How come transistor beta hFE test is only on cheap DMMs?

[turbo!]

The mass produced BT-830B or whatever that DMM was called has a dedicated terminal block for shoving in epitaxial transistors to measure the hFE gain. There are a few slightly better DMMs with this feature, but as far as I know, there is no higher end DMM with this. Why?

[Richard Crowley]

The simple answer is that it is easy to make a "quick-n-dirty" (AmEnglish) or "cheap-n-cheerful" (BrEnglish) h_FE measurement.
All the basic functionality is already there (current sources, current and voltage measurement, etc. etc.)
So it is a matter of adding another position to the function/range switch and a 12-cent socket.

Presumably, people who buy more expensive test equipment have separate gear for measuring semiconductors and don't need a low-quality test function on their DMM.

[Jay_Diddy_B]

Hi,

I suspect that one of the other reasons is safety. The transistor socket will be connected to the one of the meter terminals, probably the COM Terminal.

I personally would NOT use a meter with a transistor tester on any circuit that is above 48V.

The cheap meters don't really care about safety.

Go ahead and get one of the cheap meters for testing transistor, but don't use for measuring high voltages.

Regards,

[Wytnucls]

The safer solution is the one found on the UT61A: An additional removable module just for that purpose.
It is only functional on the cheapest model of the range, which has an hFE range position.
I would imagine that the more serious electronic buff would rather consult the datasheet, than rely on that feature.

[wiss]

I would consider buying a cheap multimeter just to get that feature, none of my Keithley, Fluke, Solartron, Prema or HP's got that...

[tautech]

I don't care if it is "quick & dirty" it is a must have for repair work.
Often one must substitute with old equipment, and while a datasheet may be available for a device, sometimes the device is not. Substitution time and gain needs to match old datasheet.
A pair often also needs to be matched for gain.
The A, B or C of a device, each can have a large range of gain and a quick test can identify the  extremes.
The often frowned upon VC99 does just fine for this as well as the old cheapies I have.
But one must have a good meter too.

[HighVoltage]

For this purpose I have the Peak Atlas DCA
But interestingly, even this one does not have a small socket to plug in a transistor for a quick test

Actually none of my multimeters have such a socket or test capability.
If safety is a concern, they all could have it done with an adapter.

But then, imagine such a socket on a Fluke 179
It would look ugly and the nice streamline design would be disturbed.
May be that is the main reason.

[Jay_Diddy_B]

Hi,
For the original poster. If he would like to make a simple adapter to add H_FE measurement to a DMM here is a concept:



The adapter has two sockets, one for NPN and one for PNP. You can add parallel sockets for SMD and thru-hole transistors.


The results of simulation show how the meter reading varies with the gain of the transistor under test:



There is no need for a power switch, with empty sockets no current will flow.



Jay_Diddy_B

[dannyf]

Because hFE is not a constant number so high-end meter makers may think lowly of such one-point measurements,

The simplest measurement would require 1 resistor (two if a digital meter is used) + a current meter: put the resistor in the base of the transistor (make sure that the resistor is of sufficiently high value, 330k+ for a small signal transistor), and tie it to Vcc (NPN) or GND (PNP), and insert the current meter into the collector. The base current is approximately (Vcc - 0.7) / base resistor, and divided that into the current reading gives you hFE.

[Jay_Diddy_B]

Because hFE is not a constant number so high-end meter makers may think lowly of such one-point measurements,

The simplest measurement would require 1 resistor (two if a digital meter is used) + a current meter: put the resistor in the base of the transistor (make sure that the resistor is of sufficiently high value, 330k+ for a small signal transistor), and tie it to Vcc (NPN) or GND (PNP), and insert the current meter into the collector. The base current is approximately (Vcc - 0.7) / base resistor, and divided that into the current reading gives you hFE.

Dannyf,

Isn't that exactly what my circuit does? A few extra resistors adjust the scaling so the reading in mV = gain

Since posting I realized the LTspice model can be simplified, This is only a technicality the results are the same.



In 20+ years of electronics, I don't recall having to know the HFE of a transistor with any precision. I have a Tektronix 576 curve tracer which can measure all the static parameters of a transistor.

I have needed to know if they broken, I have need to identify unmarked or transistor with house numbers. I have also matched transistors, this circuit is quite suitable for all these tasks.


