JFET as variable resistor

[unkarc]

variable resistor using vcr 3p fet

Hi,

Probably your JFET is faulty or you did not use the correct pin-out of your device.
Here is the data sheet including the pin-out too:
**broken link removed**

If you are sure you correctly attached your fet into the circuit I attached, then make sure you do not touch the source pin of your fet with your fingers because it may influence the voltages shown.
In my attached circuit the resistor in series with the source electrode is the symbol for the inner 10 MOhm resistance of your DVM of course.
What diemilio suggests with using a capacitor is that you can place an electolyte cap of 47-220uF in parallel with the DVM in the circuit shown. Basically it is the same method because when you wish to check the voltage on the capacitor you end up with connecting the DVM in parallel with it.

I tested a J310 JFET in the circuit and measured the followings:

VDD----- Vpinchoff------IDD@Vgs=0
1V ----- 0.997V ------- 21.6mA
2V ----- 2V ------------ 34.1mA
3V ----- 3V ------------ 38.7mA
4V ----- 3.987V ------- 40.3mA
5V ----- 4.23V --------- 40.5mA
6V ----- 4.27V --------- 41.5mA
7V ----- 4.29V --------- 41.5mA
8V ----- 4.3V ---------- 41.6mA
9V ----- 4.31V --------- 41.6mA
10V ----- 4.32V --------- 41.7mA
11V ----- 4.33V --------- 41.8mA
12V ----- 4.34V --------- 41.8mA
13V ----- 4.34V --------- 41.9mA
14V ----- 4.35V --------- 41.9mA
15V ----- 4.36V --------- 42mA

So the pinchoff voltage of my fet is between 4.23V and 4.45V, for calculations the average value of that range can be used, let's say around 4.34V.

The J310 is relatively high current VHF/UHF amplifier wrt your 2SK168 type. In fact it warmed up a little from drain voltages of 6-7V and up and I had to wait it cool down a little to get the 41-42mA range (when it was warm, the current dropped to 38-39mA range). I observed parasitic oscillations with the J310 through the power supply (the drain current changed erraticly) and a 220nF ceramic capacitor in parallel with the VDD and the negative pole helped.

Hope these are of help to you.

unkarc

 

[diemilio]

variable resistor fet

Connect Drain to Vdd, gate to ground and the cap between source and ground. As soon as you turn on your power supply the cap would be discharged so Vgs would be 0. With Vgs=0 you would have a ids equal to IDSS which would start charging the capacitor. As the capacitor continues to charge the current would begin to decrease cause Ids=IDSS(1-Vgs/Vp)^2. When the current finally reaches 0, Vgs would be equal to Vp, so all you need to do is turn your power supply, wait a couple of microseconds and with a DVM measure the voltage between source and ground. Since the DVM has a resistor that can discharge your cap during the measurement you should probably want a relatively high capacitance. I've done this before with a 2N3819, vdd=12 and a 100u capacitance. Also remember that the value measured on Vsource is equal to -Vp not Vp.

 

[sec]

fet variable resistor circuit

Thank unkarc, I already got this datasheet, maybe it's just my fet is a fool, I'll try with another one. your reply are very constructive, I surly have other question later.

Added after 2 hours 32 minutes:

can the resistance of an MOSFET be use the same way as the JFET, can it be linearized ?

 

[unkarc]

fet as viariable resistor

can the resistance of an MOSFET be use the same way as the JFET, can it be linearized ?

Hi,

Yes. Note that usually the dual-gate mosfets are more linear without compensation than JFETs without compensation. On compensation I mean the feedback from drain to source with a resistor like in this link is shown, referred to you earlier in this thread by IanP: **broken link removed**

Here is a useful discussion on FETs as controlled resistors:
http://www.vishay.com/docs/70598/70598.pdf

Here is a good application of this compensation, see page 2 of this Design Idea:
**broken link removed**

unkarc

 

[sec]

jfet as resistor

I try to desing an active filter, with a resistor that switch electronically, is the FETs the good way. I think it's hard because the voltage always change, am I right or I miss something in all my questions ?

