BJTs: Hfe vs speed

[exe]

Hi!

Someone told me transistors with higher Hfe are inherently slower. Is this true? Could someone point me to an application note or something?

I wonder if there is a noticeable difference between, say BD139-6 (Hfe 40-100) and DB139-16 (Hfe 100-250). I'm mostly interested in using them in active region as emitter follower.

[Zero999]

I presume you're talking about DC Hfe? Otherwise a transistor with a GBWP of 1GHz will have more gain, than one with a GBWP of 100MHz at 500MHz.

To be honest, I don't know the answer to your question. It is true that RF transistors do tend to have a lower DC Hfe, but that might be because it's a lower design priority, than device physics. I hope someone with more knowledge of semiconductor physics knows.

One thing I do know is that power transistors generally are slower and have a lower DC gain, which is due to device physics: not everything scales equally.

[Benta]

This goes into transistor design and the target application for a transistor.
For transistors operating in their linear region, there is basically no correlation between h_FE and "speed".

But in transistors optimised for switching, the time from full saturation to full off and vice versa is crucial. In such devices, techniques like gold doping etc. are used, and these reduce h_FE.

So the correlation is true, but the causality is wrong. Optimising transistors for switching speed can reduce h_FE, but different flavours of BD139s will not be affected.

[David Hess]

As far as I know, there is no relationship between speed and hfe for the same transistor like the BD139.  Higher hfe transistors (of the same type) actually have *worse* characteristics other than base current but this is largely irrelevant.  For instance you would not want to use super beta transistors where output resistance (Early voltage) is important like in constant current sources.

As far as transistor fabrication, reducing minority carrier lifetime increases speed but lowers hfe so gold doped transistors are faster but have lower current gain.  So RF transistors and fast saturated switches which have low storage time are gold doped but general purpose and low saturation switching transistors are not.

[T3sl4co1l]

One of the primary determinants of hFE is base layer thickness.

When a transistor is diffused, a very small difference in time*temp results in the dopant going deeper into the material (or not).  Mind, the critical distances we're talking are microns here.  For a double diffused BJT, the device starts with the collector, from moderately doped substrate.  The base is diffused in, using opposite polarity dopant (i.e., N substrate, P base).  The concentration of P is more than N, so that P dominates; the N dopant is a contaminant that tends to reduce breakdown voltage, carrier lifetime and increase leakage current.  (Or, wait, is it just leakage?  It's been so many years since I calculated a BJT from principles!)  If that dopant goes deeper, it will be more spread out (more lightly graded junction = higher breakdown voltage), and the base's P concentration will be lower.  Then the emitter diffusion is made, at even higher concentration N (since it has to dominate over the base's P), and for shorter time (less depth), giving a sharper junction, and lower breakdown voltage.

This is fundamentally why V_ebo is low (typically 7V): the emitter concentration is high and the junction is thin, same way you make a low voltage zener diode.

(By the way, when I say "junction", I really mean "depletion region".)

hFE goes as base concentration * emitter concentration / collector concentration, roughly speaking (I forget what function is applied, if it goes as, say, sqrt(conc) or log(conc) or what).  So you want a lightly doped collector (which also gives a usefully high breakdown voltage), strong base, and very strong emitter.

hFE also goes as base thickness.  The carrier diffusion distance in silicon is on the order of 10 microns.  So a base thicker than this has practically no hFE: the emitter emits charges into the base, but they recombine before being collectored.  (Good example: the input structures of a typical 74HC gate are separated by distances this large -- the input pins' GND side (substrate) ESD diodes exhibit a small hFE between themselves, while the VCC side diodes do so between inputs of a given gate.  The two hFEs combined, of course, makes an SCR, which is where CMOS latchup comes in.  In 74HC, latchup typically occurs at quite high currents, ~140mA.)

You can't make the base too thin, because for one, it's hard to (especially in a diffused process), but also because the depletion region varies with voltage.  If the base is too lightly doped and thin, the collector junction can swallow the entire junction, and it then overlaps the emitter junction: this is called punch-through.  As collector voltage rises, base thickness falls, so hFE rises even further -- Early effect.  In the extreme, Early effect takes over completely, which looks like avalanche breakdown.

So now we get to the relevant part: a thin base is also slow.  You have a resistive sheet sandwiched by two capacitors.  If the sheet spans the entire width of the junction, uninterrupted, it doesn't go very fast: germanium alloy junction power transistors may be a good example of this, yielding fT in the 10kHz range.  Most power transistors (since the 60s, say) have an interdigitated structure, with lots of base perimeter making contact with metallization -- this keeps the maximum width of the base sheet modest, giving fT of a few MHz.

