Temperature controlled switch help

Thread Starter

Fragma

Joined Aug 31, 2014
8
Hi guys/girls,

Was wondering if anyone had any ideas on the following:

I'm designing a switch using a thermocouple and a PLC. What this is meant to do is turn on and off at particular values, i.e. at +60 it turns on, +70 it turns off, -60 it turns off and -70 it turns on (this obviously equates to some current/voltage).
The PLC software included makes use of logic gates (AND, OR etc) and arithmetic. So what I did was use an R-S flip flop to check the input and if it meets one of the criteria, set it either to on or off.

The problem however is the initial conditions, since you don't know if initially the temp is rising or falling. If for example the temp is above 60 and rising, the switch should be on initially, if its more than 60 and falling, it should be off etc.

Any ideas on what I could do to set the initial conditions?

Thanks in advanced
 

Thread Starter

Fragma

Joined Aug 31, 2014
8
Hi there,

Everything is set up using the software, the only hardware would be the PLC and the thermocouple. I thus need to set this up using the software without added hardware, hence the use of the simulated flip flop. It does what I need it to as stated, just need to use the software to set the initial conditions.

Thanks for the response though, much appreciated.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
By PLC I assume you mean the Industrial self contained unit not the generic term for assembled processor and/or logic?
I wasn't sure when you mentioned logic gates?
Max.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Iv'e not come across that particular unit, but it mentions a differential analogue input module, I don't know whether this is a n add on, but it may solve your problem?
Max.
 

NorthGuy

Joined Jun 28, 2014
611
Why wouldn't you switch from PLC to directly programming microcontrollers (Arduino is an easy path). It'll be much easier and cheaper.
 

Thread Starter

Fragma

Joined Aug 31, 2014
8
Hi NorthGuy,

I would love that, unfortunately I've been pulled into this after everything has been acquired already, so I'm pretty much stuck with what I have at the moment. This LOGIC+ program is quite sucky, u can simulate an analog input but only give it one value per simulation, so I cant even simulate a fluctuating input at this stage....
 

ericgibbs

Joined Jan 29, 2010
18,766
The problem however is the initial conditions, since you don't know if initially the temp is rising or falling. If for example the temp is above 60 and rising, the switch should be on initially, if its more than 60 and falling, it should be off etc.

Any ideas on what I could do to set the initial conditions?
Hi,
What is controlling the 'system' prior to you turning ON your PLC controller.?

What would you consider the 'fail safe' state, On or Off.?

E
 

Thread Starter

Fragma

Joined Aug 31, 2014
8
Hi Eric,

There is no fail safe state per say, everything is temperature controlled. So if I were to put the system in a room that is 65 degrees and the temperature is increasing, the system should be in the 'on' state. If the temperature is decreasing, it should go to the 'off' state. The whole purpose of the PLC control system is for data logging, so basically I just want to check when the system I'm testing turns on or off. The problem thus is, when i turn the PLC on, i need to determine which state the system I'm testing is currently in.
 

ericgibbs

Joined Jan 29, 2010
18,766
Hi F,
If adding additional sensors for your PLC data logger is not permitted, I would just log the Logger On time and wait for a the next temperature switching event and continue logging the times for On and Off from that time.
Whats the typical On/Off cycle period.?

Eric
 

Thread Starter

Fragma

Joined Aug 31, 2014
8
Hi Eric,

I think that might actually be the best way to do this. The temperature in the chamber will rise and fall 3 degrees per minute, so the initial cycle time would be dependent on what the initial ambient temperature in the chamber is I suppose.
 

ericgibbs

Joined Jan 29, 2010
18,766
Hi
The problem you have is synchronising the first event with your logger as the 'system' is already running.
You can always later edit the logged data for the 'unknown' initial period, by noting if the first event is On or Off, the preceding state must be the opposite.

E
 

Thread Starter

Fragma

Joined Aug 31, 2014
8
Thanks for all the input eric, much appreciated. Will fiddle with it a little bit more but I think adjusting the data afterward would be the best approach.
 
Top