Hello all! I'm just having some trouble when trying to get some experimental data from a 2N3819 transistor from FAIRCHILD. The setup is given by the picture below. My ammeter should display I_DSS. The current I get goes from 500 uA to 300 uA a few seconds later.

I do not understand at all what's happening, the datasheet from Fairchild predicts I_DSS to be at least 2 mA (cf. https://www.fairchildsemi.com/datasheets/2N/2N3819.pdf) . Some ideas of what's happening ?

Thank you all, Théo

 

Sounds like you have a dead/fake FET, assuming you have it wired correctly and the battery is fresh.

 

Thank you for this answer. Do you know where I could get some 2N3819 (or equivalent) working properly (audio application) ? Thanks, Théo

 

Replace your meter with a 100 ohm resistor and measure the voltage drop across the resistor.

You should also note that \( \small I_{IDSS} \) value is specified at \( \small V_{DS} = 15V\).

 

Digikey has 2,660 of them.
RS Components also has them in stock.

 

TDo you know where I could get some 2N3819 (or equivalent) working properly (audio application) ?

Jameco has them for $0.39 each in quantity 100:
https://www.jameco.com/z/2N3819-Major-Brands-JFET-Small-Signal-N-Channel-Transistor_210622.html

 

Hello dl324, I did the setup you indicate, the voltage drop across the resistor is about 40 mV (so some 400 uA through it). I'm actually living in Europe, I'm looking for some European stores selling these FETs not too expensive.

 

The RS Components link in post #5 is for their French outlet.

 

I did the setup you indicate, the voltage drop across the resistor is about 40 mV (so some 400 uA through it).

I'm not a fan of measuring current with a meter unless there are no other options. When you use DMMs to measure current, you need to know about the shunt resistance. Some say to use burden voltage, but it's specified as a maximum voltage and it's more accurate to use the shunt resistance.

For example, my Simpson 467 manual says burden voltage on the 200uA scale is 0.25V max. But they also show that the shunt resistance on that scale is 1k ohm; so max burden voltage is probably closer to 200mV. What's more informative is that knowing the shunt resistance, I can calculate the drop for any current.

My Fluke 8010A also shows 1k ohm for the shunt resistance on the 200uA range. It also gave this method for correcting for shunt resistance:

I'm actually living in Europe, I'm looking for some European stores selling these FETs not too expensive.

You didn't include your location, so I picked a source that I use. Moderators can see your IP address, so they can approximate location.

What's your definition of not too expensive? What did you pay for the devices you have and where did you buy them?

 

I have never had a problem measuring the Idss of 2N3819's. In fact, I did this very recently.

I am guessing that you are measuring the I_dss for either gain control or filter control. In either case you may have to match the transistors. This means that you might have to start with a handful of them to get just a few that match. In other words, order a bunch of them.

 

I'm not a fan of measuring current with a meter unless there are no other options. When you use DMMs to measure current, you need to know about the shunt resistance. Some say to use burden voltage, but it's specified as a maximum voltage and it's more accurate to use the shunt resistance.

For example, my Simpson 467 manual says burden voltage on the 200uA scale is 0.25V max. But they also show that the shunt resistance on that scale is 1k ohm; so max burden voltage is probably closer to 200mV. What's more informative is that knowing the shunt resistance, I can calculate the drop for any current.

My Fluke 8010A also shows 1k ohm for the shunt resistance on the 200uA range. It also gave this method for correcting for shunt resistance:
View attachment 141356

You didn't include your location, so I picked a source that I use. Moderators can see your IP address, so they can approximate location.

What's your definition of not too expensive? What did you pay for the devices you have and where did you buy them?

I paid 10 2N3819 FETs on ebay 7€ (the ones I have some trouble to get I_DSS). I paid one 2.50€ in an electronics warehouse in my town (and this one works brilliantly).

 

I paid 10 2N3819 FETs on ebay 7€

Did they happen to ship from China?

I won't buy anything on eBay that ships from China unless I assume it's counterfeit and that counterfeits are okay. I do this for tools where I'd rather pay $10-20 for a knockoff of a tool that would cost 10 times that from a reputable seller. That strategy has worked fine for crimpers, computer cables, wires, and connectors. But I wouldn't buy any semiconductors.

 

There are entire cities in China that make fake electronic parts that are sold on ebay.