# Do ordinary transformers need flyback protection?

Low voltage 12V relays driven by a microcontroller plus transistor need a flyback diode to suppress an inductive voltage spike when they are switched off.

Do ordinary power transformers (say, at a distribution substation) need some sort of similar flyback protection in case a large fraction of their load is suddenly shed (e.g. a tree fells powerlines and a breaker opens, etc.)? I would imagine that a resulting voltage surge for those who remain connected to the grid would be very undesirable. What about lower current transformers in "power" household applications?

In practice is there a good way to estimate the magnitude of the spike (besides empirical testing)?

Low voltage 12V relays driven by a microcontroller plus transistor need a flyback diode to suppress an inductive voltage spike when they are switched off.

There's a whole realm of difference between disconnecting a primary inductance circuit AND disconnecting a secondary load.

Do ordinary power transformers (say, at a distribution substation) need some sort of similar flyback protection in case a large fraction of their load is suddenly shed

No, because the primary winding is connected to the AC distribution network and that distribution voltage can be regarded as having a very, very low impedance hence, any occurance of the load being dropped on the secondary side will result in absorption of transformer stored energy directly into the AC grid supply. Sure, there might be a little current surge but that will be "clamped" by the very low impedance of the grid supply.

I would imagine that a resulting voltage surge for those who remain connected to the grid would be very undesirable.

Not at all, the voltage fluctuation due to the energy release from the transformer will be measured as a few volts. The distribution voltage supply will soak it up like blotting paper.

What about lower current transformers in "power" household applications?

It makes no difference; scale is not an issue.

In practice is there a good way to estimate the magnitude of the spike

If you know the distribution voltage source impedance then you can calculate the disturbance (a few volts) based on knowing the leakage inductance of the transformer and the peak current being taken at the moment the load is shed.

I think another scenario might help. If the transformer primary were disconnected suddenly (circuit breaker etc.) then there could be a huge voltage surge across the contact breaker's contacts and that, can cause significant damage to the CB if it isn't protected sufficiently. This would be equivalent to using a flyback diode on a DC circuit.

Footnote

All AC circuits these days are protected against indirect lightning strikes and these protection devices (in buildings and with line transformers) will clamp voltage surges. So, even if there were some kind of load shedding surge, there are grid components that "clamp" to prevent excessive voltages arriving on the plugs of appliances. You can, these days rely on surges protection devices for limiting internal household and office supplies to a peak spike of about 1500 volts.

• To expand on what Andy says above - read this writeup. Dec 6, 2020 at 14:25
• Thanks @Andy aka. Maybe I misunderstand, but I was talking about a sudden transient appearing on the secondary side, visible to the people who are still connected (as opposed to the primary side or the secondary-side people who were shed). My understanding is that the current through the secondary combined with the secondary inductance + transmission line self-inductance should produce a spike visible to the people on the secondary side. Do I misunderstand you and you are saying that exactly this effect will actually be much smaller in magnitude than I worry? Dec 6, 2020 at 16:04
• Perhaps also you are implying that there is no situation where a breaker would be configured to only shed part of the load from a transformer? That makes some sense to me, but I can still see a situation e.g. where one of multiple transmission lines is severed without causing an overcurrent event. (Of course if my first comment is wrong then this last comment isn't a worry at all.) Dec 6, 2020 at 16:07
• @KeeleyHoek the surge you are talking about (or appear to be talking about) due to a load being disconnected would appear across the disconnection device if it were an inductive load being disconnected. If you think I am still not following the scenario then you must embed a picture showing precisely where the load is disconnected and what transmission line inductance you refer to and add as much info as you can. But bear in mind there will be surge protection devices and that the transformer can and will pass excess energy to the primary side. Dec 6, 2020 at 16:14