# Control system design

#### MikeSe

Joined Oct 24, 2016
18
Hello everyone.

For a project I need to design a temperature control system.
As Iam not an electrical or control engineer I only have an idea of how to solve that.

My biggest issue is how to implement the communication between Hardware and Software.
For some projects I used Labview and modules from National Instruments which works quite good.
Now I thought about controlling the temperatur system, which basically will be som heaters and Solid state relays, with labview an only an ethernet cable.

My question now if this is possible? Can I use the wires of an ethernet cable to sent multiple ouput signals to my relays?

If not is there another way so that I don't need a I/O module?

OR how can I communicate with the Hardware.

Any advice would be really appreciated!

Best regards,
Michael

#### MikeSe

Joined Oct 24, 2016
18
Hi Max.

Do I need a PLC to use the modbus protocol?

Best regards,

Michael

#### MikeSe

Joined Oct 24, 2016
18
Generally the common way in industrial control is via Modbus, most PLC's etc have the ability to communicate this way.
http://www.rtaautomation.com/technologies/modbus-rtu/
Max.
Hi Max.

Do I need a PLC to use the modbus protocol or can I use it to toggle the wires of an ethernet Cable high an Low.?

I was thinking about something simular to the command lines of Rs232.

(sorry for the double post!)
Best regards,

Michael

Joined Jul 18, 2013
26,626
You don't neccesarily require a PLC, anything that is capable of Modbus protocol, there is a site dedicated to Modbus that give all the relevant parameters and formats, the basic communication is in a string of 8 bit bytes, either ASCII or Hex formats.
IOW the communication is in packets of data, rather than discrete bits.
I have used RS232 and then a RS485 or Modbus convertor.
http://www.anybus.com/technologies/industrial-ethernet/modbus-tcp?gclid=CKWw49mM9M8CFQOraQod5-kGcg
Nat. Ins. would certainly have a module for this communication.
Max.

#### Sensacell

Joined Jun 19, 2012
3,131
Is this a project for educational purposes? or do you just need a control system to accomplish a goal?

There are very simple cheap PID temperature controllers that can do the job.
So much cheaper, faster and more elegant than messing with computers and all that stuff.

#### MikeSe

Joined Oct 24, 2016
18
Is this a project for educational purposes? or do you just need a control system to accomplish a goal?

There are very simple cheap PID temperature controllers that can do the job.
So much cheaper, faster and more elegant than messing with computers and all that stuff.
The project should mainly accomplish the goal but should as well serve as education.

The project will involve a second control system later plus a read out of about 100 thermocouples.

Best regards,

Michael

Joined Jul 18, 2013
26,626
If you have experience with Nat. Inst and Labview I would have thought you would find the answer there for shop floor monitoring and control etc.?
That is N.I.'s Forte.
Max.

#### Sensacell

Joined Jun 19, 2012
3,131
Here is an example of a simple integrated temp controller:

http://www.omega.com/pptst/CN76000.html

Controlling a temp and reading 100 thermocouples somewhat different problems.
Without more context about your project, it's hard to help more, tell us exactly what you are trying to accomplish?

#### MikeSe

Joined Oct 24, 2016
18
If you have experience with Nat. Inst and Labview I would have thought you would find the answer there for shop floor monitoring and control etc.?
That is N.I.'s Forte.
Max.
Hi Max.

I think I would manage this task with NI hardware but you never see NI stuff in commercial systems? (I could be totally wrong here).

Best regards,

Michael

#### MikeSe

Joined Oct 24, 2016
18
Here is an example of a simple integrated temp controller:

http://www.omega.com/pptst/CN76000.html

Controlling a temp and reading 100 thermocouples somewhat different problems.
Without more context about your project, it's hard to help more, tell us exactly what you are trying to accomplish?

Hi again.

The goal is to have a sample between a cold and a warm side to measure heat flux through the specimen.

Therefore the temperature has to be measured and controlled on both sides.
I thought about implementing the PID controller in the software.

Best regards,

Michael

#### mcgyvr

Joined Oct 15, 2009
5,394
You need to FULLY define your requirements BEFORE anything else..
There are so many directions to take and spoon feeding information is not the right direction to take on the internet....

Based on your first post there is a $10 solution that would fit the information you provided.. Now based on very little additional information you are being directed towards multiple thousand dollar solutions.. NI/Labview/Omega is top of the line "high dollar" stuff.. But it may be so far overkill for what you actually need.. You could easily...easily spend$20K+ or more on a NI solution including hardware..

A much cheaper solution could exist.. But your requirements are far from clear IMO...
Asking questions without sufficient details is a recipe for disaster when you aren't aware what you are doing..
You don't have to know everything.. But try harder

#### MikeSe

Joined Oct 24, 2016
18
You need to FULLY define your requirements BEFORE anything else..
There are so many directions to take and spoon feeding information is not the right direction to take on the internet....

Based on your first post there is a $10 solution that would fit the information you provided.. Now based on very little additional information you are being directed towards multiple thousand dollar solutions.. NI/Labview/Omega is top of the line "high dollar" stuff.. But it may be so far overkill for what you actually need.. You could easily...easily spend$20K+ or more on a NI solution including hardware..

A much cheaper solution could exist.. But your requirements are far from clear IMO...
Asking questions without sufficient details is a recipe for disaster when you aren't aware what you are doing..
You don't have to know everything.. But try harder
Hi mcgyvr.

Sorry for not being able to provide more detailed information as the project is just beginning.

The warm side will only contain simple heaters which are supplied by a three phase line voltage.
My thought was to implement a software PID controller with the use of a solid state relay which is energized and de-energized to keep the temperatur constant.

If that works I could use a similar method for the cold side, just with a cooler instead of heaters.

As the whole system is slow temperatur only needs to be acquired once a minute from the thermocouples on both sides (warm and cold).

The heat flux through the sample placed between the warm and cold site could be measured with thermopiles.

And that is basicelly what the system should do.

Best regards,

Michael