Jellybean hobbyist general purpose transistors..

[Vtile]

I have started to think that I stock a few hundred
 general purpose through hole transistors (an investment of a few tens of bucks). To be sure I will have them to tinker around as to me it seems that the jellybean variety is vanishing or vanished already from THTs.

I'm mostly thinking of small signal, low freq (<200MHz), throw in buffer etc. general purpose uses of tinkering and maybe slobby repair or hack here and there.

BJTs:
BC547B(npn) & BC557B(pnp) comes to my mind.
Is there other worth to look that are still readily available.
Darlington models to maybe look at?
.
JFET
Is there cheapish (<$0.50 ) replacement for 2n4117 other than mbf4117(smd). It seems to be pretty alone in fart detector range.

MOSFET
For logic level switching use. Any ideas, in leds/signals on/off range of power. Should be also as robust as possible for ESD and ideally should have the protective diodes. 2N7000 ?? or were it JFET&BJT hybrid, hmm.

Known matched pairs (for that random differential stage) that wouldn't cost arms and legs and could still be found as THTs.    ..none or should I look from IC listings..

Any other than TO-92 cases to look at for jellybean THT parts. Metal cans seems to be premium only now.

[rstofer]

2N2222A
2N3904,5,6

[Cerebus]

BJTs: There's always the old 2N3904 and 2N3906, widely available, cheap as chips, both quite quiet (NPN 1.8 nV/sqrt(Hz)), easy to get through hole and SMD versions, 300 MHz F_t.

JFETs: In the low leakage JFET arena, nope nothing cheap. As you say, MMBF4117 is widely available at low cost but that's about it. PN4117 can be had in TO-92 but expect to pay around £1.50-£2 GBP each in small quantities. 2N4117A is also still available at similar prices. Both from Linear Systems.

And if Sir needs to ask the price of dual monolithic JFETs then Sir might find them to be *ahem* beyond his budget. I'm seriously considering a bulk purchase of some dual JFETs and laying them down, like a fine port, for the next generation.

J113 in most packages and PN/MMBF4393 in most packages are easy to find and inexpensive. Both good switches and fair amplifiers up to VHF at least.

MOSFETS: BS170 is my go to for a discrete, breadboardable N-Channel MOSFET in TO-92. BS250 for a roughly complementary P-Channel. If you can find a small signal discrete MOSFET with a separate substrate terminal (so you can dodge the body diode problem) in current production at sane prices, I want to know about it.

[T3sl4co1l]

FWIW, low leakage parts aren't jellybean, it's kind of contradictory there.

I go with:
2N3904/6, 4401/3 for more current, and PBSS303NX/PX for even more.  Change 2N/MMBT when SMT.

Lower current, and faster: MMBTH10/81, BFR92A (though this exact one I think is obsolete now, so pick another similar part instead).

Higher power: TIP31C/32C, 2SD1273 (hah, not jellybean at all, but I picked up a few before they went obsolete), MJE15020, FJPF13009 (high voltage)

MOSFETs:
2N7002, BSS84
NDT3055L
IRF510, 540, 640, etc.
IRFZ46N
STP6N60M2
etc.

I rarely use JFETs, but I've picked '4393, '5486, J105, and such before.  Others (who are actually experienced with low noise wideband design) are very fond of BF862 (which is now going obsolete, but there are newer equivalents from On Semi).

Tim

[JustSquareEnough]

while may not always hold true i've found it helpful to learn what the "jelly bean" part # of a particular component is by going to mouser or digikey, go to the components category, sort by either lowest price OR stock on hand.  doing this on the BJT will see 390x that were mentioned already at the top of the list.

[jaromir]

BC547B(npn) & BC557B(pnp) comes to my mind.

I tend to use BC327/337, as those have higher Ic (800mA versus 100mA), being more useful for switching loads like relays or common anode/cathode of MUX-ed displays.

[CJay]

BC639/BC640

BC54x/BC55x

BD139/BD140

TIP41/TIP42

BS170, 2N7000 and I've a small stock of VN10KM

As T3sl4co1l says, I don't often use JFETs but have a small stock of them, I'll be looking for an alternative before they run out, BF862 and newer equivalents/versions look very promising.

[David Hess]

I would go with the 2N4401 and 2N4403 as the general replacements for almost all small signal general purpose bipolar transistors unless working above 30 volts a lot like with higher power audio amplifiers in which case the 150 volt 2N5401 and 2N5551 would be more suitable.  But it is really difficult to go wrong with other choices like the PN2222A/PN2907A and BC327/BC337.  Be warned that the BC series uses the CBE instead of more common ECB pinout.  These are all higher current parts which can replace most lower current parts however this is also an advantage in switching applications where hfe will be more consistent.

For matched pairs except in precision applications, I match them myself by measuring Vbe at the operating collector current (collector and base shorted) and thermally bonding the pair together.

