Protection for frequency meter input

[PinheadBE]

Hi,

I scored an old ICM7216 based frequency meter, which is working very well.
I do not have nor schematics, neither precise specs.
The UHF input which goes through a prescaler (reference scrubbed off....) has a limitation of 3 V p-p (printed on the front panel)

The overall bandwidth on this input is unknown (I do not have a sig gen that goes over 60 MHz), but will probably not exceed 100 MHz

I was looking for a circuit to protect that input . There seems to already be two simple diodes, but they look so fragile (1N4148 like) that I do not trust them, and I would like to add some more beefy protection.

I was considering figure 11 on this page https://www.digikey.be/en/articles/protecting-inputs-in-digital-electronics without the cap, of course.

But I would prefer to have some sort of "Automatic Gain" circuit which automagically keeps the amplitude within the 3 V p-p range.
I suspect this must be fairly simple with an op-amp(s) based circuit, but my knowledge in analogue circuitry is too basic.

Could someone help me with this, or point me to a page with some useful information ?

Thanks a lot !

[PinheadBE]

.... or simplier : A simple 2.9 V Zener and its series resistor ? 

Would this work without affecting the bandwidth (< 100 MHz) ?

[ledtester]

Generally frequency counters have an amplifier at the input because they are usually used with very low level signals -- like the mV RMS range.

If you are concerned that a signal source may be too powerful you can always insert an attenuator to decrease it:

#91: Basic RF Attenuators - Design, Construction, Testing - PI and T style - A Tutorial -- w2aew
https://youtu.be/A5gGeV7CiQ0

Some counters have a built-in attenuator you can activate with a switch.

[Kleinstein]

Zener diodes often have quite some capacitance. So they may load down a higher frequency signal quite a bit.
The more usual way are diodes like the 1N4148 towards some voltage (zener or string of diodes with a bias current).

The diodes are ususally not that bad and tend to fail short unless there is really serere overcurrent (e.g. > 1 A for a 1N4148).

[PinheadBE]

Thanks for your answer.

I intend to use this meter also to test logic signals that can go up to 5 V.   Therefore, I would prefer not to rely on the original protection: this was already a cheap meter in its time, and since the markings are not readable on the prescaler chip, I would not like to have it releasing its magic smoke....

I will take a look at that YT video.

Have a nice day