DC SSRs

Is anyone switching battery banks with DC SSRs?

If so which SSRs are you using?

The same goes for mosfets if you know of a good one that will get the job done

iceisfun said:

Is anyone switching battery banks with DC SSRs?

If so which SSRs are you using?

The same goes for mosfets if you know of a good one that will get the job done

Click to expand...

These work, however they do not protect from reverse bias (Gets very hot, conducts anyway)

The main problem with them is I loose about 10% overall power using these after a few full charge/discharge cycles.

http://www.ebay.com/itm/371874809407


Later I might measure the current and drop across them to see more specifically how bad they are for this application.

Would a contactor not do the same job with far less current loss and heat ?

I bought one of these and ran some tests on it.

It had an On resistance of ~2mOhm (0.002Ohm). Without a heatsink it ran warm to touch with a current of 88A, around 110A would be the absolute maximum without a heatsink. You can get a 400A version of this SSR which should have a resistance ~0.5mOhm. I also checked the voltage rating by placing 60.8V across the switch when switched off for several days.

Simon

KtB said:

Would a contactor not do the same job with far less current loss and heat ?

Click to expand...

I have contactors on the way, the possibility of a mechanical failure leaving the circuit closed (worst case) is what drove me away from that idea initially, its possible for a fet to fail anyway.

karrak said:

I bought one of these and ran some tests on it.

It had an On resistance of ~2mOhm (0.002Ohm). Without a heatsink it ran warm to touch with a current of 88A, around 110A would be the absolute maximum without a heatsink. You can get a 400A version of this SSR which should have a resistance ~0.5mOhm. I also checked the voltage rating by placing 60.8V across the switch when switched off for several days.

Simon

Click to expand...

Thank you, I will also check those out and do some data logging.

Could you perhaps do a test to see if hooking up the load side backwards it stops reverse flow? If you have a current limiting power supply neg to pos and pos to neg on the load side and see if it pulls it down or acts "open"?

Part of the configuration I had in mind for this is one SSR connected to the inverter rail and one connected to the charger rail (so the D and S are flipped, G go to some logic pins and there are ADCs monitoring each individual pack.

Ideally I could connect each individual pack to the charge or inverter rail any way I like, and also unplug or plug in packs without a spark/load connected.

Once I get all my proof of concept stuff done I'll draw up a more formal design and make a better post about it, right now I'm just making sure everything does its job and the numbers come together before I design controller boards.

iceisfun said:

I have contactors on the way, the possibility of a mechanical failure leaving the circuit closed (worst case) is what drove me away from that idea initially, its possible for a fet to fail anyway.

Click to expand...

Unfortunately one of the main failure modes of MOSFETs is short circuited.

Could you perhaps do a test to see if hooking up the load side backwards it stops reverse flow? If you have a current limiting power supply neg to pos and pos to neg on the load side and see if it pulls it down or acts "open"?

Click to expand...

Just testing with a multimeter using the 'diode tester' tells me that it will not stop reverse flow. The reason and solution is in the following diagram from this article





I can't see any reason you couldn't put two SSR's back to back to make a bidirectional switch.
Simon