Troubleshooting a J310?

[FrankO]

I'm troubleshooting a recently built project that involves processing RF signals at HF frequencies. The problem appears to relate to a J310 JFET that is used to buffer the RF signal:



(Click to enlarge)

The signal-in passes through a potentiometer to the J310's gate. The J310 source connects to a capacitor and resistor, both tied to ground. The drain is connected both to +VDC (through an inductor) and to signal-out (through a capacitor).

One thing that I noticed is that, when the circuit is powered, the resistor on the J310 source (a 22-ohm SMD) is *really* hot. It still tests within tolerance, though.

Some things I'm wondering:

-- By itself, does the resistor heating up indicate a probable failure of the J310?

-- Are there tests I can do on the J310 with it still in the circuit? I wasn't sure if my Fluke 77 IV digital multimeter would put out enough voltage in resistance-test mode, and whether its results might be affected by other components.

-- Any other suggested approaches? I have some extra J310's, and if all else fails I can swap one in. But it would be nice (and a good learning experience) to test whatever I can first.

[ledtester]

What is the supply voltage? And what is the voltage across the 22R SMD resistor when it gets hot?

As for testing your J310...

Do you have one of those "cheap component testers"? They are very handy devices and will give you a base level check of your component:

https://youtu.be/7Br3L1B80ow

Secondly, here's a way to use a multimeter to perform some checks on a JFET:

https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/

Then I would rig up some simple JFET circuits like a switch circuit:

http://www.learningaboutelectronics.com/Articles/N-channel-JFET-switch-circuit.php

and an amplifier circuit:

https://youtu.be/Kt1D-iqRVtg

This is where it is nice to have some sort of adjustable signal source.

[FrankO]

Thanks for the comments. The unit's overall supply voltage is 12vdc, and I'm seeing 10.7v on the J310's drain. Voltage across the resistor is about 6.8v.

I do have one of those GM328 "what's-it?" testers, and checked all of the components including all the SMDs before soldering them in, and ran continuity checks afterwards.

At this point I'm leaning toward unsoldering the J310 and testing it outside the circuit and/or replacing it with a fresh one. The 22-ohm SMD resistor has dropped about 1.5 ohm over the course of briefly turning on the unit for testing. I wouldn't want to turn it on too much more if the resistor is going to get fried.

[Manul]

First, this is not exactly a buffer, it is amplifier. Second, source resistor is used for "self biasing" of JFET. It is way too small, because you put the JFET almost at the high edge of the usefull G-S characteristic. Depending on the supply voltage you are likely even exceeding maximum power disipation of the J310 (which is 350mW I believe). I do not see a reason to bias idle point so high.

If 22R ressitor gets HOT, you can be sure, that J310 is already destroyed (shorted).

If you actually want a buffer, source follower may be a better idea - no voltage gain, but low output impedance (current gain).

[ledtester]

Thanks for the comments. The unit's overall supply voltage is 12vdc, and I'm seeing 10.7v on the J310's drain. Voltage across the resistor is about 6.8v.

Well, the resistor is dissipating 6.8*6.8/22 = about 2 Watts - no wonder it's getting hot. How big is your SMD resistor? Even the 1206 size have a power rating of only 0.25 W.

If the drain is at 10.7V and the source is at 6.8V, then the JFET is dissipating:

    (10.7-6.8 ) * (6.8/22) = 1.2 Watts

which is way above the 310 mW rating (datasheet:  http://danssmallpartsandkits.net/J309-D.pdf )

I do have one of those GM328 "what's-it?" testers, and checked all of the components including all the SMDs before soldering them in, and ran continuity checks afterwards.

At this point I'm leaning toward unsoldering the J310 and testing it outside the circuit and/or replacing it with a fresh one.

I would desolder the J310 and test it in the GM328 - wouldn't be surprised if it's cooked though. It's also a good idea to build up a collection of test circuits you can plug components into.

[FrankO]

I went ahead and swapped out the J310. The voltage on its source is now a fraction of a volt, and the circuit is functioning normally.

Regarding Manul's points about the component selections, I should mention that this is not a circuit of my own design. It's a kit that I assembled.

In any event, I'm happy to have it resolved, and nice to have learned some things about JFETs.