Help with an oscillating amplifier with Exicon mosFET ECF10P20 & ECF10N20

BTW, your question reminded me of a very gentlemanly discussion quite some years ago between Nelson Pass and Charles Hansen (RIP) that touched on some differences between VMOS and Lateral MOSFETs. It would be useful reading for such a venture. If I find a link, I'll share it. I suspect someone else will beat me to it. And, although we've lost A LOT of the incredible expertise that visited this site over years past, there is still a lot of really good stuff around that if sorted through and studied can be helpful for budding designers as well as experts - it can save one lots time, save some PCB revs and avoid some burned up parts ! 😉
 
for the OP: might be worth it to take a step back for a moment and clearly confirm your objective(s):
What your immediate goal? Quickly build a working amp to use in your system? Or ...
start to develop design skills to design your own amp at some point in time?
What is an honest assessment of your current skill levels? If you search around here a bit, you will see that designing, building and testing a high performance MOSFET amp that is not an already "tried and tested" design or even a kit is not trivial. The Profusion Lateral MOSFETs are similar but NOT identical to the old Hitachi/Renesas Lateral power MOSFETs and, as you've found out already, they're not drop-in replacements all Lateral MOSFET designs without taking into account the subtle differences if the original designer has not already done that for you. Not trying to discourage; just want to help mitigate frustration and increase chances of success by helping you select realistic goals and better aim at them ....
Hi, thanks for you repli. I will try to answer your questions, honestly and clearly:
My first goal is to build an amplifier that works like my system. It's not the first one but in the other occasions were easy projects that it hasn't given me many problems.
At the same time I want to acquire skills, not to design, but to know how to solve problems in future assemblies.
My level of knowledge, electronics technician, mainly in repairing communications equipment. I have always been enthusiastic about the world of audio, amplifiers and loudspeakers, in fact I have built loudspeakers too.

It is true that designing, assembling and testing is not easy, but as a technician I like to try. In this case I bought the Exicon because Profusion told me that they are the replacements for Hitachi, and it has become clear to me that this is not the case. After a lot of reading and looking around the forum, I have seen some great projects with these MOSFETs, assuming that they are not all the same. That is why I am asking the question and trying to save the investment in components.

Not trying to discourage; just want to help mitigate frustration and increase chances of success by helping you select realistic goals and better aim at them ....
I thank you for your honesty, and I also reply that my goal is to enjoy a self-made amplifier.
Thank so much.
 
  • Like
Reactions: njswede
47 ohms both on positive and negativ rail.
I just checked the amp in TINA with somewhat other transistors and it worked well but only without C5.
I changed R13 and R15 to 47 ohms and R16 to 33 ohm.
It doesnt have so much open loop gain so the distortion value will not be so good but it sounds better than expected due to mosfets instead of bipolar output transistors. I used 2 of them a few years as PA amplifier for a rock band.
I think mine were about 0,05% distortion at full power.
 
  • Like
Reactions: carcotado
Hello friends, I continue with the tests of the different configurations. 😀.

You may try these mods
This option doesn´t work.


That's roughly what I've got.... cheers Frank
This option doesn´t work

Use 560R (P) and 680R (N) gate stoppers and a zobel across D-G of the Exicons (220R + 22pF).
This option work, I think well, I add oscilloscope image.
At the moment, this is the best option I see for this project. What do you think of these measures?

Image1: how Flikoman describes without C8
Image 2: how Flikoman describes without C8, and C6-C7 to 150pF.


I continue with some more tests,
 

Attachments

  • 1.jpeg
    1.jpeg
    226.6 KB · Views: 29
  • 2.jpeg
    2.jpeg
    217.2 KB · Views: 28
I did a quick and dirty simulation in ltspice, the gain and phase margins are quiet low, given the board layout it maybe partially unstable given the right conditions. Also with no output inductor, just connecting a loudspeaker would most likely cause the amplifier to go into oscillation.
The main contributor to this is the phase lead capacitor (C8) across the feedback resistor R11.
Unity loop gain is approximately 5MHz, gain is only -10db @ approximately 9MHz at 180 degrees phase.
The simulated gain/phase margin is approx 9db and 40degrees, probably worse with the real circuit board layout.

Removing C8 improves the gain and phase margin to approx 20db / 62 degrees, the loop gain at 20KHz (33db) is unaffected.
The simulated unity loop gain is now 760KHz.

Other changes:
R21/R22 - 470R
Leave R23/R24 at 270R
Leave C6/C7 at 33pf

As suggested move the gate resistors as close as possible to the mosfet gate pins. I would also add a output inductor to the amplifier approx 1uH -2uH.
This option the test is not clear, because the measurement is with C8 on, now I don't remember if I test it without C8.
I do it egain.
This is the measurement, oscillation also affects the input waveform.
 

Attachments

  • WhatsApp Image 2025-02-14 at 13.28.10.jpeg
    WhatsApp Image 2025-02-14 at 13.28.10.jpeg
    312.8 KB · Views: 21