Transistors as V clamps?

[Steve Sousa]

Hello:

I've seen the following circuits used to protect the IC's on multimeters.
I think that they use the reverse breakdown voltage of the transistor to
achieve this effect, is that correct?
Why not use a zener?
Isn't it bad to reverse bias transistors beyond their RBV?
How does circuit B work?

Thanks.


In o---------o---------o Out
|
|>
.--|
| |\
Circuit A | |
'----.
|
===
GND


In o---------o---------o Out
|
|>
.--|
| |\
| |
o----o
Circuit B | |
| |/
'--|
|>
|
|
===
GND
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

 

[Fred Bloggs]

Hello:

I've seen the following circuits used to protect the IC's on multimeters.
I think that they use the reverse breakdown voltage of the transistor to
achieve this effect, is that correct?
Why not use a zener?
Isn't it bad to reverse bias transistors beyond their RBV?
How does circuit B work?

Thanks.


In o---------o---------o Out
|
|>
.--|
| |\
Circuit A | |
'----.
|
===
GND


In o---------o---------o Out
|
|>
.--|
| |\
| |
o----o
Circuit B | |
| |/
'--|
|>
|
|
===
GND
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

Circuit B is the same thing except for bipolar signals- note the
symmetry for signal +/- with respect to GND. Circuit A clamps negative
signals at -Vbe. The transistor clamps differs from the zener in that
Vce falls with conduction, a zener clamps at a constant voltage. This
Vce fall is due to the increase of transistor reverse beta with emitter
current-and is a form of positive feedback action.

 

[henryf]

I've seen the following circuits used to protect the IC's on multimeters.

I've seen similar things used to protect the inputs of op amps.

I think that they use the reverse breakdown voltage of the transistor to
achieve this effect, is that correct?

Pretty much. For circuit A, it depends on the polarity of the
input.

Why not use a zener?

Based on a quick google search and my limited sampling of
data sheets, it seems that real zener diodes have greater
parasitic capacitance when used this way. (The zener
datasheet I checked showed 50pF to 100pF, while the transistor
datasheet indicated 5pF or less.)

Isn't it bad to reverse bias transistors beyond their RBV?

Only if you ever plan to use them again as transistors. It
messes with the gain.

How does circuit B work?

Like two back-to-back zener diodes. It clamps the input
voltages (either polarity) at Vz + Vbe.

 

[Steve Sousa]

Pretty much. For circuit A, it depends on the polarity of the
input.

So all i need to do is pick a transistor with a RBV that suits my app,
nothing else to consider?

Thanks for the explanations.

 

[Steve Sousa]

Circuit B is the same thing except for bipolar signals- note the
symmetry for signal +/- with respect to GND. Circuit A clamps negative
signals at -Vbe. The transistor clamps differs from the zener in that
Vce falls with conduction, a zener clamps at a constant voltage. This
Vce fall is due to the increase of transistor reverse beta with
emitter current-and is a form of positive feedback action.

So the clamping voltage will vary with the input voltage->current?

Thanks

 

[henryf]

So all i need to do is pick a transistor with a RBV that suits my app,
nothing else to consider?

No, it's not that simple.

First of all, circuits A and B use a transistor's emitter-base
breakdown voltage, which is typically near 7 volts -- it's
unlikely you'll find any outside the range of 5V to 10V.
Unless your app can live with that, you'll need to use a
different voltage-clamping circuit.

Second, there are other things to consider, including:
- power and thermal considerations
- packaging
- price
- availability

 

[Fred Bloggs]

So the clamping voltage will vary with the input voltage->current?

Thanks

Yes- I was assuming there is a resistor in series with the signal source
not shown. The only advantage this circuit may have over a zener is a
better leakage characteristic, but I am not 100% sure. All the breakdown
devices will have more or less reverse current through them as the input
approaches the breakdown rating. For the standard 500mW zener, this can
be as much as 10's of ua at 80% Vzener, just as an example. If you say
this protects a meter, then that level of leakage may introduce an
unacceptable inaccuracy at signal levels they do want to measure, and
increasing the zener breakdown rating to reduce leakage is not an
option. Unlike the zener with a relatively soft knee, when the
transistor starts reverse breakdown, the voltage required to sustain
larger currents is reduced and there is an abrupt increase in shunt
current- this should be a sharper breakdown characteristic.

 

[Steve Sousa]

No, it's not that simple.

I'll have to give it a try on real parts and see what comes up.
Thank you and Fred for your help.

 

[GPG]

I'll have to give it a try on real parts and see what comes up.
Thank you and Fred for your help.

Have a look at this:
http://www.reed-electronics.com/ednmag/contents/images/85637.pdf