N Channel JFET Testing in ESR Tester

[SteveK]

I am testing some MPF102 JFET'S in my ESR tester. four of them identified as "TWO DIODES". The last one identified properly as an N CHANNEL JFET. I'm not sure how to interpret these test results. Are the ones showing up as "TWO DIODES" defective? My ESR TESTER USES A MEGA 328 MICRO CONTROLLER. i've also tested a 2SK152-3 JFET which I believe to be blown and it identified as TWO RESISTORS. I believe THAT indication to be correct. Please correct me, if i'm wrong. Any help with this would be very much appreciated.

 

[CDRIVE]

SteveK, welcome to EP.

I don't know why you call it an "ESR" (Equivalent Series Resistance) Tester. That's only one parameter that these devices test. I own two (Atmega uC based) Component Testers. They've been evolving over the years. Unfortunately the two versions I have both report an MPF102 JFET to be a Resistor.

I'm going to dig back through datasheets and documentation I have to confirm this but I believe the JFET is or was not a component that it could properly identify at the time of my purchase.

This technology has been steadily evolving though. I believe that if the latest version doesn't include JFETs they soon will.

Chris

 

[CDRIVE]

Here's some reading for you. In particular find "Measurement of JFET and D-MOS transistors" in the idex. BTW the author uses the word "deferred" when speaking of the Drain - Source terminals of the JFET. I believe it should read "differentiated".

3.6 Measurement of JFET and D-MOS transistors
Because the structure of JFET type is symmetrical, the Source and Drain of this transistores can
not be deferred. Normally one of the parameter of this transistor is the current of the transistor
with the Gate at the same level as Source. This current is often higher than the current, which can
be reached with the measurement circuit of the TransistorTester with the 680 resistor. For this
reason the 680 resistor is connected to the Source. Thus the Gate get with the growing of current a
negative bias voltage. The Tester reports the Source current of this circuit and additionally the bias
voltage of the Gate. So various models can be deferred. The D-MOS transistors (depletion type) are
measured with the same methode.

Chris

 

[HellasTechn]

I am testing some MPF102 JFET'S in my ESR tester. four of them identified as "TWO DIODES". The last one identified properly as an N CHANNEL JFET. I'm not sure how to interpret these test results. Are the ones showing up as "TWO DIODES" defective? My ESR TESTER USES A MEGA 328 MICRO CONTROLLER.

If the meter is like the ones sold on e-bay and gives you a 2 diodes reading then YES your FET's are BAD.

 

[CDRIVE]

The models I have were both purchased on Ebay. The first one (MK-168) came enclosed in a plastic case that includes a 9V jack to power it off a wallwart if desired. It's LCD is an older model (smaller) with no graphic capabilities that uses ASCII art to depict components. This model is reporting that the MPF102 is a resistor network that displays it something like this (2)---[135.5Ω]---(1)---[29.7Ω]---(3). Of course it doesn't display an Ω symbol.

The other model is a GM328A (V1.111) that's supposed to be a later model with a much nicer LGD. It is sold caseless. This model is reporting a single 29.7Ω resistor between pins 1 & 3.

Not long ago we had a discussion about these testers. Steve (our moderator) owns a version. Hey Steve, what does yours diagnose for JFETs?

Chris

 

[BobK]

Mine identifies an MPF102 as a single resistor.

Bob

 

[CDRIVE]

Thanks for your observations Bob. What do you think is a reasonable timeline that these testers will properly identify JFETs. I'm betting within a year.

Chris

 

[SteveK]

Here's some reading for you. In particular find "Measurement of JFET and D-MOS transistors" in the idex. BTW the author uses the word "deferred" when speaking of the Drain - Source terminals of the JFET. I believe it should read "differentiated".



Chris

Hi Chris,
i only recently discovered this device(ESR Tester). A month ago, I never knew it existed. All the listings that I see for this device almost always refer to it as an ESR Tester. So, I accepted that term. If i'm not mistaken, it was originally meant to test primarily Capacitors. please correct me on that if I'm wrong. Today, I tested 5 2SK715W JFET'S. They all identified properly. O, btw, thank you for the reading material.

Steve

 

[SteveK]

If the meter is like the ones sold on e-bay and gives you a 2 diodes reading then YES your FET's are BAD.

Hi Constantine,
Yes, got it on Ebay. Paid about 13 bucks for it.