Darlington parts can always be built as needed.

I agree with T3sl4co1l's commendation for the MPSH10/MPSH81 and MMBTH10/MMBTH81 as faster TO-92 and surface mount parts.  Just do not expect RF transistors to perform well as fast saturated switches in switching applications.

Get a small signal schottky diode like the 1N5711, BAT41, or BAT83 for use in baker clamps if you want to improve switching.  The BAT41 with its 100mA current rating can do double duty in low current switching regulators.

For 1 amp rectifiers, I tend to standardize on the fast recovery 1N4937 but there are lots of choices.

The D44H11/D45H11 are 70W 80V 10A TO-220 bipolar power transistors suitable for regulation, audio, and switching applications with an Ft of 50 MHz.  The MJE172/MJE182 are similar 12.5W 80V 3A transistors in the smaller TO-225 package.

The situation with JFETs is annoying.

There are too many MOSFET options for me to make a good recommendation.  The problem with many of them is marginal switching on 3.3 or even 5 volts; the true logic level ones are more expensive.

[CJay]

Be warned that the BC series uses the CBE instead of more common ECB pinout. 

This is not always true.

BC639/640 are ECB for instance.

Cheap Chinese transistor tester is a very handy thing to have on the bench.

[chris_leyson]

I think at one time there were two different pin-outs for the BC184/214 so they are parts I would avoid.
 

[Zero999]

BC547B(npn) & BC557B(pnp) comes to my mind.

I tend to use BC327/337, as those have higher Ic (800mA versus 100mA), being more useful for switching loads like relays or common anode/cathode of MUX-ed displays.

The BC547 & BC557 have a higher current gain than the BC327 & BC227. I like to get the BC547C and BC557C because they have a high beta and aren't much more expensive than the B variants.

[capt bullshot]

As stated, there are too many MOSFETs.
For small signal purposes up to driving LED matrices, IMO BS170 / BS250 are a good choice (I've done few hobby projects using them in the past)
For some more current / voltage: IRF540, IRF840 (these are not logic level)
For logic level: IRLZ44
These are parts that I've used in the past, were easy to obtain and quite common back then.

[Vtile]

Thank you all for the information. I see what I can find with decent overall cost (product+shipment+handling+taxes). 
Good points that I also need to go through.

For dual JFETs the only one I do know is the 2N5909 for (serious) instrumentation. I kick myself that I didn't get two (to just see what kind of animal it is) when they still were locally available way below ebay prices. Not that I would need one, but..

The 4117 were my list just because it is what it is and it still is available in the discrete form. Unfortunately the A-subtype is not hobbyist friendly priced in any form, while MMBF4117 is. (unfortunately in SMD, but luckily SOT23 is still somewhat usable, without PCB and professional equipment.)

It is true about handpicking and doing Darlingtons and pairs when needed (when you have the stock where to pick!!!). I just thought if there is known decent general types (BJT and/or JFETs).

I'm personally accustomed to check the leg order as I traditionally have had that one piece of every transistor type pulled out from that random device (starting from ancient OC-series). So yeah a slight leg order variation doesn't put me (personally) down. I also have a bag full of old free soviet/VEF-riga parts including about ½kg of transistors my father did bring to me in late 90's when he traveled in the Baltic, but they are so wild and random. True also, if someone doesn't have the "$20 transistor tester", get one, I still wonder how I survived before.

[Vtile]

BC547B(npn) & BC557B(pnp) comes to my mind.

I tend to use BC327/337, as those have higher Ic (800mA versus 100mA), being more useful for switching loads like relays or common anode/cathode of MUX-ed displays.

The BC547 & BC557 have a higher current gain than the BC327 & BC227. I like to get the BC547C and BC557C because they have a high beta and aren't much more expensive than the B variants.

BC547/557 is interesting as the A type have the least gain, while the C type the most. I settled in B-type in my own choice (as now), because the pricing, but let see what is the final decision.

[jaromir]

BC327/337 is also binned by manufacturer by gain - see here https://snag.gy/fzlU1o.jpg taken from  http://www.mouser.com/ds/2/149/BC337-193546.pdf

[Zero999]

If you want low cost, then the BC327 and BC547 with no prefix will be cheapest but the gain is pot luck.

Use the BC327 & BC337 for switches. The Hfe is specified with V_CE = 1V, which is low enough for most applications.

Use the BC547 & BC557 for linear amplifiers. They have a higher bandwidth than the BC327 & BC337 and I believe are lower noise too.

[David Hess]

BC547B(npn) & BC557B(pnp) comes to my mind.

I tend to use BC327/337, as those have higher Ic (800mA versus 100mA), being more useful for switching loads like relays or common anode/cathode of MUX-ed displays.

