Simplistic MosFET HV Shunt Regs

[merlin el mago]

Only to be sure:

ksa1381 are at the gate?
LED the negatives side?

Could be good to stick the schematic with voltages differences in the 1st post.

 

[Salas]

Voltages noted behind the KSA's are at their bases. Led noted voltage at its cathode.
After the inter-voltage differences are cross referenced with yours and agreeing I may link it from post one, or add it to the SSHV2 introduction linked post.

 

[merlin el mago]

Sorry for delay, attached measurement of voltages. Q1 is Ixys IXTU01N100D in place of DN2540.

 

[Salas]

How come and there is less voltage on the CCS MOSFETS than on the output? Despite some absolute differences it follows the rest of trends. I would also expect little more voltage difference between Q1 Q2 source pins but its the IXYS in this case. Maybe IXYS Q1 with IXYS Q2 can work better and sound better in this build?

 

[merlin el mago]

I replaced Q1 because I burned several DN2540 in this position, IXYS is more robust.

If you have time please take a look here: https://www.diyaudio.com/forums/tubes-valves/363199-salas-sshv2-moglia-hybrid-mu-follower.html#post6416281

 

[Salas]

Took a look and I am thinking something is electronically wrong so it naturally manifests in disappointing sound. This lower voltage level on Q1 Q2 source pins than the on the reg's output thing is very fishy.

 

[merlin el mago]

Attached new measurements after warm-up.

 

[merlin el mago]

The SSHV2 CCS is set to 50mA.

 

[Salas]

This looks more logical for difference between Q1 Q2 but 181.5V output >181.1 source Q2 is still weird. There should be difference due to drop on R3//R4 + R5 keeping Q2 source higher than Vout. Can there be oscillation on output so its AC+DC average?

 

[Salas]

Or its just drift between time of taking measurements?

 

[merlin el mago]

I doubt is drift because SSHV2 powered ON all time between the 1st & the last measurement.

 

[merlin el mago]

Vout is DC as all the measurements.

 

[Salas]

Oscilloscope X10 probe AC coupling 5mV vertical shows a clean DCout line?

 

[merlin el mago]

Attached scope shot 5mV

 

[merlin el mago]

SSHV2 set to 170Vout

 

[Salas]

That's when connected to the Moglia gyrator?

 

[merlin el mago]

Yes only 100R between the SSHV2 & the Moglia's gyrator.

 

[Salas]

There is no obvious finding of bad output behavior then. Make Vin-Vout higher maybe.

 

[merlin el mago]

With this tx & CCS bias 170Vout is the max. difference Vin-Vout available.

 

[merlin el mago]

Salas 5687 has too much gain, I like DHT but is a lot of money in txs, 5687 sound is very good, some ideas: how about 6V6 as analog output stage for the noDac?