Searching \ for 'Transistor H parameters' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=transistor+parameters
Search entire site for: 'Transistor H parameters'.

Truncated match.
PICList Thread
'Transistor H parameters'
1997\08\16@234012 by Sean Breheny

flavicon
face
Hello all,

I am really wondering why most transistor datasheets do not include H
parameters! I have looked at Motorola and National Semiconductor Datasheets
for bipolar transistors, and only a small percentage include H parameters.
Very few of the audio transistors, especially the power ones, include these.
What am I supposed to do if I want to accurately design an amplifier without
guesswork?
       Even the datasheets that do give H parameters only give them for the
CE configuration. I am somewhat new to discreet transistor amplifier design
and there may be a formula to convert the H parameters for the CE
configuration to CC and CB configurations ( for those H parameters that
differ ) (and if there is such a formula, would someone please tell it to
me), however one would think that the
datasheet would take the small amount of space needed and show the H params
for CC and CB.
       Thanks for putting up with my questions which often do not concern
pics. I post them because this list contains an extreamly knowledgeable
group. Thanks.

BTW I DO use PICs, too !

Sean

1997\08\17@002353 by Bob Blick

flavicon
face
At 11:38 PM 8/16/97 -0400, you wrote:

>What am I supposed to do if I want to accurately design an amplifier without
>guesswork?

Parameters vary widely due to operating conditions and manufacturing. An
amplifier should not rely on transistor specs to define its operation. Most
transistor spec sheets reflect this - they will list the limits that the
transistor can withstand, and the design is up to you.

Manufacturers will also quite often target a series of parts for a specific
use - you should familiarize yourself with them.

Cheerful regards,

Bob

http://www.bobblick.com/

1997\08\17@132814 by Sean Breheny

flavicon
face
At 09:21 PM 8/16/97 -0700, you wrote:
>At 11:38 PM 8/16/97 -0400, you wrote:
>
>>What am I supposed to do if I want to accurately design an amplifier without
>>guesswork?
>
>Parameters vary widely due to operating conditions and manufacturing. An
>amplifier should not rely on transistor specs to define its operation. Most
>transistor spec sheets reflect this - they will list the limits that the
>transistor can withstand, and the design is up to you.
>
>Manufacturers will also quite often target a series of parts for a specific
>use - you should familiarize yourself with them.
>
>Cheerful regards,
>
>Bob

Thanks for the reply, Bob.
I guess you are saying that it is up to the actual user of the components to
measure their characteristics. For me, there is only one problem with this,
I am at this point only an electronics hobbist. I am studying to become an
electrical engineer. I don't have the money to buy the test equipment needed
to measure all the characteristics I need to know.
I do need to know the H parameters of a transistor for an amplifier because
the input/output impedances depend upon this. I don't see how an amplifier
can avoid relying on transistor specs.
Also, I realize that most parts have a specific target use. However, even if
I know that, say, a 2N2222 is a general purpose amp, this just guides me in
choosing this component for my application. I still need to know its
characteristics.

Thanks,
Sean

1997\08\17@183244 by John Payson

picon face
> I guess you are saying that it is up to the actual user of the components to
> measure their characteristics. For me, there is only one problem with this,
> I am at this point only an electronics hobbist. I am studying to become an
> electrical engineer. I don't have the money to buy the test equipment needed
> to measure all the characteristics I need to know.

I think what he's saying is that the design should be such that those
parameters DON'T MATTER.  While some transistors may be engineered so
that they behave very precisely according to a detailed specification,
it's much cheaper to make them "sloppier"; while the sloppy transist-
or will require some extra circuitry to cancel out any slop, the extra
circuitry will cost less than making the transistor less sloppy.

As a simple example, suppose you're trying to design a circuit which
will convert a positive voltage to a negative current.  There are two
ways of doing this with a single transistor:

Method 1:
          /----- out
        |/
In-/\/\/-K
        |\
          \|--gnd

Method 2:
         /----- out
       |/
In------K
       |\
         \|--/\/\/--gnd

In both circuits, the current sunk by "out" will be proportional to the
voltage at "in" minus Vbe.  In the former circuit, however, the current
is EXTREMELY sentitive to the parameters of the transistor, while in the
latter case it is MUCH LESS sensitive.A  Further, in the former case, the
accuracy of the design requires that the transistor's beta be set to a
precise value, while the latter merely requires that it be as high as
possible.

