Gain of FET circuit

Hi Guys, I have a FET replacement circuit for a tube. I want to know what the Gain of the FET circuit is.

Here's the schematic of the FET circuit.

formatting link

Can you calculate the Voltage Gain from Gate to Source. The gate will be connected to a capacitor with 1.2mV to 5mV RMS on it.

Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx
Loading thread data ...

** Giant huh ??

Source followers have no voltage gain, only a small loss.

The cct drives a DC meter FFS.

.... Phil

Reply to
Phil Allison

2.5?
Reply to
krw

It's going to depend on the mutual conductance of the FET, which is a property that will have to be derived from the datasheet.

Sylvia.

Reply to
Sylvia Else

I kinda with Phil, in that it like an emitter follower, and then with a Drain resistance the reduces it. (I think) :-)

Can you try the tube circuit it is replacing? I got 1.25.

Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

Is that a DC meter movement? Is the fet expected to rectify the AC input?

If not, the voltage gain depends on the fet specs. It will be low, very roughly 0.6.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Yes it is.

Is the fet expected to rectify the AC input?

That's what I understood about the tube circuit it is replacing.

The tube is in cutoff on neg half of the cycle, so you only get a positive pulse out.

Can you try the tube circuit it is replacing? I got 1.25.

Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

Wait! You said that the AC input is 5 mV RMS max. That circuit shows 4 volts!

The series resonant circuit multiplies the millivolt low-Z source up by Q. Neither a tube nor a jfet is going to rectify millivolts.

You might have said it was an old Q-meter.

I'd suggest you simulate it, given a decent model for whatever jfet you want to use. LT Spice has njf models.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

And try to pick a low capacitance jfet, something similar to the old toob.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

** Even the 535a tube is a special.

formatting link

Operation relies on very high non-linearity of the tube in those conditions.

.... Phil

Reply to
Phil Allison

Hey your right! No surprise. It should be 1.2 Volts to 5 Volts in both cases.

Yes, we are back to Volts, not millivolts.

1955? That's my birth year, I only feel old a some days.
--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

I haven't had much luck finding anything much different from the others. But, after I figure out what gain I'm after, I want to use something like the "Ultra high Z in unity gain amplifier" shown at the top of page two, in the this link. >

formatting link
We want as little load on the cap as possible. Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

I made a q-meter using a sampling oscilloscope probe as the detector.

formatting link

formatting link

I computed Q from the width of the resonance, not the voltage multiplication.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Yes, and I don't have a full understanding of that, I don't know if they are trying to match the reciprocal non-linearity of the thermocouple in the circuit. Also the meter it drives has a non-linear scale. (On purpose) The overall plan is to measure the injection voltage with a modern RF millivolt meter, and forget the thermocouple output. Then have a linear output from the Fet and drive a meter with a linear scale. It's all a little pie in the sky, because everything I've tried has been non-linear.

The problem I'm trying to solve is as follows. The Boonton 260A only measures Qs up to 625. but if you lower the injection voltage (now you need to measure with an RF millivolt meter) you can measure higher Qs, using a little math. Injection Voltage is the multiplier, 20mV is 1X, 10mV is 2X 5mV is 4X and 2mV is 10X. The problem we're having is, it goes non-linear as you drop the injection voltage, were it starts depends on the tube. I'm hoping the Fet is not non-linear. Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

Yes, there are multiple ways to measure Q, My 6.5" diameter by 6" length coil won't fit on your pcb! Also, if I get within 2 ft of the coil it changes the characteristic. Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

But it's not an amplifier. It's a rectifier.

The trick will be to get the rectification nonlinearity to match the nonlinearity of the tube, so the meter indicates right.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Parallel resonate it with a very good, HiQ porcelain capacitor.

Loosely couple a signal gen into it with an inductive link or a very small cap. Ditto again, out into a scope. Measure the resonance width.

Presidio and ATC have porcelain caps with Q of 20,000 or so.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Ya, I'm past that, I want to use a linear scale if that will work. Maybe a peak detector and a little hold, then scaled. Maybe need to amplify it before detecting to get linearity. That's a question! I found that the jellybean 2N3819 has a 3pf Ciss, lowest I've found. But I'm not sure how much that matters with the bootstrap circuit on the Ti site I linked to. Tell me this, the TI "Ultra high Z in unity gain amplifier" has an input of 0.25pf and 100Meg, if I use a capacative divider of 0.3pf in and 3 pf to ground, does that raise my input Resistance 10 times? Assumes incorrectly the divider will not add leakage resistance. Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

You can't bootstrap that source follower detector* since the source no longer follows the AC signal but if you really wanted to lower its already pretty low input capacitance you could precede that detector with another fet wired as a bootstrapped follower.

*That source follower circuit with capacitor bypass and high value load resistor is what was sometimes called an "infite impedence detector" popular in regen TRF receivers way back. Your 12k load seems rather low though. Fet circuits I have seen used much higher values, I think >100k. As you have correctly figured there will be a standing bias DC voltage at the source that your meter circuit will have to zero out.

piglet

Reply to
piglet

Yes, I have a several HiQ ceramic capacitors.

I have done that hundreds of times. That's not the point, I want a replacement for the hand selected 535 tube, it was hand selected for certain characteristics in the 1950s. They just aren't available anymore. Right now, I know of 4 tubes, only one of them still meets those specs.

So, my questions still stands, with an addition.

Can I use the "Ultra high Z in unity gain amplifier" shown at the top of page two, in the this link. >

formatting link
For a gain of one, then add a precision rectifier with a hold cap. If everything is perfect* I get a gain of 1.4 because I'm holding the peak voltage. I can easily scale it down as needed. I'm over my head here, remember I'm the fish monger, but I can fabricate a decent pcb. With a little help, I can get this done. Can you recommend an opamp, that will work as a precision rectifier up to 10 Mhz? Oh, I probably need some fast diodes too.

On another note, the 2n4416 has a 3pf Ciss, but it costs $6 each! I want a through hole fet here because it is to easy to over drive and pop it in normal usage. "I've been warned about that." Through will be easier to change.

  • I know they won't be perfect

Thanks for your time, Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
amdx

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.