Matching transistors for discrete op amp (audio use)

Question

I am building some GAR2520 discrete op amps that require a matched pair of BC550(CTA) NPN transistors. I built a simple emitter follower with a 15 V supply, biased at about half that. I input a 1 kHz sine wave from a signal generator at 500 mV amplitude. I measured the average DC voltage, and AC amplitude across the emitter resistor, in hopes to grade them for matching. I have made a plot of AC voltage and a scaled-for-convenience DC voltage per transistor:

How important is it to consider both the DC and AC response of the transistor when matching for use in a circuit where "hfe-matched" components are specified? I can provide more information on the test circuit if needed. The discrete op amp is going to be used in a microphone preamp, where I obviously don't need DC capability, but I am still not sure in what way to weigh these parameters in the quest for matched transistors. Thank you.

Since it has been brought to my attention that I should not see gain from an emitter follower (duh), I realize that there was in fact no gain, the reality being that a 1000 mV Peak to Peak input resulted in an emitter follower output of ~840 mV Peak to Peak, depending on the transistor of course. Obviously the transistors still performed to varying degrees, but this was not the best circuit to check such a specification. I think I will build what was suggested by @Dave Tweed, and plot those results. Thanks for the input so far everyone.

Here is the ill-fated transistor testing schematic:

Multisim doesn't have a BC550 model, and from a very cursory look at the spec sheet it seems the BC547 is a good approximation. In the simulator of course the output is just barely lower than the input, as expected for the topology in the ideal case. It should be noted that the wave generator and the oscilloscope are both from an Analog Discovery USB unit.

If the circuit (schematic??) requires hFE matched transistors, according to the designer, why would you not simply measure the hFE and match according to that? — Spehro Pefhany, Oct 16 '15 at 05:52
I gave you some encouragement points for the nicely presented transistor data, but we need more detail (i.e. schematic) on this "GAR2520 discrete opamp". A quick google search turned out some how-to article containing PCB layout but not schematics, and I'm not willing to spend my time reverse-engineering that. — Fizz, Oct 16 '15 at 10:10
And as you've discovered, hFE at DC (usually written like that with caps for FE) can be rather different than hfe, which is usually the notation for AC. — Fizz, Oct 16 '15 at 10:23
Usually matched means that the beta (or Hfe) of the transistors has a similar value. I'm not sure what you're trying to do in your measurement. I would just use a transistor tester to sort the transistors on beta. Anyway, it would be interesting to see what the effect is of the transistors not being matched. — Bimpelrekkie, Oct 16 '15 at 10:24
Finding lower hfe than hHE is normal. See http://www.aoc.nrao.edu/~pharden/hobby/HG_DS1.pdf for more. — Fizz, Oct 16 '15 at 10:43
An emitter-follower configuration is a particularly bad way to compare hFE (or hfe). A common-emitter configuration with a small (or no) emitter resistor would be a much more sensitive way to measure this parameter. On the other hand, I don't understand how you're getting greater than unity gain out of an emitter follower. — Dave Tweed, Oct 16 '15 at 11:31
@SpehroPefhany I have a circa 1990 Tenma multimeter with some kind of transistor gain testing functionality. I tested all the transistors with it, and the values were between 12 and 13.5 on the LCD readout. I thought that was kind of strange, and I didn't really like the granularity of the measurement either, so I attempted to measure them in another way. — Dave Walwark, Oct 16 '15 at 14:34
@RespawnedFluff Thank you, I spent an hour on the plot last night. Well I am grateful to receive the carefully capitalized information from you that the DC and AC gains are designated differently. Had I known that I might not have even asked this question, but such is life. The PCBs are sold alone, or with components, and though there is nice assembly information, no schematic exists outside of closed hands for this particular circuit. Upon receipt of this information as to the case-sensitive nature of the gains, I opened the assembly manual and found that hfe is what is required to be matched — Dave Walwark, Oct 16 '15 at 14:41
@DaveWalwark hFE in that range indicates either a defective meter or a transistor connected incorrectly (that might be a plausible value for reverse beta- emitter and collector swapped. — Spehro Pefhany, Oct 16 '15 at 15:14
@DaveTweed Thanks for pointing that out, I didn't realize what the actual input amplitude was, and was operating under a bad assumption. I am grateful for the suggestion of a different test circuit, and will set it up and knock 'em down within a few days. I'll probably compare the two methods in the next plot too, and just leave off the DC measurement. — Dave Walwark, Oct 16 '15 at 17:32
@ Spehro Pefhany The meter is a funky thing, maybe it means 120 when it reads 12? It has another digit to spare on the LCD, so I don't know why that would be true, but like I said, it is funky. If the transistor was placed in the terminal in the wrong orientation it read out a value of 0, so I know that is not the issue. There did seem to be some correspondence between the cheap checker's verdict and the emitter follower measurements... — Dave Walwark, Oct 16 '15 at 17:36

score 2 · Accepted Answer · edited Oct 18 '15 at 22:39

Your test setup is basically a follower, which tends to reduce the effects of gain spreads. While this is good in production, it's not what's wanted here. In your job, the idle current is known, so you can set up a common emitter stage to pull the idle current. Also, the \$V_{CE}\$ the discrete op amp provides to the transistors can be determined. So you are setting up your test transistor under the same DC conditions that it will see in the discrete op amp. Now use a standard common emitter circuit with NO negative feedback. This implies using a fully bypassed emitter resistor. Now you will see more variability when you feed in your AC from your signal generator. Picking a pair will now be clear.

Matching transistors for discrete op amp (audio use)

1 Answers1