If the accuracy obtained by "Method 2" above isn't adequate, then it may
be improved further by using op-amp feedback circuits.  While op amps used
to be expensive (esp. before single-chip monolithics came on the scene)
they are now very cheap; using a circuit with op amps and feedback, it's
possible to design an amplifier or control circuit whose accuracy depends
[for all intents and purposes] solely on the accuracy of the resistors and
caps that make up the design--not on the specific properties of the trans-
istors.

To bring things back to the world of PICs, consider the specifications of
the PIC's port pins; earlier PIC books had graphs detailing how the PIC's
ports would sag when confronted with large current loads.  If one took these
graphs at face value, it would be quite safe to hook up two LED's in series
between two port pins with no resistor; since the VDD rail would drop by
about a volt under a 20mA load and ground would rise by about half a volt,
about 3.5 volts would remain for the LED's.  Any design that was built that
way would have problems, though, if Microchip were to "improve" their PICs
by beefing up the I/O ports.  By contrast, using two LED's each driven by
its own port pin and with its own current-limitting resistor would be a
much more "stable" approach.  While the LED's might get a smidge more
current on newer PICs which have a smaller voltage drop, the difference would
not be very big.  And even if a future model PIC came out which had NO volt-
age drop on its output high/low, the circuit would still be safe.

1997\08\17@204731 by Bob Blick

flavicon
face
Hi Sean,

Sounds like you are doing spice modeling. I'm not a big fan of that method,
being a former teacher I feel it is much more important to get a feel for
the real parts before simulating them.

John Payson read me correctly when he said I meant to design the amplifier
so it depended little on the characteristics of the transistors. Whether
you are designing an amp with negative feedback, or one with a dc servo, or
with degenerative feedback, anything with multiple stages cannot be
predicted accurately enough to rely on transistor specs. This is probably
even more true with single transistor amplifiers. For instance, if you
build a one-stage common-emitter amp, you have the choice of running with
as much gain as the transistor has, or reducing the gain to a calculable
number by using a resistor in the emitter(degenerative feedback). Your
choice, but the performance with regenerative feedback will be predictable,
and it's a better amplifier, too(as far as linearity, distortion and
bandwidth).

I've strayed from PICs a little, so you might reference
http://www.bobblick.com/bob/stamp/tonedec.html because it's a project that
has a PIC-based Basic Stamp in it(see, I got PICs into this email!), and a
single transistor mic preamp of the crappy type, getting as much gain out
of a single transistor as possible, but it works well enough for DTMF tones.

Cheerful regards,

Bob


At 01:26 PM 7/8/97 -0400, you wrote:
>At 09:21 PM 8/16/97 -0700, you wrote:
>>At 11:38 PM 8/16/97 -0400, you wrote:
>>
>>>What am I supposed to do if I want to accurately design an amplifier
without
{Quote hidden}

http://www.bobblick.com/

1997\08\18@050334 by nvdw

flavicon
face
> Thanks for the reply, Bob.
> I guess you are saying that it is up to the actual user of the components to
> measure their characteristics. For me, there is only one problem with this,
> I am at this point only an electronics hobbist. I am studying to become an
> electrical engineer. I don't have the money to buy the test equipment needed

NO. By using apropiate design you can make the circuit
quite independant of transistor characteristics.

The only really important things is dissipation, minimum
gain and frequency (The last being almost neglible in audio
these days). Feedback can take care of *almost* everything else.

