Ice Cube Relay and Precision Current Transmission

[elfman]

Not sure if this is the best place for this, but I have an electromagnetic application where we are driving 2 coils with 0.2-2.0 amps with TDK Lambda power supplies. The application requires reversing polarity of the current and this is where we are using the ice cube relay. It is critical that the current is controlled down to single-digit milliamp accuracy which is why we are using the Lambdas. We have another test station setup that isn't automated and are finding that we have a significant amount of discrepancy between the automated/ice cube relays and the manual toggle switches with the manual toggle switches being much more consistent/accurate/repeatable. Are there relays out there that are really good for this kind of application, are there others that aren't? I would really appreciate any advise, council on the subject.

"To be or not to be, that is the question" - William Shakespeare via Hamlet

 

[itsmoked]

Relays always have varying contact resistance - "it'll never work"

You should be using H-Bridge controllers for accurate reversals like you're looking for.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[IRstuff]

So, there's probably two things going for the manual switches, bigger contact areas and higher contact force through the springs built into the manual switches.

What is your allowable contact resistance? Have you considered using wetted relays, which might be more consistent for low current applications?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[3DDave]

Switches often have wiping contacts that scrape any oxidation from the surface; I don't recall seeing any relays that have wiping contacts (there might be some), but there should be some instrumentation relays that are designed for consistent resistance. For certain relays are used in precision test equipment to avoid the voltage drops associated with solid state switching.

 

[IRstuff]

How is the current controlled by the tester?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[itsmoked]

Dave; Most ice-cube relays scrub because they have those long bars for the moving contact and the bar flexes requiring the contact to move back and forth to provide the flexing length. Still they're not going to give repeatable results.

Actually they'll give the worst kind of repeatably - six identical with one far different value, followed by three different ones, and then a bunch at the same value. This will drive the tester, human or otherwise, insane.

Wetted relays work but are becoming an anachronism because the best were mercury.

H-bridges are a dime a dozen, get the right one and be done with it. No more relay variability and wear-out.
Very first one I came up with on an H-Bridge search
Probably cheaper than a relay.

Keith Cress
kcress -

http://www.flaminsystems.com