Designing semi-portable power station (inverter/generator)

[artificer]

I'm in the design stage of a source of electrical power for our off-grid farm property.  I've been using a Honda EU2000i for power and to charge some batteries/inverter.  I like the economy of the generator, but I'm about to lose the use of it.  So, its either buy a new generator, or build a DIY version.

My idea for a system is to use a gas (eventually diesel) engine and an alternator to charge a couple of golf cart batteries.  225amp-hrs capacity at 12V.  I picked up a Xantrex 2000w PROwatt SW pure sine wave inverter last year to power an AC unit in a trailer being towed by our Jetta TDI.   I want to use this, connected to the G2 batteries to provide a base level of power.  The engine/alternator will be used to recharge the batteries, as well as provide power for extended high load usages.  I plan on having an autostart on the engine, so the batteries never get drained beyond 50%.  This is a combination of a number of projects that I've seen here, and elsewhere on the net.  Haven't seen all of the parts in one place yet.  Maybe I've missed it...

The system that I'm thinking of so far is:

Engine:  HF 212cc  Yes, Its probably a bad engine, but for $150 it'll get me started until I can get my Chanfa engine working to drive the system.  Cast iron cylinder at least.
Alternator: Leece-Neville 555 alternator  160 amps.  Mobile_bob thinks its good... so thats good enough for me.  :-)
Electric start:  Cub Cadet starter/generator.  The question is...  use a dedicated starter battery?
Multi stage regulator:  Balmar looks good, but its expensive.  How bad is the Sterling Power controller?  The Xantrex controller looks like a re-badged Balmar, but is it?
Control system:  Arduino with whatever drive circuitry needed, but maybe just an off the shelf auto start unit.  If I use the arduino, how hard would adding regulator voltage control to it be?  Probably not worth the extra hassle, and the Balmar units are just so sweet.

The whole thing will be on a two wheel chassis that I can move around, and tow behind the tractor, if needed.  (girlfriend likes her electric chainsaw, rather than my Husky gas one)

This thing is going to eventually cost just as much as the Honda.  Hopefully the engine will run much less, however.  $200 each for Balmar, batteries, alternator... it adds up fast.

Any ideas or comments?  Nows the time to make changes, before I start cutting and welding.

Michael

Forgot to add (since this is the micro-cogen system forum)  That I have the old egr cooler from the TDI.  I was thinking that I could run the exhaust from the power unit through it to heat water.  Its all stainless, so should work.  It has the common problem of the bushing to the butterfly is bad, so it leaks.  I'd take out the butterfly, and plug the hole anyway, so no more problem.  Another source of heat for the planned radiant heat concrete slab in the shop.

[Ronmar]

Well if it is around 5HP, following the 30A/HP rule, that should give you around 150A output to play with.  I think that alternator is rated greater than that peak, which is where the programmable controller comes in to keep the alternator peak current output within the engines power output limits.  IMO, the alternator controller/regulator is the key component to this type system...  I would not skimp in this area, as the longevity of your batteries really depends on proper charge control...

[artificer]

From what I've read here, probably from Bob, the alternator gets derated to 140 amps when hot.  I guess its one reason to go with the Balmar:  amp limiting.  The engine supposedly puts out 6.5hp, so it should be able to fully drive the alternator.

I've been thinking of trying a transformer on the deep well pump.  It should work, if just barely.  If so, then that would be the only load that would draw the full amps from the inverter.  Normally the pump doesn't run for more than a couple of minutes at a time, unless I'm filling the big backup storage tank.

Michael

[mike90045]

For this idea, I'd consider AGM batteries, with a very good charge regulator.   They can absorb much higher charge rates then flooded batteries, and you won't need to run the engine with light load for 90-120 minutes to complete the absorb stage of charge.

[Tom Reed]

If what you mean by cub cadet starter is a starter/generator, they are huge power wasters and run hot. I have one on a S7 Wisconsin engine in a Bolens garden tractor. There is a whole lot more power with out the belt on the starter. Since you want to eventually go diesel why not start with one
http://www.ebay.com/itm/NEW-6HP-SMALL-DIESEL-ENGINE-WITH-ELECTRIC-START-6-HP-/180700361816?pt=LH_DefaultDomain_0&hash=item2a1294b458

With the cost for the starter/generator you'll almost be at the same price anyway.

[vdubnut62]

Ditto on the starter/generator!  They SUCK!  Literally, they will slowly drain the battery. (at least the one I had did) When we had no other option they were Ok, but not with what we have today.
Ron

[artificer]

Starter/generator:  well... that answers that question.  I was wondering if I should just get a starter and ring gear.  Now to find something cheap I can rig up.  If the starter is powerful enough, it will be able to handle the Changfa, as well as the Listeroid eventually.  If I go this route, then do I use a starting battery, or just go with the G2's?

Engine choice:  Its a matter of $150 vs. $560, and I don't have the money for either.  I'd go with a Balmar regulator over the Sterling power before I bought the diesel engine.  I already have a Changfa that I can use, but for a quick build, it takes more to get set up than the gas engine.  I already have the starter/generator, so the price really is $410 in difference for the linked diesel engine, minus the cost to switch over to a new ring gear and starter.  I have a dozen barrels of good used motor oil, so I really want to eventually be running on diesel/wmo.

