Solid State Relays

[LAST WATT]

Anyone used solid state relays for mains power?

[Enzo]

A variety of larger power amps in pro audio use a big triac for power switching. Then the power switch itself only has to handle the gate. This seriously extends the life of the switch - no arcing and pitting.

[CBS240]

I found it works great, but for toroidal transformers the triac (SCRs) switching noise can be significant and make the transformer buzz. However if the transformer has more of an air gap and with higher leakage flux such as an EI core, the switching noise is greatly reduced or non-noticable. Using snubbers will help reduce the noise as well. I found this out, building a slow start circuit for a 120VAC, 1.4KVA toroid. The triac seems quite suitable to switch the toroid onto the line through the current limiting resistor, and then after set time, a relay closes and connects the primary directly to the line. I would bet that switching such a transfo directly onto the line would cause huge surge current vaporizing the switch/relay contacts and/or the mains fuse....actually I know for a fact it will.

In my case, the triac is driven by a logic gate and a photo-triac for mains isolation and full wave on. (minus the switching time of course)

[LAST WATT]

Tks for the info.....

I have used back to back SCRs driving annealing transformers in the past and they work great. Using them for "soft start" makes sense to help with inrush and whatnot. What about using solid state relays on rails? I.E. the type that utilizes 3-32 VDC control inputs. Any thoughts on these being used for protection circuitry? The reason I ask this is that I am sitting a boatload of them from another job and figured I would try them out on the bench supply rails.

[metalsculptor]

I have used SSR's to control ovens, just the thing to use with a PID controller and a 2 second cycle.

Just be aware that they are triac based and will introduce some switching noise and they will not tolerate capacitors in the load without some inductance to control dI/dt. They are used in industry frequently for mains loads.

As for protection on the rails they will only switch off at zero current making them slower than a fuse, a mosfet or bipolar device on the DC rails after the filter capacitor would give better protection.

[LAST WATT]

Yep, same here...

Wired some industrial ovens for "burn in" of components using the same methods you describe.

The idea I was thinking of trying utilizes SSRs designed for DC. They are 3-32 VDC controlled with a switching rating of 100VDC @ 10 amps. As mentioned, these things have been laying around here for about 15 years and I was about to toss'em until I thought of this idea. I will try them on the bench supply and see how fast they are using the sillyscope. If they can be used to help in a "current sense" mode circuit it might "tidy" up some of the required component count.

The idea came to me from another job as where we were designing "arc" guards for RF sealing machines utilizing 866A tubes. That circuit was basically a current sense shutoff and was very fast to protect not only the platens but the internal ocsillator components as well. The circuit basically shut down the oscillation by slamming the grid in the main power tube at the same time cancelling whatever "control" level at that time....eh...boring stuff. Anyway, that was the thought anyway, try something similiar using these SSRs on a bench supply to help with R&D of OPSs and whatever else.

Tks a mil for everyones input. If I try this out, "I'll report, you decide", if they blow up.. I'll take carnage pics

[agdr]

I've used DC SSRs in several projects.

Here is a circuit I've used several times, although the SSRs (6 pin DIP) would run and hide if they saw your power units. These are 60 volt low current.

This circuit guarantees that if I'm using two separate DC wall warts for the rails (two DC power plugs on the chasis) that neither rail will come up until *both* wall warts are plugged in. And vise versa both rails die together if either cord is accentally pulled out. Prevents the case of having one rail up and the other down.

D1 and D2 protect against reverse polarity (adaptor with tip negative instead of tip positive). D3 and D4 prevent circuit capacitors from back-feeding the LEDs and keeping the SSR on in the case of either supply being removed.

[LAST WATT]

agdr,

Interesting OMRON part. I might try this part and a variation of the circuit in some of the VA stage power in my projects. Tks for the info...

Getting ready to go try the SSR idea in my bench supply as mentioned earlier and see what happens....

[agdr]

Here is a revised diagram with the diode drops added that I forgot to show.

With the diode drops it would be better to source with 18V if someone was trying to start with 15v, etc. Using schottky diodes would trim some more voltage drop.

Please post your results on your circuit tests! Interesting using them for mains power.

[agdr]

...also, here is a bunch of helpful SSR application information from Omron on all types - opto/SCR, opto/Triac, opto/transistor, etc. They discuss a power opto/mosfet in this one where a photodiode array triggers the mosfet. They have a nice chart in there of inrush current for various devices (they mention transformers) vs. sizing of the SSR:

http://www.components.omron.com/components/web/PDFLIB.nsf/0/0F7403ADF33E949C85257201007DD5C1/$file/SSR_Tech_Information_0609.pdf

With the G3VM series small-signal opto/mosfets the led triggers the mosfet directly. 1 mS switching time, linear transfer with no distortion, and 2.5kV built in DC transient absorbtion:

http://www.components.omron.com/components/web/PDFLIB.nsf/0/744B2171FDFA7D9F862573D90061B11C/$file/J02C-E-02+G3VM+MOSFET+Relay+brochure+2007.pdf