Research ideas for 1 MW power station using propane generators and solar.

[Millennium]

Not to poo poo solar; it is completely feasible (you need about a 100,000 square foot clear roof area and 1-2 parking stalls for the equipment, but it isn't the right solution to your actual problem. You should talk with the utility about either an automatic tap changer (per phase) or a similar type of power conditioner to solve the root problem. Running off propane to cover nights isn't viable, although natural gas could be. (Assuming you run about 500kWh 12 hours a day, 5 days a week, 52 weeks a year the propane cost is likely on the order of $780k/year.)

In the old days I would recommend a Hitech Diesel-rotary UPS for this type of problem, although a static double-conversion UPS or an APC SymmetraMW line-interactive UPS might do the job at a lower cost.

Diurnal energy storage if your overnight loads are over ~2MWh is likely not cost effective on a reasonable timeline. I think 1MWh is readily available though which can buffer your excess solar during the day and save a little at night. The system is likely to cost around $2.5-3 million all-in.

Really appreciate the input and recommendations. This type of information is really helpful, thank you.

Not sure how you get to $780K with the propane? Are you assuming that 3x 500 kW generators are running simultaneously?

Can I pick your brain about the diesel-rotary UPS? This equipment sounds intriguing. Why would you not recommend it today?

My understanding is that the static double-conversion UPS and symmetraMW line-interactive UPS are both battery-powered, with limited run times. Unfortunately, my over/under issues sometimes last for hours each day.

My expected budget is in the range you mentioned.

 

[Millennium]

Rotary UPS was mentioned.

Simply having a 3-phase motor powering a 3-phase generator should smooth things out quite a bit. Low voltage it simply draws more current. Although imbalanced voltage on legs may be an issue for motors (I only vaguely understand 3-phase power, have seen issues with transformers.)

Double-conversion UPS is another approach. If you build a photovoltaic/battery system, it could be charged by the grid.

Either way, a large system. 1 MW is about 1500 HP worth of motor. 1000A into a 1000V battery.

Easier to deploy on a building by building basis. Besides MW scale, SMA has 6kW 120V Sunny Island, 18 kW to 100 kW 3-phase battery inverter clusters for the US market. European market they have a 75kW battery inverter. Other brands may have 277/480V suitably sized for your buildings. Sol Ark I think has 35kW.

The SMA systems I like do not condition incoming AC; they either connect to pass-through (and charge batteries), or disconnect. A pair of MW scale AC coupled battery inverters back to back could potentially do this for you, with far smaller battery than if you were trying to make it through the night. Then generator is only for power failures.

On the smaller scale, EG4 "Chargeverter" charges battery from AC input, while letting inverter form island grid.

Again, the recommendations and input from you all is greatly appreciated.

Liked this: "The SMA systems I like do not condition incoming AC; they either connect to pass-through (and charge batteries), or disconnect. A pair of MW scale AC coupled battery inverters back to back could potentially do this for you, with far smaller battery than if you were trying to make it through the night. Then generator is only for power failures."

I am considering options to condition / regulate our grid's power; this solution is definitely on the table.

 

[Shimmy]

Not sure how you get to $780K with the propane? Are you assuming that 3x 500 kW generators are running simultaneously?

Just propane consumption and assuming fairly optimal load. Don't remember what I was using for propane cost though, think it is around $5/gallon.

Can I pick your brain about the diesel-rotary UPS? This equipment sounds intriguing. Why would you not recommend it today?

If you really need the UPS functionality as well as backup power maybe. It is not a great solution for more than ~200 hours on generator a year though. Hitec was the system I used the most, but other manufacturers have products today with the advent of Bitcoin mining and the cloud. The clutch is really critical in the system, so you need to educate yourself on all the little idiosyncrasies of each vendor.

It works, but there are a lot of sensitive mechanical systems that have to work and be maintained perfectly.

My understanding is that the static double-conversion UPS and symmetraMW line-interactive UPS are both battery-powered, with limited run times.

It is a function of how much battery you have and how far off the supply voltage is. One phase dropping 20% doesn't have to put you on battery.

The prime attraction is they are commodity systems that are relatively easy to maintain.

