THD 0.00045% and 15 Watt

Jump to Latest
#22

Attachments

  • 250508 Hitachi MOSFET BJT.jpg
    250508 Hitachi MOSFET BJT.jpg
    115.6 KB · Views: 77
Last edited:
  • Like
Reactions: chermann, Drbulj, decramer and 2 others
#23
Drbulj said:
This is proven Hitachi diagram, used and cloned to humongous proportions, always delivers with so little parts
Click to expand...
Very Much so. It is a Classic
Such a great topology.

At low voltages, transistor choices become even more fun.
Using higher gain readily available transistors. I have tried B grade everyday BC556B/546B

And at " 1 watt" as posted, distortion is close to the OP post around .0003%
full power holds up to .002%
If everything is tuned to work at lower voltages. This just using standard BJT power transistors.
So normal VBE for bias is employed.

Agree, fun one to play with, with such low part count verses performance
 
  • Like
Reactions: chermann, Drbulj and lineup
#24
stigigemla said:
What more can you wish from symmetry?
Click to expand...
Agree.
As noted the low parts count verses performance often leaves you not needing much more.

I think it is only people that desire dual differential input.
To solve the 2nd gain stage current issues, the parts count goes up.

To help improve the noise issue Hitachi just regulated the power supply.
Instead of using Dual Differential for more noise rejection.

At higher power levels, high frequency THD falls quickly as well too.
Some desire the numbers game, something simulation likely encourages.
Real life, not as much a issue

If people chase it . I do understand.
There is a certain feeling of achievement if THD holds up at high power at high frequency.

Otherwise the performance, with so little parts. Also gives such feelings
 
  • Like
Reactions: Drbulj and lineup
#26
I am working as I said.
But it is not promising. No longer THD 0.00016%. Almost twice that.
And no longer 16 Watt. More like 15 Watt.

I am not happy. Maybe end this project ... 🙁
I could post my revised schematic. Is there anybody who wants this?
 
#29
stratokaster83 said:
Since the wiper and the unused end of the pot are tied together, I would assume that if the pot is an appropriate value, it would merely lead to increased dissipation and not an outright failure.
Click to expand...
Correct, but still you can run very long time at hot bias unnoticed, need to calculate ful pot R scenario.... Good quality pot will probably never fail... It is just small area to improve this design, no drama, but still
WhiteDragon said:
If everything is tuned to work at lower voltages. This just using standard BJT power transistors.
So normal VBE for bias is employed.
Click to expand...
Exactly, it is versatile. In post about darlingtons from few days ago (think you were there) I showed a picture of Sanken darlingtons, driven by Hitachi amp circuit with pedestrian BC's at input, BD139/140 at VAS and Vbe bias ... works aplomb. PS +-33V
 
  • Like
Reactions: lineup
#30
lineup said:
I could post my revised schematic. Is there anybody who wants this?
Click to expand...
Why not, if you have fun, I think many also like to read your SPICE experiments . It is cool fun.

I might have suggestion for next exercise, that I want to start playing when I have time which will never happen:
Take one simple but very fast CFB power amp with mosfets, I took first picture from internet, don't know if this one works or not, seems it does if you change mosfets to laterals instead IRF's, of course you can change BC transistors to something better. point is a simple and super fast, forget THD and DC offset:
1749146832939.jpeg

Than add one classic opamp at input and run global feedback overall with gain bit higher than poweramp itself... see what happens, I think it will satisfy THD needs at many frequencies
If you spice it, I will build it eventually 🙂
 
  • Like
Reactions: lineup
You must log in or register to reply here.
Top Bottom