How efficient is your generator?

[Symbioquine]

I wanted to start this thread so folks have a place to compare real-world numbers about their generator efficiency.

This data is from two generators which I run exclusively on propane.

Up to 2022-11-10: Champion 100263 (Nominally a 3400 watt generator)
After 2022-11-10: DuroMAX XP9000iH (Nominally a 9000 watt generator)



So the interesting potentially comparable numbers are:

$1.14 USD / kWh
~4.2 kWh / LPG Gallon

4.2 (kWh generated) / 27.988 (kWh total energy) = 15% efficiency

What sort of efficiency numbers are other folks getting?

Note #1: I've read https://diysolarforum.com/threads/natural-gas-generator-efficiency.49756 before, but as far as I recall it only includes theoretically numbers and guesses.

Note #2: Please feel free to share numbers regardless of fuel type. I think the comparison between fuel types may be interesting as well.

Note #3: I'll try to keep my data updated by updating the image above at least through the rest of this fall/winter season.

 

[Lt.Dan]

This is a cool topic. I have a smaller Ryobi 2300w with electric start. Ill see if I can dig up the stats I had on it. It was not near as in depth as yours, but I did have average run time at full load and load tests.

 

[Lt.Dan]

For Ryobi 2300w (1800w continuous), It will run at 75% output (1325~ watts) for 6 hours on 1.2 gallons.

That equates out to 6625w per gal. or $1.38 US / kWh (i think, if my math is right)

 

[Symbioquine]

For Ryobi 2300w (1800w continuous), It will run at 75% output (1325~ watts) for 6 hours on 1.2 gallons.

That equates out to 6625w per gal. or $1.38 US / kWh (i think, if my math is right)

Awesome! Maybe helpful to state the gas price you're working with. I found my most recent receipt for non-ethanol gas from a nearby marina. (Used for other small engines around our farm.) I paid $6.42/gal on 2022/10/25 which would yield a price of $0.97/kWh using your data.

A gallon of gasoline has roughly 33.41 kWh of total energy, so;

6.625 (kWh generated) / 33.41 (kWh total energy) = 19.8% efficiency

 

[Vigo]

That sounds plausible since gas engines tend to be 30% efficient with the best barely at 40, and a bottom of the barrel air cooled engine with essentially no intake or exhaust manifold and fueled by a carburetor, plus conversion losses to electric.. yeah 19-20% sounds about right. ??

 

[wattmatters]

From wikipedia, energy density numbers based on chemical oxidation:

	Specific Energy MJ/kg	Energy Density MJ/L	Specific Energy W·h/kg	Energy Density W·h/L
Natural gas	53.6[22]	0.0364	14,888.9	10.1
LPG propane	49.6	25.3	13,777.8	7,027.8
LPG butane	49.1	27.7	13,638.9	7,694.5
Gasoline (petrol)	46.4	34.2	12,888.9	9,500.0

Costs for fossil fuels vary greatly from region to region, and day to day.

Above are using SI units, so if you want to use gallons or pounds then do the conversion.
Gallons always confuse me because there are several different gallon measurements.

 

[wattmatters]

I can't really contribute my own example as I just don't use my generator all that much.

In term of energy supply hierarchy, it's:
i. Grid-tied solar + Off-grid solar & LiFePO₄ battery
ii. Grid
iii. Off-grid solar & Lead acid battery
iv. Generator

iii. and iv. are my backup.

 

[wattmatters]

Rows 2 - 19: Champion 100263 (Nominally a 3400 watt generator)

3.327 kW·h / gal

3.78541 litres / US gallon

So that's 0.8788 kW·h / litre

Here a litre of petrol is ~A$2, so that's ~ A$1.76 / kW·h.

Yikes.

 

[Symbioquine]

Gallons always confuse me because there are several different gallon measurements.

3.327 kW·h / gal

3.78541 litres / US gallon

So that's 0.8788 kW·h / litre

Here a litre of petrol is ~A$2, so that's ~ A$1.76 / kW·h.

@wattmatters My kWh / gal numbers are per US gallon of compressed propane (LPG - liquid propane gas). I suspect it's not going to be accurate to directly convert from volume of propane to volume of gasoline (petrol) as it appears you did above.

Maybe somebody else here has a Champion 100263 who runs it on gasoline who can provide their numbers...

Alternatively, taking your fuel price numbers we can compute the cost from @Lt.Dan's numbers;

A$2 (A$ / liter) / (6.625 (kWh / gal) / 3.78541 (liters / gal) ) = A$1.142 A$ / kWh

 

[Symbioquine]

I suspect it's not going to be accurate to directly convert from volume of propane to volume of gasoline (petrol) as it appears you did above.

Both on account of their differing energy densities per volume and possibilities of differing engine behaviour with each.

 

[wattmatters]

My kWh / gal numbers are per US gallon of compressed propane (LPG - liquid propane gas)

Ah, my bad I thought you had one gasoline, one propane. Don't know why I thought that given you clearly stated both were propane.

