Parallel SSR Wiring Help Needed

[TKitron]

I have a background in computer electronics and currently working as a network engineer. I have very little experience with relays. I have been tasked with building a timer circuit for testing heaters. Should have been simple, but has proved to be a bit of a challenge. I need to trigger a timer when power is applied to the heater. From the diagram with the SSR it looked like I could wire it in parallel on one leg of the AC line. When I hooked it up it did not work. I can trigger it from the temp. controller, but I need to test for heater failure. Most of the heaters fail open. Any help would be appreciated.

 

[itsmoked]

Yeah you need to check the current not the voltage. Going parallel to the heater will show you the voltage.

The much, much, better way to do what you are trying to do is to pitch the SSR and use a current coil sensor. There are some that put out a go no go signal for any current over X.

So, A current coil will tell you that current is flowing thru the loop formed by the heating element, the switch, and the power source. If the heater opens you get zero current.

Need more info to provide any more useful help.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[TKitron]

Thanks for the quick response. I had looked at the current sensors, but was worried about response time. At temp the heater turns on an off about 3 times per second. Would a current sensor be quick enough. I will also start looking at specs on the sensors.

Thanks again for the help.

 

[Mike2468]

Yes, there will be some accumulation timing errors if you want a precise total. The current coil will have some minor on and off delays if that is the device feeding the timer circuit. Over the life of a heater burn out test, this might be significant.

If you need higher precision, then parallel the controller output to a high speed accumulation timer circuit and use the current coil sensor as the end-of-test sensor. No current being sensed for greater than a few seconds = stop the accumulation.

I assume that you are a testing agency or a heater manufacturer, not a production company who is using the heaters.

 

[ScottyUK]

Current sensors are available with bandwidth up into the low MHz region, probably higher if using a Rogowski coil rather than an iron or ferrite cored sensor. There should be no timing delays of any significance if you choose the right sensor for the application.

3 time per second is pretty quick for a heater to respond to: what is the thermal time constant of the heater? Chopping the power on-off at a rate much faster than the time constant of the load is the basis for PWM control; you may be better with an averaging sensor which will more accurately represent what the conditions the heater is being subjected to.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[TKitron]

Thanks for all the help. I work for a heater manufacturer in the IT department. Lately I've been learning PLCs, prox sensors and now relays and current sensors. We are testing new products life. It should be >4000 hours.


I ended up going with an NK current switch. It does have a delay on startup, but it also has one shutting off. I hoping it will be a wash. To test the delay I plan to run a 10 hour test with one timer triggered by the temp controller and one timer triggered by the current switch. Then check to see how much difference there is.

 

[TKitron]

Thanks ScottyUK...good points. I had to look up PWM control and I see what you are talking about.

In response to your Thermal Time Constant question, I had to think a little.

Our Temp Controllers are
At power up the power in on constantly until setpoint is reached. Actually it over shoots a little and then cools below setpoint a little. This happens a few times until zeros in on the setpoint. To maintain the set temp the controller bumps the heater 2-3 times a second to maintain temp. Which I guess would make it a PID controller. This is the setting of the controllers sold to our customers. The heaters are fairly small cartridge heaters 1/4" x 4" to 3/4" x 24".

The key I guess is that the customer wants to know actual energized time.

I appreciate everyones patients as I learn or re-learn these things. It has change from a dredded project into a fun one.

 

[itsmoked]

You have it it easy! My customers demand that they find out on the precise power line cycle that the heater fails so they can alter their process parameters and try to save their product-in-process..

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[TKitron]

Wow that is precise!

How do you do it? Sounds like you would need to use LabView and record sine wave.

 

[itsmoked]

http://www.flaminsystems.com/page/page/2424482.htm

This thing has the feature built into it. I don't actually recommend it for your case and I'm not actually actively even producing them at the moment. So I am not trying to sell you these!

Essentially the controller,(PID), sends a Turn-ON command to the SSR and using a snap-on current coil with an internal loading circuit in it, it looks for the expected current result. It's all nearly instantaneous from a 50/60Hz point of view.

In my case I am looking for a blown heater or a blown SSR as either is a problem for the user. I also look at the opposite condition of commanding the SSR OFF and discovering that it didn't actually go off! There is a further SSR failure mode of half cycling. They can stick on on all of a certain polarity half cycle or fail and lose a particular half cycle polarity. You should probably look at that in your testing too. You should actually average your monitored current to catch a partially toasted SSR. If you just look and say I turned on the SSR and YES there is current you will completely miss the fact that the SSR is stuck on and the heater will be dead in another 30 seconds from over heating. Then you discover "OH the heater failed" and assume it wore out when in reality it was murdered.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com