Unfortunately, my over/under issues sometimes last for hours each day.

That is where an automatic transformer tap changer shines, if the utility has the generation capacity but not the distribution capacity.

If the utility has a generation shortfall then you really want to solve it with additional generation. If the issues are daytime then solar is great. 1-2MWh of battery will give you a robust option.

Making the best decision requires talking to the utility.

 

[Millennium]

Just propane consumption and assuming fairly optimal load. Don't remember what I was using for propane cost though, think it is around $5/gallon.

I was told that a 300 kW propane generator burns 9 gallons per hour at full load, so I interpolated that a 500 kW pro. gen. burns about 15 gallons per hour at full load. Assuming we run it 12 hours a day, 365 days a year, and that propane costs about $4/gallon, I got $263K, which is still a lot of money.

That is where an automatic transformer tap changer shines, if the utility has the generation capacity but not the distribution capacity.

I like this option, but unfortunately, I believe our utility has both generation capacity (they buy all their power from others and are still short) and distribution capacity issues.

If the utility has a generation shortfall then you really want to solve it with additional generation. If the issues are daytime then solar is great. 1-2MWh of battery will give you a robust option.

What battery types are available at 1-2 MWh of capacity?
Is there anything available, beside LiFePo (very expensive), on the market that's more economical but has comparable discharge and charge efficiency number?

 

[KauaiMolokai]

I think the SMA recommendation was a really good one. I've submitted an email to their Engineering and Storage Application departments.

I have a good friend who uses "Tesla" everything, but they're so overpriced, and I hate paying 'brand name' premiums.

I'm not looking for cheap equipment or labor, but definitely want someone who's worth their salt.

I would have no issues paying a savvy technical person, who has the experience and knowledge to avoid possible faults, defects, malfunctions.

Yes, I'd be interested in finding that one DIY genius out there who's looking for such a challenge. Any recommendations?

Engineer775. He's on youtube and on the east coast so that might be challenge.. But he knows what he's doing.

 

[AntronX]

That will be a lot of Sol-Arks.

 

[Shimmy]

What battery types are available at 1-2 MWh of capacity?
Is there anything available, beside LiFePo (very expensive), on the market that's more economical but has comparable discharge and charge efficiency number?

A Tesla Megapack will run you about $400/kWh before the IRA incentives and is 3.6-4MWh with inverters integrated in a ~26' shipping container. There are smaller competitors (at least in terms of mindshare) that should be less capital outlay. I would also see if your utility has any grant money available for energy storage. Not to sound like a broken record, but the best time/value you will spend is talking with your utility, the problems you face, and finding ways that you can solve your problems for mutual benefit.

Things like a Bloom Energy fuel cell might make sense for your needs as well (as stupid as I might think they are). Coupling natural gas microturbines with an absorption chiller might also make sense. It all comes down to the detailed needs and challenges. Mixing and matching might give you a great balance as well-- 300kW of PV, 300kW(electric) of microturbine with 100 Tons of absorption chillers, and a 300kW fuel cell as an example.

Whatever you decide, please do report back here though!

 

[Millennium]

Not to sound like a broken record, but the best time/value you will spend is talking with your utility, the problems you face, and finding ways that you can solve your problems for mutual benefit.

Whatever you decide, please do report back here though!

I have, multiple times, over the years. The guy I had the most contact with is no longer working for them.
He was telling me about all their problems; he basically verified our suspicions about their issues with distribution and generation.

I have sent multiple emails, to multiple individuals, requesting an update on a service upgrade, and no one has responded.
I'm frustrated. I'd say more, but I'd prefer not too, so I don't come off as bitter.

Just know that they are not regulated, like other utilities, so they do as they please.

I'll let you know what direction we decide; lots of options on the table right now.

 

[Supervstech]

I have a buddy that designs and implements ups systems for the major internet systems, and I can pass on your generator requests… send me message with your digits.

 

[AntronX]

The location in question is serviced by a poorly run electric company and we've found that the legs are very often out of balance, especially during the summer months, which is destroying the life of our motors and electrical equipment.

How much are you paying per kWh ? How many motors and power rating of each? It may be cheaper to buy VFD for each motor so phase voltage misbalance will no longer be an issue.

