What does "ganging" two transistors mean?

Thread Starter

PartVIII

Joined Sep 26, 2015
2
I have a homework problem that asks "What's the result of ganging two "npn" transistor amplifiers together?"

From what it looks like (and intuition tells me), it seems like ganging up two npn transistors is connecting them in either series or parallel, which (I think) would have dramatically different outcomes. I was wondering if someone could please help me understand what "ganging" transistors is?
Also, my answer to the question is:

Ganging two npn transistors together will result in amplification of the already amplified signal. This will cause it to heat up, which in turn will increase the current flowing through the transistor, resulting in thermal runaway. However this can be prevented by adding low-value resistors to each emitter terminal. Another outcome of ganging two transistors together is the inversion of the already inverted signal. The 1st transistor shifts the phase of the input signal by 180° and the 2nd shifts the phase 180° again, resulting in an output signal with the same phase as the input.

Any help would be tremendous. I'm taking a graduate level analytical chemistry course and as a mere chemistry major with no prior knowledge of circuits, I'm definitely in way over my head.
 

WBahn

Joined Mar 31, 2012
29,979
I think you are doing pretty good.

But you seem to be mixing the effects that you would expect to see for series and parallel connections.

The context is a bit hard to discern because there is a difference between doing something to transistors and doing that same something to transistor amplifiers.

Usually, the term "ganging" means to put things in parallel. We "gang" lots of things -- circuit breakers, switches, resistors. The word basically means a group of things but, in this context, usually means a group of things that are all working the same way to accomplish a task.

If you are putting two BJTs in parallel, then the thermal runaway considerations you mention are pretty much spot on. But the problem is a bit more specific than you mentioned. With just a single transistor it will heat up but that doesn't result in thermal runaway. So why do paralleled transistors exhibit this? I think you understand the cause, just need to word things a bit better. But if they are in parallel then the notion of the phase shifts adding up doesn't make any since. Your recommended solution will work (though it's not the only solution).

If they are in series (which isn't even well defined for a three-terminal device), then the phase argument may make sense, but now the thermal runaway explanation doesn't apply.
 

Thread Starter

PartVIII

Joined Sep 26, 2015
2
I think you are doing pretty good.

But you seem to be mixing the effects that you would expect to see for series and parallel connections.

The context is a bit hard to discern because there is a difference between doing something to transistors and doing that same something to transistor amplifiers.

Usually, the term "ganging" means to put things in parallel. We "gang" lots of things -- circuit breakers, switches, resistors. The word basically means a group of things but, in this context, usually means a group of things that are all working the same way to accomplish a task.

If you are putting two BJTs in parallel, then the thermal runaway considerations you mention are pretty much spot on. But the problem is a bit more specific than you mentioned. With just a single transistor it will heat up but that doesn't result in thermal runaway. So why do paralleled transistors exhibit this? I think you understand the cause, just need to word things a bit better. But if they are in parallel then the notion of the phase shifts adding up doesn't make any since. Your recommended solution will work (though it's not the only solution).

If they are in series (which isn't even well defined for a three-terminal device), then the phase argument may make sense, but now the thermal runaway explanation doesn't apply.
Thank you for your quick reply, you cleared that up for me quite a bit.
 

WBahn

Joined Mar 31, 2012
29,979
Could also apply to Darlington arrangement?
Max.
Possibly, if it is referring to transistors -- though I still wouldn't think that the term "ganging" makes much sense. But what does it mean to put two transistor amplifiers into a Darlington arrangement?

Putting two amplifiers in series is generally referred to as "cascading" them. I'm pretty sure that "ganging" them would only make sense if they were being put in parallel.
 

Alec_t

Joined Sep 17, 2013
14,280
Yet another example of the low question-setting standards that seem to apply in academia nowadays ;). Perhaps the profs/lecturers/tutors should be sent on a clarity course (if you can find a course tutor with the necessary high standards!)
 

WBahn

Joined Mar 31, 2012
29,979
Yet another example of the low question-setting standards that seem to apply in academia nowadays ;). Perhaps the profs/lecturers/tutors should be sent on a clarity course (if you can find a course tutor with the necessary high standards!)
In general I agree. I don't know if we have a English-translation issue at play here or not. But it is definitely the case that TAs and tutors are frequently simply not prepared to act in those capacities, nor are any efforts made to prepare them for any kind of a teaching role before hand. They are simply awarded a TA as part of their financial support, given a schedule of courses, and told to have fun. Most of them have never taught and have absolutely no desire or interest to teach -- that's not why they are there.
 

vk6zgo

Joined Jul 21, 2012
677
In general Electronics parlance "ganging" of components refers to (& has for around a century),mechanically controlling a number of passive devices from the same action-----for example,"ganged switches","ganged potentiometers",or "ganged variable capacitors".

