If you do SRPP you end up in the same troubles as the ray samuels A10.

Same sort of thing as mikhail's original disaster.

Ehm, no, I wasn't talking about using tubes as the plate current source. I was about using mu-follower wired solid state plate current source, output connected to drain of the source's transistor, as luvdunhill said.

Ah, it is only called SRPP is the current source is made using same tube?

I think the SRPP should be viewed as a modulated CCS like Pass Aleph http://www.passdiy.c...df/zen-ver2.pdf - the 50% split of AC current seems likely to optimize efficiency of bias current/power dissapation with little danger of destabilizing gain swings that the higher modulation levels of the mu follower extreme

Pass shows enhancment mode power fet with bjt regulator and AC coupled modulation - with depeltion mode power fet and tapped resistor string the circuit is a lot simpler

I tried talking about these ideas in the T2 thread when the good Doctor asked/speculated about alternatives

Edited September 21, 201114 yr by jcx

Actually because the load is not a constant impedance with frequency

the term SRPP does not apply anyway. It degenerates into a simple

current source. And solid state seems to be a much better idea for this.

And the final mad thought..

What crossed my mind is an such remake of SRX circuit:

Input stage is modified by using a mu-follower/CCS/modulated CCS.

Output stage modified same way as well.

Interstage coupling capacitors replaced by voltage divider connected to highly negative voltage (say, 1.5 kV). Input stage CCSes handle the DC current through that divider.

It is simple and straightforward, compared to active batteries etc.. Drawbacks I see is really poor PSRR and loss of gain - though not very high.

Or should I banish that thought in deep night?

I also added Pro bias to one of the sockets and it sounds pretty nice driving a SR-007Mk2....

To the Birgir-mobile!

It's what I do...

To the Birgir-mobile!

I'd like to think whoever made that pic was tripping balls while watching HeMan and drinking eggnog.

^Awesome GIF, one should be warned not to drink while watching it.

Actual reason for the bump though is that i'd like to know what could be causing unsatble offset and balance readings with my the KGSS, mainly the left channel keeps bumping up the offset to around 7-8V and the trimmer is already at the end if it's rotation range and is clickin. the balance is 2-3V but i'm able to adjust it easily while the offset is a lot less responsive. Also the readings fluctuate pretty wildly in a scale of 1-2V with me doing anything, No phones connected and no input signal presnt. Is that supposed to behave like this ?

Not supposed to happen. What are the psu voltage readings?

Hi spritzer, Aplogize for the delay, I've not had the time to do any measuremts until today.

I'm getting 366V in the postive rails, 355V on the negetive rails. After some time (a few hours) another 3-4V are added on both rails. Bias is 584V.

And i have 300VAC across the primaries on the trnasformer btw, It was commissioned accrding to your specs from sumR.

Is there anything else i should check?

The voltages being off isn't a good sign. AC in is a bit high but that will only cause a bit more heat so nothing to worry about. What I'd check are the zener strings and measure each step against ground. Personally I'd just replace them to eliminate a possible cause for the voltage difference.

Well i did just that not long ago, after i got both the FETs blow up on me... Unfortunately it took off all the traces between them and the zeners so i had to adjust:

No vert pretty but gets the job done, basically at least... I have tested all the zeners and they pass the diode test, but the 1N4757 at the positive rail (The beaten side:) has double the resistance, 3.9Mohm at the voltage drop vs 2Mohms of the parallel negetive rail. Reversed it's half the resistance from the other rail 12Mohms vs 24Mohms. Is that significant for the voltage pass to the mosfet?

damn that looks nasty

must've been a pretty scary explosion if it took out that many traces...

not commenting on anything else, but I think the jumpers could have been done better.. and if the diodes are in circuit, you are probably not measuring the diode, but measuring everything around them.

I don't know how useful this is, I checked a few points on my kgsshv psu, in ohm test, the negative and the positive rail at the same component gave me near identical readings (just poked some random ones.)

Well i did just that not long ago, after i got both the FETs blow up on me... Unfortunately it took off all the traces between them and the zeners so i had to adjust:

No vert pretty but gets the job done, basically at least... I have tested all the zeners and they pass the diode test, but the 1N4757 at the positive rail (The beaten side:) has double the resistance, 3.9Mohm at the voltage drop vs 2Mohms of the parallel negetive rail. Reversed it's half the resistance from the other rail 12Mohms vs 24Mohms. Is that significant for the voltage pass to the mosfet?

Just a suggestion. You might want to consider nylon washers under those mounting screw/washer combos that overlap the board traces. Solder mask is quite thin and easily subject to tears/penetration. Very easy to create unseen shorts just by torquing down the screws a bit too tightly.

I would install nylon washers on those screws as that is just an accident waiting to happen.

The 1N4757 could be the culprit as the uneven PSU voltages could be causing your problems. The caps might also be damaged from the time it all went arcy sparky...

Since you're back onto the KGSS, I do have a question if I may.

I've just put together a KGSS using Justin's pcb's, (I think they were his last from a while ago).

How hot do the 50 mm heatsinks, on the amp pcb, normally get?

I'm measuring 62 to 70 degree C after a approx 30 min's warm up.

PSU sits at 361 and 362 +/- vdc, and the +/- 15 vdc are stable and ok.

Offset and balance vary a fair bit on warm up, approx 20 to 30 Volts

Just seems too hot and maybe the offset/balance shows something is not quiet right,

The psu heatsinks run at approx 40 Deg C

Boards not cased yet, in open test rig.

Allan

I just measured mine through the top panel and got 62-70°C. This is after about 30 minutes in chassis which is not the best in terms of cooling.

Large drifts in offset and balance could be tied to how the output devices are attached to the heatsinks. What interface did you use?

I just measured mine through the top panel and got 62-70°C. This is after about 30 minutes in chassis which is not the best in terms of cooling.

Large drifts in offset and balance could be tied to how the output devices are attached to the heatsinks. What interface did you use?

Hi

I'm using sil pads at present, was thinking of mica and goop.

Still, it does seem a bit hot,,,, The devices will be even hotter.

Does the offset/balance vary much while it's getting up to temp?

Allan

Just a thought

I could drill same vent holes just under the heatsinks

The went holes will help but it is also a good idea to lift the sinks off the board a bit.

I already have the nylon spacers under the heatsinks

Might get the taller ones..

AAVID THERMALLOY - 530002B02500G - HEAT SINK

HEAT SINK

Packages Cooled: TO-220

Thermal Resistance: 2.6°C/W

External Height - Imperial: 2.5"

External Height - Metric: 63.5mm

External Width - Imperial: 1.65"

External Width - Metric: 41.91mm

External Length - Imperial: 1"

Same dimensions just taller

That's what I've been using for years now. Runs a bit cooler.

But on the build at the start it recommends the 50 mm ones.

I'll try the 63.5 mm one's too..

And maybe perforate the pcb, while have the heatsinks off,

and use mica and goop

The sinks are not grounded so you can just use the shoulder bits for the screw and plenty of goop.

but then the heatsinks would be live.

Voltage is a bit high to have live heatsinks.