Transistor Testing Ports In Multimeters: Junk?

[Ginsly]

I have a Crenova 890z multimeter that seems to work well, but when using the little NPN/PNP ports to double check hfe the readings are wildly off of what a pair of TC1s say. I wound up with two TC1s last year, so in certain situations it's good to try both - they're typically very, very close to each other.

Some recent TC1 readings had me wondering what a different tester would say, so I broke out the Crenova. It was a little tricky to make good contact in the ports, but once I did the readings were WAY off what both TC1s said. I of course made sure the pinouts matched. Tried a handful of basic silicon transistors and each and every one was off by a LOT.

I wish I had a DCA55, as it would be the final word. Are multimeter hfe ports known to be bad, or is mine just a lemon?

 

[hamerfan]

Hfe depends on the current in the circuit (and on the testing device).

 

[jwin615]

TC1 and T7s even vary, testing at different voltages and thereby currents.
If you'll notice, on better datasheets, multiple hfes at different currents are listed. Sometimes they even give a plot.

 

[Ginsly]

Hfe depends on the current in the circuit (and on the testing device).

I put fresh batteries in the multimeter since I had considered something along these lines. This is good to know.

The question is, which would be more reliable for determining hfe for our purposes? The answer must be the TC1, but still, it usually reads something like “3.8v Battery” when testing… The multimeter runs on two AAs, so the max for that device would be 3v, likely less. Hmm.

 

[Ginsly]

I guess the question is: what is the optimal voltage/current for a testing device when trying to determine hfe and leakage for pedal building purposes? Obviously this multimeter's transistor ports won't be giving be useful information.

The TC1 seems well-liked around here, but may be a bit dodgy in certain Ge situations (?). Not sure how the rechargeable battery's variable power (until it dies) may affect readings.

The DCA55 seems to be the go-to. There looks to be a couple different versions of that model, though - one with a 12v battery, and one that is a "AAA Battery" model. Seems like each of them may come up with different readings based on the difference in power supply...

 

[jwin615]

I guess the question is: what is the optimal voltage/current for a testing device when trying to determine hfe and leakage for pedal building purposes? Obviously this multimeter's transistor ports won't be giving be useful information.

The TC1 seems well-liked around here, but may be a bit dodgy in certain Ge situations (?). Not sure how the rechargeable battery's variable power (until it dies) may affect readings.

The DCA55 seems to be the go-to. There looks to be a couple different versions of that model, though - one with a 12v battery, and one that is a "AAA Battery" model. Seems like each of them may come up with different readings based on the difference in power supply...

The DCA 75 will plot hfe across voltages when using the peak software. That's the "gold standard".
You can't get leakage measurements from the TC testers at all. I thought there was at least some correlation between Iceo and leakage but there's not. At least in the limited comparison I've done.

 

[Ginsly]

I thought there was at least some correlation between Iceo and leakage but there's not.

Oh! Really? I had read on some posts here that Iceo was leakage on the TC1… hmm that’s disappointing. Good to know, though.

So even the DCA55 isn’t really gonna cut it? Seems like that’s the one everyone uses…

 

[jwin615]

Oh! Really? I had read on some posts here that Iceo was leakage on the TC1… hmm that’s disappointing. Good to know, though.

So even the DCA55 isn’t really gonna cut it? Seems like that’s the one everyone uses…

I'm not 100% but I think the 55 and 75 are functionally equivalent except the 55 doesn't work with the software/is standalone only.
But yeah, I thought similar on Iceo but revisiting it with my recently acquired 75 proves otherwise. Had some transistors I thought were super leakey that turned out fine per the DCA75. I haven't seen any true relationship between the two but haven't really tried to find one either.
But yeah, for germanium, I think it's a DCA, building a keen tester or acquiring a vintage tester. I have an old heathkit IT-121 that tests leakage. Seems to work fine. I'm going to compare it to the peak and once confirmed it's 100%, I'll probably put it up in the trading post.
If you, or anyone, decide to go the vintage tester route, I'd advise doing some research first. Some of the older ones were better than others. It's also hard to find one that's been tested any further than a power on.