The BC547 & BC557 have a higher current gain than the BC327 & BC227. I like to get the BC547C and BC557C because they have a high beta and aren't much more expensive than the B variants.

The BC547/BC557 have higher current gain at lower collector currents, much lower collector currents.  Where they are useful is low input bias and low collector current amplifiers where the input impedance is higher.  The 2N equivalents are the 2N5087/2N5089 which are designed to operate at an even lower current.  You can identify transistors like these by the test current used for their hfe and noise specifications:

BC639/BC640 - 150mA Test Current
2N4401/2N4403 - 150mA Test Current
BC227/BC327 - 100mA Test Current
2N3904/2N3906 - 10mA Test Current
BC547/BC557 - 200uA Test Current (from noise figure)
2N5087/2N5089 - 100uA Test Current

If I was doing low level audio work or building log/antilog amplifiers, then I would add the BC547/BC557 or 2N5087/2N5089 to my list of jelly bean parts.

Use the BC547 & BC557 for linear amplifiers. They have a higher bandwidth than the BC327 & BC337 and I believe are lower noise too.

The BC547/BC557 have lower current noise at low collector currents so may be used in higher impedance circuits.  The BC327/BC337 would be more appropriate in low impedance circuits where voltage noise needs to be lower at the expense of current noise although they were intended as output drivers.

[Vtile]

Thanks..

Now I can navigate in this transistor jungle much easier.

These coding methods started to hunt me, here is a simple explanation for Bxxxx types.

http://www.audiobr.com.br/old/forum/kb.php_mode=article&k=267.html
or
http://www.radio-electronics.com/info/data/semicond/bipolar-transistor-bjt/numbering-codes.php

ProElectron Coding

An Introduction.     

Once upon a time, long ago, British and European transistors were commonly coded as valves: 0 = no heater, C = triode. However it was quickly realised that everything would end up as an OC something or other (likewise with diodes, which would all become OAxx) and could be confused with existing cold-cathode valve types. So the ProElectron organisation developed a new system of coding semiconductor devices and the one in use today. Basically, British and European transistors are issued with a unique combination of letters and numbers.

The first letter identifies the semiconductor type:

· A = Germanium
· B = Silicon
· C = Gallium Arsenide
· D = other compound semiconductor material

The second letter indicates the intended use:

· A = Small signal diode
· B = Varicap diode
· C = Small signal LF transistor
· D = LF power transistor
· E = Tunnel (Ersaki) diode
· F = RF small signal transistor
· K = Hall effect device
· L = RF power device
· N = Optocoupler
· P = Radiation sensitive device (e.g photo transistor)
· Q = Radiation emitting device (e.g LED)
· R = Low power SCR
· T = High power SCR or triac
· U = High voltage switching transistor
· Y = Rectifier diode
· Z = Zener diode

So the European system actually have some information coded in, compared to JEDEC 2Nxxxx codes.

[T3sl4co1l]

the 150 volt 2N5401 and 2N5551 would be more suitable.

Or the similar MPSA46, or whatever they are, from that family.  MJE350 and complement are also quite popular for audio (driver stage), though the datasheet is sorely wanting.  I think On Semi makes a detailed datasheet?  Or, there are equivalent parts with good data out there, shop around.

I agree with T3sl4co1l's commendation for the MPSH10/MPSH81 and MMBTH10/MMBTH81 as faster TO-92 and surface mount parts.  Just do not expect RF transistors to perform well as fast saturated switches in switching applications.

Also, beware of oscillation, especially as you go over fT > 1GHz.  Keep ferrite beads handy for use on the base or emitter pin.

Get a small signal schottky diode like the 1N5711, BAT41, or BAT83 for use in baker clamps if you want to improve switching.  The BAT41 with its 100mA current rating can do double duty in low current switching regulators.

I'm partial to BAT85 and BAT54 (especially BAT54S, handy ESD clamp for logic I/O!), but they're all fine.

And BAV99 for analog / low leakage ESD clamping.

UF4004/7 for fast, higher voltage diodes, or SiC schottky for higher current and high speed, or Si schottky for low voltage and high speed.  Protip: junction diodes are good for snubbers as they have lower capacitance.  The forward recovery voltage (and thus peak voltage overshoot) isn't usually a problem in such application, but the capacitance is.

There are too many MOSFET options for me to make a good recommendation.  The problem with many of them is marginal switching on 3.3 or even 5 volts; the true logic level ones are more expensive.

FWIW, this is physics: MOSFETs rated 30V and up are physically different from those below.  The Rds(on) tempco is large (usually 1.8-2.5 times higher at 150/175C than at 25C), and the gate threshold and turn-on-ness (i.e., transconductance) isn't very good.  You can always adjust threshold (from negative Vgs(th) depletion mode, to enhancement with Vgs(th) up to 5V), but you can't adjust the slope of how fast it turns on.  Point being, you can get "logic level" power FETs, with Vgs(th) around 0.8V, and they're usable at 5V drive, but they still turn on harder if drive is stepped up to 9V, say.  They're also very slow at 5V, because drive current is limited by series resistance.

