Generator and current spikes

[Veketti]

Whenever facing power outage and running vital equipment like fridge and freezer from small ~2000kW "suitcase" generator with load varying rpm, it's bit annoying that when the fridge/freezer compressor starts, it takes so big current spike that the generator almost choke before it can react with higher rpm. Fridge and freezer when running draws very little current so the startup current is the only problem.

The question is, is there any purpose built capacitor that could be plugged in between which could handle these current spikes? Or anyone built one or handled the issue some other way? Bigger generator is not an option...
Was thinking that if building such thing, would need hefty resistors and diodes that would allow the capacitors to charge with smaller current to prevent current spike when charging them but still allowing limitless output...

Thank you in advance

[Marco]

If they are really fancy they might be inverter driver and able to work off rectified AC, then you could add capacitors. If they are AC you could use an UPS.

Either way, you'd need a (solid state) relay to disconnect them from the generator when they start up.

[Andy Watson]

Put an NTC thermistor in series with the load. For example:
https://www.bourns.com/products/circuit-protection/ntc-thermistors

Bear in mind that, by design, they run hot.

[microwhat]

Motors don't like low voltage. Adding a resistor in series will change the V/Hz ratio, and the motor will now draw more current and run even hotter.

If the motor is run by a VFD, it might be possible to adjust the acceleration time, but I doubt any fridge that will run off a 2KW generator will contain such hardware.

With single phase motors, there is no good solution short of a bigger generator.

[Veketti]

Then maybe portable power station like DJI Power 500 or 1000 in between would take the current spikes and work?

[Jeroen3]

Your problem is that your generator does not have enough capacity in it's magnetic flux to provide this current, and then it does not have enough mass to provide the sudden energy.
There is no magic option to resolve this except a larger, more poles, generator.

You can add an inverter, with large DC-Link capacitors or even batteries to replace this lack of oomph.
But caution, feeding cheap inverters will cause harmonics and poor power factor, which will derate your generator by two or three and worsen the voltage quality.
Also, check whether your power station actually uses the output inverter when you have the AC-in connected. It may not, or may not be able to be output when charging.

Modern fridges are inverter driven and should not cause significant inrush anymore. So upgrading your fridge may also be an option. It may even be more efficient in the long term.

Invest wisely.

[Zero999]

Whenever facing power outage and running vital equipment like fridge and freezer from small ~2000kW "suitcase" generator with load varying rpm, it's bit annoying that when the fridge/freezer compressor starts, it takes so big current spike that the generator almost choke before it can react with higher rpm. Fridge and freezer when running draws very little current so the startup current is the only problem.

The question is, is there any purpose built capacitor that could be plugged in between which could handle these current spikes? Or anyone built one or handled the issue some other way? Bigger generator is not an option...
Was thinking that if building such thing, would need hefty resistors and diodes that would allow the capacitors to charge with smaller current to prevent current spike when charging them but still allowing limitless output...

Thank you in advance

A synchronous motor, with a large flywheel connected to its shaft. It will need a clutch and possibly some form of starting circuitary, so it doesn't stall your generator, when initially connected.

[synccond]

I would look into a TSC (Thyristor-Switched-Capacitor).  Plenty of articles on the internet discuss the basic topology.
Your problem is that the compressor motor is immediately switched on, which applies a highly reactive load to the generator.  The only way to get reactive power out of the generator is by increasing excitation via the DC magnetic field, and this takes TIME.  The overall time is approximated by a 2nd order transfer function consisting of the DC rotor field time constant (L/R) and the open circuit time constant of the armature (T'do).  The load is applied immediately, yet the generator's armature output rises slowly, hence you end up with low voltage output from the generator for a good amount of cycles.  Perhaps the generator catches up eventually, but not before your other connected loads have experienced the same voltage drop and are complaining or even dropped out.

Do not pursue synchronous motors or flywheels - you do NOT need extra watts (which a flywheel would provide) you need extra VARs.  This is only possible with magnetic excitation and/or passive capacitance.  The TSC is a passive solution, and allows cycle-by-cycle adjustment of the amount of capacitance.  I suggest you might want 2 or 3 stages for optimal voltage regulation.  One stage is pretty aggressive, and risks overexciting the generator when load drops.  As well, a synchronous motor in theory is a proper solution for getting VARS, but it suffers from the same problem as the generator, which is the 2nd order lag transfer function.  They are simply not fast enough to supply what is needed for an across-the-line motor start.  Thyristors are there just as fast as you can detect the start and respond with an 'on' signal.

