common transistors and/or ICs?

[eleanor296]

hey!
Just wondering if you guys could give me a list of some more "common" transistors and ICs, that show up a lot in designs...
so I can "finalize" my next futurlec order.
Thanks!

~Andy

[Hanglow]

http://www.aronnelson.com/gallery/main.php/v/DRAGONFLY-LAYOUTS_0/The-Mod-Zone-_/TRANSISTOR_SUBSTITUTION_CHART.gif.html

for dual opamps, the most popular are ones like TL072, NE5532 . If you do a search there are lots of suggestions for substitutions and so on

No matter what though, you are bound to forget something ! sod's law.

[demonstar]

Opamps
TL70x
TL06x
LM741
KA1458 (dual opamp)

Transistors
2N3904 (BJT NPN)
2N3906 (BJT PNP)
BS170 (N-channel enhancement mode MOSFET)
J201 (N-channel depletion mode JFET)


Other ICs
555 (timer) or 556 (dual timer)
4069
LM386 (power amp)

I think I listed the most common ones assuming your in Europe. I find they're the ones I use most often although I do use a selection of others and there are other alternatives available. They're slightly different for the U.S.

[eleanor296]

thanks for the substitution chart!
Actually, I'm a European living in America (as of right now), but still thanks a lot!
I'll be able to put this to use!

Any other suggestions?

Thanks
~Andy

[drk]

don't forget FETs! like 2N5457, J201, MPF102...

[demonstar]

I think now is a good time to reference R.G. Keen's Laws. One and six are good in this scenario...

1. Use what you can get, subject to some rough and ready rules. See Resistors, Capacitors, Inductors, Transistors and Op amps below. If you have a circuit that needs a specific NPN transistor, and all you can find is a 2N3904, try it. Chances are that if it's not a high power application, it will work mostly. If you need a 3.3K resistor and don't have any of those, notice that 3.3K is about 1/3 of 10K and make one up by paralleling three 10K's that you do have, or a 2.2K and a 1.1K in series. There are places where this kind of approach will not work, but it does work surprisingly often.

6. Transistors: When in doubt, use a 2N5088. Or a 2N3904 - or a 2N4401 - or a BC549 - or... find cheap, available NPN and PNP devices that you can get easily and use them. In the USA, 2N3904 (NPN) and 2N3906 (PNP) are easily available and can be found almost everywhere. I personally prefer 2N5088 (NPN) and 2N4250 (PNP) because they have slightly better gain and lower noise. 2N4401 (NPN) and 2N4402(PNP) work almost as well. Find good parts you can get, and then figure out why they **won't** work, not how to find magic parts from Timbuktu.

I can't be sure but I seem to remember hearing that in the US folks often replace the 2N3904 with the 2N5088 and the 2N3906 with the 2N5089 just because they're more easily available and cheaper. Maybe someone else may be able to shed some more light on the validity of my claim.

[eleanor296]

Thanks for the suggestions guys!  I'm compiling a list as we speak.  Wow.. for me this has almost "sticky importance".

Just one afterthought... thanks for the suggestions demonstar, but what "TLXXX" opamps are most often used, other than TL072 which seems to pop up a lot?

[asfastasdark]

I think the 741 (or something in that direction) OP amp is used in a bunch of schematics... I'm not exactly sure anymore.

[demonstar]

Just one afterthought... thanks for the suggestions demonstar, but what "TLXXX" opamps are most often used, other than TL072 which seems to pop up a lot?

Really I just swap in whatever opamp I have lying around thats suitable for the job but out of the TLXXX opamps you refer to, the ones I have in my parts are some TL06X and some TL07X. I went for some single, some dual and some quad opamps. 

I just got myself a decent range of different opamps from different product ranges to try out when I made my first oder. So sure you could live easily with just the TLXXX but I'd give a couple others a go too. Now unless I need a specific one for a specific job if I'm just generally buying I go with the parts that have got a good deal at the time.