Devices 20V and under, however, have a shallower Rds(on) curve, and much more transconductance.  12V devices are effective at 2.5V logic level and below.  There's one part out there with Rds(on) in the microohms (for battery management)!

So if you have an application that's on the borderline between choosing 20 and 30V devices, try to use 20V.  To keep the circuit safe, put more effort into controlling peak voltages (add snubbers, use slower commutation, or a resonant design).

Tim

[orolo]

The BC547 & BC557 have a higher current gain than the BC327 & BC227. I like to get the BC547C and BC557C because they have a high beta and aren't much more expensive than the B variants.

I also stock the C series for the same reason, and the 'low noise, higher voltage' 550/560s since there is no great price difference.

But there is a catch, that I first heard of from Bob Pease: the higher the beta, the lower the output impedance. It can be seen in this datasheet: the 547A has 55.5k typical output impedance, the 547B 33.3k, and the 547C 16.6k, and it could go as low as 9k!. This can be very important when building current sources/sinks, active loads, etc. The beta-output impedance relationship seems to be (rougly) inversely proportional. This has led me to think seriously about stocking low beta versions of the transistor.

The Vishay datasheet for the PNP parts quotes the same output impedances for the A,B,C series. IIRC, in the Art of Electronics, data suggested that PNP parts often have even lower output impedance (lower Early voltage).

[T3sl4co1l]

But there is a catch, that I first heard of from Bob Pease: the higher the beta, the lower the output impedance. It can be seen in this datasheet: the 547A has 55.5k typical output impedance, the 547B 33.3k, and the 547C 16.6k, and it could go as low as 9k!. This can be very important when building current sources/sinks, active loads, etc.

Only relevant to pure current mirrors without degeneration.

Tim

[bd139]

Jellybeans for me. I don't do SMD really. Well I do, but it's done with TH parts

2n3904 (gp npn)
2n3906 (gp pnp)
2n2222 (plastic - switch/high Ic)
bd139/bd140 (complimentary / med power npn - make nice compliance range current sources/sinks!)
j113 (jfet)
IRF510 (medium mosfet)
bs170/2n7000 (same part! tiny mosfets).

BD139 is my favourite if you didn't guess. They are literally the swiss army knife of transistors. Nice package, high Ft (150MHz+), Reasonable Ic, High Vce, quite difficult to blow up, cheap as chips.

One thing to note, your average crappy 2 cent Fairchild 2N3904 has a better NF than any BC transistor, including the low noise 549 until you start getting into RF transistors. Keep the gain reasonable and there's a huge mileage in that part. Literally 90% of my designs use just the 2n3904/2n3906.

Edit: also I've got a tube each of CA3096 and CA3028. CA3096 is a few well matched transistors in a DIL package. Great for current sources, compensated feedback for log amps. CA3028 is a can package with a diff amp in it which can be reconfigured in more ways that I care to imagine. Makes nice cascode amps etc. Shame they're both obsolete.

[David Hess]

the 150 volt 2N5401 and 2N5551 would be more suitable.

Or the similar MPSA46, or whatever they are, from that family.  MJE350 and complement are also quite popular for audio (driver stage), though the datasheet is sorely wanting.  I think On Semi makes a detailed datasheet?  Or, there are equivalent parts with good data out there, shop around.

Did you mean the MPSA43/MPSA93?  They are a little slower.

The Motorola/On MJE371/MJE521 are the same way.  They are recommended for specific applications but lack a full set of specifications.  I ran across them when doing a search for fast TO-126, TO-225, and TO-220 parts to use in high performance regulators.

The BD135 through BD140 are like that also.  The ST and Fairchild datasheets say nothing about dynamic performance but there are SavantIC Semiconductor (who?) datasheets which say 190 MHz for the NPNs and 160 MHz for the PNPs.  What?

[CJay]

The BD135 through BD140 are like that also.  The ST and Fairchild datasheets say nothing about dynamic performance but there are SavantIC Semiconductor (who?) datasheets which say 190 MHz for the NPNs and 160 MHz for the PNPs.  What?

The BD139 makes an excellent little RF power transistor, 2-3W output across the HF bands and probably into low VHF (never tried it above 30MHz), it's not been a secret but it's also not that widely used, people seem to prefer the long obsolete 2SC Mitsubishi devices

[bd139]

Indeed. Most transistor selection appears to be religion.

Recently I stopped someone paying £8 for an OC transistor in a circuit they didn’t understand just to copy it verbatim. A 2n3906 would do the job, arguably better as well!