Multi Configurable Testing PCB for Papa’s SIT-PMosfet, SIT-NMosfet MUFF and DEFiSIT Amplifier

[JeyDee]

🤘🙂Rock on!!

 

[wwwtttwww]

Amp boards assembled.

 

[Adams_Leo]

Amp boards assembled.

would you please share the mouser or digkey 3W,0.22R resistor link?thanks

 

[wwwtttwww]

nothing at Mouser
at digikey
https://www.digikey.ca/en/products/detail/panasonic-electronic-components/ERX-3SJR22V/3982890

 

[Adams_Leo]

yeah,thank you

 

[Nelson Pass]

Sometimes you have to work with what you can get.

 

[wwwtttwww]

Sadly, LuFet got go to make place for real SITs. It was my main amp for a year.

 

[wwwtttwww]

Power on without SIT and NMos. Bias preset.
48VDC on, no smoke. Adjusting RV1, G of SIT terminal block around 22.5V. Applying 1.3VDC on S of NMos terminal block, simulating 2A bias, adjusting RV2 the voltage G of NMos terminal block to around 5.5V.

 

[wwwtttwww]

Connecting SIT and NMos, and power on again. Adjusting RV1 and RV2 again to get 24V Vds on SIT and 2A bias. It is very stable. After fine tuning of two modules, they are in the 5U case.

 

[wwwtttwww]

From the beginning, considering the case I have and copying Papa’s Sony V-fet, a single 48VDC power supply and around 2A bias is in mind. Checked all Mean Well 48VDC adapters, found the V- is not connected to Safe earth in its medical model GSM. All GST model including what is used in the Sony VFet has it V- connected to the earth. Two GSM220A48-RB7 are selected for the power. They are expensive.

 

[wwwtttwww]

This is my first time to use Papa’s optocouplers bias circuit. Excellent. Once set, almost nerve change. Cover off or on, it doesn't really matter.

​

 

[wwwtttwww]

Listening..............
It is the time to satisfy or disappoint that a big effort you put towards the building.
Papa's speaker slow on/ fast off circuit is excellent. When properly passed to V+ voltage and set, there is no thumb at power on/off.

 

[wwwtttwww]

when started this build, one goal is clearly set: the amp has to have same measured signature as Papa's SIT model. that means similar THD and 2nd is dominant(negative, as Pa' often says)
picture 1 to 4, speaker negative terminal connected from top(close to S of NMOS) to bottom (close to GND).
Rs/Rm = 0.11/0.565ohm, 0.22/0.445ohm, 0.33/0.335ohm, 0.44/0.235ohm, Iq=2A, SIT Vds=24V at 2.17Vrms output(max I can get with Focusrite) and 8ohm load.
It is very clear that THD is 2nd at +89. The 3rd is 20db lower than 2nd. the speaker- terminal goes from up to low, THD reduces slightly. the minimal 3rd is around 50%/50% Rs/Rm.
The THD number is very close to Pa's number.

 

[wwwtttwww]

the influence of bias current
from left to right, Iq=1.8A, 2A, 2.2A, Vds 24V, Rs/Rm=0.44/0.235ohm, 2.17Vrms at 8 ohms.
THD reduces slightly with a decreasing Iq. 2A Iq has minimal 3rd at the testing conditions.

 

[wwwtttwww]

the effect of SIT Vds.
from left to right, Vds=22V, 24V, 26V. Iq=2A, Rs/Rm=0.44/0.235ohm, 2.17Vrms at 8 ohms.
Increasing Vds reduces THD, but still minimal 3rd is around 24V Vds.

 

[wwwtttwww]

The DIY of Papa's SIT MUFF has got what I wished, at least in design and engineering the PCB board I used.
The pw is very excellent for the amp, too.
As promised, I post the zipped whole KiCad project for the amp and power PCB here.

 

[wwwtttwww]

correction, picture #4 in the post #33 was wrong. It should be this one.

 

[wwwtttwww]

Frequency response

 

[Adams_Leo]

fantastic！did yoy try more v+ and more base?like 30v/3-3.2a base

 

[wwwtttwww]

No. Considering my 5U store chassis, it cannot handle 2*30V*3A heat.