News:

Forum may be experiencing issues.

Main Menu

transistor hFE measuring & matching mayhem

Started by jessenator, January 08, 2023, 02:28:57 AM

Previous topic - Next topic

jessenator

So to get ahead on a project I'm working on, I decided to get a transistor tester going for me.

I opted first for the $7 Harbor Freight special.


I know you get what you pay for (foreshadowing?), and I don't need it for anything other than hFE measurement, since I have better tools for those other jobs. I have no clue what sort of current is getting sent to the base, because, for example, a 2N3904 is measuring well over the DC current gain spec maximum of 300 (@10mA, per the spec sheet)

I wondered if something was f*@#y, so I looked up a video for using some breadboard to measure current two different ways, then doing some math to get the hFE value. I thought, hey, I'll do this to check the value and see by how much the cheapo meter is off.



I have the same setup on my bench: Same 9V (battery) source, same transistor, same value (470k) resistor, and I'm not getting anywhere near his numbers for current. With and without power to the emitter, I'm getting a mere 1.2uA, compared to his 18.2uA. I double-checked the rails, the resistor, the breadboard path, tried multiple samples of the same transistor, and still not getting anywhere close to the values it should, measured on either of my good meters.

I could try to suss it out, which I'll do only if it doesn't lead to more madness : P  Alternatively, shy of shelling out $100 for one of those Peak tools, is there 1) a way I can determine if my el cheaptastico multimeter is relatively close, or 2) a better tool for the job that's on the lower half of the price of that Peak tool?

Thanks

PS for now I'm really only concerned about Si jobbies; not fussed about germs.
[NOBODY CARED ABOUT THAT]

Haberdasher

I've got a cool thing called a MK-168 that was around $30.  If it broke, I'd definitely buy another one. And there are other similar gadgets out there.
Looking for a discontinued madbean board?  Check out my THREAD

FABBED PCB's FOR SALE:
Now carrying Matched JFETS

jessenator

Quote from: Haberdasher on January 08, 2023, 03:20:59 AM
I've got a cool thing called a MK-168 that was around $30.  If it broke, I'd definitely buy another one. And there are other similar gadgets out there.

sold  ...er, bought   ;D

Thanks for the recommendation.
[NOBODY CARED ABOUT THAT]

jimilee

Quote from: jessenator on January 08, 2023, 03:39:20 AM
Quote from: Haberdasher on January 08, 2023, 03:20:59 AM
I've got a cool thing called a MK-168 that was around $30.  If it broke, I'd definitely buy another one. And there are other similar gadgets out there.

sold  ...er, bought   ;D

Thanks for the recommendation.
I was just looking at those too. Pretty cool.


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

Haberdasher

Kind of a poor man's Peak, but I use mine all the time!
Looking for a discontinued madbean board?  Check out my THREAD

FABBED PCB's FOR SALE:
Now carrying Matched JFETS

matmosphere

I have a couple of these, and they work well for my needs.

https://www.taydaelectronics.com/esr-meter-transistor-tester-lcd-diode-multimeter-capacitance.html

You can find a version with a simple case on Amazon for a few bucks more.


jwin615

Quote from: dan.schumaker on January 08, 2023, 09:07:20 PM
I like this one.  Best part is that it is USB powered, no screwing around with batteries.

https://stompboxparts.com/tools-supplies/lcr-tc1-multi-function-tester/

I use this guy extensively.
I'd recommend a set of Dupont to test clips if you don't have them. They're good for a lot of other uses as well but make testing tranny's a bit easier.
https://a.co/d/06ljork

Just don't forget to always discharge caps before testing them. I put a piece of Al tape on the top of mine for doing so.

peAk


jessenator

Got the MK-168 and finally had some time to break it in. I really like this tool, because I can put the component in any way, as long as I'm not repeating pin #s. Nice to just pop in, test, and move on to next one! Slick.

Decided to test the batch of transistors with both the cheap-o multimeter and the MK-168. The meter one was nearly consistent in how off it was, averaging about 39 higher than the MK:


Another thing I had proved to me was "you get what you pay for" with discount parts.

I have about 6 remaining Fairchild/Onsemi 2N5088s and tested those with the new tester (MK-168):


Then I tested the batch of the same transistor that came from Tayda Electronics in Thailand (no idea if they produce them or source them from inside their borders or beyond):


Yeah, I know the Fairchild sample size wasn't that large, but the spread in the other batch speaks for itself, I think. Just for kicks I ran the test on some other transistors and the known-brand samples were similarly and exceptionally close to the same values, only differing maybe 1 unit.

But like I say, these aren't going into a transistor ladder for like a phaser/flanger or frequency filter, so 1:1 pairing is less critical. Still, just amazing how wide the spread is...
[NOBODY CARED ABOUT THAT]

jessenator

Got some NOS Motorola 2N3904s (3903s incoming) and measured those. Much more decent of a spread, with a good portion sitting bang on at 141 hFE (weird number, but I'll take it)



My youngest spawn got a kick out of them being so uniform (he loves numbers) while we tested them.

Not that it's anything definitive, but I'm going to keep all of this stuff in a Google sheet for reference; this and anything else I feel that's worth recording values. Although the measurement exercise is strangely satisfying and I did a handful of other components just for kicks  ;D
[NOBODY CARED ABOUT THAT]

jessenator

Quote from: Haberdasher on January 08, 2023, 03:48:25 AM
Kind of a poor man's Peak, but I use mine all the time!

Hey, do you happen to know (or have figured out) what the test current for BJTs is on your MK-168? I've been told the Peak Atlas is 2.5 mA—from a small sample size, it seems the current is higher on our beige plastic rectangles, but if you've been able to figure out what it is, that'd be super useful.
[NOBODY CARED ABOUT THAT]