Maker Pro
Maker Pro

Protecting inputs of transistors of a low noise amp

Hi
I have designed a low noise trransimpedance amplifier which has a GaAs
FET as an input transistor. I want to protect the input (gate) of these
fets from high voltage transients due to a RF source in proximity.
If i add Transient voltage suppressor/ zener - is that going to
deteorate my noise performance. Or can there be leekage issues in the
zener which can suck my signal current away from the FETs.
 
J

John Woodgate

Jan 1, 1970
0
Or can there be leekage issues

Watch out for onionization, too.
in the zener which can suck my signal current away from the FETs.
Probably not, since your FETs probably don't draw appreciable input
current.
 
J

John Larkin

Jan 1, 1970
0
Hi
I have designed a low noise trransimpedance amplifier which has a GaAs
FET as an input transistor. I want to protect the input (gate) of these
fets from high voltage transients due to a RF source in proximity.
If i add Transient voltage suppressor/ zener - is that going to
deteorate my noise performance. Or can there be leekage issues in the
zener which can suck my signal current away from the FETs.

Why not regular diodes?

How much capacitance can you stand? How much leakage?

John
 
J

Joerg

Jan 1, 1970
0
C'ld use if they are more suitable!

What would be unsuitable about diodes?

No more than couple of pFs. The leakage has to be less than few nAs.

If you run them the against the rails and the bias voltage is far enough
away from there diodes should be able to fulfill that.
 
J

John Larkin

Jan 1, 1970
0
What would be unsuitable about diodes?



If you run them the against the rails and the bias voltage is far enough
away from there diodes should be able to fulfill that.

Or just back-to-back diodes to ground, if the TIA input is nominally
at zero volts.

John
 
J

Joerg

Jan 1, 1970
0
Hello John,
Or just back-to-back diodes to ground, if the TIA input is nominally
at zero volts.

That is often done in radio applications but leads to poor 3rd order
intercept, harmonics and all kinds of nasties. I'd pre-bias them in that
case.
 
M

Mike Monett

Jan 1, 1970
0
Joerg said:
Hello John,
That is often done in radio applications but leads to poor 3rd order
intercept, harmonics and all kinds of nasties. I'd pre-bias them in
that case.

But the op is not using a 10 element beam on 20 meters - it's a TIA.

Wouldn't pre-bias increase the leakage currents and reduce the SNR?

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm
 
J

Joerg

Jan 1, 1970
0
Hello Mike,

But the op is not using a 10 element beam on 20 meters - it's a TIA.

True but he wrote "high voltage transients due to a RF source in
proximity". That can modulate in and make a pudding out of the signal to
be measured. Unless it's a pulsed source and he can measure in between
pulses.

Wouldn't pre-bias increase the leakage currents and reduce the SNR?

Very little, at least at room temperature. For the BAV99 it's a few nA.
The nice thing is that with a few volts reverse bias the typical
capacitance is around 0.5pF. Not bad for such a low cost part.
 
M

Mike Monett

Jan 1, 1970
0
True but he wrote "high voltage transients due to a RF source in
proximity". That can modulate in and make a pudding out of the signal to
be measured. Unless it's a pulsed source and he can measure in between
pulses.

OK, I just got the "high voltage transients" part and skipped the RF
source. But that doesn't say much about what kind of transient he wants to
protect against.
Very little, at least at room temperature. For the BAV99 it's a few nA.

Isn't that about the same order as the signal he is detecting? I thought
TIA used gigohm feedback resistors to get some kind of useful signal from
very small currents from a photodiode. He mentioned

"Or can there be leekage issues in the
zener which can suck my signal current away from the FETs."

So leakage noise might be an issue.
The nice thing is that with a few volts reverse bias the typical
capacitance is around 0.5pF. Not bad for such a low cost part.

I wonder if the input of a 74AC04 CMOS gate might be useful as a low-
leakage clamp. The spec claims the max leakage is pretty high, say about
100na. But everyone has seen how a floating input can hold a voltage on an
input pin constant for a long time. This might work if the fault current is
not too high.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm
 
B

Bill Sloman

Jan 1, 1970
0
Joerg,
Thanks, Also i found this one, with very low reverse leakage currnet.
http://www.vishay.com/docs/85541/85541.pdf

That isn't a a very low leakage current. Check out the Calogic PAD5 - 5pA
leakage - and the cheaper JPAD50 with 50pA leakage current.

Farnell stock both.

The PAD5 offers a worst case capacitance of 0.8pF at 5V reverse bias, while
the JPAD50 can go up to 2.0pF.
 
J

Joerg

Jan 1, 1970
0
Hello Mike,
OK, I just got the "high voltage transients" part and skipped the RF
source. But that doesn't say much about what kind of transient he wants to
protect against.
If it's a radar pulse then often all you can do is gate it out and
protect against it so nothing will be fried.

Isn't that about the same order as the signal he is detecting? I thought
TIA used gigohm feedback resistors to get some kind of useful signal from
very small currents from a photodiode. He mentioned

"Or can there be leekage issues in the
zener which can suck my signal current away from the FETs."

So leakage noise might be an issue.

It can be but he wrote in the answer to John Larkin that it should be
less than a few nA. So the old BAV99 might still fit the bill.
I wonder if the input of a 74AC04 CMOS gate might be useful as a low-
leakage clamp. The spec claims the max leakage is pretty high, say about
100na. But everyone has seen how a floating input can hold a voltage on an
input pin constant for a long time. This might work if the fault current is
not too high.

Maybe that could be done. I don't know his circuit but maybe he could
even re-design it for a MOSFET tetrode from the BF9xx series. Those are
very low capacitance and often include protection diodes because their
market segment is TV tuners which need similar protection. IIRC the
BF998 contains diodes and those are pretty cheap.
 
W

Winfield Hill

Jan 1, 1970
0
John Larkin wrote...
That was a very shallot thing to say.

It's been rather peasful around here
with John Woodgate away.
 
Top