The clips that Rick suggested should work. I made a Idss test rig and used clips with no problem. My clips were from an electronics multi tester (T7).
My JFET Idss test rig is hardwired - very simple with an old linear supply wall wart for power (this particular power supply had a negative inner pin). Since the JFETs draw little current, I added a resistor in parallel with the power supply to make sure that power supply voltage was stable. I also added a ZIF connector so that individual JFETs may be inserted for testing. The clips are used when the JFETs are packaged together in a cardboard strip.
To measure Idss, I provide power to the JFET for a second or so touching the loose ammeter probe to the power supply wire to complete the circuit and quickly read the value on the ammeter display.
The JFETs in the picture are J113 and their Idss were all between 20mA and 27mA, with most between 20mA and 24mA.
My JFET Idss test rig is hardwired - very simple with an old linear supply wall wart for power (this particular power supply had a negative inner pin). Since the JFETs draw little current, I added a resistor in parallel with the power supply to make sure that power supply voltage was stable. I also added a ZIF connector so that individual JFETs may be inserted for testing. The clips are used when the JFETs are packaged together in a cardboard strip.
To measure Idss, I provide power to the JFET for a second or so touching the loose ammeter probe to the power supply wire to complete the circuit and quickly read the value on the ammeter display.
The JFETs in the picture are J113 and their Idss were all between 20mA and 27mA, with most between 20mA and 24mA.
Attachments
Thanks, Ben. I never realized these ZIF connectors were sold. I ordered some.
I am thinking I have been doing this wrong, as I waited for things to stabilize, while you do a 1-second reading.
So much to learn, I still have.
To measure Idss, I provide power to the JFET for a second or so touching the loose ammeter probe to the power supply wire to complete the circuit and quickly read the value on the ammeter display.
I am thinking I have been doing this wrong, as I waited for things to stabilize, while you do a 1-second reading.
So much to learn, I still have.
Yes, I'll have 1/2W resistors on Wednesday, and I'll have tight contacts everywhere then. I might still try putting everything on a veroboard to see if the readings are different then.
As always, I bet I am going crazy looking for precision where only so much can be found.
As always, I bet I am going crazy looking for precision where only so much can be found.
El Arte - wish the wires were not thin - but the idea would be to use as short a length as possible. At these currents I think we would be OK.
The tester I bought has wires that are seven inches. Being able to use this sitting at a table the wires can be four inches.
I think when the clips are this small the wire will be thin. I am assuming the ones from AMAZON are using the same gauge.
The tester I bought has wires that are seven inches. Being able to use this sitting at a table the wires can be four inches.
I think when the clips are this small the wire will be thin. I am assuming the ones from AMAZON are using the same gauge.
El Arte - wish the wires were not thin - but the idea would be to use as short a length as possible. At these currents I think we would be OK.
The fact they are thin is only a problem if you touch them and if you use a breadboard (because they wiggle too much in the breadboard sockets). You are right to say they are plenty thick enough for milliamps.
I am sure the wires that came with your Peak device are great. Hopefully, the ones you get on Amazon will be as good.
My Idss rig is just an example of a hard wired test circuit. The Vpinchoff circuit can also be hard wired. Both circuits are simple and a hard wired/soldered circuit eliminates possible poor connections. I find these permanent test circuits handy as next time I need to test some devices, I do not need to assemble a test circuit again. 🤓
In the end, this did the trick reliably. First ZIF to sit the JFET. Second ZIF to sit the resistor. 20AWG solid core wire to link the parts.
Short(er) test clips.
All JFETs have been selected. They go in tomorrow.
And it sings...
As you can see, there's a headphones out and a preamp out. The headphones out uses the push-pull headphones buffer @ra7 also designed.
I called the previous SCG "lush". I call this one "delicate". It is not warm, per se, but its harmonic profile and resolution gives acoustic music pinpoint accuracy.
It is also a H2 machine.
If I power my most demanding headphones at 105 dB SPL (hearing damage occurs after 1 hour):
THD = 0.91%
H2 = 0.9%
Remarkable.
If I power my most demanding headphones at 100 dB SPL (hearing damage occurs after 2 hour):
THD: 0.5%
H2: 0.5%
I am not of the school that teaches distortion is bad. But, to me, this is incredible.