 

[Shimmy]

I have, multiple times, over the years. ...

See if you can find an account representative or senior planning engineer, FWIW. Indicate that power quality is bad enough that you are prepared to move off grid if your system reliability cannot be improved. If it doesn't work... time to get off grid.

 

[DIYrich]

Sol-ark makes a 60k 3-phase unit. You can parallel up to 12 for 720kW. Maybe make two 540kW plants and divide your site.

Don't know how to make 800v battery for it. If you can, charge the batteries from the grid, and it gives you a double conversion UPS to clean up grid power, and reduce cost of generators to stand-by.

You need enough batteries to store the solar produced during the day for use during the other 24 hours. Commercial net metering is probably much worse than residential. You can do a cost/benefit of storing solar vs running off grid (via double conversion) to optimize battery sizing. You can always add more bateries.batteries. BTW, you are looking at 1,200 amps at 800v to produce 1mW. I think you are looking at 1,200 LF300 cells per hour of storage. About $180,000

 

[OffGridForGood]

ESS for a 1MW system!...need a couple of lakes close together with a thousand feet of elevation difference between them.
Solar pumps water to upper lake, hydro turbine(s) create the power needed.
AZ - maybe water is not an ideal solution.

 

[curiouscarbon]

AZ - maybe water is not an ideal solution.

rocks on rail carts?

with obligatory mention of gravity batteries being not optimal in every situation

 

[Skypower]

I was told that a 300 kW propane generator burns 9 gallons per hour at full load, so I interpolated that a 500 kW pro. gen. burns about 15 gallons per hour at full load. Assuming we run it 12 hours a day, 365 days a year, and that propane costs about $4/gallon, I got $263K, which is still a lot of money.


I like this option, but unfortunately, I believe our utility has both generation capacity (they buy all their power from others and are still short) and distribution capacity issues.



What battery types are available at 1-2 MWh of capacity?
Is there anything available, beside LiFePo (very expensive), on the market that's more economical but has comparable discharge and charge efficiency number?

That power company seems flaky and some may be leaving a sinking ship. There’s the possibility that the company model is to take the income and distribute it as wages without any consideration for maintenance or upgrades. When the system totally collapses, the phone won’t answer and you’ll find the office door ajar with not a soul in sight, corporation records gone. Ran into the ground. Personally I’d take care of yourself with an off grid solution because if I’m right it’ll be a long time till your community has reliable power again.

 

[Millennium]

Sol-ark makes a 60k 3-phase unit. You can parallel up to 12 for 720kW. Maybe make two 540kW plants and divide your site.

Don't know how to make 800v battery for it. If you can, charge the batteries from the grid, and it gives you a double conversion UPS to clean up grid power, and reduce cost of generators to stand-by.

You need enough batteries to store the solar produced during the day for use during the other 24 hours. Commercial net metering is probably much worse than residential. You can do a cost/benefit of storing solar vs running off grid (via double conversion) to optimize battery sizing. You can always add more bateries.batteries. BTW, you are looking at 1,200 amps at 800v to produce 1mW. I think you are looking at 1,200 LF300 cells per hour of storage. About $180,000

$180K for one MWh off batteries is much lower than my initial estimate but getting 12 hours of battery power would require $2.2MM (based on your numbers). It might work. Does your battery cost estimate include all the necessary accessories, besides inverters and solar panels?

I can't wait till a cheaper solution replaces LiFePo.

 

[Millennium]

ESS for a 1MW system!...need a couple of lakes close together with a thousand feet of elevation difference between them.
Solar pumps water to upper lake, hydro turbine(s) create the power needed.
AZ - maybe water is not an ideal solution.

I have thought of this. I know a very good working system, but I'd need a mountain to store water at elevation.

 

[republicmachine]

About 10 years ago I was in Detroit and we took a ride out to the old Packard plant that closed in the late 1950's. In one of the buildings, it was called the powerhouse and a guy we were with explained that some old large factories had their own power stations that made their own electricity. I was immediately hooked on the idea of being independent from the man. Years later I started playing with solar but between then I have been looking at steam, turbines, gasifiers, biogas, incinerators, and other ways to make power. Asynchronies induction motors and all that. If I owned a large company with a 30k per month electric bill, I'd be building my own electric system. Small towns still have power statins many powered by GM train diesel engines run on Natural Gas. After years of trying to understand how to make factory/small town sized power, it seems that having multiple ways to generate electricity is the way.

