Hello All,

I would like feedback for a project I am attempting to get pushed through at my job.

Background:
We currently employ 3 phase contactors to control 3 heaters each (wired in delta) as part of a zone. As an example, Zone 1 has 10 total heaters, therefore 4 contactors. The voltage is 480VAC 3phase. Each leg from the contactor reads approx 10A during operation. Each heater is 3kw. They are resistive elements similar to an standard domestic stove.
The Oven has 6 zones. A thermocouple closes a contact which activates (energizes the contactor coils) the entire zone or deactivates it depending on the temp set point and allowed variance. (present schematic attached)

The Vision:
My idea is to replace the 18 contactors that operate the zones and run each zone with an SCR, using the same heater config. The SCR would be controlled from a micro controller running a PID loop and outputting a 4-20mA signal.

My question concerns the benefit of this scheme versus the present. I hope to improve control and perhaps save some energy in the process.
I welcome ideas and feedback. I am still learning about these devices.

 

By "contacter", do you mean relay? I get the idea of using a controller to tighten up the operation, but why not use the existing relays? It would simplify things quite a lot if you could control those instead of trying to replace them.

Switching big currents at 3kW is not trivial.

 

Hi,

Instead of an scr, you could use triacs to drive the heaters using triac optoisolators. It is just a matter of selecting the proper triacs that will handle the 480V.
Use a triac opto witha zero crosssing circuit. The input to the opto can be driven with 4-20 mA DC signal using one resistor on the input. the output of the opto needs one resistor to drive the gate of the triac. The triac will also need snubbers across it, and a heatsink, too.
An option that might be cheaper rather than buiding these circuits is to buy some solid state relays that are heavy enough to handle your line voltage and load. I believe you will only need 2 relays for each set of 3 phase heaters, but they also make 3 phase SSR's too.
In the factory where I work, we bought a machine with 3kW of heaters in it and the manufacturer set it up with a mechanical contactor and PID control. When it was at setpoint, all that you could year was this heavy "thunk" "thunk" every second or so. After one month, the contactor welded "on" and the temp went up to almost 1000 degrees F (there was no overtemp safety on it either, a really poor design for a $10K machine). I ripped that out and installed a SSR and an overtemp circuit. It is dead quiet and has great control, too. This machine was only single phase 208v.
We also run Blue M ovens that are 3 phase, and they have two SSR's driving the 3 phase heater elements, and they are very reliable. Hope this helps.

 

I can also get you some part no's for the ssr's tomorrow when I'm at work. I think the ones we use are only rated for 480v max, so you might need a different #. I'll check on it and repost as soon s I can.

 

wayneh,

Thanks for the response! Yes I do mean relays. Mechanical ones at that. As EdGs mentioned at setpoint we get a lot of "banging" in the cabinet and the control is terrible. Also the life of the relays is substantialy shorter due to the cycling on and off so much. I am trying to improve the system, because currently it is a cabinet of relays and timers with spaghetti wiring. Other items in the system are a conveyor and 8 fans. Since I want to go with a PLC type of control, I thought that it might be better to go with an SCR or soemthing comparable.

 

I've done a lot of work on ovens. SSRs are the way to go IMO. Omega sells some nice controllers for them too.

 

SSRs are the way to go IMO. Omega sells some nice controllers ...

+1 on both points

 

What is the coil voltage on the relays/contactors that are currently driving the elements?

Building your own SSR's with triacs or SCR's would definately be cheaper in terms of the parts used, but the time used putting them all together would definately eat up the savings, too. You would probably be looking at $30 - $40 each for single SSR's, and $100 or so each for 3 phase ones, off-the-shelf.

Does each zone of elements have it's own controller? You also have to look at the available current to drive the SSR's and make sure there's enough headroom there too.

Most of the baking processes at my job are just 1 setpoint at one specific time length, or a series of different bakes, depending upon the job. It seems that you have some sort of conveyor oven for your process. PLC/computer control definately sounds like the way to go, too. There's plenty of products out that would fit your needs, it all depends how much money and time the bosses decide to spend on it.

I don't have the part #s in front of me for the relays, but you can check online with Mouser, Digikey, Newark, Arrow, Allied, etc.

Good luck, and let us know how it goes.

 

It sounds like you want to combine all the zones to be controlled from 1 controller. Not a good idea. I have personally done this on a rubber die press. it had 4 zones; 1 thermocouple, 4 heaters, and 1 PID heater controller per zone. 2 of the heater controllers went out, so I piggy-backed the 2 effected zones SSRs (solid state relays, that is what you are going to want to use by the way). The results were less than ideal; heating was uneven, resulting in the product being overheated in areas and underheated in areas, unpredicably. The zones are there for a reason. It's nigh impossible to evenly heat up something big all at once.

That being said, my advice would be to
#1 keep the zones.
#2 if you don't like the contactors (yes, I would call them contactors, not relays) then upgrade to Solid State Relays (basically big triacs). But keep in mind that you will need 2SSRs per zone which will get expensive pretty fast. SSRs are not immune to failure either; at least with contactors you can replace the contacts, whereas with SSRs if they malfunction they have to be replaced. + you already have the contactors, So, I would say keep them.
#3 You mentioned that a thermocouple activates the contactors. How? there would have to be some kind of controller to do that. PID heater controllers normally do this job; if you already have PID heater controllers, keep them. If not, updrade to them. Eurotherm/Barber Coleman sells good heater controllers, also Omega as mentioned.

