Modern power relays for 120VAC?

But I do. I invented a way for CMOS to drive a relay without the need for a protection diode... for Chipcom, 12 years ago.

...Jim Thompson

or

if

not

I was

same old.

at

and

at

Sam Adams makes an Apricot Ale now??? Wow, thanks! I used to be addicted to the Pyramid Apricot Ale until they seemed to have changed the recipe around the year 2000 or so.

=3D)

Michael

If you have the test gear for it, measure force versus pole position and current, then you can calculate just about everything you need.

...Jim Thompson

AC relays and contactors do that automatically. Once the armature seats, the inductance goes way up and the coil current drops. I recall ratios like 5:1 or something.

One of my customers marks our stuff up 6:1 to their end-users. And keeps pounding on us to cut our costs.

was

same old.

It's quite the opposite for me although I naturally grew up with Pilsener. The Anchor Steam OBA at your favorite saloon was one of the best beers I had in months. Unfortunately unobtanium around here so I'll have to swing by again.

Oh-oh, something like a kid's prank gone terribly wrong? :-)

Well, I don't have any of the relays and the decision has to be done by tonight. And the Concorde don't fly no more.

Normally one would expect such stuff to be in the datasheet. I can't be the only one PWM-ing relay coils. It's just like the contact resistance spec which is the pits. Like a weather forecast of this "quality": The temperatures tomorrow will be under 150F.

Most relays specify must-pull-in and must-drop-out voltage. If you use PWM on the coil, you can fine-tune it later.

Can you use an AC coil relay, with an opto-triac pilot switch?

The must-pull-in is usually only about 20% or so under the nominal. I'd need 50% to save on dissipation. On solar inverter relays they actually spec that in great details because of some silly "green laws" or whatever, but in my case this spec is of major help. It can take forever to cajole that sort of information out of a traditional relay company.

Yes, absolutely and it would even be nice, but those are typically massive hunks of metal. And both would have to be cheap, low single-digits Dollars.

Just looked at the Potter&Brumfield T92 series. Absolutely ghastly, 4VA on the coil for the AC versions, no pulled-in drop-back to write home about, and over 10 bucks. Yikes. If they are all like that then it has to be DC.

Well, you could cheat and use a low-power relay or phototriac isolator to switch 120VAC to a big relay with an AC coil...

If you aren't putting spade lugs out the top, that means you have 120VAC on the circuit board, nicht wahr?

I've looked at 120VAC relays. Too big, too expensive, and way too wasteful in coil energy.

Yup. Asi es como sera :-)

If it's important in your application, you'll also need to cajole the definition* and the deterioration of functional vibration and shock resistance with hold current. That sets the requirement for excess hold current.

Solar inverters probably don't care much about that spec.

• Sometimes the spec allows closed contacts to open to some extent and they're often not too clear about what voltage and temperature the spec applies to (usually nominal only, but they don't say so).

I might be a bit old fashioned, but IMHO running 20 A through ordinary PCB tracks is a quite a lot.

If 20 mOhm is too much in the actual relay contact, have you calculated, how wide and how thick PCB tracks are needed to have a comparable (20 mOhm) resistance from loop input to output in addition to the actual 20 mOhm relay contact resistance ?

Joerg wrote in news: snipped-for-privacy@mid.individual.net:

In which case, put a triac on your board and switch that on/off. With the specs you asked for, all relays are monsters.

the heat from the forward voltage on a triac and 20A will quite a bit of space too

if its ac maybe two fets back to back, but it'll need an isolated supply

-Lasse

With 2oz copper that'll be in the very low single-digit milliohms.

Well, a relay is a requirement in this case. There has to be physical gap when off, not just a semiconductor junction.

to

Just to throw another curve at you, the contact resistance might not stay in spec if you PWM the coil current and thus the contact normal force.

?-)

as

old.

You're not going to find anything with those specs, and the contact resista= nce will rise rapidly with number of closures depending on the type of load= ing. The low coil power dissipation doesn't make an enormous amount of sens= e given the relatively large kW loads you're switching. You must have some = kind of heat dissipation issue going on with the board. Maybe you need to r= econfigure with just a small relay on the board driving an external contact= or.