Help with heater control

Mousemania

Member
Join Date
Jun 2016
Location
Boulder
Posts
14
Hello everyone,

So I've been put on a project improving the electrical design of a given system, but I'm pretty new to the field of panel design and have mostly worked solely with PLCs. My biggest question is the best way to design heater and motor controls. Motor control seems much easier; I can just buy a motor controller and that covers everything hardware wise. But I'm uncertain about heater design. What they have right now is an SSR (solid state relay) controlling a heater, and upstream from that a contactor. But the SSR failed at one point, and it failed on, causing some catastrophic failure, so is using an SSR a bad design, or was that SSR just designed poorly? Is all I need for heater control a contact and a relay, and is there any reason to use an SSR instead of an EMR? I've also heard about PWM control, but my system doesn't need fine tuned control so I'm not sure I would need to use that. Any tips and advice would be greatly appreciated.

Thanks in advance!
 
The contactors make noise and the switching period can not be short which produces a certain oscillation of temperature, and the switching generates electrical noise if you do not care about all this then use EMR.

But for me, it is better to use SSR with zero crossing switching and an upstream contactor that opens in the event of over temperature controlled by a separated temperature alarm controller with a separated sensor.
 
Last edited:
That basic design is pretty typical. Contactor provides positive shutoff, and SSR is used to modulate heat. Contactors wear out if cycled under load too often, SSRs do not. SSRs can fail, but that is because they are a semiconductor, which typically fail due to overheating. Are they sized and fused properly? Is their cooling mechanism sized properly? Are there fans on them? Lots to investigate.
 
It sounds like the real problem was a lack of over temp shutoff. The contactor/SSR setup is generally how it's done. You just need to add a way to shut things down before it gets to the catastrophic level.


Bubba.
 
The absolute best option for heater controls is a Silicon-Controlled Rectifier or SCR paired with a shut-off Contactor. An SCR is essentially a very, very fast SSR.

Most SSRs work by taking a sample time of a certain number of seconds, and then shutting off the power for a certain percentage of that period. The problem with this is that this can cause changes in temperature of the heating element which eventually leads to element failure.

An SCR works by cutting out the flow evenly at the individual cycle level. To illustrate, this is what the sine wave looks like with an SSR being controlled at about 50%:

Code:
_/\  /\  /\  ____________________/\  /\  /\  /\  _
   \/  \/  \/                      \/  \/  \/  \/
This is, of course, assuming the SSR has zero crossing. If not, then it could be interrupting and re-establishing current flow at non-zero voltages, which might explain why it failed. Zero crossing makes sure it only transitions when the voltage is zero. That way, you're not dealing with field collapse and all kinds of noise problems that develop when you're switching AC voltages.

Here is what an SCR with zero-crossing does:

Code:
_/\  ____/\  ____/\  ____/\  ____/\   ____/\  ____/\  _
   \/      \/      \/      \/      \/       \/      \/
Most SCRs are controlled by a direct analog 0-10 or 4-20 signal. It does a much better job at splitting the time the element is on and off, which reduces thermal shock and increases the life of the element. Also, if you get an SCR with Zero crossing, it should last quite a while since there are no mechanical parts to break down and you won't see voltage spikes from field collapse.

Finally, solid-state electronics tend to fail closed rather than open, so you need a contactor, preferably upstream of the SSR or SCR. You tie in your E-Stop and Hi-Limit and whatever other circuitry you want in with the contactor.
 
Last edited:
It sounds like the real problem was a lack of over temp shutoff. The contactor/SSR setup is generally how it's done. You just need to add a way to shut things down before it gets to the catastrophic level.
Bubba.
Agreed. Size SSR at a minimum 3.5x the current rating of the heater, and it will last forever. Proper heatsinking is critical. I've seen several with SSR direct to heatsink or panel, without thermal compound or thermal pad.
Minimum is a two channel controller. First channel controls SSR using DC (transistor) output. Channel 2 is relay output for alarm. Relay is wired normally open. If controller is working and happy, relay closes, pulling in contactor.
Some will prefer or mandate a second, dedicated overtemp controller. It will have a separate T/C, and work the same as the second alarm channel as above.
Set sensor failure (loop break) for upscale. If T/C is open or disconnected, controller will assume maximum temp, turning off SSR. In addition, alarm channel with turn off contactor.
 
About 12 years ago we made an installation for a kiln with around 20 separate zones of electric heating, each with contactor + triphase SSR on radiators, mounted in a spacious cabinet with air conditioned. It had also intensity transformers, on the SCADA screen it was possible to see the ampers of each heater.

Total power : 1200 kW

We never changed a single SSR
 
Last edited:
I have a similar application where the system has what is called a policeman cotrolling the large contactor block.

If the SSR or the temperature probe fails, the policeman will trip on high temperature.
 
Thanks for all the feedback!! Much appreciated (y)

The SSR that failed does have zero switching. Although, they did an interesting design where they used two single phase SSRs, so each line going to the heater had it's own SSR. I'm not sure this could be considered the cause for failure, but at the least it's really strange. Also, the spacing isn't terrible but they're a little cramped vertically.