After trying this, I realized that I basically instructed you to do nothing, which makes me wonder, what is the problem with the system that you have?


EDIT: Re-read your post, I see you want 1 SSR per zone. that's good, but you will need 2 per zone to switch 2 of the 3 phases. I still am curious what's controlling those contactors. IMO off the shelf heater controllers are going to be your best bet rather than trying to write PID into a microcontroller or (emphatically) a PLC.

 

If its noisy and 'seems' like it could be made to work better, STOP and ask yourself this.--Does it do the job it is supposed to do? Does it cause any problems with the product it processes?

If you said YES and NO. Then there is not a "problem" with the device.

The first rule is: Don't fix it, if it isn't broke.

I'm with Strantor on this one. If it makes your boss money and doesn't 'ruin' the parts it processes. LEAVE IT ALONE.





(Maybe suggest soundproofing the cabinet the 'noisy' relays are in.)

 

Here's a link to some SSRs. Personally, I would choose one with a current rating at least twice what your load draws. As you can see, the prices are very reasonable so the extra expense to gain a safety margin is not a deal breaker.

http://www.mpja.com/products.asp?dept=133

Digikey also sells them.

 

Here's a link to some SSRs. Personally, I would choose one with a current rating at least twice what your load draws. As you can see, the prices are very reasonable so the extra expense to gain a safety margin is not a deal breaker.

http://www.mpja.com/products.asp?dept=133

Digikey also sells them.

That's the cheapest I've ever seen for SSRs. I wonder how they get away with not using a heat sink for such a large load. this is the kind I use for our plasic extrusion heaters (see the one on the right, huge heat sink) They run about 350$ for a 40A 480V model. Part of the reason why I suggested agains SSRs, but hey, if you can get away with using a 7$ part instead then more power to you. Just don't want to see your heroic upgrade plagued with constant SSR failures.

 

That's the cheapest I've ever seen for SSRs. I wonder how they get away with not using a heat sink for such a large load.

I see that they're Symmetrical DIN rail mounting. That's why they have integral heat sinks. They don't really have any choice about it. That said, the price difference is astronomical. Far beyond the cost of heat sinks. The SSRs at MPJA are intended to be heat sinked by the end user or when mounted to a large heat dissipating surface using heat sink compound.

 

I see that they're Symmetrical DIN rail mounting. That's why they have integral heat sinks. They don't really have any choice about it. That said, the price difference is astronomical. Far beyond the cost of heat sinks. The SSRs at MPJA are intended to be heat sinked by the end user or when mounted to a large heat dissipating surface using heat sink compound.

Good point. Starting to look like a better idea after all. And you give me information that I can use to save money. next time one of mine goes out, I might try ripping out the DIN rail and replacing with one of those.

 

Here's a Digikey 240VAC @ 25A model and as you say... expensive. I've purchased quite a few parts from MPJA with no complaints but I can't testify to the worthiness of these SSRs.
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=PB554-ND

 

Another reason I am not partial to SSRs (yes, I realize that they are superior to contactors/relays but that doesn't mean that I have to like them) is that when they fail, they do weird stuff. I have seen a situation with a 3ph heater with 2 SSRs draw the appropriate amps on all 3 legs but produce no heat. I disconnected 1 leg at a time and was not able to isolate the problem. decided to try replacing one of the SSRs (though I had no reason to suspect they were bad) and it started heating again. maybe half of the triac was bad, allowing only half-wave to pass? IDK, but it put a bad taste in my mouth.

 

Another reason I am not partial to SSRs (yes, I realize that they are superior to contactors/relays but that doesn't mean that I have to like them) is that when they fail, they do weird stuff.

An Excellent point here, too. In our manufacturing plant, we currently have about 60 SSR's controlling various heating processes, and there have been about a dozen times where an SSR has failed in some way, mostly "on" but a couple times where they were "partially on". This usually was due to dead shorts from a defective element. However, these failures are over a 15+ year period, so really not bad considering how hard we run things in our plant.

I can say much for the cheaper SSR's. We use mainly Crydom or Continental SSR's, so I have no experience with the cheapies.

If you do decide to try SSR's, Might I suggest you do one zone only, taking time to make sure that you have the means to drive them properly. Make them up in a separate cabinet, so that you dont have to undo too much in the original cabinet, and so you can put things back easily if you need to, just in case. You know how Murphy's law is....lol.

Hopefully, you will have time to install and test things a bit when your processes aren't running. Many times I have had to do repairs and installs in the evenings or very early mornings, just to make sure thins are up when they need to be. Good luck.

 

Rather than trying to re-invent the wheel, maybe something like this .......
http://iseinc.com/SCR_Power_Controls.htm
would work for you. This is only one example of what a Google search turns up for SCR Temperature Controls

 

I remember an brand new supposidly precision oven failed 3 times during its 1 year warrenty period, because it used an air bulb type of temperature regulator. The 3rd time I was replacing the contact element the company allowed me to convert it to a SSR/Omega controller, which fixed the problem. There are much, much worse than SSRs, frankly I love them.