1
\$\begingroup\$

I am working on a control circuit for a heater. The circuit has a temperature controller (not included in the drawing), an overtemp controller, I/O going back to the PLC providing "heater ON" and "heater overtemp" status to the operators in the control room, and a strobe light on top of the heater panel.

Green on the strobe light indicates that the heater is ON. Amber indicates that the level of water in the tank that the heater is heating is low; this is controlled by a float switch in the tank. Red indicates that the heater is in overtemp condition.

One of the permissives required for the heater to run is that the drive controlling the pump should be in run mode: NO contact CR in the drawing. This is so that the heater doesn't run dry.

The circuit is working the way it's supposed to but I feel like it could be improved...and that's why I am posting here. I'd appreciate if you guys pitch in to simplify my design while satisfying all the requirements.

For the curious ones, the main temp controller (Yokogawa UT35A) has a 4-20mA signal going to a step controller which then controls 4 heater banks based on PID. I didn't include that because that was designed and installed by the manufacturer. I only did the overtemp and status indication part of the project.

Stack exchange wouldn't let me upload a DWG or a pdf file, so I am attaching a snapshot of the dwg. Hope it's legible.

Heater control circuit

Heater control circuit Rev1

\$\endgroup\$
1
\$\begingroup\$

It appears that the overtemp controller is acting as a supervisor to the main temperature controller. I offer the following comments:

  1. I would be inclined to hard-wire a contact of the over-temperature controller's output, CR28, into the heater circuit so that in the event of a PLC malfunction that the heaters would be switched off.
  2. I would consider adding an NC contact of CR28 in series with the LT26 reset button to confirm that CR28 has dropped out correctly before reset can be carried out. This is common on e-stop safety circuits and, at least, tests the circuit on power-up.
  3. I would check the overtemp controller's reset to establish whether or not it can be held in reset mode by holding the button in with a wedge, etc. The controller should be using an internal one-shot to prevent this but if it doesn't then you may need to address that risk. Providing the reset one-shot through the PLC might be a solution.
    To test this hold the reset in (or jam the button in the depressed state using a toothpick or match), create a fault, clear the fault and see if the overtemp auto-resets or not. (It shouldn't.)
  4. Check that you have open-loop protection in your setup. This means that if the controller doesn't see the expected response in the sensor feedback within a specified time that an alarm is generated and this can be used to cut power via an additional relay. This can protect against situations such as continuously heating while the sensor is hanging out in the air or a short-circuit failure of the heater control relay / SSR.

The need for each additional protection depends on the risk to personnel, property and product which you will have to assess.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Thanks for your feedback. But just so I am sure I understand, could you please clarify: \$\endgroup\$ – vasiqshair Feb 25 '19 at 3:22
  • \$\begingroup\$ Thanks for your feedback. But just so I am sure I understand, could you please address the following: #2) LT26 is an illuminated push button. In an event of an overtemp, the light turns ON and the user would have to push the button to reset the controller (assuming the overtemp condition is no longer present). I am not sure I follow when you say LT26 [should] drop out correctly. #3) Please elaborate. #4) When you say sensor feedback, are you referring to the thermocouple? Sorry if my questions seem too basic; I'm new to industrial controls. \$\endgroup\$ – vasiqshair Feb 25 '19 at 3:30
  • \$\begingroup\$ Updated. #2) should have read 'confirm that CR28 has dropped out'. Fixed. #3) See the update. #4) Yes, thermocouple. \$\endgroup\$ – Transistor Feb 25 '19 at 7:24
  • \$\begingroup\$ I revisited this problem today and have been thinking of the significance of adding an NC contact of CR28 in series with the reset pushbutton (#2 from our discussion). I am sure it's important but I am having difficulty in seeing how (mainly because I am not too familiar with industrial controls). Could you give me an example where not having this setup can cause problems for the system? \$\endgroup\$ – vasiqshair Apr 8 '19 at 12:42
  • 1
    \$\begingroup\$ Most industrial machinery requires a reset on power-up. You need to do a risk analysis on your machine and weigh safety, of people, property, equipment and product (probably in that order) and decide the level of safety you require. If auto-start of heating on power-up or after a brown-out are requirements then that steers you in a certain direction for design. Can you come up with another scheme to cycle test CR28? \$\endgroup\$ – Transistor Apr 8 '19 at 18:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.