SSR latching?

[BetterSense]

I have this SSR

http://www.amazon.com/gp/product/B00843IUG6/ref=oh_details_o00_s00_i00?ie=UTF8&psc=1

I'm driving it with a microcontroller at 5V, hooked directly up to the inputs from the microcontroller output pins. The micro is adjusting the element duty cycle with PWM.

When I turn the SSR all the way up to 100% and leave it there for a few seconds, it 'latches' on. If I turn it down, it stays on 100%! I can make it return to normal operation by 1) cycling the main power to the whole box (micro and all) or 2) unplugging the element and plugging it back in (micro appears to stay working normally).

Is there anything I can do to prevent this? Is this normal or is this a bad SSR?

 

[AllanMar]

how are you determining the % of the SSR?

 

[BetterSense]

100%=on all the time
50%=on half the time
0%=not on

PWM frequency=1/2Hz

 

[AllanMar]

No I know that, I mean how are you measuring it? Are you just watching the voltage cycle with a regular multi-meter? Normally thats not good enough for PWM, but I guess if its only 1/2 freq.

If you disconnect the 5V side of the SSR, it is still stuck on?

 

[BetterSense]

I'm trying to figure out what you would have me measure.

I have an indicator lamp on the SSR output. I also have an LED that blinks whenever the microcontroller switches the SSR input. That's what I'm watching.

The SSR works at all duty cycles up to 100%. I have had it boiling water for an hour at 50% duty cycle. However, if I turn it up to 100, and leave it there for at least 5 seconds or so, it will not turn down again. The output lamp stays latched on, even though I can see the microcontroller is back to cycling at a lower duty cycle. If I disconnect the heating element and plug it back in, it goes back to cycling at the microcontroller frequency. The microcontroller does not seem to have rebooted or frozen up; it continues cycling the whole time.

I can try unplugging the microcontroller output from the SSR input and see what that does. However, I'm not sure what that will tell me.

 

[reynolds5520]

I can try unplugging the microcontroller output from the SSR input and see what that does. However, I'm not sure what that will tell me.

That will tell you whether the relay is locking internally, or if something external is telling it to stay on. A useful test.

Is it possible that one of your indicator light circuits is feeding back to the input and latching it on?

Hmmm, you did say that disconnection the element will reset things. Is it possible the element is back feeding the control in some way?

Can you show us your schematic?

 

[BetterSense]

I can post a schematic in a bit.

 

[BetterSense]

Schematic:

Here is a video of the behavior. The small LED is SSR input. The big lamp is SSR output. It works great except I turn it all the way up, it latches on.

http://www.youtube.com/watch?v=jiqjL5y6pPo&feature=youtu.be

 

[AllanMar]

What sort of lamp is that on the SSR output? In your youtube video desc your said it was a 120V light? It is in fact wired across the 2 hot legs though?

It would still be useful to know if it remains latched when you disconnect the low voltage side of the SSR.... Also, does the LED on the SSR follow this behavior?

 

[BetterSense]

The 240V SSR output indicator is a 120V indicator lamp with a 130k resistor so it doesn't pop with 240V. The 120V indicator is also an incandescent panel indicator. The SSR input is the only one that is an LED.

 

[AllanMar]

I suspect the SSR is not latching on, your just getting a false indication that it is. Check the voltage across the output to the element with the element connected, see if it stays on.

 

[BetterSense]

My boil kettle disagrees with you. When it latches it's obviously 'real' because the boil goes crazy (5500W element in 5 gallons).

Just now I tested it like in the video, but after it latched on, I took the lid off the box and disconnected the leads from the microcontroller so the input wires to the SSR were dangling free and it stayed latched. There was nothing connected to the SSR input except an LED and series resistor. It also did this before I added the LED; I added it to troubleshoot this issue. It un-latched when I cycled the power, as usual.

Bum SSR? Should I try a pull-down resistor across the SSR inputs? I would expect it to have an internal one.

 

[AllanMar]

I've never seen one behave like that (latching on isnt uncommon, but the instant reset is) Its not overheating is it?

Doubtful on a pulldown, but you could try...these are usually optically coupled LED input's. The quality on these SSR's is obviously not great so you could be right that its just a bad one.

 

[reynolds5520]

That does sound like a bad SSR is most likely. Just to eliminate one last very remote possibility, remove the indicator light on the high voltage side and test again.

 

[fbold1]

Try putting a 10k resistor across the relay coil.

 

[BetterSense]

An 8k resistor made no difference. Time to pursue a replacement/warranty I guess. Crazy thing is this is already my second SSR. Third time is a charm, at least I hope so.

Can anyone recommend an economical 30A hollow-state relay?

 

[AndrewT]

I have this same issue with my HLT. I have tried several SSRs. They all do this as well. Mine are rated for 60 amps and is controlling 5500w element. I have to cycle the power to get them to release. What I have noticed with mine is how hot they get. I'm thinking of installing a fan on the heat sink to see if that makes a difference.
I watch this thread or post back once I have had a chance to intall the fan.

Andrew

 

[BetterSense]

Ok, so Amazon won't exchange this relay and will only refund my money. That means I can choose a different relay. Is there a non-crap SSR out there? Does anyone know of a particular brand of SSR that is HBT-approved?

I will also consider mechanical relays at this point, since I'm kind of fed-up with solid-state ones. I can buy Bosch automotive 30A relays but I will need a transistor to switch it on, and then I'll have to live with the clicking...

 

[theknub]

i just installed two of these and they (so far) work fantastically. you are maybe paying $10 more for them but well worth it IMO.

 

[BetterSense]

?

 

[theknub]

?

helps when you hit ctrl-v when posting

http://www.auberins.com/index.php?main_page=product_info&cPath=2_30&products_id=30

 

[AllanMar]

IF you can find them cheap on eBay or something Crydom are top notch, thats all I use for work. For my brewery I use the cheap ones you got, haven't had any problems (only testing so far tho).

The Auber ones otherwise are probably a good middle of the road option, and probably the most popular on HBT.

 

[fbold1]

http://www.physics.unlv.edu/~bill/PHYS483/ssr.pdf more than you want to know.

 

[BetterSense]

Well, I finally got my Auber 40A SSR and it seems to work fine based on the limited testing I have done.

Summary: Don't buy cheap chinese SSRs off Amazon; buy slightly less cheap chinese SSRs from Auber.

 

[jalthaus]

Just in case someone else had similar problems as this thread (I did)... I found this document very helpful

http://www.crydom.com/en/tech/newsletters/solid statements - ssrs switching types.pdf

SSR ≠ Electromagnetic relays (duh.)

For the record, I was suffering from a zero-crossing issue myself.