA42s emitters and they were all roughly the same (1.69-1.73V) between the good and bad board, which from ~17.5mA on the CCS seems to make sense.

I guess my next step is to measure the GDS or BCE voltages to ground on the 10M90s, DN2540s, but more importantly, the C2Ms and the PZTA42s. I'm also going to replace the 2 150K resistors since I did this quickly and am not satisfied cosmetically with them. The 2 on the good board look fine.

If a DN2540 was leaky, wouldn't it just affect that side and not both? I do have a spare pair of these, but no 10M90s. I do have more C2Ms though.

If I hadn't checked the PZTA Vce, I would have never guessed that something might be wrong, since everything adjusted correctly. No indication whatsoever.

[Pars]

OK, didn't do much on this this weekend, but did take another look today. Pulled the board off the heatsink and examined it. Other than a bit of heatsink compound, didn't really find anything. I resoldered a bit on the PZTA42s and a couple of other SMD PZTAs. Hooked it back up and no change.

The bases on the PZTA42s looked the same between boards (-393Vdc). The one thing I noticed is the drains of the C2M devices on the good board were more or less the same at -3.93 V or so (to gnd), but on the bad board they varied from device (-1.9V to -5.xV on the other one). I think since I have them, I will change out the DN2540s since I'm not finding anything else. I was starting to think it might be a front end issue, but since the bases of the PZTA42s seemed the same between boards, I can probably rule that out.

[mwl168]

Did you measure the PZTA42 Vce between the two channels before making the mod?

[Pars]

No, I did not measure the Vce prior to making the mod.

[Pars]

I remeasured the drains of the C2Ms to ground and they looked OK between the boards. I did a diode test of all the SMD transistors and the C2Ms between boards and they all matched (that took about an hour). I replaced the DN2540s to no avail. The offset went up a bit (18mA), but readjusted that to 17.5mA. I guess I'll replace the C2Ms next, and maybe go over the resistors once again. I'll also take a closer look at the PZTA42 voltages. I don't have any more of these, but could get some. I also don't have any 10M90s, and can't get any that I am aware of for quite some time.

[luvdunhill]

23 hours ago, Pars said:

I also don't have any 10M90s, and can't get any that I am aware of for quite some time.

These old things? 

[Pars]

Yeah, dem der tings. I'm really doubting I need any now, but might hit you up for a couple if I determine that is the problem. You're taunting me, right?

 

Edited March 23, 2023 by Pars

[luvdunhill]

Only if you promise no one else will bother me about them

[Pars]

Took some more measurements, but still need to verify resistors (again):

Also measured the DN2540 tab to ground, and noticed a difference in these which could account for the Vce difference? Coincidence? All voltages to ground.

Edited March 23, 2023 by Pars

[simmconn]

The good board C2M measurement doesn't look right. The VG should be higher than VS, and the Gs are tied together so the two VGs should be the same. Did you swap the G and S when filling in the table? If so, chances are you modified the VG resistor divider on the good board but not the bad board.

 

[Pars]

I'll recheck those, could have swapped them. I didn't do any mods other than changing the 174K resistors for 150K, and that was done on both boards. I just redid the bad board the other night.

EDIT: you were correct, those measurements were flipped. Rechecked and verified.

Edited March 24, 2023 by Pars

[Pars]

I rechecked all resistors and everything was correct, though I noticed that the 1M resistors ohmed out at ~750K or so on the bad board vs. around 500K on the good board. I went ahead and changed out the C2Ms. After that, the 1M resistors now ohmed out at ~500K in board. Fired it up, and the tabs of the DN2540s were now around 3-ish volts. Vce on the PZTA42s is now ~13+ volts, so SUCCESS!

[Reima]

I finally got the last parts that I needed from Mouser last week. Now I have everthing except the chassis and volume pot to build my Carbon. I had thought about using the RK50 that I have for my DIY T2 but when I saw the price ($1,252.27) to replace it I nearly had a heart attack.

I have a few questions:

My boards are .6G so I need to raise the resistors, how high off the board should they be? Is 2mm enough?

I see some builds have the amp boards at the back of the chassis and others at the front. Is there any benefit (sonic or otherwise) to putting them at the front or back?

I will start with building the GRLV next weekend.