Exploring using Diesel Gensets with renewable energy and storage (efficiency question) 1

[PrimalPete]

HI,

I work on renewable energy microgrids for commercial and industrial. Typically this includes solar panels and battery storage.

I am exploring adding a renewable microgrid to an existing building which has a diesel generator backup. There is an ATS that switches to diesel during a grid outage. There is no emergency panel, the genset picks up the entire building load. The battery and solar would be added on a load side tap, downstream of the ATS.

I am thinking of using the diesel generator during a utility grid outage situation to form the a grid/frequency. The renewable inverters would then follow that voltage and provide up to 80-90% of the load, diesel would handle the difference. I am aware backfeeding and so the renewable output will be following the load in real time, which is why there is a 10-20% cushion between renewable power vs diesel power. The idea is to explore how long we can increase the diesel's full reserve when the renewable system is actively providing 80-90% of the load.

Now here comes the question at hand. I am not too familiar with diesel gensets. Based on data sheets I have looked at, they typically give you fuel efficiency of 25%, 50%, 75% and 100% of rated load (output power). The problem is, I am looking at several sites with 150kW diesel generator, and an average building load of 20kW. This average load is already less than 25% of rated output power of the genset. I fear that anything below 25% will not change the gal/hour specified for the 25% level in the datasheet, which would ultimately mean that there is an almost complete waste of renewable power in such a system.

Is there any way to calculate fuel efficiency at low levels of 2-5% rated output for diesel generators (given that I know the exact model number and can obtain data sheet)? Or is this simply a terrible idea?

 

[davidbeach]

From the efficiency numbers you might be able to come up with a rough estimate of how much mechanical power it takes just to keep machine spinning. You’re going to burn that fuel regardless. Then there’s then the running unloaded problem and consequences such as wet stacking.

It would be better to let the renewables and storage carry the load by themselves as much of the time as they can. Then run the diesel at full load when needed to top off the storage.

 

[BrianPetersen]

Use the quoted efficiency numbers at various engine loads to calculate fuel consumption in kilograms per hour (or whatever units you care to use) at those engine loads, graph them, and interpolate with a curve fit (it will probably be fairly close to a straight line that doesn't intersect at zero). It will be close enough for what you're doing with it.

Engines of all types require a certain amount of power just to keep themselves spinning and warmed up. Efficiency at very light load will be terrible.

How frequently are your grid outages? If it's anything like here, "not very much". It might not be worth doing anything about the lousy efficiency in that specific set of circumstances. Or maybe it's worth getting a smaller genset that's better matched to the actual load. Only you are in a position to evaluate this; we're not!

 

[FreddyNurk]

There is now commercial equipment that will do what you are asking, although at 150kW I'm not quite sure of the size and arrangements available. The other factor would be that you're in the US, which may change availability somewhat as compared to a lot of the 50Hz equipment that is available.

Some of the issues include managing minimum generator loading (electronic engines are much better than the mechanically governed counterparts these days, but maintaining adequate combustion temperature is still critical), allowing for changes in output due to changes in renewable resource, and what particular equipment manages the frequency reference. Control capability with whatever inverter systems you choose will be critical to maintain stability, and you're better off running diesel off rather than running the set just for a frequency reference.

The real question though, is what are your requirements, are you looking at being able to utilise renewable power most of the time, but not have the penalty of an outage when the grid drops off, or are you looking at running permanently islanded and doing away with a utility connection completely?



EDMS Australia

 

[waross]

A rule of thumb for loaded generators is 13 KWHr per gallon of fuel.
Look at the fuel consumption per hour at 100% load and calculate the KWHrs per gallon.
Take 24% of that and subtract from the fuel consumption per hour at 25% load.
The difference will give an indication of the fuel used to keep the engine running and hot.
It may make more sense to run on solar until the batteries get low and then pick up the load with the generator and charge the batteries from the generator. When the batteries are charged, shut down the diesel set.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Primal sounds like he's been burned by PG&E recently.

Supplying power to a lightly loaded generator can cause stability problems with its voltage regulator that could be rather disastrous to the supplied loads.

With local storage you can UPS the building starting the generator when the allowed depth of charge has been reached. The generator then charges the storage at the generator's near maximum ability while also providing energy to the load. That uses the generator around its highest efficiency. Use the solar only to charge the storage when in emergency mode.