So I'd start with what is the most reilable and cost effective. Probably a steam turbine and new certified boiler run off natural gas and or trash incinerator, I'd piggy back a huge diesel genset unit and run it on some type of bio or recycled oil which I'd process in house gaining partners at other businesses to collect said oil. I'd then have wind and solar on the roof pumping into my own grid. You'd have to employ a team of people to make this happen and get engineering in house, a team of legal to make sure you were epa and maybe get some grant money and subsities to pay for it then you could sell a bit of this power back to the surrounding company and neighbors. Tie it all in. Like establishing your own little civilization. Start watching youtube videos on old small town and factory powerstations and modern steam units. Make it full carbon neutral. It'd be very cool. You could even turn your land into fields and process grain into fuel. So many directions. Start with a 500k budget to get reliable boiler units and power now.

Side note. I love solar and wind. It's certainly easy for a small percentage of homes and small businesses to gain independence but the problem is that it simply doesn't scale up well. The cost and after life problem far outweighs the current system we are using. What do to with 100 acres of pannels after they're cooked. Well you've got a big problem. And how reliable and trouble free will it be over its life. Compared to other more heritage options. Steam runs somethings for a 100 years continuous with nothing more than pig fat smothered on it every week.

 

[SajgZi]

That power company seems flaky and some may be leaving a sinking ship. There’s the possibility that the company model is to take the income and distribute it as wages without any consideration for maintenance or upgrades. When the system totally collapses, the phone won’t answer and you’ll find the office door ajar with not a soul in sight, corporation records gone. Ran into the ground. Personally I’d take care of yourself with an off grid solution because if I’m right it’ll be a long time till your community has reliable power again.

Who is John Galt?

 

[SajgZi]

Otherwise very good conversation.

Also look up the big natural gas generators. If you are lucky enough to be near natural gas lines (they don't even need to be super big), you can just feed them natural gas and they crank out power. Lots of big units used in the oilfield to power electronic submersible pumps (ESPs). Company I work for has ~150 ESPs in the Permian basin sucking down 400-500 kW each. Many ran off generators. Some natural gas (burning natural gas that would otherwise be flared off), some diesel. They're designed for high reliability.

Let's run through the math for a Cat CG127-12 TA engine (used for electric generation and gas compression).

400 kW at 480V is 100% capacity so you'd need 2-3.

Fuel consumption at full load = 4906 SCF per hour (this shows at 905 BTU/cf, which is a bit below what usually comes out of pipelines). Going to use 1000 BTU/cf, which is a bit more realistic and easy for math purposes, so that's 4906/(1000/905) = 4439 SCF per hour x 24 = 106558 SCF per day. That's 107 Mcf (oil/gas uses M for thousand) per day. Current henry hub spot price is $3.07 per MMBTU. Since we used 1000 BTU per CF, that convienently results in the same price per MMBTU as MCF. So you're using 107 MMBTU * 3.07 $/BTU = $328.49 per day of gas. You're also generating 400 kW x 24 hours = 9600 kWh. That's a cost of $0.034 per kWh (3.4 cents per kWh).

You won't get spot pricing, but you'll get close to it at 100-300 mcf/day. Amortize the generator cost over X years and you're good to go. These generators are also commonly available for rental/lease if you want to test things out.

Also the efficiency goes up slightly as you back off from full load. If you run 3 of these at 83% each (3x333 kW = 1 MW), you'd be near peak efficiency.

Here's a list of CAT engines in this product line - https://www.cat.com/en_US/products/new/power-systems/oil-and-gas/land-production-generator-sets.html

Edit: there is a lot of interest in this area to burn natural gas to generate electricity in the oilfield to power bitcoin miners. The gas would otherwise be flared. Up to you to decide if that's a valuable venture or not (hint: the mined bitcoin has 2-3x the value of the generated electricity).