 

[Ginsly]

I still wonder if there’d be a difference in readings between the 12v DCA55 and the AAA battery version…

I suppose I’ll start savin! Those things are pricey.

 

[bronomo]

For anyone cross shopping the 55 vs 75, I believe the 55 lacks testing for most useful JFET parameters, they aren't exactly equivalent:

Peak Atlas

I just purchased a Peak Atlas DCA55, I figured that I didn't need the PRO model and didn't want to spend the extra money. Is this just as good for measuring transistors or did I make a boo boo?

forum.pedalpcb.com

 

[chris.knudson]

I'm not 100% but I think the 55 and 75 are functionally equivalent

I don't think they're that close. I have the 55, but I don't think it's accurate enough to match FETS.

 

[traumprinz]

Iceo IS the leakage people typically refer to. Not sure how accurate the TC1 is, but Iceo is what you are measuring when you measure leakage via (for example) the RG Keen method. Leakage is very tempersture sensitive and will increase if you handle the transistors, so make sure they've settled back to ambient temp. If you're measuring at the same temperature and get notably different leakage/Iceo results, one of them is wrong.

 

[Ginsly]

Iceo IS the leakage people typically refer to. Not sure how accurate the TC1 is, but Iceo is what you are measuring when you measure leakage via (for example) the RG Keen method. Leakage is very tempersture sensitive and will increase if you handle the transistors, so make sure they've settled back to ambient temp. If you're measuring at the same temperature and get notably different leakage/Iceo results, one of them is wrong.

Gotcha, thanks! It sounded like the TC1 wasn’t all that accurate with Ge leakage measurements, and possibly hfe. I try to measure in a 67 degree room, more or less, and always use tweezers.

Does anyone here count on the TC1 as a reliable measuring tool for anything Ge? For all I know it might be close enough, but it seems like that might not be the case.

 

[Ginsly]

Update: outside of the hfe ports, the Crenova is also giving me much different DC voltage readings on basic resistors than both TC1s I have. I think it may not be calibrated correctly or something... Time for a new mm, possibly. Argh. Suggestions? Can't shell out for a Fluke, I'm afraid.

 

[jwin615]

Same 2N404. All tweezers. 65F in the basement.

 

[darwin999]

Right, hFE is dependent on the parameters used in the measurement (e.g., Ic, etc).
Hence both meters might be correct if they are accurately measuring using different conditions.
(or, they could be in error...)

 

[Ginsly]

@jwin615 On one hand, it makes me feel like I'm not the only one... but yikes!

@darwin999 I'm starting to understand that, and the question I have is - which one is most relevant to pedal circuits? How would one pick a set of results if they had to (which they often do if they're selling them, I'd imagine).

If there's that much variation in testing/results, then almost all of the hfe/leakage figures out there would kind of be a crap shoot. Even the RG Keen method could be erroneous if the multimeter isn't great (my case, possibly) or the resistors used are out of spec. I'm surprised I don't see this mentioned more often considering how different the results are from different testers/tests. I guess any Peak Atlas would be the most reliable? But why?

 

[chris.knudson]

I have the DCA55 and the TC1, and they rarely agree.

 

[Ginsly]

I have the DCA55 and the TC1, and they rarely agree.

Gah! In that case, the "better" more expensive tester would be the one for builders to trust I'd think, but again - why?

 

[chris.knudson]

Gah! In that case, the "better" more expensive tester would be the one for builders to trust I'd think, but again - why?

If I had to do it over, I'd just buy a DCA75, a Fluke DMM, and a nice 2-channel oscilloscope with a waveform generator. Every tester I bought was to test for a very specific measurement, and each of them was great for what I bought them for. To be honest -- you don't need a lot of test gear if you're just building pedals or even doing basic troubleshooting. You'll get more mileage put of an audio test probe than any bench test equipment. However, when you get into stuff like matching FETs or analyzing a PWM signal, there's other gear that comes in handy. I used to do a lot of work with analog video, and the high-end scopes were a must. On the audio side, your ears are are the best piece of test equipment you have.