Nic

1997\08\18@232823 by Sean Breheny

flavicon
face
At 11:02 AM 8/18/97 +0200, you wrote:
>> Thanks for the reply, Bob.
>> I guess you are saying that it is up to the actual user of the components to
>> measure their characteristics. For me, there is only one problem with this,
>> I am at this point only an electronics hobbist. I am studying to become an
>> electrical engineer. I don't have the money to buy the test equipment needed
>
>NO. By using apropiate design you can make the circuit
>quite independant of transistor characteristics.
>
>The only really important things is dissipation, minimum
>gain and frequency (The last being almost neglible in audio
>these days). Feedback can take care of *almost* everything else.
>

Thanks to all who replied on this topic. I always knew that feedback could
alter gain and input impedance, but I never knew that this was true to the
extent that it made the actual parameters almost irrelevant. I have seen
many transistor amplifier schematics (everything from audio to RF small
signal to RF power) and many do not use feedback (especially the power amps,
but I guess that the designer just makes sure that the drive would not push
the amp too far.

Also, can feedback alter output impedance? How much do input and output
impedance vary among various transistors (in the same basic configuration,
i.e. CE,CB, or CC)

By the way, to Bob who asked if I was using spice, I am not. I have been an
electronics hobbyist ever since I was about 6! (Of course, not really
seriously until I was about 10). I have tried for a long time to learn about
transistor and IC analog design. I have made a lot of progress, but the fact
that the textbooks I have been using talk a great deal about H params. and I
cannot find them on most datasheets has been I stumbling block to me until I
thought to ask it on the PICLIST.

Thanks again,
Sean
KA3YXM

1997\08\19@164737 by Harold Hallikainen

picon face
       Without getting TOO involved, here is the approach I used for
transistor circuit analysis when I taught it at the community college
level.

       For a silicon NPN transistor, assume Beta of about 100.  Assume
the base current is negligilbe.  If this is a common emitter amplifier
with voltage divider bias and an emitter resistor, determine the base
voltage using the voltage divider formula.  Assume the emitter is 0.7
volts below that.  Determine the emitter current based upon the emitter
voltage across the emitter resistor (assuming the other end of that
resistor is grounded).  Assume the collector current is the same as the
emitter current.  Use Ohm's law to determine the collector voltage.
       Assume there is an internal emitter resistance of 25 mV/Ie (if
emitter current is 1 mA, Re=25 ohms).  Gain is Rc/Re if the emitter
resistor (RE) is bypassed, Rc/(Re+RE) if not.  Output resistance is Rc
(in parallel with the very high resistance of the collector current
source, or still about Rc).  If the emitter resistor is bypassed (with a
bypass capacitor), the input resistance is Beta*Re.  If the emitter
resistor is not bypassed, the input resistance is beta*(Re+RE).
       A differential pair has an input resistance of 2*beta*Re.  If
only one output is used (single ended-unbalanced output), the gain is
Rc/(2*Re) and the output resistance is Rc.  If a balanced output is used
the gain is Rc/Re and the output resistance is 2*Rc.
       When negative feedback is applied, generally "stuff gets better"
by AB+1 where B is the "gain" of the feedback network and A is the open
loop gain of the op amp.  For example, a noninverting op-amp based
amplifier with a gain of 10 has B=1/9.  If the chip's input resistance is
20 meg and the gain (A) of the chip is 200,000 (at DC), the circuit's
input resistance becomes 444 gig ohms.  The higher input resistance can
be justified by recalling that a "happy op amp" has about the same
voltage at the two inputs.  The voltage a t the inverting input is very
slightly below that at the noninverting input (assuming a positive
output), so the input current is that very slight voltage across the
chip's input resistance.  Since there is a very small voltage, very
little current flows.  The circuit driving the non-inverting amplifier
"sees" that it is supplying a voltage yet very little current is being
drawn.  looks like a high impedance.
       Output resistance similarly "gets better" by AB+1.  In this case,
the 25 ohm output resistance of an op-amp would be divided by AB+1.
       This series of approximations seems to work pretty well, but does
not take into account frequency response limitations.  It also does not
deal with nonlinearity (though clipping levels can be determined).  The
problem I see with H parameters is that they change too much with biasing
conditions.  Using the above approximations, however, we might assume
hie=beta*25mV/Ie, hfe is beta, and hoe is about infinity.
       Finally, I've had a lot of fun with the evaluation version of
pSpice, available at http://www.microsim.com .

