I'm looking for some pointers regarding G59/2 (assuming this is the correct spec).

Our company develops 3 phase to DC power converters and we currently using dissipative electronic loads during testing. We are looking to build a couple of higher power test systems capable of exercising 10kW power supplies. Putting 10kW into the air in the lab is undesirable so regenerative DC to AC test loads look to be a good choice. The resulting AC to DC to AC system will be less than 100% efficient and so will be a net user of energy. Question: do the rules on back feeding energy onto the grid apply here?

Thanks,
Craig

If the DC supply is derived from the mains, then as you point out there is no possibility of backfeeding, the losses would make such impossible.

If the units contain large capacitors, then backfeeding is possible very briefly but is unlikely to be a concern in practice, no worse than an induction motor running briefly as an alternator.

If however the equipment contains backup batteries, then backfeeding is possible by discharging the batteries, and in such cases the requirements for grid backfeeding would have to be followed.

Great care needs to be taken over the connections to avoid risk of shock.
Consider for example the risks of obtaining the DC supply from a transformer/rectifier on circuit "A" but backfeeding into circuit "B"
There would be no net backfeed of course due to losses, but what if someone isolated circuit "B" to work on it ? They could receive a dangerous shock.
Likewise consider the risks of the transformer/rectifier being on a different phase to the grid tied inverter being used as a load.
Any failure of the grid supply to the phase being backfed could result in it becoming live, perhaps at the wrong voltage or frequency, and almost certainly at the wrong phase angle relative to the other phases.

Going off on a tangent, as I do!

I went on a tour of the Worcester Bosch Boiler factory in Worcester and they showed us some boilers running on a "Z test" meaning some had been running flat out, continuously for many years. I passed comment that they must never be short of hot water in the cloakrooms and was told it was all run to waste and dumped into the air. I do wonder about peoples logic at times!

Could you not have a dump load of immersion heaters in large water storage tanks or simlar, could be easier than trying to feed it back to the grid.

Andy

I am involved with a heritage museum, where we demonstrate a mercury arc rectifier. To load this I built a load consisting a domestic cold water tank, fitted with a 3-phase 18kW immersion heater. The rectifier supplies 200V dc. The heater is rejigged so that the three elements are connected in parallel. Water is bottom fed into the tank, which can overflow into a drain at the top. At 200V it draws 60A (12 kW).
If the tank is filled without overflowing the load can be run for about 3 hours before it reaches about 50 degrees C (depending on water input temperature). If run with a small overflow it can be run indefinitely.

David

Thanks for the reply broadgage. The rectifier in question uses a 3 phase input, draws an actively balanced / power factor corrected load and switches off in the event of a phase loss. Likewise, the regenerative test loads source balanced high power factor AC. The combination of the two should stop A feeding B and mitigate the valid risk that you describe.

While net back-feeding is impossible for the system as a whole, I'm still unclear as to whether each element should be considered separately with regard to the regs.

Thanks for the water heater suggestions. Unfortunately we need far more control than heater elements will provide. We need to exercise our products dynamically across their full performance range - no load through to over load. For static testing, a water based system would indeed remove many of the problems associated with regenerative loads.

There do seem to be some test systems that use water cooling. These would allow us to use a heat exchanger outside of the building to dump the waste energy. Not very green - but if regulation proves difficult to overcome we may be forced down this road.