Need input on load sharing between multiple generators

[Westcliffe01]

So as most of you know, Surplus Center has a pretty good deal going on the small Yanmar 2 cylinder 2TNV70 engine, as well as the larger Kubota D903 and Perkins/CAT C1.1.

These engines range in power from:
4.5kW @1600rpm for the 2TNV70
7.0kW @ 2200rpm for the 2TNV70
9.5kW @ 3000rpm for the 2TNV70

8.0kW @1600rpm for the D902
12.5kW @ 2200rpm for the D902
17.0kW @ 3000rpm for the D902

11.0kW @1600rpm for the C1.1
16.0kW @ 2200rpm for the C1.1
20.0kW @ 3000rpm for the C1.1

My application with cabin + barn/workshop "addition" would not require a lot of power the majority of the time.  Anytime I am actively engaged in working in the barn, using my lathe, mill or welding equipment, I would have no problem running a second genset to provide the additional power needed.

To reduce engine wear, noise and try to operate the engine at a decent efficiency point, I am interested in converting the 3 phase generator output into rectified and smoothed DC (using a wide range AVR system) which is then supplied to a large inverter, from which I then draw power to the house and barn.  I could add a substantial amount of capacitor buffering on the high voltage DC bus to absorb transient loads and stretch the system time constant out long enough that the mechanical load regulation of the engine would be able to keep up with the demand.

So basically this would be the diesel equivalent of the Honda/Yamaha inverter generator, with the engine adjusting speed with load and the output being clean from the inverter.  Speaking of inverters, I was looking at the Sunny Boy SB 8000US inverter http://www.sma-america.com/en_US/products/grid-tied-inverters/sunny-boy/sunny-boy-5000us-6000us-7000us-8000us.html

Now if I take this system, and on the DC bus side I add a second generator, same as the first, as well as couple the second engine to the controller for load sharing, is there any reason this would not work ?  If the second generator is inactive, no current can backflow through the diode bridges to the windings of the alternator, right ?  On the outlet side of the inverter, one could isolate the circuits one considers "expendable" with a contactor / relay, which is active only when the load is below a specified threshold. This would have to be managed by the controller, to allow for proper delays after isolation to avoid damaging equipment through being repeatedly being turned on and off in quick succession.  This would be a primitive form of load shedding to avoid browning out the gensets.  Once one adds in a series connected battery bank on the HV DC side of the bus, the load shedding could be more sophisticated yet, based on battery state.

So, since this seems a little complicated compared to just getting a big ol genset, here are my thoughts:  This setup permits one to start simple.  To begin with, one can go with a single engine/generator, constant speed.   If I use the Yanmar engine, one is looking at $800 for the engine + $870 for the YHG-8.2KW 3 phase head from GA Generator.  The 4 pole head is made for 1800rpm, but one could probably crank it to 2200rpm and adjust the AVR to get a few more hp (from the engine) in the most basic configuration.

Then once the primary construction phase is over and cash flow improved somewhat (does that ever happen ?), one could go out and add the second genset, this time perhaps the D902 with the exact same YHG 8.2kW 3 phase generator.  At 1800rpm the engine and generator are a better match than the 2 cylinder Yanmar.  Then the inverter, control system etc.   I figure that when one lives in a location that is likely never to see utilities and has a relatively modest income, one needs to operate the systems that one has in the most efficient manner possible.   I would love to run a more modern diesel (PD or common rail) but one is looking at over $3k for the engine, they have more HP generally that what I would need from a utility point of view.   I do however plan to get one of those beats under the hood of my 93 Toyota Pickup.

Feedback desired.
Keith

[mobile_bob]

what voltage are you thinking for the inverter dc buss?

i assume 48vdc nominal?

how many amps can you take from the genhead at your desired voltage?
this is usually the deal killer in my opinion. an 8.2kwatt head is good for about 35amps at 240 3phase
and will be the limit in amps even if you regulate down to some other rectified voltage below 240volts dc




bob g

[Westcliffe01]

Bob, when I said High voltage, I meant 208/230 rectified and smoothed, so that would be higher than the "rms" AC voltage.  The current would not be particularly high, which is why the inverters should not be too expensive.  If one looks at regular variable frequency drives (which have the rectifiers and capacitors built in on the input side) http://www.driveswarehouse.com/Drives/AC+Drives/Variable+Torque+VFD/L300P-110LFU2.html they are cheap at $890 for 11kW.   I think that by adding additional capacitance to the DC bus, one will be able to absorb larger transients through the stored energy.

[Westcliffe01]

Looks like I found another piece of my puzzle.  A 400V 12 000ufd capacitor which at rated capacity stores 960 Joules.  @$45 each one can add quite a bit of energy storage to a high voltage DC bus.   I would have to do some work to review load balancing with high capacity electrolytics, but I assume a shunt in series with each capacitor that has a higher resistance than the internal resistance to help balance the flow of current. 

Even 4 of these caps will make a substantial difference to the surge when starting an air conditioner or well pump, or in my case, the motor on my lathe, mill or surface grinder.  Since a Joule is 1Ws and for 230Vac the P-P voltage is ~322V this is likely to be the voltage on the DC bus.  Thus for the actual application, energy stored is closer to 622J / cap since the voltage is lower than the rated 400V.  So 4 capacitors provide 2488W.s and 8 take it to 4976Ws or the ability to "carry" a large load that is suddenly applied for several seconds while the governor has a chance to react.  Provided one provides space to expand, one could start with 2 or 4 and add capacity as funds permit.   I know there is a lot of discussion regarding safety with high voltage DC systems, and they certainly need to be laid out in the appropriate manner, with safe means to discharge any high energy device.  Shielding to prevent contact (inadvertent or otherwise) with high voltage and or high energy circuits is essential.  My father spent his life working on high voltage traction system, so I do have a mentor in that regard and otherwise have to rely on common sense and good system design.

Idealy, I would like to use a current transformer coupled to one or all of the output phases of the inverter to power the fuel solenoid which will govern the engine.  The advantage of  system like this is extreme simplicity, once the design is perfect.  Quite a bit of experimentation would be required to figure out the load imposed by the fuel rack, together with whatever mechanism one comes up with and then developing the force and stroke of the custom solenoid (or possibly the other way around using existing solenoids once the forces are known).  I currently have access to Maxwell magnetic analysis software at work, so designing a special solenoid would be a lot less tedious that a custom wound transformer (I just did one last week).

http://www.amazing1.com/capacitors.htm (scroll down to the section headed "High Energy Storage Electrolytic Capacitors"