Modern power relays for 120VAC?

Oh-oh, explosive effects?

Like one of our first bosses used to say about some of the more spicy foods,: It burns twice :-)

but I

None

promises to

by

be

resistance

stay

rethinking it I guess a pnp and and a diode could do it, I was hung up on using the same protected automotive fet for everything

-Lasse

but I

None

promises to

by

be

resistance

stay

Using a FET is normally more economical these days, they have become so cheap.

What, no Joerg-effects? ;-)

Nah, these would be too disgusting :-)

Thank you.

Good point. It would excite JF and AW.

but I

None

promises to

by

be

resistance

stay

Aargh. What was i thinking? Thanks for discussing what i should have remembered.

amp

but I

None

promises to

position

by

be

resistance

The

stay

force.

As for that, just keep track of e ~=3D -L di/dt and semiconductor device protection.

?-)

Maybe not. An alternative seems to involve higher on time ratios. The off time current decay rate is much faster. To a certain extent it plays better in discretes.

?-)

I really don't understand what you are saying. What has on-time to do with tunr-off? What difference does it make whether on chip or in discretes?

Not funny, when you're driving across US6, East of Tonopah, and all that the only available rest area has is a 50-gallon drum with a seat, on top of a *very* deep hole.

I thought I was going to have to use Lunar Crater. :-(

The

plays

discretes?

Catch zeners instead of forward biased catch diodes; and L * di/dt?

?-)

That would not be such a good thing in a PWM because it'll burn off tons of energy ... phssss ... *POP*

you

turn

The

plays

discretes?

It is always a tradeoff which parts get the stress. The energy always = has to go somewhere. If it cannot be directed to the load or recycled it = must be dissipated. It is a design tradeoff.

?-)

Joerg schrieb:

Hello,

but if you want low contact resistance, you need a relay with large contacts and enough force from the coil to close those contacts with high pressure. 20 A is a bit too much for a small relay on a little circuit board.

Bye

Oh, they make relays bigger than that for PCB mounting. It's no problem to need lots of power for a few ten milliseconds to pull in but hold power should be low. It is possible, I have designed such things but for injectors. For example, it could close over a cam. But progress in the relay market is slooooow.

That's where design smarts come in. It is very much possible to recycle such energy so it is not lost. Less stress on parts, less dissipation, more green :-)

There is this relay from Omron that is similar to the P&B automotive = relay.=20 It's 30A SPDT and a 5V coil at 900 mW, but hold-in current is 10% so you = can=20 use PWM. Contact resistance 100mOhm maximum, probably a lot less:

It does seem odd that a 20A relay would dissipate 20*20*0.1 =3D 40 = watts. The=20 resistance is measured with 5VDC at 1A, so I'm sure it will be much = lower=20 when there is some power available. Probably 10mOhm, and even that's 4=20 watts, so I think it will be much lower than that in real life.

You might be able to use two MOSFETs in series, and drive the gates with = a=20 little DC-DC converter. Something like this, although it's only 200V and = a=20 bit pricy, but it has only 10mOhm Rds.

Or maybe two IGBTs, but they have 1.2V CE at 20A, or 60mOhm equivalent, = (and=20 they are even pricier):

Paul=20