PID with an electrical heater???

RPax

Member
Join Date
Feb 2004
Posts
125
Hi all, I was wondering if I could get a few suggestions on how to control the following scenario.

We currently have a heating system where we can heat a jacket to over 120°C and as a result can control the Pot to a similar temperature. However, we have recently got a new product in house that requires a temperature of 200°C in the Pot. To deal with this we are installing an electrical heater on the recirculation loop to the Jacket. This heater will have a digital output (to turn the Heater ON or OFF) and an analog output to control the power set to the heater.

To allow us get to the 200°C we will heat our jacket to 120°C then pressurise the Jacket and then switch on the electrical heater while manipulating the analog output to bring the vessel contents to circa 200°C. There will be no cooling media available to the jacket during the heating operation as we will need to maintain the pressure and will be solely relying on the electrical heater to get to our setpoint.

Our main concern is the control of the electrical heater with regards to delayed response and with residual heat transfer to avoid overshoot. We have never used an electrical heating element in-house before.

What we were wondering is whether we can control this electrical heater using normal PID or will there be an element of trial and error by manually manipulating the electrical heater at different stages during the heating process. Could we combine an element of the two?

I suppose my question narrow downs to the best way to accurately control an electrical element while also minimising overshoot due to residual heat. We have tuning facilities inhouse and can also set up trends to manually tune the loop but not sure it's as straight forward as that. Appreciate if somebody could point me in the right direction or suggest some reading material on heater element tuning\control. Thanks - Rpax
 
Time Proportioning Control with PID

You can use a time proportioning control that takes the CV output of a PID and scales it into a value which represents the amount of on time vs. off time for a digital output. Depending on the PLC that you are using will depend on the ease at which this can be done. Rockwell has a very nice example of this control.
 
I suggest that instead of PID you look at a simple deadband controller - essentially a thermostat. You can vary the set temperature, the width of the deadband, and include time delays for length of heater on cycle. This might give you a system that is more stable and easier to tune than PID while providing control that is as good or better.
 
You can combine PID with a deadband control, with PID you get proactive reaction from the derivation control. You tie the deadband into the CV (output) of the PID.
 
Typically, I will use a Watlow EZ-ZONE PM controller for the control of the electric heater, but if you are wanting to use an existing PLC with PID control and send a 4-20 mA signal out to control the device you can use another Watlow device called the DIN-A-MITE controller. This is an SCR Power Controller that accepts a 4-20 mA input. I have used these with Micrologix 1500 PLCs and Wago 750-841 PLCS. They work well and you can get Single Phase or Three Phase controllers. I have had these in for about three years now with no issues to date. Here is a link to the website:

http://watlow.com/products/controllers/power.cfm

Regards
 
Typically, I will use a Watlow EZ-ZONE PM controller for the control of the electric heater, but if you are wanting to use an existing PLC with PID control and send a 4-20 mA signal out to control the device you can use another Watlow device called the DIN-A-MITE controller. This is an SCR Power Controller that accepts a 4-20 mA input. I have used these with Micrologix 1500 PLCs and Wago 750-841 PLCS. They work well and you can get Single Phase or Three Phase controllers. I have had these in for about three years now with no issues to date. Here is a link to the website:

http://watlow.com/products/controllers/power.cfm

Regards

I don't mean to Hijack this thread, but my question is similar:

I also have a Watlow DIN-A-MITE controller that is used to drive SCRs for a 75KW 3-phase 480VAC heater. This controller works great, but I always wondered about the way it operates - it controls the temp by varying the OUTPUT ON time of the power.

For my application, we are heating natural gas fuel for a gas turbine engine. I need the full 75KW during startup of the unit until the lube oil heats up (then the hot lube oil preheats the fuel gas before the electric heater). After that, it typically runs at something like 1.5 seconds ON, 0.5 seconds OFF, repeating.

This represents about 100-150A of our typical plant use at the time of 300-400A @ 480V. This is very noticable on my plant's incoming power MCC switchgear, because the ammeter needle swings in tune with the SCRs.

I always get asked about it by visitors and engineers, asking why the needle swings like that...all I can tell them is it's due to the 75KW heater, and it is controlling as it should.

Is this normal? I always thought "yes", but I'm not sure....

I have been asked about disconnecting a portion of the heating elements in order to have the heater remain on at 100% of the time to do away with the swings, but then I don't have the full 75KW at startup.

Isn't this the normal way of controlling SCRs using this controller - that is, you have to control the firing of SCRs using discrete outputs and cannot use an analog output to control "how much" power the SCRs let through, right?

Kent
 
I don't mean to Hijack this thread, but my question is similar:

I also have a Watlow DIN-A-MITE controller that is used to drive SCRs for a 75KW 3-phase 480VAC heater. This controller works great, but I always wondered about the way it operates - it controls the temp by varying the OUTPUT ON time of the power.

For my application, we are heating natural gas fuel for a gas turbine engine. I need the full 75KW during startup of the unit until the lube oil heats up (then the hot lube oil preheats the fuel gas before the electric heater). After that, it typically runs at something like 1.5 seconds ON, 0.5 seconds OFF, repeating.

This represents about 100-150A of our typical plant use at the time of 300-400A @ 480V. This is very noticable on my plant's incoming power MCC switchgear, because the ammeter needle swings in tune with the SCRs.