Harold

1997\08\19@214759 by Andrew Russell Morris

picon face
>
>Thanks to all who replied on this topic. I always knew that feedback could
>alter gain and input impedance, but I never knew that this was true to the
>extent that it made the actual parameters almost irrelevant. I have seen
>many transistor amplifier schematics (everything from audio to RF small
>signal to RF power) and many do not use feedback (especially the power amps,
>but I guess that the designer just makes sure that the drive would not push
>the amp too far.
>
>Also, can feedback alter output impedance? How much do input and output
>impedance vary among various transistors (in the same basic configuration,
>i.e. CE,CB, or CC)
>
Sean;

I have been following this thread from the beginning, and haven't spoken
until now, because what they have been telling you has been accurate.

I graduated in 1976 with a BSEET degree from Va. Tech. EET (Electrical
Engineering Technology) is Electrical Engineering with more focus on
practical application as opposed to theory. I have a textbook which I (and
everybody else I know who used it) believe is one of the best books ever
written on this subject. It covers the details of designing with most
discrete semiconductors (bipolar transistors, FETs, diodes, etc) from a
practical standpoint. It gives the formulas necessary to do an accurate
design, with explanations of what is going on, but without the detailed
theory that is not really nessary to do the design. It explains things in a
simple, easy to understand way. In the case of the information you
specifically requested, they have a table that shows equations for
calculating voltage and current gain, input and output impedance for all
three transistor amplifier configurations.

Try to find it if you can find a used bookstore that has technical books.
BTW If anyone out there knows of such a place, please let me know. The book
is called Applied Electronics" by J.F Pierce and TJ Paulus. It is no longer
in print ( I tried to get another copy of it) and was published by Bell &
Howell. If you have access to a fax machine, I will fax that table to you.
Please, don't everybody on the Piclist ask me to fax it to you.  The ISDN
number of the book, BTW, is 0-675-09139-x. The Library of Congress catalog
card number is
77-184058

Hope this helps!

TTFN :-)

1997\08\19@224459 by Mike Hagen

picon face
Great for Harold!

I watched this thread wondering why a student would be so concerned with H
parameters for transistor amplifier design.

H parameters for transistor were not even around (available) back when most
of the transistor circuits were designed.  The use of models containing the
magic 25ohms/ma was where it was at.   Learning to design an amplifier should
be first be handled just as Harold described, and then go to other means if
required.  Basic understanding using good old Ohms Law always comes first.

Mike

1997\08\19@225955 by Sean Breheny

flavicon
face
At 10:34 PM 8/19/97 -0400, you wrote:
>Great for Harold!
>
>I watched this thread wondering why a student would be so concerned with H
>parameters for transistor amplifier design.
>
>H parameters for transistor were not even around (available) back when most
>of the transistor circuits were designed.  The use of models containing the
>magic 25ohms/ma was where it was at.   Learning to design an amplifier should
>be first be handled just as Harold described, and then go to other means if
>required.  Basic understanding using good old Ohms Law always comes first.
>
>Mike

Thanks again all of you for the info on this topic. The reason I was "so
concerned"
with H parameters is that I have several poor EE texts. The only one which
actually gave practical formulas for transistor design says "Look at the
manufacturer's datasheet for the H parameters for the particular
transistor." and then proceeds to use H parameters to obtain the gain and
input/output impedance for several simple transistor amplifiers.
       I guess that this book led me wrong. I am very glad that I asked
this question on this list. I have been successfully designing transistor
amplifiers for about a year and a half, but I could never quite predict what
the gain and input/output impedance would be, mainly because I had to guess
on the H parameters and did not use any form of feedback.

It is strange that you say that H parameters were not around when most
discreet transistor design was done. The text I got the formulas from was
from around 1968!

BTW, I am 17 and just going into my freshman year at Cornell University
majoring in EE. I think I will probably get better textbooks there!

Thanks again,
Sean

More... (looser matching)
- Last day of these posts
- In 1997 , 1998 only
- Today
- New search...