Hi there,
I am repairing a guitar tuner and suspect one FET is toast. It's 2sk433 (https://www.idc-com.co.jp/product/Search/Pdf/en/4/2SK433), rated -50V and 10mA. The problem is this exact number is very hard to come by, and ridiculously priced. I am thinking of getting an interchangeable FET from Farnell or similar. Question:
Can I use a similar n type 3 legged FET with higher amperage rating? Or higher voltage? What should I look for?
Many thanks for your help..

 

Yes, I would think any N junction FET with similar characteristics or somewhat high voltage/current specs should work.
If you post the one you are thinking of buying, we can give a better opinion.

 

You can contact the preferred vendor and ask for a cross from that part
to an equivalent vendor part.

Regards, Dana.

 

Discrete JFETs have never been very popular, so where they are used there is usually a special reason. They are sometimes used in situations where specific IDSS (drain current with gate shorted to the source) is important. Some applications require two JFETs that are quite closely matched in characteristics or where one differs from the other in a specific way. But there are also uses where, as with bipolar transistors, most of the variability of a particular type number of FET is largely "designed out" out of the circuit by selection of the surrounding passive components. Where JFETs are used for switches (signal selection, integrator reset, etc.) there is often moderate flexibility. Occasionally the gate (to source or drain) junction of a JFET is used as a very low leakage diode.

Without knowing the actual circuit where the FET is used it is hard to predict how a substitute, or even the same part number that has not be "selected" will perform. But it is pretty unlikely any damage would result from using one that is fairly different, so an experimental test with a low cost easy to get sub is probably worth it. I recommend looking for a sub with similar IDSS - note the original comes in four different subranges of a really wide range. The original has moderately high transconductance (3 µS), which may or may not be important. You probably aren't going to find very many possible subs because there just aren't a lot of JFETs on the market.

 

Looking at the datasheet (see Rds(ON) test circuit), I would say you would need to look for a depletion mode NFet, or possibly even a PFet with the source oriented appropriately.

 

Many thanks for your help. I am attaching the diagram. I've circled in red the FET, as well as the opamp which also went toast.
Any thoughts?

https://ibb.co/mRstCS

 

The failure of the FET should have no effect on the opamp.
Why do you think it's bad?

 

Please post the file here so we don't have to go to a site that will steal as much info as possible about every visitor.

 

Hello,

Here is the downloaded picture:



Bertus

 

The failure of the FET should have no effect on the opamp.
Why do you think it's bad?

Actually the opamp is toast 100% (burn marks w black residue), while the surface of that FET nearby is a bit 'bumpy', so I suspect it's toast too. The device powers on, but all the lights are on and no single button works, it's not responding.

 

Thanks, bertus.

That FET is just used as a high input impedance "source follower" from a 5 volt supply, Probably just about any N-channel JFET will work, though the DC operating point may shift a bit. This isn't likely to be a problem unless signal clipping results. I can't see where "M" comes from, but I assume it is some filtered bias voltage.
EDIT: found M - filtered 2 V

===
"... you would need to look for a depletion mode NFet ..."
JFETs are depletion mode devices.

 

Ah ok thanks. Any idea what could have caused that opamp (Ba10358f) to burn? Is opamp failure a common thing? My worry is that I will be treating the symptom, while it might be receiving overvoltage or something