Jay_Diddy_B

[krivx]

Because hFE is not a constant number so high-end meter makers may think lowly of such one-point measurements,

I think this is why. The very cheap meters (308 etc) don't even specify the currents used to test hFE.

[macboy]

I have a couple of cheap meters that I keep around just because they have the hFE function. The test current is unknown, so the hFE reading itself needs to be considered very approximate, but they are useful to find the pinout and the type of transistor (PNP/NPN), or to identify dead transistors.

[dannyf]

The very cheap meters (308 etc) don't even specify the currents used to test hFE.

For practical uses, you don't really need to know the precise values of hFE. So no problem (for me).

As others have pointed out, it can be used to identify dead / defective transistors, determine pin-outs, or types of transistors or if it is a transistor or not.

[madires]

Please check https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/ :-) Let's see if I'm able to list all BJT related features:
- NPN/PNP detection
- pinout detection
- hFE
- V_BE
- detection of freewheeling diode
- detection of freewheeling diode on same substrate (creating a parasitic BJT)
- detection of base-emitter resistor
- I_CEO

[janoc]

Just received one of these from eBay for about 15EUR delivered, there are tons of them posted from China for a few bucks. Not really worth building it - I can't make it for that price myself.

It works a treat for quick-and-dirty test or identification of an unknown part. Just try to buy one with the regular two line character display module, so that the firmware can be upgraded. Some have the more fancy graphical LCD and the standard firmware won't work with it.

[krivx]

The very cheap meters (308 etc) don't even specify the currents used to test hFE.

For practical uses, you don't really need to know the precise values of hFE. So no problem (for me).

As others have pointed out, it can be used to identify dead / defective transistors, determine pin-outs, or types of transistors or if it is a transistor or not.

hFE can vary by orders of magnitude with Ic, how do I make sense of a hFe measurement if I don't know Ic? Even a very coarse measurement is only useful if I have a known good component to compare with.

[dannyf]

how do I make sense of a hFe measurement if I don't know Ic?

Simple: if the approach doesn't work for you, it doesn't work for you.

However, the fact that it doesn't work for you doesn't mean it doesn't / cannot work for others. You just need to recognize that we have different needs and what doesn't work for you may work for me, and vice versa.

[krivx]

how do I make sense of a hFe measurement if I don't know Ic?

Simple: if the approach doesn't work for you, it doesn't work for you.

However, the fact that it doesn't work for you doesn't mean it doesn't / cannot work for others. You just need to recognize that we have different needs and what doesn't work for you may work for me, and vice versa.

No, I get it that it might work fine as a dead/maybe-ok test. I just would prefer if it was called something like that, when I started out I expected that my cheap meter that had a hFE mode would actually measure hFE in some meaningful way.

[dannyf]

in some meaningful way.

What's meaningful can be subjective: for what you consider meaningless measurements, others may consider it meaningful; or vice versa.

To me, it gets 80% of what I want in a simple / simplistic test. So that's pretty meaningful by my book,

[wiss]

Well, one would need an oven for proper hfe measurement then...

1 mA Ic is enough for this purpose.

[SeanB]

I have a few meters with Hfe measurement, and I have almost never used it, using instead a simple diode test that I find is more useful. A simple test jig that allows you to change the base bias in 2 steps is a lot more useful, as this allows you to get an idea of leakage current through the device. The simple test does not distinguish between a high gain or a very leaky device at all.

If I wanted to match transistors I always made a test jig that allowed me to use a few different base resistors so I could bin at a particular current then select through the biggest bin to find devices that matched at a different lower current then finally select the best match at the third lowest current. That way they are matched at least over 2 decades of current. Typically for a small signal device it would be at 50mA, 5mA and 1mA. For some power devices it was a cup of ice water and 3 leads, with the test done at 10A, 5A and 2A in ice cold water.

[mzacharias]

Working on consumer audio, I fairly often find signal transistors which exhibit a very low or non-existent gain and cause actual failures due to this type of fault. Although I exclusively use Fluke or Agilent meters for actual troubleshooting, I still keep a cheapie Soar manual range model with hFe socket and extender cable and use it fairly often, sometimes just to confirm my diagnosis of the fault prior to an extensive re-assembly.
Parenthetically, the cable was for a Viz (RCA) model that used the same socket. I knew that Soar had made the Viz model and when I saw the Soar rotary switch model on eBay which had the same socket, I figured it would work. It did, and I retired the Viz.