Rupej: outside of the spacing there is no other cooling inside the panel (but the parts did come with a heat sink attached from the manufacturer so I assume that was attached properly). Again, not sure if it would cause the SSR to fail after only a few months but it's probably not helping. As far as the fuses go, the heater pulls around 28 amps (6 kw / 208 vac) and there are 45 amp fuses. They're pulling 2 of the 3 phases, so would this mean each line is pulling 28 amps, or each line is pulling 14 amps? Fuses should be sized 1.2x the expected current right?

keithkyll: I didn't realize SSRs should be sized 3.5x the current rating, is there a reason to size them that much higher? That seems like a lot.

As a side note, the failed SSR was rated for 30 amps, and with the heater pulling 28 amps would that cause an early failure? There's another SSR in there rated for 35 amps on a same sized heater, is it likely that one will be failing soon too?
 
The attached is from Watlow literature, but the concept is pretty generic. As stated above a mechanical contactor cannot typically handle the cycle rate for a heating application. Mercury contactors have been used for years, but are now an environmental hazard and difficult to dispose of. Solid state relays will fail in the shorted condition about 95% of the time. The contactor provides a type of overtemperature safety function.

Most likely reason for SSR failure would be improper fusing. You need to use Solid State protection fuses, fast blow.

Hope this will help.
 
Minimum is a two channel controller. First channel controls SSR using DC (transistor) output. Channel 2 is relay output for alarm. Relay is wired normally open. If controller is working and happy, relay closes, pulling in contactor.

Are you saying there is an SSR with those outputs included on it, or using a separate controller to monitor the SSR?


Solid state relays will fail in the shorted condition about 95% of the time. The contactor provides a type of overtemperature safety function.

Most likely reason for SSR failure would be improper fusing. You need to use Solid State protection fuses, fast blow.

Hope this will help.

Thank you! I did not know there were fuses specifically for solid state protection, that's good to know.
 
Are you saying there is an SSR with those outputs included on it, or using a separate controller to monitor the SSR?
No. The temperature controller. This one has 11-15V pulse output to drive your SSRs. One of the relays is configured as an overtemp alarm to run contactor. Others can be used for additional alarms for your process.

Two SSR's is fine. Not strange. No need for extra expense for 3 SSRs, when 2 will do.

I've always oversized to allow for surge current, line surges, etc. 3.5X for small heaters up to 10 Amps. I would use a 50A or 75A for your application.
Crydom says you can size for actual current, and ignore surge currents, based on NiCR based elements.
"NiCr wire has the advantage of a very low thermal coefficient of resistance, meaning that its hot and cold resistance are very nearly the same, thereby minimizing any cold temperature inrush currents."
They recommend a 20% safety margin. Your 30 Amp SSR doesn't have any safety margin. Link to Crydom white paper.

45A fuses may have been sized based on inrush currents. I don't know what type heaters you have.

Solid State fuses are great, but very expensive. Many times, they cost more than the SSR they're protecting. See this link.
 
While I don't completely disagree with keithkyll's post, there are a couple of things to consider.

If you use an alarm contact on the temperature controller and have a sensor failure the contactor is not going to open.

In the sample diagram I sent earlier there is a separate limit controller and sensor that operates the contactor. This provides a redundant system to prevent a run away situation.

There are temperature controllers that have a built in limit controller and two sensor inputs.

I do agree solid state fuses are not inexpensive. The advantage is the fuse will blow and stop current flow. If the SSR fails it will usually short and continue to pass current.

Just my two cents.
 
We are not in disagreement. There are systems that demand maximum protection from failure by using a dedicated overtemp controller. Semiconductor Industry, for example.
There is a way to set up a system with a single controller, and be confident typical failure modes are covered.

Today's controllers will detect an open sensor. You program it to go full scale in this condition, shutting off the SSR, and triggering the High Process and Overtemp alarms. Overtemp drops contactor.

Failed temp controller CPU. We program the CPU to turn on the alarm relay in the normal condition, and drop out in the case of alarm. In this condition, the CPU in the controller needs to be running and "happy" with all conditions before it will activate the alarm relay and pull in the contactor.

Shorted SSR has been covered.

Getting back to the OP's question. There is two problems.
1) The contactor should have dropped out when the system overtemped from the shorted SSR. Why didn't it?
2) The SSRs are undersized.
 

Similar Topics

to all PLCS gurus, pardon a question from a newbie.. 1.) A conveyor will be controlled by a motor speed controller, to be specific a gear motor...
Replies
7
Views
4,828
Good morning all, It's been a while since my last visit. It's been busier at work, new boss, etc... Anyways, I'm a maint. tech with some...
Replies
21
Views
5,060
Can anyone help out with this Panelview Plus Performance Email. I keep getting an "unable to activate connection "smtp-mail.outlook.com". I've...
Replies
0
Views
1
Hi!! I'm looking for Temperature rise calculation software from Rockwell, I just download "Product selection toolbox 2022" but this software is...
Replies
1
Views
113
Please see attached file. I need this program in Function Block form but I am totally lost on this. Any help would be appreciated. Thanks!
Replies
8
Views
260
Back
Top Bottom