I need to switch supply power from one source to a different one without losing power. I am designing a control system for a customer to run an induction heater. They want to run the system off a generator but it draws to much power during the initial heating phase. Once the item is heated they want to switch over to the generator. My initial thought is to have two contactors, one NO and one NC, feed by the same control wire. That way, in theory, they could never both be on at the same time.

Since they are not perfect ideal devices there is a good chance that they will both be on at the same time or there will be a delay. The control system is also powered from this so I really don't want a voltage spike but I also don't want to lose power and cause the controller to reset.

My question is what other components do I need to add to keep this safe and not lose power?

  • 3
    \$\begingroup\$ I don't see why a heater would need uninterrupted electrical power fed to it. Time constants in heating are in the range of seconds at least. Switching one source off and the other one on half a second later would sure do no harm. \$\endgroup\$ – Janka Oct 19 '18 at 22:37
  • 1
    \$\begingroup\$ @Janka . . An induction heater normally has a short restart time of a few seconds, which could easily disrupt the process. \$\endgroup\$ – Marla Oct 20 '18 at 1:06
  • \$\begingroup\$ Are we talking about a PID controller for this system? So that it is at 100% at first and not really within PID so much, but then goes into more of PID control once it gets close? I think we'd need to know a lot more about the induction heater system and its response function. Simple ON/OFF switching from supplies sounds pretty dangerous/ignorant to me. I think there is some math here that needs doing. But maybe that's just me. \$\endgroup\$ – jonk Oct 20 '18 at 4:23

To switch between two sources a single contactor with changeover contacts would normally be used. However, doing this with a mains AC supply is not recommended. Since there is (most likely) no synchronisation between the two supplies you could end up switching while one supply was at maximum positive peak voltage and the other at peak negative voltage. This would generally be considered a bad thing.


simulate this circuit – Schematic created using CircuitLab

Figure 1. A possible solution.

  • SW1 is the control.
  • With SW1 up RLY2 is energised and the circuit is powered by mains.
  • When SW2 is switched down. RLY2 is switched off. When it drops out RLY1 coil is energised and RLY1 is energised and the generator powers the load. Whether or not the resultant time delay is adequate is up to you to determine.
  • Switch back is a similar sequence.

  • The RF generator suffers a brief interruption in power during the transfer.

  • A small UPS powers the controller during the transfer and prevents the controller reset.

If it's an induction heater it's going to be running a large current anywhere from tens to 100's of kHz (or whatever the design frequency is). We need some clarification on whether the control system and the power section is powered separately, but in the worst case you are looking for an closed transition transfer switch (CTTS), with the generator having the ability to synchronize to the utility phasing.

You might just run a quick test with a power kill and see how long the system stays up and running (the power section will likely die fast but it the control section has a decent power supply it may have some decent hold up time) to switch to the generator. You might get away with just adding a UPS to the control section, but that might mean working with the internal circuitry (don't know your experience level).

Costs for a CTTS will depend on the size of the system, and often the utility company will want to check out the wiring due to safety reasons to make sure the generator doesn't back-feed into their system (it shouldn't since it's not being plugged into the building bus) but for a brief time the generator will be online- this is a safety hazard for utilities if the generator gets put on the utility grid and back-feeds the system (those step-down transformers work just as well stepping up), hence the utilities' interest (it's not an economical one at that point).


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.