Relay selection for Ultrasonic signals

Hi all,

I want to select a relay. The relay will be used for switching ultrasonic signals(25Mhz frequency). The contacts also can see 900V spike signals through it. Also, the BNC cable which will be used to carry ultrasonic signals will be having 50Ohm impedance. How to select a relay which has all the above requirements

Coil Voltage: 24V
Contacts(changeover contact): Carrying ultrasonic signals of frequency 25Mhz and has sometimes seen a voltage of 900V spike


Can anyone suggest me a relay or the selection procedure to follow?

Thanks in advance.

If you do not switch a 900V signal, most of the small signal relays available on the market support at least 1000V between contacts (isolation) and can handle a 25MHz signal.
But if want to switch a 900V signal, you enter in a different category of high-voltage relays.

https://www.panasonic-electric-works...relay_1060.pdf
http://components.omron.eu/ProductsB...Signal--Relays

Thanks for the reply.

I selected ARJ2024 from Panasonic for our application. Let us check the performance



Do you know any relays having a faster switching(frequency of around 100Khz) and having a load current capacity of 100mA

I don't see a necessity to use GHz relays for 25 MHz signal. I would probably select a slightly bigger relays with higher contact breakdown voltage.



Impossible with mechanical relays, even reed relays don't work faster than a few kHz rate. Also consider limited lifetime with high frequent switching.

And also, because ARJ2024 is a microwave relay the isolation is given only in dB.
30dB isolation doesn't tell you anything about high-voltage isolation capabilities.
As was mentioned, any of the signal relays from the links above will do the job at 25MHz.

there are electro-optical "squibs" that can be used that are wicked fast. They are semiconducting pieces of material, with two electrodes. Nothing happens when the rf is applied, UNTIL you shine a laser light beam on the material, which generates charge carriers, and suddenly the "switch is closed". I think you can get sub microsecond switching times...but you need the complexity of laser diodes to turn them on/off.