Protection for Electrical Heater

[rajsiyer 7]

Hi Friends, These days, I am in the midst of designing a control panel. The panel has around 20 of heater loads, all 3 phase 480V with capacities from 7-45kW. I plan to use SSRs for each heater. To protect the SSRs & the heaters, I plan to use OMRON's K8AC-HXXX burnout detector. The panel also has 5 small motors all less than 5HP. This control panel has to comply with the requirements of UL508A. Coming to my design...In the circuit of each heater, there will be a MCCB (rated at 125% of heater's FLA), for Short-Ckt protection of the feeder. Downstream to this CB, is a contactor interlocked with the burnout detector. After the contactor, comes the 3-phase SSR. The outputs of these are taken to the terminals for heater connections. There are CTs in 2 phases, between SSRs & terminals to sense the load current for the Burn-out detector. The burn-out detector is powered up from the control voltage (120VAC) . This device is intended to detect the SSR's Open-circuit /Short-Ckt failures as well as the heater's opn-ckt & abnormal currents. Should any of these get detected, the contactor will open. My questions are ... 1. Can a single common control-power transformer be used for all contactors and the power-requirements of the burn-out detectors?. This question is because the OMRON K8AC** provides compensation for fluctuations in input voltage, which is as seen at its own power-supply terminals> 2. Must the panel be of a compartmentalized type? 3. Must the control power transformer (approx rating 1kW) secondary be earthed?? 4. What is the correct way to protect the control power transformer. 5. How good/reliable is the OMRON device?? Are there any other makes for similar burn-out detectors?? Regards, thanks a million, Raj S. Iyer

[Powered By Dodge V8 0]

1= without doing any research into the tech, i dont see why not - simply size appropriately. 2 = no 3 = yes 4 = class D breaker or time delay breaker where instantaneous trip is 10 to 15 times nameplate size. break the secondary as well, same breaker type / fuse type. 5 = honestly, i wouldn't bother with any of it. i'd use a 3 phase SSR for the heaters (for fast switching) and then strap contactors with thermal overload decks to them, with the control power for the SSR and the contactor going through the normally closed contacts of the thermal overload deck. no need for burnout detection, as overamperage will show on the overload deck. i also assume you will fuse / break each of these accordingly.

[rajsiyer 7]

Thanks Dodge, Yes, I plan to put a class C or D breaker (nameplate rated at 125% or min 15A) on each heater's circuit + there will be a contactor sized for AC1 duty as well. I had my doubts about the ability of the overload relays to protect the SSRs in time. I am really grateful for your confident suggestion that the burnout detector devices are unncessary! This could be a huge saving as they cost $500-700 each!. Now if you were to select OLRs for heaters, how would you size them? Aren't their inverse-time characteristics made for motors? Regards and thanks for your suggestion. Raj S. Iyer

[machinemaker 0]

One of the legs should always be connected to the heater. This advice comes from a heater manufacture. I use a Bander Ground Fault unit to detect any ground faults. It is possible to have the heater ground and with our system 277 volts will flow through the heater. It is a single unit that the main wires coming in go through a large hoop to detect the ground fault. I simply use current sensors to each heater and some plc logic to detect shorts or opens. I will not get into the logic. I also have a large contactor to supply power to the heaters. Besides the PLC controlling to shut down the contactor, a hard wired switch in in series to the coil. Since using these methods, I have not had any problems with shorted, open or any other bad condition. Before we had ssr's short and the heater was on, had one side of heater short to ground and the heater came on again. And YES ground that secondary of the transformer.

[Powered By Dodge V8 0]

If you are using single phase power (which a single element heater is... there are multi-element arrays of heaters that accept 3 phase power at their pigtails, but let's assume you're not using that)... so if you're using single phase power, and one of the legs is neutralized (bonded to ground at the supply transformer), then THAT leg and only that leg can be connected to the heater at all times. If you are using 3 phase power, and just pulling 2 legs off of a 3 phase 480 line, and then feeding that 2 legs into a heater, and neither of those legs are bonded to ground, then it may not be a code violation (it may or may not, I'm not sure off the top of my head), but it's bad practice for sure. Anyway... thermal (or electro-magnetic variety) overloads have inverse time characteristics, but so do plain old circuit breakers. Basically, it's a trip curve - a relation of instantaneous current draw versus time that let's you predict at what current and what time that current is held for that a breaker / fuse / overload / or any other device will trip out. Straight thermal overloads will require that they cool off before they can be reset - they physically won't stay held in until they cool off after a trip. They're older technology and generally cheaper. Electro-magnetic overload units are direct replacements for thermal overloads (and we still call them "overloads" or "thermals" or "heaters", because they're doing the same job), but can be reset immediately after a trip, and generally cost two to three times as much. For a 3 horse motor running on an appropriately sized IEC din rail starter, a thermal overload might run you 15 to 25 dollars, whereas a electro-magnetic one might be 50 to 100 dollars, which will vary widely based on manufacturer and ratings, but that's a general ballpark. The thermals also have a limited lifespan - you can only crap them out so many times (a lot of times, but still it's limited) before they're shot. Electro-magnetic ones have lifespans of 50 to 100 times that -- basically, they'll only go bad if they're defective (usually). To specifically answer your question... heaters are purely resistive devices - not inductive. So your overload factor can be dang near zero. Most overloads, right out of the box, will trip at 125 percent of their dial setting. So will this protect your solid state relay? Yes. At least if you size the SSR's correctly. If you're going to run 10 amps typical draw on a single heater, then you should use an SSR with a nominal current carrying capability of no less than 125% of 10 amps (which would be 12 and a half amps). When the heater starts to go bad, it'll increase amperage draw, and the overload will trip... the SSR will still be in its nominal operating range. If you see a dead short, the breaker will trip. Everything must be sized appropariately, this goes without saying. Here's one example XFMR --> C breaker (5 to 10X nameplate trip) @ 150% typical heater draw --> contactor+overload @ 125% typical heater draw --> SSR @ 125% typical heater draw --> heater XFMR --> 15 amp breaker --> 10 amp contactor with 10 amp overload setting (will cut out at sustained 12.5 amp) --> SSR @ 12.5 amp or better --> 10 amp heater ... use 14 awg THHN, MTW, or THWN conductors. You're never going to (at least not without a lot of wasted time and money) protect a 10 amp SSR from the failure of a 10 amp heater, but you can protect an over-rated one from it.

[rajsiyer 7]

Thanks again Dodge, your V8 energy in describing the phenomenon empowers me. Yes indeed. Any bimetallic OLR will sleep indefinitely below 25% over-current. Yes actually now I have more than 50% overcurrent rating for the SSRs as OMRON's SSRs are 15,25,35, & 45 amp sizes. As against this my loads will draw only 9A, 12, & 24A . As I gather the heaters are all 3 phase star connected. The source is 480VAC 3wire. Still need to ground?? at the heater side. I don't see why. Thanks a million for your very helpful suggestions. Regards & best wishes. Raj S. Iyer