Cello One. Good Amplifier 15 Watt with TMC and Laterals

[lineup]

@thimios
Please do.
In my schematic are single die, I think.
All information is good.

 

[mason_f8]

As long as PS is ±20V, I don't think a protection diode is necessary, but if you want to add one, add a small signal diode in series and in the opposite direction, since the capacitance between the electrodes of the Zener diode is very large.

 

[minek123]

As far as I know Exicons have gate protection diodes built in, but it won't hurt...

 

[brian92fs]

I know you are designing this for a ±20V supply, but with the components designated, I don't see any reason this wouldn't with with a ±42V supply from 30V secondaries.

 

[lineup]

As far as I know Exicons have gate protection diodes built in, but it won't hurt...

I will check out if they are built in.
-----
"I am told by Profusion that the Exicon mosfets have 14V 1W protection zeners"
https://www.renardson-audio.com/mk5-comps.html

 

[Bonsai]

Good progress on this amp 👍👍

 

[brian92fs]

For the 500Ω trimpot R4, might consider adding a 100n bypass cap.

For the 5pF cap C7, maybe add 100Ω in series with it?

I agree with mason from post #122. Add some 1N4004 diodes in opposite direction in series with the current limiting zeners. You can see that Rod Elliott's P101 uses this configuration.

 

[lineup]

I know you are designing this for a ±20V supply, but with the components designated, I don't see any reason this wouldn't with with a ±42V supply from 30V secondaries.

Sure.
It is very possible to use other supply voltages.
It is just because I do think 10W or 15W is all that is needed for a power amplifier.
Many hunt for power. Even when it is not necessary.

 

[brian92fs]

I have a layout started in EasyEDA. I'll share it if I manage to get far enough with it. I'm designing it to work flat against the heatsinks of a Dissipante 2U chassis.

 

[lineup]

Good brian!
We can remove the gate protection zeners:
https://www.diyaudio.com/community/threads/paralleling-exicon-lmosfets.313792/post-5221218

Have a look at ECX10N20 features:

 

[lineup]

For the 500Ω trimpot R4, might consider adding a 100n bypass cap.

For the 5pF cap C7, maybe add 100Ω in series with it?

I agree with mason from post #122. Add some 1N4004 diodes in opposite direction in series with the current limiting zeners. You can see that Rod Elliott's P101 uses this configuration.

I will test this in simulation, thanks

 

[Bonsai]

Here is the current circuit.
I have done numerous changes:

Input filter now has 82pF cap
Moved the DC-offset adjustment to the top at the current mirror
Changed emitter resistors in the mirror
Added baker clamp diode
Added overcurrent protection
Added mosfet protection zener diodes
Changed gate resistor values
Output inductor is now 6uH
Increased current in the input LTP

View attachment 1244429

If you are using the BAV21 you probably don’t need U13 or vice-versa

 

[thimios]

@thimios
Please do.
In my schematic are single die, I think.
All information is good.

I will do this later today.

 

[lineup]

@Bonsai
I will remove U13

 

[Mark Tillotson]

As far as I know Exicons have gate protection diodes built in, but it won't hurt...

Yes, that is the case according to the datasheet - however other laterals may not have so its probably wise to provide protection to allow any brand of lateral to be used safely?

 

[lineup]

I added C9 and R20 and removed U13.
R19 is now 10 Ohm.

THD is now 0.000058%

 

[Bonsai]

Lineup, I don’t see any supply rail decoupling. I would suggest 220uF per rail along with 2 off 1uF X7R 100V per rail.

🙂

 

[thimios]

Here you are
The N channel.

 

[lineup]

Yes, that is the case according to the datasheet - however other laterals may not have so its probably wise to provide protection to allow any brand of lateral to be used safely?

I think those who will build this amplifier will use ECX10N20 and ECX10P20.
There is no recommendation to use anything else ...

 

[thimios]

The P channel.