I'm trying to automate an exhaust fan by using this ESPHome-driven relay: enter image description here

I've got the relay setup working and now I'm puzzling over the actual AC motor control unit. The unit is Sentera STR-1. The transformer inside is STC-1-10L10. The unit has a rotary switch (Gottak GTK 46RCTH), which basically shorts different paths to the transformer to get the required voltage (if I understand it correctly).

My plan was to simulate different states of the rotary switch by connecting one relay per rotary knob position to the transformer's terminals. motor control unit disassembled

I think that the lower left terminal on this is connected to the coloured wires in the top row in different positions. Here in this gif I cycle through all knob positions from OFF to 1 to 5 and you can see the mechanics behind the switching – the terminals are connected to the copper plate when the knob is turned. The plan is to connect wires to the appropriate terminal/rotary knob terminals and let the ESPHome switching do the magic.

However, before I blow up my motor or the relay, I want to ask:

What happens if I energize multiple inputs (e.g. 80V and 110V) on the transformer at the same time?

I'm using the interlock feature provided by ESPHome switch component. The ESPHome documentation has an angry red box warning that the software interlock might fail. I'm not sure what would happen with the relay/motor then. Hence the question.

Here's an image of what I think the current scheme looks like on the top, and what I would like to do on the bottom: (Disclaimer: the schematics are probably wrong, but I hope they give an idea of what I'm trying to do) enter image description here Sorry for butchering the schematic, but several online tools didn't support all the necessary features, so I did an "MS Paint" on it. I hope it's understandable.

Ideally, I would appreciate two answers:

  1. Would the transformer blow up if the interlock feature fails (it might fail momentarily – for a few dozen milliseconds, perhaps).
  2. What would be a proper, reasonably cheap solution instead of separate relays?
  • 1
    \$\begingroup\$ Welcome! Can I suggest drawing out the circuit of the fan? And also what you're proposing. You'll find you get much more helpful answers. \$\endgroup\$
    – jonathanjo
    Commented Oct 18, 2022 at 7:59
  • 3
    \$\begingroup\$ Driving multiple different voltage inputs simultaneously is a Bad Idea (tm). It creates a nominal short circuit, with the only thing limiting the current being winding resistance and leakage inductance int he transformer. At best, the transformer will simply get hot, or the supply fuse blow. At worst, it will smoke and maybe catch fire. \$\endgroup\$
    – Neil_UK
    Commented Oct 18, 2022 at 20:35
  • \$\begingroup\$ Thanks for the welcome and the suggestions, @jonathanjo. I've added a circuit-ish thing to my initial post. \$\endgroup\$ Commented Oct 19, 2022 at 6:41
  • \$\begingroup\$ @Neil_UK with resistances typical for such a transformer, would <1s short circuits also be dangerous or would the danger come from prolonged heating of the coil? The short would likely always be between adjacent taps, so the winding taking the hit would be minimal. \$\endgroup\$ Commented Oct 19, 2022 at 6:46
  • 1
    \$\begingroup\$ It means you short circuited some turns and that is like having a small winding that is short-circuited. Which is like powering a short circuit through a transformer. \$\endgroup\$ Commented Oct 21, 2022 at 22:45

2 Answers 2



simulate this circuit – Schematic created using CircuitLab

Figure 1. Series connection of the available changeover contacts ensures that only one tap can be energised at any time.

  • \$\begingroup\$ Oh, wow! This sounds exactly like something I was looking for. Thanks so much for this! \$\endgroup\$ Commented Oct 20, 2022 at 11:38
  • 1
    \$\begingroup\$ @Lightheaded, I've redrawn the schematic to make it clear that the input voltage should be applied across the whole winding and the load connected to the tap-off points. \$\endgroup\$
    – Transistor
    Commented Oct 20, 2022 at 12:11

The schematics you have drawn are wrong.

The transformer is an autotransformer which is basically a single winding with multiple taps:


simulate this circuit – Schematic created using CircuitLab

An autotransformer is basically a step-down transformer but with no isolation. You apply voltage (presumably 230V) across its L-N terminals only and take output from each tap (e.g. 170V output will have 170V across this tap and N). So, normally, no voltage is applied to the taps externally.

The relay box gets its inputs from autotransformer taps and directs only one of them to the output via relays. The main application of this is to drive an AC motor with different speeds.

Now the potential problem is having two relays direct their inputs to the output at the same time. This is probably what happens if interlock fails. In this case some taps will be shorted and therefore an unexpected voltage will be applied to the load. For example, if 140V and 110V switch relays are activated at the same time then L3 part of the auto transformer will be shorted. Therefore the voltage ratios will change so the output voltage at the output will be different. You can calculate the ratios and estimate the output voltage for different failures (i.e. tap shorts).

  • \$\begingroup\$ Thanks so much for this explanation. I went back and forth between which way the autotransformer works – are the taps inputs or outputs? This clears it up. Also, as suspected, using software-based interlocks is probably a good recipe for a house fire. Just wanted to confirm it with people who know what they're doing :) Thanks again so much! \$\endgroup\$ Commented Oct 20, 2022 at 11:36
  • 1
    \$\begingroup\$ @Lightheaded the taps are normally output, as shown in the schematic in my answer. The entire multi-section bobbin (autotransformer) works as a voltage divider so when you apply 230V across L and N terminals the tap voltages will be lower than 230V. A tap can be selected via a rotary switch or a relay bank to direct to the load. \$\endgroup\$ Commented Oct 20, 2022 at 11:37
  • 1
    \$\begingroup\$ btw: the way @Lightheaded drew the multiple taps is fine - the "wrong drawing" is referring to the fact that there is only one winding (and the output is another tap on this same winding) not two \$\endgroup\$ Commented Oct 21, 2022 at 22:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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