 

[OffGridInTheCity]

Wow - I've been using 4kwh/gal of propane for my prepping basis. At 3.276kwh/gal to charge the batteries - 15% loss inverter -> AC that's 2.7856kwh/gal of useable power out of my solar system with propane driven generator power. Yikes! (much worse than I was thinking)

On a semi-related note, I put in a whole house (top of line SEER) heat-pump a couple of years ago and have been tracking electric (heat-pump) vs natural gas (flame) for heat in Zone 4 climate (mild). My electric is ~11c/kwh and at 29kwh/therm the my gas is ~5c/kwh. Assuming the 29kwh/therm - the gas is taking twice the kwh for the same amount of heat as heat-pump but since it's 1/2 the price of electricity they are similar in home-owner cost.

Recently discovered kerosene is safe to store long term etc + kerosene heaters are simple and safe to run indoors. Looks like 1 kersone = 1.348 therms = ~39kwh. So I'm heading toward thinking kerosene (or propane) flame-heat circulated by furnace fan as more practical than generator -> electric heat-pump for dead winter when heat-pump is not so efficient. Let's say
- 2000kwh of gas (flame) heat / 39kwh per gallon of kerosene = 51.3gal of kerosene for a month of heat.
vs.
- 1000kwh of electric heat-pump heat / 2.8gal propane(generator/inverter) = 357.1gal of propane for a month of heat implies a 51/357 or 1/7 ratio. It's not quite 1/7 because the furnace fan needs electricity to spread heat in the house so for this napkin discussion let's go with...

1 gal of kerosene (flame heat) vs 5 gal of propane (electric heat-pump heat)?! Looks like kerosene or propane flame is a no-brainer for heat compared to generator / electric heat-pump as a backup against lack of solar. **My reality is Zone 4 climate with many days of 35F lows which makes heat-pump less efficient then 45F lows as an example - e.g. every DIY situation is different.

 

[Symbioquine]

Wow - I've been using 4kwh/gal of propane for my prepping basis. At 3.276kwh/gal to charge the batteries - 15% loss inverter -> AC that's 2.7856kwh/gal of useable power out of my solar system with propane driven generator power. Yikes! (much worse than I was thinking)

You should measure with your specific generator and report the numbers back here. So far we only have two datapoints - from Lt.Dan and myself. (Maybe 2.5 since there's a little bit of data from my second generator included.)

Looks like 1 kersone = 1.348 therms = ~39kwh. So I'm heading toward thinking kerosene (or propane) flame-heat circulated by furnace fan as more practical than generator -> electric heat-pump for dead winter when heat-pump is not so efficient.

Bit off topic here, but you also have to consider how efficient your kerosene heating is. If you're using it in some sort of fixed furnace that's externally vented, you'd want to see how warm the gasses exiting the vent are. Otherwise, if you're using a portable one, there's some efficiency loss associated with needing to provide extra ventilation which usually also lets in cold air. Semi-related, here's a pretty good write-up about the portable types: https://www.sevarg.net/2022/11/06/keropunk-part-3-kerosene-heaters/

 

[Checkthisout]

I don't keep numbers that good.

But,

Yamaha 3000 ISEB $0.86 per Kwh @ $5.64 a gallon for Ethanol free.

 

[Symbioquine]

Yamaha 3000 ISEB $0.86 per Kwh @ $5.64 a gallon for Ethanol free.

Wow, that's awesome! Seems like we might be starting to see a trend in terms of running gas being cheaper per kWh.

Converting that into a comparable numbers:

$5.64 ($ / gal) / $0.96 ( $ / kWh) = 6.558 kWh / gal
6.558 (kWh generated) / 33.41 (kWh total energy) = 19.6% efficiency

 

[Checkthisout]

Wow, that's awesome! Seems like we might be starting to see a trend in terms of running gas being cheaper per kWh.

I think it might actually be better than this. I'll take some better measurements but not for at least a couple months.

 

[wattmatters]

How are each of you measuring:
- the volume of fuel actually used?
- energy generated?

What load level are you using for the generator?
This will have an impact on efficiency.

My generator (Yamaha EF300ISE) tank has a rated capacity of 13 litres. I don't think I want to waste 13 litres of fuel on a test.

I guess so as not to waste too much of the energy I could pick a crappy solar day with battery at low state of charge, switch off the PV array and charge the battery at a fixed current level from the generator. With the charge current set based on a suitable load level. 40 A would be approximately a 75% load level for my generator.

I would need to run the generator dry, and then use a fixed well measured amount of fuel. I can't fill the tank as the generator on a full tank will supply more energy than my battery can accept.

The specs claim on a full tank the unit will last 20.5 hours at quarter load, and 8 hours at rated load.

Manufacturer Specifications
Yamaha EF3000ISE

Rated load (continuous)   2800 kVA

13 l of fuel @ 9.5 kWh/l = 123.50 kWh

2800 W x 8.0 hours =        22.40 kWh
700 W x 20.5 hours =        14.35 kWh


Efficiency @ rated load:    18.1%
Efficiency @ 1/4 load:      11.6%

Real life? No idea.