On the other extreme, RF transistors are all perimeter and hardly any sheet width: heavy doping is used to keep carrier lifetime down (giving Vceo typically 10-30V, Vebo 2 or 3V).  This also reduces diffusion distance, forcing smaller feature size.  The base and emitter are heavily interdigitated, giving low base resistance (Rbb') and a short time constant.  They're made of so much perimeter that the vertical (sidewall) and surface parts of the junctions contribute significantly to performance, not just the buried, planar part.  (This necessitates different SPICE model designs, offered in the enterprise grade simulators.)  DC hFE is typically lower, 20-100, but it extends to far higher frequencies: as far as I know, plain Si BJTs run fT up to ~10GHz (with SiGe HBTs taking over above, as well as PHEMTs and the like).

fT is normally assumed to be a dominant pole effect, so that means the transistor considers "DC" as anything up to (fT / hFE), or >100MHz for such an example.

Obviously(?!), this applies to other kinds of BJTs as well, not just diffused.  Triple diffused types can have lower resistance (the collector isn't a thick slab of lightly-doped resistance).  Epitaxial types have total control over layer thickness and doping profile (so, don't need to be subject to compromised mixtures of dopants -- lower leakage, better figure of merit). 

Tim

[exe]

Wow, that was very detailed, thank you! Now I'm starting to understand.

I have another question: which transistor is faster: BD135 (45V Vce maximum) or BD139 (Vce 80V max)? From what I understand, BD139 should have thicker depletion region to withstand more voltage. So, BD139 should be faster, but less Hfe, right?

[David Hess]

Datasheets do not normally show a variation.  The datasheets for parts intended for high hfe at low current like the 2N5088/2N5089 show a variation in hfe with collector voltage though.  Datasheets for gold doped fast saturated switches seem to show a pattern of speed versus collector voltage.

I doubt the variation is significant for more general purpose parts but if you want to find out, buy a big batch of the same transistor and measure the hfe, collector voltage, and speed for all of them.

[Benta]

I have another question: which transistor is faster: BD135 (45V Vce maximum) or BD139 (Vce 80V max)? From what I understand, BD139 should have thicker depletion region to withstand more voltage. So, BD139 should be faster, but less Hfe, right?

That's wishful thinking. Process parameters are so wide that there's no way to determine that. Basically BD135, BD137 and BD139 are all the same and grown on the same wafer and just assembled and stamped with whatever part number the market desires right now. I'd expect them all to be BD139s.
In earlier days, they would be tested for maximum Vce, but I doubt that it's done nowadays (Vce testing means increasing Vce until a certain noise level is reached, meaning you're close to avalanche).
For a power transistor costing a few cents, it's cheaper just to produce them to the max. spec.

For expensive high-end transistors it's a different story.

[T3sl4co1l]

Higher voltages tend to be slower, but within a family, that may not be a strong correlation.  Both hFE and Vceo vary independently across a family (though they may not be sold that way, i.e. in all 9+ combinations thereof).

Processes are also more consistent today than yesterday, so as Benta says, they may sell the same die in all parts.  Testing is a bit part of production cost and the less they have to test, the better (low leakage is an especially bothersome one that drives up cost).

1N4001-7 are definitely made this way; 1-4 are the same, 5-7 are the same, and the specs are divided in two groups.  You're not guaranteed to get a 400V-capable 1N4001, mind -- but it does seem likely based on reading between the lines.

Tim

[rfeecs]

hFE goes as base concentration * emitter concentration / collector concentration, roughly speaking

It should be more like hFE is proportional to emitter concentration / base concentration.

You can refer to equation (8.4.5) here:
https://people.eecs.berkeley.edu/~hu/Chenming-Hu_ch8.pdf

A typical good βF is 100. D and W in Eq. (8.4.5) cannot be changed very much. The
most obvious way to achieve a high βF, according to Eq. (8.4.5), is to use a large NE
and a small NB. A small NB, however, would introduce too large a base resistance,
which degrades the BJT’s ability to operate at high current and high frequencies.
Typically, NB is around 10¹⁸ cm^–3.

That chapter also describes other ways to improve hFE, like using graded base doping or a heterojunction or poly-Silicon emitter.  This can allow increasing the base doping and still getting a high hFE.

So in general hFE and "speed" are not directly related.