Michael

(just found a Ford ring gear for $1, shipping included. I'll take a chance on it, and see if it shows up.  Should be farily common, so a used starter will be cheap.  Must have been a mistake, because the other one is now $18)

[BruceM]

I second Mike's suggestion of an AGM battery for this application.  While expensive, they would pay for themselves in fuel savings, as they have MUCH better charge efficiency, at much higher charge currents, and  they also can handle higher discharge currents better as well.

Wet lead batteries are lucky to achieve 60% charge efficiency for the last 10% of the charge, when new.  I have had very good service life experience with the AGM's that I use; one for my home and shop 12V supply, that includes driving the 12V circ pump for the heating system, and one in my alternator-less PV panel charged diesel car.

My main bank is wet batteries as I just couldn't afford AGM's, and because it's 120V, not 12V, charge and discharge currents are very low.

[artificer]

BruceM:  would you suggest getting a separate starting battery, even if I went with AGM's?  I already have (3) 80amp-hr 12v wheelchair batteries connected to a Xantrex DR1012 inverter/charger.  They've been going for several years, which is fairly impressive since they were takeouts, and I bought them for $10/each.  I have no idea of what their condition is, since I just selected the AGM charging setting, and have let them go ever since.  I  fire up the generator to power the saws/drills/mixer, and that's been enough to charge the bank.  I should get a state of charge meter on them.  I like that I never have to equalize them.  A good think, since the charger can't.

Thanks for the comments.  A throttle control is one of the open ended, very many different solution problems.  Electronic, analog/digital, electrical, mechanical, hydraulic...  the choices are too many to list.  One of the sillier ideas I had was a current driven motor connected to a flyweight which actuates the throttle.  Don't know how well it would work on the rolling seas, however.

Michael

[BruceM]

No, you don't need a separate starting battery.  You should never, ever, take the battery down that far. 

Reliability and longevity are improved by a single battery of larger capacity over parallel 12V batteries, but of course a real bargain like what you found can trump that, for folks like us on a tight budget.

I think there are some decent battery state of charge meters out there in the commercial market.  Avoid a unit that has no current in/out sensing capability.  They can't possibly do a decent job with nothing but voltage monitoring, in my opinion, despite marketing claims otherwise. 

[artificer]

I wasn't worried about battery charge when using the starter, since thats the whole purpose of the autostart:  Keep the batteries charged over 50%, and 80% would be better.  The question is about current draw.  If you add a high load to the circuit, the inverter is going to be drawing a lot of power.  The controller senses this, and decides that its better to provide the power from the alternator.  It then starts the engine.  You now have two large draws on the house battery.  Does it make sense to have a separate small starting battery, or am I over-thinking this again?

Michael

[Ronmar]

What would charge the separate start battery?  If the system is going to be 12V, it is probably better to just use that source as start power...  It is certainly simpler.  A charged battery bank is going to be able to provide for surge loads better than the generator will. IMO you would probably be better off configuring thengenerator to meet sustained loads and recharging the bank when low.  To do this you will probably wany some time delay built into the start/shutdown control to keep the generator from starting for every load that pulls the battery voltage below the setpoint for a moment, such as the startup load of an appliance.  So it will take either a sustained load that pulls the voltage below your desired setpoint, or the battery state of charge falling low enough to cause a sustained low voltage to start the engine and recharge the bank or help provide for the load...

[BruceM]

The AGM batteries should be able to handle the start surge; they can handle higher current loads better than wet batteries.  Ron's good advice about delayed start would also help with that- the generator wouldn't try to start at the same time as a surge load on the inverter. 

[squarebob]

Would the start surge in your system be great enough to drop the volts to the point the inverter kicks out on low voltage? This has happened to me using the inverter battery as a starting battery also.

Bob

[artificer]

I'm trying to think of as many things that can go wrong as possible.  One scenario is if I end up getting a transformer to bump the inverters 110V to 220V to power the deep well pump.  The inverter should be able to handle it, depending on transformer losses.  The problem is, the draw of the pump on the generator is just under 2000 watts.  It goes down a bit after the starting surge, but the inverter is still going to be drawing about 140 amps.  The pump should only run for a minute or two, and then shut off.  If it stays on, however, then the engine needs to start, and provide the main source of power.  This is what the Arduino is going to be used for: time delay, control, and sensing of battery state in order to determine when to start the generator.

The original idea was to use the Cub Cadet starter/generator to start the engine, and then charge the starting battery.  Since everyone agrees that the starter/generator is just a fuel hog, and not a good idea, I'm going to go with a Ford truck starter.  Power only used when its operating, no extra drag, except for a bit of wind resistance to the gear plate. 

The plan was to use two CG2 6v batteries for about 225Ah-hrs of capacity.  Cost would be under $200 total.  If I go with a single AGM, as suggested, then I'm looking at close to $600.  How much gas will $400 get me?  This is also incarnation #1, so I'm not sure how well its going to work in the future.  I'll eventually get more house batteries, and probably go to 24V, if not 48V.  How will that effect the system?  (The alternator should be even more efficient...)

I'm still waffling on the AGM vs. golf cart batteries.  Yes, I should just bite the bullet and get AGM's, but its like the question of the Balmar regulator...  my wallets latched onto my feet, and is causing them to drag when I go towards that decision. 

Bob:  The dropping of the inverter during starting is the big question.  I might have to just plan for a space for the start battery, but try without first.  It makes for a simpler/cheaper system if I don't need the start battery.

Thanks everyone for the suggestions.

Michael