DC SSR Won't turn off

[MatCat]

I am using a DD220D10 DC SSR being driven by MAX4820 at 5V for control, and the supply is 72V 1400ma constant current supply, and the load is 100W LED, when all is powered reguardless of control signal (even directly grounded) the SSR will stay on.  Any ideas how to get it to work properly?

[Zero999]

There are two possibilities:

1) LEDs can light with a very small current and solid state relays leak a considerable current, even when off.
2) The SSR has failed short circuit, which is the most common failure mode.

Test the solid state relay with an incandescent lamp. If it works, then it's #1 and the solution will be to connect a resistor across the LEDs. If it still doesn't work, then the solid state relay is bad and needs replacing.

[wraper]

Part number suggests it came from ebay or aliexpress. I won't be surprised if it's not even DC SSR internally. FFS don't buy them from china, they ARE fake crap which is not safe and not nearly meeting specs. If you want it cheap, build your own from discrete components. In many cases SSR is not even needed, and simple MOSFET is more than enough.

[kosine]

Yep, at that power level you've probably cooked it. They tend to fail close circuit. Would need a big heatsink and ideally a fan for what you're doing. Something like a CPU cooler should work.

I've bought quite a few of the cheap Chinese "fotek" SSRs over the years (both DC and AC), and not actually had any bad ones yet. (Tempting fate here, aren't I...)

A couple of them have been used in factory machines and are still going strong. Maybe I just got lucky, but I must have used about 20 from different suppliers by now.

[MatCat]

I know the relay wasn't failed before, I used the same relay to switch 50 watts with no issue (36v 1400ma), but with 100 it's no go, I will test the relay and see if it got cooked.  The LEDs seem to be coming on full tilt, not partially as if A really low current.

[MatCat]

Any good recommendations on a NON-SSR relay?  I would love to use something in a similar form factor (screw terminals, not socketed / pinned), but that seems impossible to find.  I have 2 open auxiliary relay ports I can use to drive H Bridge for a latching relay.

[Cerebus]

Any good recommendations on a NON-SSR relay?  I would love to use something in a similar form factor (screw terminals, not socketed / pinned), but that seems impossible to find.  I have 2 open auxiliary relay ports I can use to drive H Bridge for a latching relay.

The SSR form factor mimics that of the bases for plug-in relays, so you're kind of putting the cart before the horse there and also saying that you don't want a horse or a cart. 

[BergRD]

Just within the last few months, a guy who added on to his Water Heater had a spectacular failure.. Beware!

I have a few of those Fotek ones to play around with but not sure I would use one of this .. 'calibre' in a critical application without LOTS of testing.



Now, yours is different but figured I would post regardless.  Good luck!

[thm_w]

Just within the last few months, a guy who added on to his Water Heater had a spectacular failure.. Beware!

I have a few of those Fotek ones to play around with but not sure I would use one of this .. 'calibre' in a critical application without LOTS of testing.

youtube.com/watch?v=FV9t1GFVbhU

Now, yours is different but figured I would post regardless.  Good luck!

Big clive took a 25A unit apart and found a 8A (?) rated part inside:

So they are clearly copies or falsely labeled.

100W should be fine from a trusted seller, but I wouldn't even consider running a ~2000W heater.

edit: as said below the clearance between pri/sec is not good (8:20 in the video), so unless the real deal is a lot better, wouldn't use it either.

[wraper]

They are also dangerous because of inferior insulation between control and power side. I would not use them even under tiny load if it's anything mains related. Actually I would not use them in any situation unless I'm on uninhabited island and have no other choice.

[rstofer]

They are also dangerous because of inferior insulation between control and power side. I would not use them even under tiny load if it's anything mains related. Actually I would not use them in any situation unless I'm on uninhabited island and have no other choice.

You would be better off to just twist the load wires together.  Yank them apart if necessary.

[Circlotron]

There are two possibilities:

1) LEDs can light with a very small current and solid state relays leak a considerable current, even when off.
2) The SSR has failed short circuit, which is the most common failure mode.

Test the solid state relay with an incandescent lamp. If it works, then it's #1 and the solution will be to connect a resistor across the LEDs. If it still doesn't work, then the solid state relay is bad and needs replacing.

Third possibility, the DC SSR is connected reverse polarity and it’s mosfet body diode is conducting.

[Cerebus]

There are two possibilities:

1) LEDs can light with a very small current and solid state relays leak a considerable current, even when off.
2) The SSR has failed short circuit, which is the most common failure mode.

Test the solid state relay with an incandescent lamp. If it works, then it's #1 and the solution will be to connect a resistor across the LEDs. If it still doesn't work, then the solid state relay is bad and needs replacing.

Third possibility, the DC SSR is connected reverse polarity and it’s mosfet body diode is conducting.

They are normally constructed with back to back MOSFETs with complementary gate bias precisely to make this a non-issue. I've never seen a polarised DC solid state relay, although I'm sure that there's someone out there who's cheap enough to have made one.