I always get asked about it by visitors and engineers, asking why the needle swings like that...all I can tell them is it's due to the 75KW heater, and it is controlling as it should.

Is this normal? I always thought "yes", but I'm not sure....

I have been asked about disconnecting a portion of the heating elements in order to have the heater remain on at 100% of the time to do away with the swings, but then I don't have the full 75KW at startup.

Isn't this the normal way of controlling SCRs using this controller - that is, you have to control the firing of SCRs using discrete outputs and cannot use an analog output to control "how much" power the SCRs let through, right?

Kent

It's A normal way, but I wouldn't say it's THE normal way. There are several methods of controlling power with SCRs; and yours is very valid. We also use some DIN-A-MITEs in some of our equipment. We ordered ours with the variable time base control input. This spec sheet gives some details on the method. http://watlow.com/downloads/en/specsheets/windmc0309.pdf

It is basically doing the same thing you are, but on a much shorter time scale. 75% power would be something like 9 cycles (60 Hz power cycles) on, 3 off.

You can also control power with SCRs through a method called phase angle control (also mentioned in that spec sheet). This method is required when driving heaters through a transformer. It is not generally needed when driving heaters directly, and it is more expensive.

It sounds like your setup is just a contactor input (AC or DC) hooked to a time proportioned PID output. If you wanted to reduce the needle jump (not really hurting anything IMHO), you could look into swapping your controllers to the direct analog input and have it do the time proportioning on the much faster timescales. You would likely have a hard time seeing the needle jump then.

Brian
 
Hi, firstly many thanks for your replies. The Watlow module definetely looks like something that would do the Job. Unfortunetely, I'am using a DCS system (not that that make much odds) but the wiring\documentation has already been done upfront and can't see them going back on that. But will bear in mind for the next time so thansk for this.

Just that i understand what yee have suggested

You can use a time proportioning control that takes the CV output of a PID and scales it into a value which represents the amount of on time vs. off time for a digital output. Depending on the PLC that you are using will depend on the ease at which this can be done. Rockwell has a very nice example of this control.

Gscure, I've never used the above but am i correct in saying that I would run a normal PID but use the CV output to calculate a set time e.g If my CV was 66%, then I would turn on the heater for 66% of a certain lenght in time, say 40sec (of total lenght 1min), and turn off for 20 sec. Once this ON\OFF for the 1min is completed, I then take another snap shot of the PID CV and do the same for the next 1min and so on??


I suggest that instead of PID you look at a simple deadband controller - essentially a thermostat. You can vary the set temperature, the width of the deadband, and include time delays for length of heater on cycle. This might give you a system that is more stable and easier to tune than PID while providing control that is as good or better.

Tom, would this simply be a matter of loading a set analog output to the electrical heater at certain temperature bands while the temperature is rising e.g. At a temperature between 120°C ~ 140°C i'd Output 100% to electrical heater, at 141°C ~ 160°C i'd Output 50% to the heater, at 161°C ~ 190°C Output 20% to the heater etc. At around the Setpoint, i'd create a deadband to turn off a 200°C and back on a 190°C. Of course there would be initial testing to find out the most suitable bands etc. I'am I correct on the concept that you are suggesting??


You can combine PID with a deadband control, with PID you get proactive reaction from the derivation control. You tie the deadband into the CV (output) of the PID.

Harryting, is what your suggesting that I use normal PID until I get up to around the set deadband on my Setpoint?. How would I tie in the deadband into the CV. Would it be time portional control as Gscure suggested?


Again, thanks for your replies and help - Rpax
 
Tom, would this simply be a matter of loading a set analog output to the electrical heater at certain temperature bands while the temperature is rising e.g. At a temperature between 120°C ~ 140°C i'd Output 100% to electrical heater, at 141°C ~ 160°C i'd Output 50% to the heater, at 161°C ~ 190°C Output 20% to the heater etc. At around the Setpoint, i'd create a deadband to turn off a 200°C and back on a 190°C. Of course there would be initial testing to find out the most suitable bands etc. I'am I correct on the concept that you are suggesting??

That would work, but it is actually more complex than I was suggesting.

What I had in mind was a simple thermostat. Heater on at a set temperature, heater off at a different set temperature a little higher. The difference between the two temperatures is the deadband. This is the fundamental control logic, used by most furnaces and A/C units.

To prvent short cycling you could add a time delay so that when the heater came on it would stay on for a minumum time - say 5 seconds. You could add another timer so that when the heater turned off it would stay off of a minimum time - say 2 seconds. These are arbitrary times, and you could adjust them as needed.
 

Similar Topics

I saw in a 24 V solenoid valve which has 3 wires (2 reds, 1 green); so 1 red is 24V power, another red is .... a common ??? or the green is the...
Replies
16
Views
4,220
I have S7 1512C controler for controlling 48 PID temperature loop, the output is PWM. Please I need the best, most efficient way to write the...
Replies
5
Views
213
Hi all, I'm having trouble solving a problem I've been working on for several months, and thought you might like a stab at it. The machine runs...
Replies
22
Views
840
How can I connect PID Output to a valve. In ladder logic program is there any logic do I want to add between valve and PID? PV=SP What will be the...
Replies
7
Views
383
I'm setting up a PID for a 50hp RO pump in a water treatment plant. When I go to manual mode on hmi I set the swm bit, I also ons the pid output%...
Replies
5
Views
524
Back
Top Bottom