Hope you get what you're after!

[JDoyle]

The 2N508X subs for the 2N309X originals is because the 2N50XX parts have higher gain and lower noise...

For Silicon BJTs, pick one NPN and one PNP - I like the 2N50xx series myself - buy in bulk to save money and forget about any others - simply put them, and any worry about your choice out of your mind. Modern silicon transistors are jelly beans these days and are pretty much the same. Find a 'low noise, audio amplifier' and buy a pile - very rarely will you run into a situation that these, no matter what you pick, will NOT work in a situation that calls for a SI transistor. Most schematics of concepts/ideas/new designs don't even mention a part number for small signal BJTs for exactly this reason.

Metal Can Silicon BJTs - here you may want to look a little more at the datasheets, these can most likely dissipate more heat than their plastic/TO-92 counterparts and will also most likely have a lower gain range. I and others have found the 2N2222 metal can type to be nice, and I have some 2N2907 PNP metal can types I like as well. Spend a little more time than the plastic ones, though not a whole lot, finding a NPN/PNP pair you like, then do the same as above. I do find that these can be less noisy because of the better shielding (I think) of the case - though not in a drastic, 'I am never using a TO92 again' way.

In short, silicon is so popular because it performs so consistently in the manufacturing process, other than power dissipation, gain range, cut-off frequency, Vce(sat) and Vbe(on), (none of which are all that important to us, or even audio in general) most of the rest of the specs are going to be pretty much the same down the whole line of Si BJTs - so don't waste too much time on part numbers here, one will most like work just as well as the other. Another factor is that the transfer curve and gain of Si BJTs is so steep/high as to make one that has a gain of 200 seem almost identical to one that has a gain of 800, especially for audio. Even if it is used for a distortion - emitter degeneration doesn't take overall stage gain into account like negative feedback so a BJT common emitter with a gain of 10 (10k collector resistor, 1k emitter resistor) is the same whether it's hfe is 200 or 800. And running open loop, well, I'm of the opinion that a square wave is a square wave and is musically uninteresting, but your results may vary...

Bottom line - I think you will find noise to be the biggest determinant of your selection. Again, I like the 2N50xx series.

For Germanium BJTs, as we all know these are surplus/rare/most likely already picked over so you are going to be testing them anyway, specific part numbers are pretty much useless because who knows if you can find them. If you have the luxury of being able to choose, you are looking for low leakage, low noise, and preferably breakdown voltages above 9V - you can't guarantee this last bit with the older parts.

JFETs - Exact opposite of  BJTs, every JFET is different. Did you know that the J201, J202, and the J204 are ALL MADE FROM THE SAME DIE? They are made, tested for Idss/Vgs(off), and THEN given a part number. The manufacturer doesn't even know how many J201s vs. how many J202s he is going to get out of a run until they are tested.

In general (or - my general guidelines when I go looking to purchase parts) with JFETs you are most likely going to want to look for 'amplifier' types, specifically 'audio' if you can find them, with Idss values under 10mA (if only to keep current draw in a 9V battery operated circuit to a reasonable level, not to mention that with 9V V+ and a 1K source resistor, you'll never get to Idss if it is over 9mA) and Vgs(off) values of 7V or under (again going on the assumption of a 9V V+). The Vgs(off) spec doesn't rule out all that many, but the 10mA Idss does. JFETs labeled as switches or RF/VHF/UHF types are ones I generally avoid - they are going to have really high Idss to overcome the Miller capacitance and improve overall speed for their purpose. I personally think that an Idss of 100mA is useless for FX purposes on a 9V battery, though that isn't to say that I am at all right. One thing about the switch/RF/high Idss types - the transfer curve has a wider 'slope' as Vds approaches Idss, which is really good for getting that 'JFET distortion', but again, you've got to suck almost 100mA (while not dissipating more than 200mW or so) to get there.

