relay switching speed

The "on" switching time is depending on the speed with which you can push current into the relay coil, and would rather quickly get into the range to burn out the coil, if you want high switching speed. The "off" switching time however depends two things, first how fast you can kill the current in the relay(rather difficult), and secondly on a metal spring in the relay to pull the relay "open".

The spring is what is limiting the ultimate speed.

An ordinary relay will be ~50-200 hz, and very small relays, like a reed relay for example might be pushed a bit higher, like 200-400 hz

A quick test would be to put the NO contact in series with the coil, and make it self-switching, acting like a buzzer.

You can increase the supply voltage a bit, but you risk burnout of the coil. Also the contacts wont last long.

1000Hz? unlikely. 10000hz? wishful thinking. If you really want fast mechanical switching , use a rotary switch, driven by a motor.

Using FET switches, you can get into the Megahz range.

Ok I found this part at Digikey.com CLA280-ND it is a static relay, that can handle the required switching but it is not a SPDT, and there are no SPDT in the 6 amps or higher regions.... are these high power relays done in static technology?

K

Some of the small telecom-type relays, like a Fujitsu FTR-series maybe, can pull in or drop out in under a millisecond if you drive them right, so could juat about handle 1 KHz. Most bigger relays are a lot slower.

Solid-state relays can be faster, but not always.

John

Firstly, if the coil current is modulated at 1000 Hz, this is NOT DC current we're talking about. On every cycle, you'll generate some core heat due to magnetization losses. This may be important. Smoke is possible. Do not expect that 'normal' operating voltages are sufficient, either

Secondly, the 'ON' position of the relay is achieved by hitting a pair of parts together , and they either progress to hammer each other out of shape OR they engage in elastic deformation. That elasticity means there is bounce, and the 'ON' status is very much compromised until the bouncing stops. This is the one, the only, item that a relay manufacturer is likely to specify about the short-time character of the relay.

And third, the 'OFF' position usually is a long way from the 'ON', so that sparks don't occur. If you expect 5 kV operation, or 50 mV operation, the suitable delay-to-OFF time will be quite different. Some small-signal (telephone) relays have impressive contact lifetime with circa 50V stress; lower voltages have dirt-related reliability issues, and higher voltages have wrath-of-Zeus related reliability issues. At 1 kHz your prospective device will hit millions of times per hour, and contact life will be a major issue. Practical vibrators (I seem to recall) worked at tens of Hz.

The literature on this kind of application is heavily influenced by Motorola and relates to early car radios (which Motorola made); the oscillating 'relay' was called a vibrator, and failed as frequently as any of the vacuum tubes.

Most likely induction is to high in the coil for that kind of speed, you'll never get enough current to pull the contacts.

Assuming this is a 24 Volt AC coil, it has been wound and designed to create enough current with in the cycle time of 50/60 hz..

Ect...

"

typical DC relays are good to something around 100Hz (AC relays have slower switching rates)

you may be able to coax higher rates by closed loop control ISTR getting about 300Hz in this way once. (i wired the NC contats in series with the coil and adjusted the supply for the highest pitch)

if you want to go higher you may need to consider piezo actuators instead of magnetic, or go solid-state instead.

Bye. Jasen

Forget about it. As mentioned in all of the above posts, relays are far too slow for your application. AC relays are even slower because they have a copper "shade" on the pole piece to deliberately keep the field from collapsing fast. This keeps the relay from chattering on AC drive.

Your application requires electronic switching for the speeds you mention. Transistors, FETs, SCRs, etc. should be considered depending on the exact requirements.

So you've been extensively lectured about why relays won't work.

Share what you're trying to switch so fast, and maybe someone can help you find something that will work.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

:Ok I found this part at Digikey.com :CLA280-ND : it is a static relay, that can handle the required switching but it :is not a SPDT, and there are no SPDT in the 6 amps or higher :regions.... are these high power relays done in static technology? : :K

Did you look at the datasheet?

$file/CPC1709.pdf

Max T-on is 20mS and max T-off is 5mS. This limits the maximum (reliable) operating speed to 40 operations per second but you would need some safety margin. Say around 37 ops per second guaranteed.

A reed relay will typically have operate and release time of around 1mS but they can't handle the current you are talking about.

:Did you look at the datasheet? :

$file/CPC1709.pdf : :Max T-on is 20mS and max T-off is 5mS. This limits the maximum (reliable) :operating speed to 40 operations per second but you would need some safety :margin. Say around 37 ops per second guaranteed. :

And that is assuming that an on time of say 1 - 2mS is sufficient to perform the required function.