MPSA18 transistors work in every orientation, why?

[patrick398]

Got a few MPSA18 trannies of the bay for another vox repeat percussion i'm doing. I noticed that which ever way i orientate the transistors it sounds fine. When i say fine, i mean it works. I have a hunch that it doesn't sound as good as the last one i made but i don't have that anymore to compare it to.



I can have them orientated as per the layout, or flip each transistor, and even have them both pointing the say way and it still works...i'm suspicious.
How can that be?

[Kipper4]

Got a scheme for that Pat?
It depends on how the jelly bean is being used.

[pinkjimiphoton]

well, darlingtons are a bit different from a "normal" transistor,
but its cuz you have discovered that it can be reverse beta'd. many transistors will work either orientation, tho the gain will be seriously different. germanium, in particular, sometimes is so leaky you can literally plug it in either way, and can be hard to get the orientation right.

ya getta better fuzztone in some cases with one or both BACKWARDS.
peeps look at me funny when i say that, but  its true.
AMZ, aka Jack Orman has a page about reverse biased transistors i had pointed out to me recently:

http://www.muzique.com/lab/reverse.htm

i did it inadvertantly the other day when working up my 'weeti pie project. works great.

in the case of the repeat percussion, you're using them more like switches i think than gain stages, so the beta and orientation don't matter so much. its a cool circuit, i built one a while ago but didn't keep it.... not for me.

but no, you're not crazy, and yes, it will work. it may work better, or it may work worse. depends on the circuit!

https://en.wikipedia.org/wiki/Darlington_transistor

look at how darlington pairs are put together



so, it will have MORE gain reverse beta'd than a standard transistor i'd imagine... probably why it works, but i hope someone who actually can answer this much better chimes in, i am far far far from an EE.

good luck!

[pinkjimiphoton]

[pinkjimiphoton]

https://www.edaboard.com/showthread.php?188129-What-is-the-reverse-beta-of-BJT

[patrick398]

Got a scheme for that Pat?
It depends on how the jelly bean is being used.

This is the only schem i have Kip, not the largest or best quality.

well, darlingtons are a bit different from a "normal" transistor,
but its cuz you have discovered that it can be reverse beta'd. many transistors will work either orientation, tho the gain will be seriously different.

It's funny, i read about this or watched a video about it either earlier today or yesterday...It's scary sometimes just how little actually goes into my brain.

But that makes a lot of sense Jim, thanks so much for the replies. I'll get stuck into the suggested literature.

I think i may stop buying trannies and ICs of the bay, i'm just convinced all my parts are fakes 

[patrick398]

'However, the most interesting effect is when this reverse booster circuit is driven into clipping - it clips in a matter totally unlike that of the basic common emitter circuit! The overdriven reverse booster generates a mess of harmonics different from other clipping circuits and configurations. '
From: (http://www.muzique.com/lab/reverse.htm)

I often thought it strange that some Death by Audio schematics seemed to have transistors in backwards. Enlightening.

[pinkjimiphoton]

my old friend devi evar too uses them to great effect often.

check my "weetie pi" project. second stage uses a reverse biased 5089 similar to what jack has on his page, but arrived at independently <kinda, as i read his whole site years ago so i am sure i'd seen it there in the past>

it has a completely different sound from other ways of doing it. it "chimes" more when ya whack it hard, too.

in your case, if ya look at the transistor at the bottom of the schem, you can see how it could go either way just by the way the q is drawn.. look at the arrow. looks almost like a kind of fet, doesn't it?

[patrick398]

I was following that thread actually, i'll have to have another read through with my new enlightened mind haha.

Isn't that 2N2646 a unijunction tranny though? I'm guessing that's a whole different ball game. I'm sure R.G wrote something about them, i'll have a dig

[pinkjimiphoton]

no freekin clue. you needs someone with some intelligence, i'm just a monkey with a breadboard.

[Rob Strand]

Isn't that 2N2646 a unijunction tranny though? I'm guessing that's a whole different ball game. I'm sure R.G wrote something about them, i'll have a dig

It's a UJT for sure.   It generate pulses at the programmed rate.  2N2646 was the NE555 of the olden days.