This does not imply "components in parallel"!

Latterly, some peripheral groups have began incorrectly using the term to refer to multiple transistors connected with their bases in parallel,their collectors in parallel,& their emitters in parallel,in effect,making a higher power transistor,at the expense of matching problems & greatly increased base current.

"Cascade" does not refer to transistors "in series"--it just means one stage following the other,as in conventional multistage amplifiers.

The only amplifier circuit which could vaguely be referred to as transistors "in series" is a special arrangement
called a "CASCODE" (note the "O").
 

WBahn

Joined Mar 31, 2012
29,979
Where did I say that cascade refers to transistors in series?

Note that I started from the term "ganged" meaning a group of things all working together to accomplish a task. I described ganged switches as being "in parallel" only in loose terms (but didn't make a point of that, probably should have) in order to try to move towards what the instructions might have meant. This is because we are forced to interpret what was meant by a poor use of the word. I then tried to extend that to what it could possibly mean to "gang two transistor amplifiers" -- and note that the OP did not say ganging two transistors, but ganging two transistor amplifiers. I pointed out that the very concept of putting two transistors in series is poorly defined for a three terminal device). How do you put two transistor amplifiers into cascode? Nothing comes to mind. But if you cascade them, then you have the signal coming into the first and the signal out of the first going into the second and then the final signal coming out of that. From a signal standpoint, they are in series.
 

dannyf

Joined Sep 13, 2015
2,197
connecting them in either series or parallel, which (I think) would have dramatically different outcomes.
In parallel.

It typically doesn't yield "dramatically different outcomes", :). In most cases, it is not recommended without some form of load balancing. Parallel can be useful to lower current noise, and you may find it more commonly done on mosfets or jfets.
 

vk6zgo

Joined Jul 21, 2012
677
"
How do you put two transistor amplifiers into cascode? Nothing comes to mind. But if you cascade them, then you have the signal coming into the first and the signal out of the first going into the second and then the final signal coming out of that. From a signal standpoint, they are in series. "

No! they are in cascade!

"Cascade" has been used for many years to uniquely describe this sort of connection,& it is only recently that people have started incorrectly using the term"series" (which has its own precise meaning),in this context.

I suppose you could in circuit analysis represent the stages as "in series",but that is "drawing a very long bow,indeed",& any phase reversals between stages may be a bit messy to accommodate.

Admittedly,it can be hard to determine where the stages of some circuits start & finish,but in most discrete
amplifiers they can be separated into individual units.

Yes,I confused things by saying "transistors" where I should have said amplifiers,when referring to "cascading them".
I did it again,in reverse,re amplifiers in "Cascode"--obviously we can only do that with the transistors themselves,making one amplifier.

The OP was obviously thinking in terms of transistors,rather than complete amplifier stages,
But that was no excuse for me getting it wrong,too.
I Googled, found a reference to "Ganging" multiple transistors,& compounded the confusion.
There was no reference to this term in any other context,except that,& the normal "Passive devices controlled by the same mechanical action".

I agree with you,that the question was probably referring to running two or more amplifier stages in parallel,as is widely done in Solid State RF Power Amplifiers,with,as dannyf says,load balancing (& input balancing).
Report
 

WBahn

Joined Mar 31, 2012
29,979
"
How do you put two transistor amplifiers into cascode? Nothing comes to mind. But if you cascade them, then you have the signal coming into the first and the signal out of the first going into the second and then the final signal coming out of that. From a signal standpoint, they are in series. "

No! they are in cascade!
I agree. I definitely don't recommend referring to cascaded stages as being in series. But the fact remains that the TS is being offered those two choices -- parallel or series. Although, going back and looking at Post #1, I'm not so sure that is really the case or rather that is just an incorrect interpretation that the TS is placing on it.
 

vk6zgo

Joined Jul 21, 2012
677
Actually,after Googling the OP's question,I found on AAC's very own "e-Book",a reference to "ganging"of multiple transistors,so if he did the same,that might be where it came from.

According to him,the problem asks "What's the result of ganging two "npn" transistor amplifiers together?",but his thread headings says "What does "ganging" two transistors mean?",which,as you suggested,is a quite different question.

