0
\$\begingroup\$

I have one question related to this schematic: enter image description here

Scenario:

  • Source 1: AC source 220VAC, 50Hz, 120º
  • Source 2: AC source 220VAC, 50Hz, 0º

If I want to switch from source 1 to source 2 or from source 2 to source 1, which is the best moment?

  • When the voltage in the inductor is 0V or when the current in the inductor is 0A.

The key is to reduce as much as possible the voltage peaks, so as the load is an inductor, I don't know what will happen, because the volatage over the inductor will change due the 120º between two phases of the powerline when the switch changes the source.

\$\endgroup\$
7
  • \$\begingroup\$ Start from Lenz's law. \$\endgroup\$
    – user16324
    Commented Sep 28, 2020 at 16:23
  • 2
    \$\begingroup\$ There is no one moment that is best. There is a good time to disconnect and there is a good time to connect but, they are not the same in time. \$\endgroup\$
    – Andy aka
    Commented Sep 28, 2020 at 16:37
  • \$\begingroup\$ I think disconnecting at 0V is a lot worse than connecting at 0A. \$\endgroup\$ Commented Sep 28, 2020 at 17:28
  • \$\begingroup\$ The best solution, you will find it, if you expand equations of common problem - transformer inrush calculation. Perhaps making a MATLAB simulation and then visually inspect the best result. I have no other idea, it's quite difficult math. problem. \$\endgroup\$ Commented Sep 28, 2020 at 17:32
  • \$\begingroup\$ This reminds me of a cool paper co-authored by Carson Taylor. The idea was to disconnect a generator that was slipping a pole, and reconnect it to different phases and regain the lost angle. \$\endgroup\$ Commented Sep 28, 2020 at 20:44

2 Answers 2

5
\$\begingroup\$

This picture appears in the original question. It shows that when the sinusoidal voltage is at zero (black waveform), the current in the inductor (red waveform) is at a peak (positive or negative). It also shows that when the voltage is at a peak, the current in the inductor is zero: -

enter image description here

So, the most appropriate time to disconnect an inductive load from a sinusoidal voltage supply is when the current is small and that only happens when the applied voltage is at a peak. So you open circuit at that point and naturally, there will be very little stored energy in the inductor that could create a spark and erode the switching contact.

The most appropriate time to connect an inductor to a sinusoidal voltage supply is when the voltage is at a peak because the developing current will naturally begin at the right point and there will be little or no inrush current peak. If you connect when the voltage passes through zero, the current peak (for the first cycle and diminishing thereafter due to losses) will be twice the steady-state maximum and, of course this is to be avoided to prevent excessive current flow in the circuit and possible core saturation problems.


Regarding what the op says in comments on another "answer", these misconceptions must be properly addressed: -

For me, this answer make sense. During initial conditions, (inductor current =0), if we apply VAC=0, then the current on the inductor will be 0A. But, using this way, there will be a short time, where the Load (inductor) will be 'stopped'. (No voltage/current in the Load) – Kotik_o

  • "During initial conditions" - there are no initial conditions in the proposed circuit and this exacerbates the problem if you are trying to locate "the best moment" for switching.
  • "if we apply VAC=0, then the current on the inductor will be 0A. But, using this way, there will be a short time, where the Load (inductor) will be 'stopped'." - No that will not happen as you believe - the inductor current will immediately start to rise or fall to a value dictated by the formula \$V = L\frac{di}{dt}\$ and will potentially rise to a peak value that is twice the normal running peak current.
\$\endgroup\$
6
  • \$\begingroup\$ Hmm a down vote - has anyone got the guts to own up to it? \$\endgroup\$
    – Andy aka
    Commented Sep 29, 2020 at 12:23
  • \$\begingroup\$ With pleasure, it was I. your answer doesn't make any sense. \$\endgroup\$
    – Moty
    Commented Sep 29, 2020 at 13:33
  • \$\begingroup\$ A total red-herring. I think you are struggling to justify yourself here. \$\endgroup\$
    – Andy aka
    Commented Sep 29, 2020 at 13:48
  • \$\begingroup\$ It's theoretically impossible to have less downvotes than Moty ^^ \$\endgroup\$ Commented Oct 1, 2020 at 14:28
  • 1
    \$\begingroup\$ Yeah oops, meant that \$\endgroup\$ Commented Oct 2, 2020 at 5:55
-5
\$\begingroup\$

Disconnect when phase 1 is 0V, connect load to neutral, wait for the phase 2 to be 0V and connect to it. Do the same when switching between phase 2 and phase 1.

\$\endgroup\$
5
  • 2
    \$\begingroup\$ Can you edit to explain why you think this is a good solution? What will the current be when Phase 1 = 0 V? \$\endgroup\$
    – Transistor
    Commented Sep 28, 2020 at 17:50
  • \$\begingroup\$ Have you downvoted my answer @Moty \$\endgroup\$
    – Andy aka
    Commented Sep 29, 2020 at 12:23
  • 2
    \$\begingroup\$ Anyway, just so that people know, this answer is totally wrong when dealing with an inductive load. \$\endgroup\$
    – Andy aka
    Commented Sep 29, 2020 at 12:37
  • \$\begingroup\$ For me, this answer make sense. During initial conditions, (inductor current =0), if we apply VAC=0, then the current on the inductor will be 0A. But, using this way, there will be a short time, where the Load (inductor) will be 'stopped'. (No voltage/current in the Load) \$\endgroup\$
    – Kotik_o
    Commented Sep 30, 2020 at 15:13
  • 3
    \$\begingroup\$ @Kotik_o when the voltage is zero, for an inductor, the current is maximum. You should really think about this and not be so easily and incorrectly convinced by a simplistic and erroneous explanation. You only need to look at the top left waveform picture in your question to realise this. \$\endgroup\$
    – Andy aka
    Commented Sep 30, 2020 at 20:48

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.