Is this an SSR problem or something else? 1

[smah]

I seem to be having a problem with a recurring Crydom D53TP50D ssr that I'm trying to diagnose. This problem is outside my scope of knowledge, so any advise is appreciated. The application & history is: This is a drying oven where the load of the SSR is a 208V 18KW heater assembly controlled by a CAL 3200 controller. The unit operated fine for approximately 6 months & then had a catastrophic failure of the ssr. Thinking that there may have been insufficient cooling in the control cabinet, we added an additional fan. About a year later, over a weekend (when no one was around), there was a car accident that snapped a power pole & took out the power to the building. When we arrived the following Monday the ssr was burned up again which we assumed was somehow due to the power disruption. The next replacement lasted only a few months then just seemed to stop working - no visible damage/failure. Once again the ssr was replaced & was working fine for a few months. This one now 'seems' to be dying as well. The oven is now unable maintain the set point temperature. The controller appears to be telling the ssr, to activate & the ssr seems to be doing so, but the load side of the relay shows strange reduced voltages. I should also say that the load (heaters) have checked out each time the ssr was replaced & the controller re-tuned using the auto tuning program, line voltage is correct & well balanced. Any guidance on how to next to proceed?

 

[itsmoked]

Your SSR is not up to the task... But then you're probably coming to that conclusion.

An SSR will live a long, long, time if it is kept below its maximum operating temperature and not subjected to other switching abuse.

SSRs are extremely sensitive to temperature. As the ambient (or heat sink) temperature rises you must decrease the current thru the SSR or it will fail rapidly.

Your particular SSR will be dissipating about 120W while running your oven! This is a LOT! It all has to leave out the back of the SSR thru the small area. If there are any defects or even flatness issues with the heatsink your SSR is being mounted too, (including hold down screw torque) Your SSR will toast!

Your power level is far beyond the slap it in a panel, screw it down, blow some air on it type of application.

If I were to install an SSR in a product that had to handle this much power, it would be at the top of the "What-do-we-design-the-system-around" items.

Picture how hot a 120W light bulb gets. You should strive to keep the base plate thermally touchable for decent reliability.

To wit, I would procure a large rectangular cross section aggressive heatsink. I would cut holes that matched the heatsink in the top and bottom of your enclosure. I would mount a matching pan fan at the top of the heatsink to draw air up thru the heatsink. I would then drill and tap holes to mount the SSR. I would carefully countersink these holes so no burr could hold the SSR from uniformly flush mounting. Then I would use heatsink grease or thermal padding to maximize contact between the two. And firmly screw down the SSR using the largest screws that would fit thru the SSR holes.

Then I would hook up the wiring and run the oven while using an contact-less IR temp gun to confirm the heatsink temperature is hand cool immediately above the SSR.

This will stop the rapid degradation and failures you are experiencing.

 

[itsmoked]

Here's a heat sink example:

http://www.aavidthermalloy.com/products/forced/index.shtml

You mount a fan on one end.

Remember though that that much heat cannot be handled by dumping it inside a standard enclosure. The enclosure will need lots of ventilation holes.

Another solution would be to drop the Crydom SSRs and use something else.

You could put in three of these. They are excellent! They include temperature protection. If you going to abuse them they shut down preventing damage.

http://www.ciicontrols.com/products/int75.htm

 

[smah]

Thanks for the info. We're using the 'stock' matching heat sink with thermal paste. There are fans at each end of the heat sink- one pushing, the other pulling. The heat sink is slightly above room temperature, but certainly not what I'd call hot. I'll take a closer look at the cci product that you've suggested, but right now I don't know if there's enough available space for three of these.

 

[jraef]

When you replaced the SCR after the power accident, did you replace the dV/dt protection board (or components) as well? they protect the SCR from rapid changes in voltage, and without it the SCR can be falsely gated by line noise and other transients. The dV/dt system was most likely damaged by the power problem at the same time as the SCR.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[itsmoked]

Well ..If you say you can put your hand on the sink... Try touching the SSR's plate or an IR-gun and check the max. Because obviously if the SSR isn't coupling to the sink the sink will of course not get too hot.

Second, if as I mentioned this is all trapped in a closed box it will still heat up, only to cool rapidly when you open the doors.

Third jraef makes a good point. SSRs at around 50A transition from being triacs to being opposed SCRs. SCRs (and in fact triacs) can be triggered in the normal manner buy sending current into the gate OR by having extremely rapid voltage fluctuations applied to its main terminals. Unfortunately this method of triggering instead of starting in the gated manner as a wave sweeping the junction, starts in tiny "spots" like rocks tossed into a pool. This damages the gate structure every time it happens. Damage then leads to less resistance to it next time.

This can also lead to slowly increasing deterioration and ultimately will lead to failure. Also matching your description. This is the dV/dT that jraef is referring to.

If there are large spikes coming into your plant or originating in your plant this could be happening. Also I have seen cases where the heater element being coiled and long can cause inductive kick back hence generating the damaging dV/dT.

By the way functionally your present SSR has lost one of its channels and your heater output has dropped dramatically hence your lack of set-point attainment.

Also in some of your failures the SCRs have started failing. Instead of just frying open they instead change from an ON voltage drop of 1.5V to maybe 4V so instead of:
30A x 1.5V = 45W you now have:
30A x 4V = 120W. This is off just a third of the SSR!

This rapidly rising temp cranks the voltage drop up on the other SCRs up, so ultimately the fall down the elevator shaft results in a blackened smoking mess .

 

[Mike2468]

Your 18kw load is probably (18,000 / 208)/1.73 = 50 amps per leg. Times 1.2 for typical electrical installations = 60 amps per leg is the absolute minimum solid state relay that you should use. The Crydom that you picked is 50 amps max, when at 25C (room temperature) = undersized for your application, especially when at factory temperatures (35-40C).

Something like three of the 75 amp solid state contactors already mentioned above or something like

http://www.power-io.com/products/cfamilydc.htm

or

http://download.eurotherm.co.uk/downloads/DL/7300s_029162_1.pdf

or just google on solid state contactors and look for something larger than what you have. Depending upon your application, you might be able to use 2 leg break or 3 leg break contactors.

 

[itsmoked]

I forgot the 1.73 Mike. Doh!

His SSRs did pretty well. [Kinda like Custer]

Those SS contactors look good too.

smah: Depending on your system you may need only two poles of control not three.

 

[smah]

Thanks for the info & help. I will check out the products that you've suggested & try to figure out how to incorporate them into what I have. And just for the record, I didn't pick this - I just inherited it.