DIY SSR question

[DarkZero]

Hello All!

i bought one of those cheap electronic thermostats to control space heater in my shed but relay gets useless in about 2 weeks with my space heater so i got tired of changing relay and decided to make SSR so i came up with circuit attached below.

now it works but my heat sink for the triac gets to like 70C. My space heater is rated for 2000W at 240Vac and my heat sink is 50mm x 50mm x 10mm.

am i doing something wrong or its just the way it goes with such things because i have seen a lot of SSR modules that are smaller and rated for 25A

[Kalvin]

You can use also a combination of relay and SSR connected in parallel. The SSR will protect the relay contacts and the SSR needs to be active only very short period of time. First, turn on the SSR, then turn on the relay, then turn off-the SSR, and the relay will hold the heater on. When turning off the heater, turn on the SSR, turn off the relay and finally turn off the SSR.

[retrolefty]

You can use also a combination of relay and SSR connected in parallel. The SSR will protect the relay contacts and the SSR needs to be active only very short period of time. First, turn on the SSR, then turn on the relay, then turn off-the SSR, and the relay will hold the heater on. When turning off the heater, turn on the SSR, turn off the relay and finally turn off the SSR.

 That does sound like a through solution if not somewhat complex. I know something similar was done in a refinery where I worked where the control house for a plant had a large UPS that powered all critical loads. It had a output transfer switch that allowed one to transfer the load to and from the UPS to a alternate AC source. This was for maintenance/isolation/load testing of UPS/etc. It utilized SCRs and large relay 'sealing contacts' to take advantage of the best properties of both methods, zero-crossing turn-on/off and minimum voltage drop of the contactor relay. However I suspect the supporting circuitry control is a design challenge without micro-controller support?

[capt bullshot]

Complexity depends ...
If you need to protect the relay at turn on only (which is often enough), it's rather simple:
Use a 230V AC relay, place the contact in parallel with the SSR and the coil in parallel with the load. This ensures the turn on sequence, but holds the relay infinitely on. Use a small auxiliary relay (or a small SSR), contact in series with the main relay coil (not the load), coil parallel to the primary side of the SSR to turn it off again.

[David Hess]

am i doing something wrong or its just the way it goes with such things because i have seen a lot of SSR modules that are smaller and rated for 25A

Solid state relays are intended to be mounted on a suitable heat conducting surface to handle their losses.  You just need to do the same thing with your circuit or use a solid state relay which has less loss.

[DarkZero]

so if i draw 10A and drop is ~1v so im dissipating ~ 10W into the heat sink ?

some nice ideas to have in mind but like i said im using that cheap chianese electronic thermostat where i have on off 12v
the thing still works and temp is siting on 70C i wonder if this is too high temp and how long will it work and how will it fail in what manner ... i dont have enough knowledge to predict that stuff
 

[SeanB]

You could use the small relay to control a better industrial contactor ( use a mains voltager coil in the contactor) that will handle the heater load comfortably. A cheap industrial contactor will last 100 000 cycles of use at full current ( 16A resistive) and in your use case with the contactor should last a good number of years, they typically wear out after a half million or so cycles for any reasonable brand ( Omron, Square D, Schneier, etc) and often are cheap to free in a scrapyard in scrapped industrial equipment.

[Zero999]

so if i draw 10A and drop is ~1v so im dissipating ~ 10W into the heat sink ?

some nice ideas to have in mind but like i said im using that cheap chianese electronic thermostat where i have on off 12v
the thing still works and temp is siting on 70C i wonder if this is too high temp and how long will it work and how will it fail in what manner ... i dont have enough knowledge to predict that stuff

Yes, that's true.

If you want it to dissipate under 1W, then you need a switch with a resistance of less than 10mOhm. You won't get this from a MOSFET. In order for a MOSFET to switch AC, two need to be connected in series, back-to-back, which will double the on resistance. You will not get a MOSFET, rated to 400V, with an on resistance of 5mOhm, for a reasonable price.

[floobydust]

Your SSR needs a heatsink. Without one, it can dissipate a few W at most before running too hot, for short life.

For SSR heatsinking, a 1W heat dissipation per amp rule of thumb, so 10W for 10A with triac is what I tell laypeople.