Temperature Control: PLC or Temperature Controller?

Hello friends

I need to automate a machine, I thought Micrologix 1400 and PanelView Component C1000.

This machine has 2 heating tanks, 1 conveyor, some sensors and actuators.

Actually, there are not PLC and HMI, heating tanks are controlled by simple temperature controllers (Durant TC48, 1 relay output, 1 alarm output, No Remote SetPoint)

In the panelview, Operator will choose a recipe, which one would have configured temperature setpoint for every tank and speed for conveyor, between another

conditions.


I thought 2 options:

Option 1:
Replace temperature controller for another with Remote SetPoint (Omron E5CC-RX3A5M-007)
According with the recipe, PLC will send the RSP for each tank and Speed Reference

Cons.
Actual temperature reading will be displayed just in Temperature controller, not in panelview.
http://imageshack.com/a/img716/2723/6kxb.jpg


Option 2:
Eliminate the actual temperature controller, connect the thermocouples to an 1762-IF4 card and control temperatures through PLC.

http://imageshack.com/a/img706/562/lyxz.jpg

which you think is the best?
Do you recommend another PLC or controller?
How difficult is to control temperature with PLC?

Any comment is appreciated.



Thanks in advance.

Wwhich you think is the best?
Do you recommend another PLC or controller?
How difficult is to control temperature with PLC?

It depends on how complicated your recipes are. I would think that doing all the functions to carry out a recipe without a PLC would be difficult. PLC temperature control works well, but you do have to write the program and setup the PID instruction correctly. If you have experience doing that, then a small PLC with analog inputs and outputs would be a good option. Otherwise stick with temperature controllers.

You could start with a hybrid system. There are temperature controllers that have analog outputs that can be used to send the temperature to a PLC (and operator panel). Let the temperature controller control the temperature, and let the operator panel and PLC control the remainder of the process.

If you do not need all the functions and versatility that is available from the PID instruction in the PLC then I would go with a separate temperature controller. The PID function gives you a huge range of options and a lot of versatility, but it can be a pain. I have an application that I wish I would have done a separate controller for.

I think that the answere depends on how acurate your temperature should be. If the temperature should be very acurate go for the external controller, if the temperature may vary something, the PID in your controller will be good.
If you can write the recipe setpoint to your temperature controller, I'm quiet sure that you can read the actual PV.

Jack

Hello

I have checked the parameters of the actual temperature controller and it has not doing PID function, it is just working on/off. Proportional band = 0, ti=0, td=0

It has connected a thermocouple J, a relay to activate a resistance and a relay to stop the conveyor when the temperature is to high.

So, I could use a Micrologix 1400 and just turn on/off the resistance to reach the temperature? It is not necessary the PID?

What do you think?


Best regards:nodi:

It's a pretty loaded question, there is no right or wrong answer, you could do it either way. I have applications that do it both ways. In my opinion if setting up and tuning PID's is not your strong suit then a temperature controller might be the solution for you. There are many of them out there and some of them are really inexpensive. The better ones seem to have better auto tuning functions in them though.

We use Eurotherm quite a bit.

In a lot of our systems we use 2400 series controllers communicating on either Modbus or Devicenet to a PLC. We send only the PV and SP between the controller and the PLC and let the controller do all of the control. That way you can do all the reciping in the HMI and PLC.

Or you can do everything in the PLC. Some people prefer one method over the other but I am basically at the point that I don't really care which way I do it, I weigh the pros and cons of each and decide from there.

asteroide said:

Hello

I have checked the parameters of the actual temperature controller and it has not doing PID function, it is just working on/off. Proportional band = 0, ti=0, td=0

It has connected a thermocouple J, a relay to activate a resistance and a relay to stop the conveyor when the temperature is to high.

So, I could use a Micrologix 1400 and just turn on/off the resistance to reach the temperature? It is not necessary the PID?

What do you think?


Best regards:nodi:

Click to expand...

Sorry I didn't read this post before I responded. If that is all that is required to maintain an acceptable temperature then by all means just hook it up to the PLC. I would still use a PID instruction though. I would also have the PLC drive a solid state relay for the heater, the relay will last longer.

My assumption is that the heater is being controlled with On/off type control. Meaning the heater is always either on or off and you are not controlling the voltage going to it. In this case to use a PID I would set up the PID so that the CV output turns on an output for the heater within a specified range, for example 70%-100% using a simple GRT instruction. You can then fine tune the PID and your "on" range to suit the application.

For probably a better way of doing this search the forum for Ron Beaufort and PID or temperature control. He probably has better ways of accomplishing this.

On way is to use the PID to set the Percent Time ON for a timer that controls the PLC output for the heater.

we have decide to control everything with plc, I just need the approval from the boss.

Thanks everybody

i will post to this only because i have been going through that same thing for the last few weeks here at work and with help on this site have something we are testing.
What i have found out is that yes, with a pid etc. i can control the temperature to a setpoint and maintain it on a test setup +- 1 degree.
What i ran into is that i was gettig false reading sometimes on the hmi i am using that displayed the temperature. I tried to really figure out the best way to deal with it due to the fact that about 10 years or so ago another division in my company hired a integrator to convert a lot of machines, about a million dollar project and after a year they had to go back and convert the temperature control back to using Fuji controllers. From what i found out they never could get past the false readings which was the same i saw on the bench test. They used a automation direct plc and i am using a micrologix 1400. i found that when some sewing equipment near my lab turned on and started running i saw the readings jump around. i had to go in and change the filter on the thermocouple card to 10hz and then make sure i had the plc grounded good to the power cord ground. that really helped a lot. What i then had to do was change over to a ungrounded j type thermocouple. That really fixed the issue at least on the test bench. Drove me nuts for a few days because our new director who is the same guy who approved the previous change to using plc's now says plc for temperature control only. Come to find out it is not cheaper. 8 automation direct temperature controllers 90 bucks each = 720. With plc, it was about 1000 for the two thermocouple cards another 250 for a output card to control the ssr's and another 375 to go to a bigger size panel to accomdate the current 1400 plc in a panel that is already too full to make room for the 3 extra plc cards.
i will attach the program. most likely not the best way to control the temp but best i can come up with at the moment. there is a lot of extra stuff in it to communicate with a red liion (emerson) hmi i have been working with but hopefully it may help. just change the extension on the file from txt to rss and you will be able to open it. To me, examples of something that worked helps me a lot, especially with the time frames my director thinks something should be done it. just could not understand why it tooke me 3 days to write this program and test it when integrators tell him it is a piece of cake. just have to deal with it and hope information on here helps.