Oh - one last thing, if you want to do a quick sort of a pile of the same part numbered JFETs, don't bother doing both Idss and Vgs(off) tests, only one is really needed - the two values will track. Within the same part number and manufacturer, small Idss valued parts have small Vgs(off) values and vice versa; two J201s with the same Idss will have very close Vgs(off) values. (it also follows that gm/gfs will track as well, smaller Idss/Vgs(off) will have smaller overall gm/gfs). So sort by whatever value is most important to the application and worry about the other value for ultra-critical (is there such a thing in stompbox circuits???) operation. But if the part is ultra-critical for the entire design, it is an EXTREMELY bad idea for that part to be a JFET. Period.

I am still a J201/J202/J204 for n-channel, J270 for p-channel fan. Though I have had good results with the 2N3820 as a p-JFET.

OP AMPS - There are so damn many of these made for every purpose and optimized for every possible specification that it is overwhelming. Understandably so, especially when you consider not a SINGLE ONE has a spec named 'Tone'. Duals are the FX industry standard, mostly because of the JRC4558 and it's substitutes, and ALL have the same pinout. Quads are common but there are two pinouts possible so check the datasheet. Singles are also common and allow for the user to 'tap into' the circuit through compensation and  offset adjustment pins - though there isn't a law saying that they HAVE to be used for that purpose. Input and output stage design means a lot here, input design as to the best signal source type to interface it with and output design on output impedance/voltage output swing/entry and recovery from overload/output drive etc. For input stage design, BJT inputs offer the best noise performance for low impedance sources (in other words - if the input is buffered); JFETs offer the best noise performance for high impedance sources (signal straight from the guitar, from the collector of high gain transistor amplifier stages). Side note - there isn't any 'standard' for when EEs speak of 'high impedance source' or 'low impedance source', at least that I could find written anywhere, but from what I understand, anything over 10k is considered 'high' and anything under 1k - 2k is considered 'low' in-between is just that and rarely spoken of.

While input stage design matters most to noise level/rejection, in my opinion, output stage design of an op amp matters the most to the actual tone of the chip, especially when used to clip. Almost all op amps have Class AB output, with the 'PNP' half being a 'quasi-complimentary' stage where the PNP transistor is ganged to a NPN to multiply their gains and bring the PNP 'side' up to the specs of the NPN 'side' - but it will never be perfect and most certainly never be the same. It should be noted that op amps with the 'quasi-complimentary' output stage can not be forced into class A operation. Some op amps have Class B output stages, though this is pretty rare, and of course, some Class AB outputs are closer to Class A, while others are closer to Class B. It is very rare for an IC op amp to have a Class A output - simply because it is a waste of efficiency at the end of a very efficient and accurate circuit and what self respecting chip designer would want to do that? One that cares about audio, would be my response, but I digress...

Output flavors are either BJT or MOSFET, never JFET. In general: BJTs are faster due to the lower device capacitance, but integrated MOSFETs are better are sourcing current without burning up the whole chip. Is faster better? Who knows. It gets you out of clipping faster, but also brings you into it faster. And as op amps are always used with negative feedback, if one wants to keep them 'operational', there isn't the natural 'slowdown' in gain as the device approaches the limits as there is in discrete, single device stages.

I like the 4558, TL07x, LF411, LF353, NE553x, CA31x0 and TLCXXXX series of chips - though I should note that I rarely use op amps in the signal path these days.

You will most likely want to experiment in each circuit to find out if one sounds better or not. A good test is to put in a 1458 type (a dual 741, BJT input, about as old as you can get) and then switch it out with a JFET-input TL072. If there is an audible difference in how the circuit sounds, you may want to keep going and testing other chips, if there isn't a difference, move on, nothing to improve here...

That should be enough to get you started, I hope.

Regards,


Jay Doyle

[eleanor296]

wow.... that was an awesome read.
Thanks so much for the info, it's very appreciated!

~Andy

[demonstar]

Jay has done it again, that will help a lot of people. I've certainly learnt from it. Thankyou very much!