Steve

 

[SteveK]

The models I have were both purchased on Ebay. The first one (MK-168) came enclosed in a plastic case that includes a 9V jack to power it off a wallwart if desired. It's LCD is an older model (smaller) with no graphic capabilities that uses ASCII art to depict components. This model is reporting that the MPF102 is a resistor network that displays it something like this (2)---[135.5Ω]---(1)---[29.7Ω]---(3). Of course it doesn't display an Ω symbol.

The other model is a GM328A (V1.111) that's supposed to be a later model with a much nicer LGD. It is sold caseless. This model is reporting a single 29.7Ω resistor between pins 1 & 3.

Not long ago we had a discussion about these testers. Steve (our moderator) owns a version. Hey Steve, what does yours diagnose for JFETs?

Chris

Hi Chris,
Mine indicates for as an example, for a 2SK715W, I=0.81ma and Vgs=0.57V. as well as the Pinout with GSD identified.
For one of my MPF-102's, shows a Diode between 1-3 and 2-3.
I tested a 2SK152-3 which identified as 2 Resistors.
Mine uses a 328 Chip. I'm not sure of the firmware version.
Steve

 

[(*steve*)]

It really depends on how good the tester is. maybe the "two diodes" have low tra

Not long ago we had a discussion about these testers. Steve (our moderator) owns a version. Hey Steve, what does yours diagnose for JFETs?

When I'm closer to home. I'm over 5000km (that's 2 states) away by road at the moment. Hard to get to the shed at the moment

 

[BobK]

Two states away, 5000Km? Here in Massachusetts, 2 states away is about 150Km

Bob

 

[(*steve*)]

Two states away, 5000Km? Here in Massachusetts, 2 states away is about 150Km

Yeah, I cheated a bit. There's no direct road route. it's only 3450km as the crow flies. It's not quite the two most distant points in the country, but it's pretty close!

 

[CDRIVE]

Hi Chris,
i only recently discovered this device(ESR Tester). A month ago, I never knew it existed. All the listings that I see for this device almost always refer to it as an ESR Tester. So, I accepted that term. If i'm not mistaken, it was originally meant to test primarily Capacitors. please correct me on that if I'm wrong. Today, I tested 5 2SK715W JFET'S. They all identified properly. O, btw, thank you for the reading material.

Steve

The history of the device is included in the PDF file I posted. The author released his schematics and source code to SourceForge as freeware years ago. He's also well aware of the existence of the many Chinese clones.

Chris

 

[HellasTechn]

All the listings that I see for this device almost always refer to it as an ESR Tester. So, I accepted that term.

Yes that's how most sellers on e-bay call it.

 

[CDRIVE]

Well its author/designer calls it a Transistor Tester, which it is and as he modestly states " and a little more". Personally, I think it qualifies as an invention because I know of no other device that automatically finds the pinout and device family of 2 & 3 pin solid state devices. At least not one that the average electronics enthusiast could afford.

Chris

 

[SteveK]

Actually, it does (to me anyway) a lot more! A very handy little "toy". I just received an order of 2SK345, and 2SK291 JFET'S. I was curious so I tested each type (5 of each) to see if they would identify correctly. Every one correctly identified. It really is a great little "gadget".
Steve

 

[SteveK]

And Chris, your right! You can install the FET backwards, it figures out the pin out and displays the result, VERY NICE!!
Steve

 

[CDRIVE]

And Chris, your right! You can install the FET backwards, it figures out the pin out and displays the result, VERY NICE!!
Steve

That's a primary feature. There is NO correct or incorrect connection to the tester. Any 3 pin device (that it's capable of testing) can be connected to it in any pin-out orientation. The uC analyzes the device under test and determines which device pin is connected to each of the three tester input terminals. You can mix and match the way you connected it and it will sort it out regardless of which device pin is connected to which tester pin.

Chris

 

[CDRIVE]

I found this during a Google search.
http://socrates.berkeley.edu/~phylabs/bsc/PDFFiles/bsc4.pdf
It's from Berkeley, so it's quite thorough. Though I'm surprised that within this document I didn't see any reference to the Drain / Source ambiguity encountered when determining the pin-out of JFETs.

If and when these Atemega or other uC based testers can properly analyze them I would imagine that a negative source (for Gate / Source bias) would be required. Even with that, JFETs may confuse the hell out the tester as this simulation demonstrates.

Check out the Drain / Source hookup between these two circuits. I inverted Q2 connecting the named pin "Source" to V+ and the named pin "Drain" to (-) GND. As you can see in these (Vg vs Ids) plots they're identical when wired either way.

The schematic symbol of the JFET in itself provides a clue to it's ambidextrous Drain / Source properties, as it's not differentiated symbolically.

Chris