Last option would be a STATCOM, which is insane for a 2 kW system.  Those are essentially PWM drives run off a DC bus which recreate the AC waveform.  By adjusting phase angle, they are able to absorb or supply reactive power.  Not impossible, but really quite the overkill.  TSC is simpler and easy to implement with off-the-shelf products.

[microwhat]

And by the time you've paid for all that screwing around, you could have bought a larger generator.

[synccond]

You might be visiting the wrong forum if the share of knowledge and pursuit of creative ideas are to be superceded by what is simplest in buying larger than is necessary (not to mention the cost of a 10 kW generator in order to run a 1 kW load).

[MF-jockey]

I've had good experiences with soft starters on electric drives, but I have no experience with cooling units or compressors. However, the sudden demand for reactive energy to magnetize the motor must cause problems. I would try a standard two-pole Soft Start Current Limiter ($10).

The following video on soft start and hard start might help to solve the problem:

[indeterminate]

Your suitcase generator is running in echo mode it will have a switch that allows it to run flatout all the time.
if you take the wiring from the echo switch to a relay that is controlled bye the fridge thermostat and have a delay before another relay switches the fridge compressor on you will circumvent your problem.
the sequence of events

fridge thermostat closes
generator switches to full power mode
delay 3 seconds
fridge compressor starts
delay 3 seconds
generator switches to echo mode mode

[antenna]

When the power goes out, don't switch over to the generator right away.  Give it an hour.  If the fridge had been running, there is pressure at the compressor head.  Let the pressure dissipate and the compressor cool before trying to run it on the generator.  Side note: varying RPM means varying frequency, which throws off anything that has power factor correction.  A bigger generator is cheaper than replacing everything else you own, so don't play stupid games.  It's just too small.

[Veketti]

Excellent points.. Fridge and freezer are both maybe 25 years old, but Swedish made Electrolux so most likely will still after these years, outlast any modern devices…

Regarding the wait time before switching to generator. There’s a downside on that. If both units have waited for longer time, then they both most likely are already triggered by thermostat and then when the generator is connected, they both start their compressor same time. Faster the generator is connected the risk of simultaneous switch on is smaller.

About the eco -mode that user indeterminate mention, that’s true. Power need is all the time so small that it runs with small rpm and only when the compressor starts it chokes... So having it in full throttle before thermostat closes would be one solution. I could of course leave it to full throttle all the time, but it’s not good for fuel consumption and especially noise.

Also the soft starter might be interesting to test and maybe the easiest to implement.

Thanks for the suggestions. I need to think about all these.

[Doctorandus_P]

Have you measured the startup current and verified the fridges work OK? Maybe they are (slightly) broken which may cause a bigger startup current.

I agree with antenna, when the fridge starts right after it stopped, staring is probably more difficult, but a few minutes is probably enough.

For the generator. Adding a flywheel to the generator itself should also help. And maybe the generator is faulty or not adjusted properly.

[Jeroen3]

You could try this: https://www.h-tronic.de/en/Inrush-Current-Limiter-SCL-3680/1114730
But you risk that the compressor won't start at all with inrush limiting.

(And it needs a minimum load, fridge may be too small)

[tggzzz]

Have you measured the startup current and verified the fridges work OK? Maybe they are (slightly) broken which may cause a bigger startup current.

With a fairly new freezer, maybe 5yo, the rated consumption is ~150W.

The startup transient varies considerably, apparently randomly, and without any visible reason. At worst it can be >1kW for ~1s.

I am sceptical that there is a quick fix for that involving a few added components.

[antenna]

The startup transient varies considerably, apparently randomly, and without any visible reason. At worst it can be >1kW for ~1s.

I think the reason is a combination of pressure in the compressor head, counteracting the energy needed to demagnetize and magnetize the core if the power was cut on a voltage peak of opposite polarity as that applied during restart, and the added resistance if the windings are still warm.

[tggzzz]

The startup transient varies considerably, apparently randomly, and without any visible reason. At worst it can be >1kW for ~1s.

I think the reason is a combination of pressure in the compressor head, counteracting the energy needed to demagnetize and magnetize the core if the power was cut on a voltage peak of opposite polarity as that applied during restart, and the added resistance if the windings are still warm.

It would be difficult to measure the relative magnitude of any such effects.

The net result is that you can get a startup transient of a factor of 10.

[indeterminate]

An ozi magazine published a kit for a soft starter that may be of interest
https://www.siliconchip.com.au/Issue/SC/2023/February/Active+Mains+Soft+Starter%2C+Part+1