[Premium-Rx] Tubes versus FETs

[Gary Geissinger]

Richard,

I have had to fight this war a couple times in the past in FET
instrumentation around nuclear accelerators.  An arc in a vacuum chamber can
be very fast and very high amplitude.  My experience is that the trick is to
guarantee that the transient limiting device is faster than the device it
protects.  FETs can be very fast to blow indeed.

I agree with the low pass filter concept that Ahmet proposed.  Here is why.

A fast Gas Tube Arrestor takes a finite time to conduct.  This generates a
narrow (but high amplitude) pulse that is conveyed into the gate of the
device.  The gas tube limits the energy, but the residual energy on the
leading edge can be more than enough to punch through the gate.  One
suggestion is to put an RC (or maybe RL) circuit in front of the gate but
after the arrestor.  This is the low pass filter that Ahmet described.
Since the gates of FETs have capacitance to the source and drain, this may
be nothing more than placing a low value resistor in series with the gate.
This filter will integrate the spike down to a voltage level that can be
handled by the device.  Usually it is possible to set the filter such that
it doesn't knock out the desired signal, but knocks down the spike. 

I remember reading an application note some time ago from Marconi
Semiconductor concerning their static protection circuits.  They consisted
of:

1.  A multisection spark gap,
2.  A 2 K series resistance, and
3.  A distributed diode/resistance network to power and ground.

Their SOS logic could take repeated hits from an ESD source with no
damage...none of this whimpy "human body model" stuff!!

Good Luck,

Gary


Hello Richard,

Interesting subject, for me at least.

There are several way to protect transistors especially FETs being more
sensitive ( if you are using MOSFET they are even a lot more sensitive
to high voltage spikes ).  If the configuration you are using is the
Common Source ( grounded Source of the FET ), then you can try to
put a Choke Coil say at 1mH value between the Gate of the FET and
the Ground. Basically you are Shorting the Gate of the FET for DC and
not AC.  However lightning is a DC pulse and it contains AC components
that will still not be completely shorted out by using the choke.
You can use a Low Pass filter ahead of the Gate to help this.
An addional safety measure is to use Gas Tube Arrestors and similar
low loss devices.  They are very very effective. In addition companies
such as CP Clare sell transient protection devices which can be
used on the FETs.
Can you tell me what FET you are using ( part no and mfg ).
Also can you tell me how you are matching the antenna to the input
of the FET as well as how you are getting output matching to 50 Ohms.
FET 's are not too easy to match unless a negative feedback is used.
I am curious about the circuit.  Are you getting amplification ? Sometimes
this is very difficult to measure for obvious reasons. An input and
output matched amplifier is of course easy to measure however.
Another thing, some Bipolar transistors have much better Noise Figures
as well as Gain as opposed to FETs.  They are also by their nature
much more resistant to High Voltage Spikes.  You may want to experiment
with these as well.  They are much much lower cost.  I can also recommend
a company who makes very nice amplifier modules ( 50ohm matched )
with a 20db Gain and they claim a 2.5db NF.  All for less than $ 40.00
including the case and connectors. Very nice device.

However I really do not think that you need to go to tubes as there are
many preamplifiers mounted outside operational in perfect shape for years
without any "poof"s.

Regards
Ahmet

Richard Reich wrote:

> Greetings.
>
> For a few weeks now I have been messing around with receiver antenna
'impedance
> matching' using a super-linear JFET. Effectively the antenna wire goes
onto the
> Gate of the FET, which is biased in its linear region. The output of the
FET is
> then matched down to a lower impedance, which is great for most comms
receivers
> at 50 or 75 Ohms. The problem is this: poof!  A little bit of static, or a
> lightning strike nearby, and the FET dies. The Gate is already 'leaky' to
ground
> - but I suspect not enough to give protection, and reducing it any more
defeats
> the object. The FETs I have been destroying are $15 each! So I was
thinking,
> perhaps a tube would be better for this application. Does anyone have any
prior
> experience of this? I am not really looking for voltage gain in itself -
but
> obviously low noise is important. I have a box full of tubes which I could
> experiment with, but don't want to embark on a fruitless excerise. Anyone
have
> any info on this?  Many thanks - Richard
> ***************************************************************
> Richard Reich
> Principal Engineer Hardware
> SAAT Technology Ltd
> Web: http://www.saatt.co.uk
>
> DDI: +44(0)1420 545388
> Fax: +44(0)1420 87259
>
> Privileged, confidential and/or copyright information may be contained
> in this e-mail. This e-mail is for the use only of the intended addressee.
> If you are not the intended addressee, or the person responsible for
> delivering it to the intended addressee, you may not copy, forward,
> disclose or otherwise use it or any part of it in any way whatsoever.
> To do so is prohibited and may be unlawful.
>
> If you receive this e-mail by mistake please advise the sender
> immediately by using the reply facility in your e-mail software.
>
> This message is subject to and does not create or vary any contractual
> relationship between SAAT Technology Limited and you.
>
> Registered Office: Test House, Mill Lane, Alton, Hampshire, GU34 2QG
> Registered in London. No. 4106589
>
> _______________________________________________
>
> Premium-Rx Mailing List
> To Post: premium-rx at ml.skirrow.org
> For Info: http://ml.islandnet.com/mailman/listinfo/premium-rx


_______________________________________________

Premium-Rx Mailing List
To Post: premium-rx at ml.skirrow.org
For Info: http://ml.islandnet.com/mailman/listinfo/premium-rx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qth.net/pipermail/premium-rx/attachments/20030305/03bd83e7/attachment.htm