 

[unkarc]

fets as variable resistor

Hi SEC,

I think it's high time you write what your problem really is... you started with this:

can anyone help me, am trying to desing a circuit with a JFET that
act like a variable resistor, or if you have any other idea to making a resistor change by changing voltage

and I think you got several good answers how to, now you ask this:

I try to desing an active filter, with a resistor that switch electronically, is the FETs the good way. I think it's hard because the voltage always change, am I right or I miss something in all my questions ?

I can only guess now that maybe you wish to tune the pass/or cutoff of whatever frequency of an active filter with a changing DC voltage and you wish to use FETs as variable resistors instead of mechanically ganged potentiometers to accomplish this? IS THIS WHAT YOU WISH or I am far from it in guessing? Please specify exactly. I do not fully understand your last sentence (the voltage always change).

unkarc

 

[sec]

fet as a voltage variable resistor

Yes, you got all right. Here is my circuit. What I mean by the voltage change is, the signal in input will pass into the fet. The signal is near zero so it will be under the Vref
I want to change the resistance of R3

 

[unkarc]

jfet resistance variable

Ok,

1) What is Vref and where is it in your circuit?
2) How much DC voltage do (will) you have available for controlling a FET placed as R3?

 

[sec]

6v variable resistors

Vref is the ground show in my schema, voltage is between 4 and -4

 

[unkarc]

using a fet as a resistor

Hi,

So you seem to need a floating (=ground independent) controllable resistor to replace R3 in your filter circuit. Although there are research papers on MOSFET based voltage controlled floating resistor circuits, they need additional auxilary circuits and rather complex, now I would suggest using optocouplers with FET output like the H11F1 type (see data sheet link http://www.fairchildsemi.com/ds/H1/H11F1.pdf and its Page 6 for applications) or there is a Canadian firm making so called resistive optocouplers, see
http://www.silonex.com/ and an application here:
**broken link removed**

Finally here is a practical circuit you can use without any further problem in place of your R3: http://www.zetetics.com/bj/papers/phd/appxa.pdf and go to the very last page, 15, where you can see the H11F1 optoFET is shunted with optional Rp resistor, and also an Rs optional resistor is connected in series with it, these two resistors make the resistance change more linear and can restrict/trim the from-to range of the resistance change as is appropiate in your filter.
This circuit as you can see works for a 0-5V control voltage but you said having +/-4V, you can use the 0-4V positive part of it only. And of course you do not need the relay at the output, you can directly include the two legs (pin 6 of the optofet and the upper leg of Rs) in place of your R3.
You can use the silonex resistive optocouplers in the same control circuit like the H11F1.

Hopefully these are of help for you and solve you original problem.

unkarc

Added after 29 minutes:

Yes, you got all right. Here is my circuit. What I mean by the voltage change is, the signal in input will pass into the fet. The signal is near zero so it will be under the Vref
I want to change the resistance of R3

You do not need to worry about the signal level because it is some tens or max a few hundred mV at the input of your filter and the gate control voltage of a FET is in the range of a few Volts, and if this is not so and you have a Volt input voltage there is the resistance R1 in series at the input of your lowpass filter, forming a voltage divider with R3, so further reducing input voltage no real interaction can develope. And using an optoFET this problem is solved.
Basically what circuits I have just showed you in the links they all can be viewed as a refined versions of that of Gundam001, also commented by Eugen_E.

unkarc

 

[sec]

h11f1

I would like to know more aboute MOSFET and the ohmic region, does anybody got explanation for me ?

 

[unkarc]

fet variable resistance

http://en.wikipedia.org/wiki/MOSFET

**broken link removed** page 15

http://www.techlearner.com/Apps/MOSFETBasics.pdf page 7 on ohmic (linear)region

**broken link removed**

**broken link removed** some math formulas

 

[eegchen]

variable resistor using mosfet

nice comment