MMBT3904/6 as general purpose BJTs

[blueskull]

I've seen those BJTs listed by Nexperia as switching transistors, but I wonder if they can be used as general purpose pass transistors for linear regulators. Any suggestions?

I can see 3904 has lower max noise figure, higher current and higher fT, so I wonder what prevents it to be used as a general purpose BJT?

IMHO, MMBT3904 is not built based on a process that is strictly optimized for switching applications like PBSS series, so why it's branded such way?

Any ideas are welcomed and appreciated.

[soubitos]

I think MMBT3904/3906 and the lot, ARE general purpose transistors....do a google search "MMBT3904 general purpose transistor" and you will find them like that from mouser and other listings up to the manufacturers datasheets like on and the like

[T3sl4co1l]

Hmm, 3904 probably predates PBSS style optimizations...

A BJT is a BJT is a BJT; all obey the basic rules on hFE and transconductance.  The only differences are optimizations for certain features (like speed, saturation, fT, SOA..), and how the part is characterized in testing.  There might be no difference at all between [assortment of 2Nxxxx parts], except that 2N3904 is characterized for general purpose and switching use, 2N5088 for noise, 2N2369 for switching, etc.

Some manufacturers (e.g., ON Semi) go ahead and measure everything, but this should not be taken as an endorsement of other manufacturers' parts under the same numbering, because anything that meets the JEDEC 2N3904 (and such) spec can be sold as such, and those specs are quite basic.  Anything above and beyond, is manufacturer specific.

Likewise, there is very little variation in JFET characteristics, the overall V/I characteristics being nearly identical, except for two axes (pinchoff voltage and channel width), and various small optimizations (mainly for capacitance, hence BF862 being better than MMBT4392).  The primary difference then is characterization, with some being very gross (2N3819's 10:1 Idss spread), some selected for switching (Rds(on), like PN4392), some for RF performance (like 2N5486), and some for AF or general low noise (LSK389, etc.).  And here again, you'll see real differences -- parts that are not qualified for noise, may sometimes be unexpectedly quite noisy.

Tim

[BrianHG]

For premium grade, cheap high f matched pair transistors, I personally had great results with MMBTH10 / MMBTH81.  Note that they come at a premium of 4 cents a piece in quantity, 20 cents each for 10.

I know they are considered UHF transistors, but, I still use them in the audio band as excellent single ended class A line drivers and their sub ns delay from base to emitter make them great as an opamp buffer, using the output emitter sent to the feedback to drive hard a line level signal.  Note that 3906/3904 are also still very good, have a 4x maximum collector current, but, are nowhere near as fast or consistent gain VS frequency.

If you are making a single ended output driver, obviously go for only the MMBTH10.

Note that I have used 3906/3904 everywhere, even to buffer and line drive NTSC/simple VGA video.  They are really good and clean up to around 10MHz.  H10/H81 blow them away once past 15MHz.  I've used H10 driving the 75 ohm long distant in RGB 1080p video and even higher with excellent results.

[danadak]

Not an expert here, going out on a limb -

1 - A pass transistor in a regulator is essentially an emitter
follower. Literature discusses capacitive loading on emitter
followers. Contributing to this problem is ESR performance
of the output caps. You can derive this by examining reflected
input impedance, point at which it turns negative. There is a
paper on this "Analysis and Design of Emitter Followers at High
Frequencies", IEEE, that is useful.


http://www.ti.com/lit/an/sbaa045/sbaa045.pdf


2 - Of course the 2N3904 can only function as a pass transistor at
relatively low current/power, say compared to a TO220 packaged
type regulator.

3 - Ft does matter, in a regulator, generally speaking, one would
use a low Ft device because of (1) above. But there is a gotcha
in low Ft devices, excess phase shift occurring closer to rest of control
loop freq behavior. Not generally a problem in regulators as the
Ft >> Control loop dynamics. Usually just a minor consideration.

4 - Common technique of killing oscillation in emitter follower is
to add small amount of Re Z in base, like a 10 - 50 ohm type R.
Or use of ferrite beads on base lead.


Regards, Dana.

[fenugrec]

Haven't much to add to this discussion except some anecdotal evidence that 3904/3906 are indeed "general purpose". My Tek 2215 scope had a defective final stage in its vertical deflection amp; I used 2n3904's as a temporary replacement for the BFR96 pair (high-fT , low noise etc.). I still had sharp edges and full deflection with a 10MHz square wave, which is as far as I could test it.

[David Hess]

They are calling the MMBT3904/6 a switching transistor because they are specifying and guaranteeing the switching characteristics.  But they are really general purpose transistors and there is no reason they cannot be used in both switching and linear applications.  The datasheets for these parts are very complete.

I think it would be more useful if they also listed the transistors by process and fabrication but this data is not always available.  For instance fast saturated switches are gold doped and have much lower storage time than general purpose or even undoped RF parts.  Another kind of process and construction yields very low saturation voltage which is useful for high efficiency saturated switches in inverters and switching regulators.  Based on the datasheets, I think the PBSS parts are this type and as far as I know, Zetex was the first one to produce bipolar transistors like this.  They can be used in linear applications where their high hfe at high collector current is an advantage but are slower and more expensive for a given power rating.  Are they the lower power version of ring emitter transistors?  I don't know but it not seems unlikely.

[T3sl4co1l]

Last time I was shopping for BJTs I saw plenty in tiny DFNs like that.  I don't have any recommendations for that value specifically, but even if it doesn't have "3904" in the number, there should be plenty to choose from.  Consider Panasonic also (but check life cycle as they have soooo many part numbers in soooo many markets, and they're EOLing some all the time).

Tim