If "ganging"means what we surmised-----Amplifiers in parallel,what is the significance of "NPN transistor" in the question?
Why would it make any difference,if the two amplifiers were PNP transistors,FETs,or even vacuum tubes?

Surely in each case,the result would be,in the absence of external impedance matching circuits, lower input & output impedances of the combined amplifiers,& increased power output,if the amplifiers are power stages to start with.
 

JoeJester

Joined Apr 26, 2005
4,390
Ganging must be a contemporary misuse of the word, or we are having a translation problem in this context. The proper terms would be cascade and cascode.


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

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

The only one close to my recollection of vacuum tubes being cascaded is the graphic labeled "HV Stack".

If "ganging" were to be taken in a manner similar to "ganged capacitors" used to tune both the Local Oscillator and the RF Amplifier in superhetrodyne receivers or multiple RF Amplifiers in the Tuned Radio Frequency receivers (no Local Oscillator or IF Amplifiers. Literally, you controlled multiple circuits with one action.

.
 

vk6zgo

Joined Jul 21, 2012
677
Ganging must be a contemporary misuse of the word, or we are having a translation problem in this context. The proper terms would be cascade and cascode.


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

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

The only one close to my recollection of vacuum tubes being cascaded is the graphic labeled "HV Stack".

If "ganging" were to be taken in a manner similar to "ganged capacitors" used to tune both the Local Oscillator and the RF Amplifier in superhetrodyne receivers or multiple RF Amplifiers in the Tuned Radio Frequency receivers (no Local Oscillator or IF Amplifiers. Literally, you controlled multiple circuits with one action.
Joe.if you read back,we've been through all that!

Wiki "blew it' in your first link!
The second paragraph confuses "cascade" with "cascode",before resuming the correct description of "cascade" in the last paragraph.----time for an edit?

As has already been seen above,WBahn & I agree on the definitions of "cascade' & "cascode"respectively.

"Cascade" is where the output of one stage becomes the input of the next,& so on.
A "cascode" amplifier configuration is normally regarded as a single stage made up of two active devices.
connected as correctly described in your second link.

Vacuum tube "Cascode" stages were widely used in the RF stages of TV tuners,before Solid State devices
became common.
It's half a century ago,but I think the claimed advantages were lower noise & improved stability.

We were effectively "trying to read the Examiner's mind" to see if in using the term "Ganging",he meant what we assume,or something else entirely.
 

JoeJester

Joined Apr 26, 2005
4,390
I was paying more attention to the graphics. I did find something better ... in my searches. Cascode is not in the current websters dictionary, however cascade is.

from www.m-w.com
Full Definition of CASCADE
1 : a steep usually small fall of water; especially : one of a series

2 a : something arranged or occurring in a series or in a succession of stages so that each stage derives from or acts upon the product of the preceding <blood clotting involves a biochemical cascade>
b : a fall of material (as lace) that hangs in a zigzag line

3 : something falling or rushing forth in quantity <a cascade of sound> <a cascade of events>
2.a. is the applicable definition. I hadn't checked Oxford's dictionary yet.

As far as cascode ...
from http://findwords.info/term/cascode
1939 Hunt & Hickman in Rev. Scient. Instruments X. 16/2 This tube connection we have called the ‘cascode’... It is generally useful for d.c. amplifier work in which it is inconvenient to supply the additional bias voltages for a pentode. The dual triode, as connected, may be shown to be equivalent to a single triode. 1948 H. Wallman et al. in Proc. Inst. Radio Engin. XXXVI. 701/1 Search for a concise name for the grounded-cathode, grounded-grid combination led to the designation ‘cascode’, after a somewhat similar arrangement employed by Hunt and Hickman. Ibid., The low-noise property that forms the feature of the present circuit is, however, entirely unconnected with the original use of the cascode as a d.c. amplifier in a voltage stabilizer. 1955 Electronic Engin. XXVII. 323 The features of the cascode type of amplifier. 1958 Chambers's Techn. Dict. 964/1 Cascode, thermionic valve circuit in which a grounded-cathode triode followed by a grounded-grid triode provides a low noise amplifier for very high radio frequencies. 1960 Electronic Engin. XXXII. 44 A transistor cascode amplifier does not require neutralization.
I hadn't tracked down the references cited.

I have repaired equipment that had used cascoded amplifiers, as they were designed in the 40s and 50s.
 
Last edited:
Top