I think this is another extremely fine example of the Pass way, and I can only praise @ra7 for sure a great circuit.
As you all know, I am not the most experienced builder, but this build was EXTREMELY easy. Matching the JFETs was what required the most care.
I also had to adjust a bias resistor to get the quiescent current I wanted from the headphones buffer, but that took no more than 5 minutes to adjust.
Many smiles!
As you can see, there's a headphones out and a preamp out. The headphones out uses the push-pull headphones buffer @ra7 also designed.
I called the previous SCG "lush". I call this one "delicate". It is not warm, per se, but its harmonic profile and resolution gives acoustic music pinpoint accuracy.
It is also a H2 machine.
If I power my most demanding headphones at 105 dB SPL (hearing damage occurs after 1 hour):
THD = 0.91%
H2 = 0.9%
Remarkable.
If I power my most demanding headphones at 100 dB SPL (hearing damage occurs after 2 hour):
THD: 0.5%
H2: 0.5%
I am not of the school that teaches distortion is bad. But, to me, this is incredible.
I think this is another extremely fine example of the Pass way, and I can only praise @ra7 for sure a great circuit.
As you all know, I am not the most experienced builder, but this build was EXTREMELY easy. Matching the JFETs was what required the most care.
I also had to adjust a bias resistor to get the quiescent current I wanted from the headphones buffer, but that took no more than 5 minutes to adjust.
Many smiles!
What a beautiful, clean build! Great work, ElArte! I'm so happy it makes you happy!!!
I'll take this opportunity to thank Papa. He is the inspiration, the teacher, the planter of seeds. Thank you, Papa, for igniting this passion that makes life more interesting and fun.

I'll take this opportunity to thank Papa. He is the inspiration, the teacher, the planter of seeds. Thank you, Papa, for igniting this passion that makes life more interesting and fun.


Not really. Vpinchoff for my lot was in the -7.5Vgs to -8.0Vgs. So, one would pick the -7.5V part.
Another one! I see you have changed the compensation capacitor to 100 pF.
Another place to use the WIMA or have you found something you like better?
Another place to use the WIMA or have you found something you like better?
What is the other board?
I think you speak of the headphones buffer. It is a precious beast. That board turns out to be more expensive than the marvelous preamp board! You know why? Because we live in an age where the primitive terminal block, heat sink and capacitor cost more than the mighty transistor, much more.
This build was super easy overall.
The biasing network on the headphones buffer is a bit finicky because LTSpice fails to help you anticipate what you will need. It suggested using a 15.5K R105 with those MOSFETs I picked, but in practice I needed 13.5K instead, or I would be limited to below 40 mA quiescent current.
Also, the power supply board needs M2.5 standoffs and screws, while everything else can use M3.
I opted not to connect the preamp board to PE because I know signal ground is referenced to PE at the source. My choice eliminates one opportunity for ground loops, but it also means there is some noise when the source is powered off or has no reference to PE. One day, I will become intelligent enough to understand what those TH_Chassis footprints really beg for.
The biasing network on the headphones buffer is a bit finicky because LTSpice fails to help you anticipate what you will need. It suggested using a 15.5K R105 with those MOSFETs I picked, but in practice I needed 13.5K instead, or I would be limited to below 40 mA quiescent current.
Also, the power supply board needs M2.5 standoffs and screws, while everything else can use M3.
I opted not to connect the preamp board to PE because I know signal ground is referenced to PE at the source. My choice eliminates one opportunity for ground loops, but it also means there is some noise when the source is powered off or has no reference to PE. One day, I will become intelligent enough to understand what those TH_Chassis footprints really beg for.
Folks, I was suggesting 50-60V as the supply voltage sweet spot earlier. Turns out it is closer to 60V and above. I'm checking and will report back. Meanwhile, you can try another operating point by turning the pot to set the Vds or buffer midpoint to 35V.
From my experience with higher voltages I did worry you were cutting it a bit close. I have never had one of "the little guys" go away.
I did have one of the FETs explode during testing with the 9 volts battery, though.
I did have one of the FETs explode during testing with the 9 volts battery, though.
- Home
- Amplifiers
- Pass Labs
- Schade Common Gate (SCG) Preamp