[Keppy]

2N2646 is the same UJT used in the Phase II. Even though it looks like it can be reversed, it can't. The Ludwig factory schem showed it the wrong way, causing the animation not to work in the clone until I looked up some generic UJT timer circuits and figured out the problem.

R.G. did indeed write a bit about the UJT in the documentation for that project. He went as far as recommending replacing it with a more-available PUT (programmable unijunction transistor). That... also required debugging. Memories.

Sorry for the hijack, I know that's not the transistor the OP asked about.

[patrick398]

I've never really understood how the repeat percussion works but from what you're saying i guess that the
MPSA18s are used more like switches and the 2N2646 is creating pulses that essentially turn them on or off to create the tremolo?

[Ben N]

well, darlingtons are a bit different from a "normal" transistor

Buht eet eez nuht a Darlingtuhn, Jeemee!

https://alltransistors.com/pdfdatasheet_update/mpsa18rlrag.pdf

[R.G.]

The MPSA18 is not a darlington, merely a very high gain bipolar transistor.

The Vox repeat percusssion circuit is an odd circuit.

Bipolar transistors may be operated either as usually intended, with the base-collector junction reverse biased, or reversed, with the base-emitter junction reverse biased. If the doping of the two junctions were identical, they would work equally mediocre-ly each way. Modern transistors specialize the base-collector junction for high voltage and high power operation, and the base-emitter for low voltage operation and high gain.

Current gain in the normal mode is called "beta" because current gain in the reverse mode was called "alpha" in the days when transistors were new.

The first transistor in the Vox repeater is deliberately used in reverse mode because it is used as a variable resistor, and bipolars do better as variable resistors in the reverse mode. The second transistor is used as an amplifier to regain some of the signal loss from the resistor divider action of the first transistor. Not much gain is needed. The gain of a bipolar in reverse mode (alpha) is quite low, but in this circuit, it's not so low as to prevent things from working at all, and the reverse voltage of the base-emitter junction for your MSPA18s is apparently high enough not to zerner with 9V on the "collector" for the second transistor as it's used there.

So they work either way because the demands are not very high, and there is some luck involved with transistors. You're right that it would work better with them oriented properly.

[pinkjimiphoton]

2N2646 is the same UJT used in the Phase II. Even though it looks like it can be reversed, it can't. The Ludwig factory schem showed it the wrong way, causing the animation not to work in the clone until I looked up some generic UJT timer circuits and figured out the problem.

R.G. did indeed write a bit about the UJT in the documentation for that project. He went as far as recommending replacing it with a more-available PUT (programmable unijunction transistor). That... also required debugging. Memories.

Sorry for the hijack, I know that's not the transistor the OP asked about.

cool!!! lol i've just recently got the parts order in to build one finally, i can use this in it perhaps

sorry, always thought that series were darlington, as usual, i'm an idiot and should just stick with breakin' stuff

[Ben N]

Jeemee, Jeemee, zaht iz because MPSA13 & MPSA14 are Dahlinktons, Dahlink. So you are nuht an idiot!

[pinkjimiphoton]

ok, i split the difference with ya, bubbula, i'm half lit lol



literally.

lithouanian. they say a litvak is a polak with its brains shot out. lol.

sure seems to fit most of us lol!!

[PRR]

> Current gain in the normal mode is called "beta" because current gain in the reverse mode was called "alpha" in the days when transistors were new.

That was before your time, I see.

Nothing to do with "reverse".

Alpha is the Common Base current gain. In Point Contact devices it can be more than one (which is why we might want to know Alpha). In Junction devices at sane conditions it is always less than one.

If Beta is 100, Alpha is 0.99; Beta 1,000 Alpha 0.999, if Beta were 20 then Alpha would be 0.95.

For any useful Junction device, Alpha is so close to "1" that we pay no attention.

Yes, biasing a device with Alpha more than one was wonky.

[R.G.]

I yield to your superior age, sir.