 

[Checkthisout]

How are each of you measuring:
- the volume of fuel actually used?
- energy generated?

What load level are you using for the generator?
This will have an impact on efficiency.

My generator (Yamaha EF300ISE) tank has a rated capacity of 13 litres. I don't think I want to waste 13 litres of fuel on a test.

I guess so as not to waste too much of the energy I could pick a crappy solar day with battery at low state of charge, switch off the PV array and charge the battery at a fixed current level from the generator. With the charge current set based on a suitable load level. 40 A would be approximately a 75% load level for my generator.

I would need to run the generator dry, and then use a fixed well measured amount of fuel. I can't fill the tank as the generator on a full tank will supply more energy than my battery can accept.

The specs claim on a full tank the unit will last 20.5 hours at quarter load, and 8 hours at rated load.

Manufacturer Specifications
Yamaha EF3000ISE

Rated load (continuous)   2800 kVA

13 l of fuel @ 9.5 kWh/l = 123.50 kWh

2800 W x 8.0 hours =        22.40 kWh
700 W x 20.5 hours =        14.35 kWh


Efficiency @ rated load:    18.1%
Efficiency @ 1/4 load:      11.6%

Real life? No idea.

It's hard to measure exact because the loads fluctuate.

Top your tank off to the Itty bitty top and then run your normal routine for a couple hours and then measure the quantity of fuel it takes to get back to Itty bitty top.

It won't be much fuel so just pour the gas out of your normal jug into a graduated container of some sort and just measure what you put back in the Gen tank.

That's not exact but its good enough.

 

[Symbioquine]

How are each of you measuring:
- the volume of fuel actually used?
- energy generated?

My use case is pretty similar to what you described @wattmatters. I'm using the generators to charge my batteries when there is insufficient solar to keep up with the essential loads we've got.

The generator load that I've configured through my Victron Multiplus is 1800 - 1900 watts which is right around my guess at the peak efficiency point of 70% for the Champion 100263. (Of the 2700 watts the generator could produce on propane.) I'm hoping to get the second multiplus hooked up soon for split phase which will allow a charge rate closer to the peak efficiency for the larger DuroMax generator too.

I have data from the multiplus recorded in InfluxDB and I'm using the 1 minute average datapoints to take the sum of the watts going into my batteries plus the VA going from the generator to serve my AC loads. (Happy to share the hacky script that does that if anybody is interested.)

The volume of fuel used is pretty easy since I have dedicated propane tanks for the generator so each time I get a tank filled, the volume I'm charged for is the amount used since the last fill. (I get the tanks filled at the same place and they seem to be pretty consistent between fills. (Judging by the final weight of the tanks.) There's about a 1% margin of error there though since they only charge to the nearest tenth of a gallon and I'm using 9.2 gallon tanks.

 

[wattmatters]

The generator load that I've configured through my Victron Multiplus is 1800 - 1900 watts which is right around my guess at the peak efficiency point of 70% for the Champion 100263

So about 70% load is the sweet spot?

The volume of fuel used is pretty easy since I have dedicated propane tanks for the generator so each time I get a tank filled, the volume I'm charged for is the amount used since the last fill.

I guess you could weigh the tank yourself.

 

[Symbioquine]

So about 70% load is the sweet spot?

That's my guess, but I'd love to have more data to support/disprove that!

I guess you could weigh the tank yourself.

Unfortunately, I don't have a scale in the 100 lbs range than is going to be more accurate than the volume measurement from the propane seller.

 

[Checkthisout]

Inverter generators, if they are well designed, should always operate near 100% engine load. I.E. the throttle plate should be open fairly far for the given RPM.

So, this differs from a non inverter which runs at a fixed RPM and then varies throttle plate opening based on load.

Best overall efficiency will be produced at the torque peak so maybe look up similar engines and their power curves and see what RPM is commanded at that given electrical load for your particular gen.

 

[crazydane]

I have a MEP-802A diesel generator. It burns 0.5 gallons of fuel per hour with a 5,000 watt load.

I also have a Kubota 17 kW generator. It has a 4 cylinder turbo charged diesel. Next time I load test it, I'll see about collecting some fuel consumption data.

 

[schmism]

My 1800w inverter genny runs about a quart an hr on 500-800w.

 

[Symbioquine]

I have a MEP-802A diesel generator. It burns 0.5 gallons of fuel per hour with a 5,000 watt load.

Wow, that's awesome! Seems like you're doing a bit better than spec which seem to be "0.57 gph @ rated load" for that generator.

Converting that to comparable numbers;

5 (kWh / hour) / 0.5 (gal / hour) = 10 kWh / gal
$5.718 (USD / gal) / 10 (kWh / gal) = $0.57 (USD / kWh)
10 (kWh generated) / 40.7 (kWh total energy) = 24.6% efficiency