UDNeSS, or You don't need Semisouth's

MOSFET Matching

Reports from Alpha Tester :

"The Vgs values of 25x IRFP9240 devices were measured at 23Vdc and 1.2A.
Values ranged from -4.51V to -4.60V so quite consistent.
Can get 8 very good pairs out of them."


Patrick

.
 

Attachments

  • 191002 UDNeSS Matching.jpeg
    191002 UDNeSS Matching.jpeg
    106.9 KB · Views: 1,178
First test

First test successful.
Everything worked as intended.

🙂

Heatsink temperature 40°C after 1 hour.
DC stable to 20mV after trimming.
So servo can be optional.


Patrick

.
 

Attachments

  • 191009 UDNeSS Proto1 Test 2s.jpeg
    191009 UDNeSS Proto1 Test 2s.jpeg
    54.7 KB · Views: 1,007
  • 191009 UDNeSS Proto1 Test 3s.jpeg
    191009 UDNeSS Proto1 Test 3s.jpeg
    232.6 KB · Views: 969
A short update.

A lot or work has been done over the weekend, without saying.
We have to make some adjustments to the circuit, as the Spice models for the MOSFETs and the JFETs do not seem accurate.
See e.g. :
JFET buffer with J113

For example, we need to adjust the R11/R13 ratio to get close to equal current sharing,
We have also measured the distortion spectrum and it is predominantly H2, with H3 20dB lower.
H4, H5 are below noise level.
But the H2 level is still a bit high for our liking, so we have a bit more fine tuning to do.
Will take a bit longer, but we are pretty sure we know how to get there.


Cheers,
Patrick