Trying to run grid-tied to a tiny grid island during E-service will probably be problematic.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

During a power failure, if the solar output is greater than the load, it may try to motor the generator. Not good.

Is there any way to calculate fuel efficiency at low levels of 2-5%

There will be a dramatic change in the percentage efficiency between 2% load and 5% load. The actual fuel consumption increase will be related to the 13 KWHr per gallon to produce the added 3% output.
Consumption is more closely related to the weight of fuel than to the volume. There may be as much as a 10% difference in specific gravity between summer diesel fuel and winter fuel.

Suggested mode of operation:
Run on solar until the batteries become low.
Start the generator and charge the batteries at the highest acceptable and possible rate.
Curtail the inverter output so that the generator is fully loaded.
You will achieve best efficiency if the generator is fully loaded for the shortest time.
You may consider the time of day.
For instance, if you are charging the batteries at 5:00 AM, you may anticipate good solar production in a few hours and charge to less than 100%.
You may decide to terminate charging at 50% (or some other percentage of your choosing).
If you have just finished charging to 100% with the generator and the grid power returns, you may have wasted fuel.
With grid power available, you don't need the batteries at 100%.

This is an interesting problem with more than one interesting solution.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[PrimalPete]

HI all,

Thank you for all the responses. I have limited access to my computer today (and yesterday) but I look forward to reading and digesting all comments.

Just to clarify a few things, the renewable system was designed and built for two reasons. One is obviously for reduction in energy costs which will play out over time. The 2nd is resiliency. This facility needs to be operation during a grid outage and so the idea was to use the renewable system to island them, with the diesel generator as the backup.

A few clarifications:
1. This system is already built and operational (USA)
2. As of today, if grid is lost, diesel takes over and renewable stays idle
3. If the facility wants to island using renewable, both diesel breaker and main breaker must be manually flipped first by personnel onsite
4. I am tying to automate the process for the renewable to island without any manual process
5. I understand this can be done with hardware (ie ATS upgrade, replacing main breaker with motorized switchgear breaker, etc) but I am exploring "cheap" alternatives
6. This is why I am asking the question of the genset. I want to explore if its viable for the renewable to follow the genset. I was hoping that renewable following genset and providing most of the load, that the fuel would last 5-10x more than if renewable was not present (I need help to calculate this out, how far can I theoretically extend fuel reserve by enabling the renewable system as I described)


Also, once the ATS is in emergency mode (ie diesel), I cannot fill the batteries, turn off diesel, then island with renewable. The problem is that the renewable is on a load side tap. If I'm islanding using the inverter with diesel off, the ATS will not know better and will assume the diesel is providing the voltage/frequency. Then once utility grid returns, it ATS will switch back to source 1 (utility) and I will have inverter grid colliding with utility grid. (apologize for not having a diagram handy, but I think most of you get the gist of this)

I have been exploring hijacking the ATS controls and manipulating it in such a way to safely isolate the renewable from utility, and having the renewable controller handle all decision making (ie it would manipulate ATS controller, be master decision maker, and I would need to add additional hardware to the existing ATS circuit). This would be a code violation without proper approval and help from ATS manufacturer (they quoted $20,000 for oversight and approval, which is insane).

 

[BrianPetersen]

How frequently do grid outages occur? If they are as infrequent as they are here, the economics of doing what you propose are never going to be viable ... just deal with burning fuel in that time.

 

[itsmoked]

1) Brian's point is a good one.
2) Lightly loaded generators are soon to be damaged generators.
3) Driving energy into a generator as you propose can cause voltage instability, (i.e. over voltage, brown-outs, tripped protection, failed generators)
4) Depending on the load size, solar array size, actual battery storage size, and generator size, this scheme could be completely unworkable or might need major adjustments or could maybe work. It needs to engineered.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

$20,000 Wow.
When I was seeing a lot of standby sets i several times had occasion to modify the operation or controls of an ATS.
I always got approval over the telephone with a followup e-mail.
I put in a lot of their sets.
Check out what accessories are available for the existing breakers. You may be able to retrofit a motor operator to an existing breaker.
You may be able to use a sync check relay to allow the grid to return in parallel with the renewable.
In several jurisdiction that I have worked in the AHJ would inspect the equipment and do a one time approval for that job site.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[crshears]

Bill: "This is an interesting problem with more than one interesting solution."

Agreed!

Keith's recommendation as elaborated on by Bill seems to me to be the most sensible [and elegant] approach, as this seems to be the way to best avoid davidbeach's concern about "wet stacking" and what I've usually referred to as engine fouling.

Lightly loaded generators are soon to be damaged generators.

I would modify this to read "Lightly loaded generators Diesel engines are soon to be damaged generators Diesel engines"; I haven't ever encountered any generators damaged by light loading.

Then again, I haven't seen everything...


CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[itsmoked]

Hi CR; Two days ago I was reading about damaged small gasoline generators that people leave running all night just to run, say, a 50W CPAP machine.

But I certainly take your point that it's always the engine that takes it in the pants not the generator on light loads. (Not related to this thread)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[crshears]

Hi Keith,

Politely, I must disagree; I think it IS related to this thread, as the reason I promoted endorsed your [and Bill's] recommendation in the first place was precisely to ensure the Diesel engine does in fact get up to proper operating temperature whenever it is run [other than for occasional (monthly? quarterly?) start/serviceability confirmation tests, of course].

 

[FreddyNurk]

Depending on your solar system, the OEM may have some product available that might manage the interaction between renewables and generator, I'm aware that SMA has something as I've seen it operating in installations.
The catch is, is that most solar is designed to:[ul]
[li]Shutdown in the absence of a grid as a voltage / frequency reference[/li]
[li]Push as much power back into the network as possible, as its normally connected to a stiff network and reverse power of nearby sources isn't an issue[/li]
[/ul]
There's nothing wrong with using the OEM equipment, the general rule in my experience though is that the client wants their own control of the system rather than a proprietary OEM system.
It might well be that it would take a change of control systems, remove the ATS and replace it with motorised breakers, synch check relays and controllers, and integrate the solar inverter's system for grid regulation and off it goes. Controllers & motorised breakers also gives the advantage of seamless transition to and from the grid (i.e. advanced notice of grid loss doesn't mean site outage). There may be a small issue of getting the utility / AHJ to agree to it.

The payback in terms of diesel fuel saved, given this is a backup source rather than prime power, will likely be very long. If its a backup generator, you want it to be reliable and capable of taking significant load, at least overnight, and running at 10% load infrequently likely won't get the result you want.



EDMS Australia

 

[waross]

I have had two experiences of brand new diesel generators Pumping all of their oil out and shutting down on loss of oil pressure the first time that they were run.
I installed a new 275 KVA set that would run at 10% to 25% loaded most of the time. It needed the full capacity to start one motor. When that motor was starting the set would lean on the motor mounts and blow black smoke.
The vendor insisted that I do a 5 day load test to ensure that the rings were well seated to avoid pumping oil.
I boiled off hundreds of gallons of water in 5 days.
Of the two sets that lost their oil:
Interestingly, one set pumped the oil out and shut down. The building's handy man saw a little red light and a push button beside it and pushed the button. The set restarted and ran until the oil pressure circuit timed out and it stopped again.
As far as I know the set was re-started three times before I was called.
Funny thing, we filled it with oil and it never pumped oil at light loads again.
We got lucky but that is NOT the way I recommend seating in the rings.
Another set shut down once. I was called immediately. I load banked it for a few days to seat the rings and no more trouble.
In older sets with a lot of wear and the tendency of the rings to become weak over time oil pumping may start.
Two issues. The engine may run low on oil and oil in the exhaust is a fire hazard. The engine may eventually start puking out gobs of burning oil from the exhaust stack.


PrimalPete If you post the make and model of the transfer switch and the make and model of your invertor we may be able to pull up some drawings and/or instructions.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

You're correct CR it is this thread too. I was, at the time I wrote that, thinking more about instability bringing down the generator\protection more than the grossly under-loaded engine.

Anywho I think the OPs job can be done but it needs to be engineered not batted around with little to no data.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[FreddyNurk]

Something that may be of interest to people here, This is an example of a controller available on the market, that works with Generator control from the same make in order to regulate renewables. I've not used one, although I have used a number of their generator products over the years, thus I can't comment on how well it integrates with solar inverters from other OEMs. What I can state though, is that the technology and options have come quite a long way in the last 10-15 years, and the concept of a regulated microgrid capable of running both islanded and grid connected is much easier than it used to be.

EDMS Australia