Petrol battery charger

[zoltan]

I thought some on here might see the value of a petrol powered battery charger. Most useful for overlanders who want to be parked up for a while and don't want to use up fuel running the vehicle for hours a day to recharge house batteries.

I saw this in a yachting forum initially and decided rather that spend out on a suitcase genny to make 240v then convert back to 12 volts that I would replicate what the other guy did using a small light fuel efficient Honda motor to keep fuel consumption and noise to a minimum.

Still got a way to go in the control stakes as a pair of run down 110ah batteries pull too much voltage for the little 22cc motor but with some field voltage control added I should be able to remedy that. The guy on the yachting forum used a PIC microcontroller and servos to achieve a good working setup but I'm no-where near that at the moment. I was lucky to find two race car brake disc bells at work to make the joining of the alternator and the motor simple. I wanted it direct coupling as that shape fits the side lockers best on my truck.

I opted for a Lucas ACR alternator with a 45 amp output as I have a stack of these to plunder for parts when I kill them in testing. Seddons had the GX22 engines on offer for £115 incl vat.

I started with the two brake bells, faced off the alternator casing in the lathe to get a nice flat surface for mounting the bell, added some cooling holes

The fan needed to be turned down to fit inside

The motor has a centrifugal clutch and it seemed like too much bother to remove all of this so for now I have kept it so I made a clutch drum in steel

On the other (engine) side a different bell with 4x 6mm holes added to the pitch of the engine mount. Both bells have the same hole PCD where the disc rotor would have mounted. In theory that should keep things concentric and get my alternator running true

Voila! The two joined together waiting for a mounting cradle and some sort of controls. The Honda GX22 is a 1hp engine and uses a pretty minimal amount of fuel per hour.

I'm not happy with my budget ammeter and voltmeter so I will try and sort some better digital ones. I was in a rush to get something knocked together as I took my truck off on a ski trip. Now I have some more time I can concentrate on a better control arrangement as well as a handle. The skid frame is on some little cotton reel rubbers so it doesn't bounce all over the place

It works well on a single battery but with no field current control, a pair of run down 110ah batteries just stall the engine as the alternator tries to knock out a large dollop of current. I'm learning as I go with this, electrical things aren't my forte.

I think losing the centrifugal clutch is required now as I'm sure I'm getting excess and unnecessary noise from this area and one goal is a quiet unit. Eventually it will be housed in an insulated ammo box type affair assuming I can get cooling sorted.

Still a work in progress but I thought I ought to post it up. Of most interest to the camping/overland fraternity but useful if you have a dead vehicle miles from a 240 supply and you don't want a hernia from lugging a genny out there

[minimaquinas]

hi zoltan

that looks a great peice of kit that you are building there

[Stevie D]

I'll watch with interest - I've a much bigger ex mod 24v version with a 5.5hp 4 stroke engine, but it's too big for proper portable use - think full size 240v genny - i us it for my off grid log cabin to supplement wind power. i'd passed up on a much smaller 12v kubota version . It has a manual control circuit rather than an auto control circuit - being mod it has the circuit engraved onto the plate - i can try to dig it out and get a photo if you're interested.

Mine will also struggle to charge a pair of 12v 110ah batteries that are fully discharged.

Steve

[FridgeFreezer]

A very lazy way of limiting the load would be to put a headlamp bulb (or some other dumb load) in series with the battery, that would knock the charging current right down. When the bulb gets dim, you know the battery's approaching charged.

Like I said, very lazy and a bit wasteful, but super easy.

[zoltan]

Would that not be just like adding to the current load though like having batteries that are even deeper discharged?

I've kind of made my own problem by using an engine which isn't powerful enough but somehow by controlling the field current with the rotor I should be able to limit the output voltage. If I add a variable resistor on the voltage feed to the regulator will that limit the strength of the coil in the rotor and hence the output from the alternator?

[Bowie69]

Would that not be just like adding to the current load though like having batteries that are even deeper discharged?

No, as you put it in series, not parallel, with the battery, it acts as a resistance limiting the current to ~5 amps, for a simple solution it's actually very effective, not sure it would glow much though, even with a mildly discharged battery, half the voltage = quarter the light output.

[FridgeFreezer]

In the old days it's how they synchronised generators - put a bulb between the mains & the generator, when all 3 lights go out you're in sync and throw the switch. Or, wait till all 3 are on and throw it when the apprentice is stood next to the genset and it tries to jump off the platform

[zoltan]

Thanks, I'll have a go and see how it works out

[TheRecklessEngineer]

In the old days it's how they synchronised generators - put a bulb between the mains & the generator, when all 3 lights go out you're in sync and throw the switch. Or, wait till all 3 are on and throw it when the apprentice is stood next to the genset and it tries to jump off the platform

Some of us still do this with the bulbs. Standing next to the light switch and switching the lights off when the breaker goes in provides some amusement.

Very nice looking build there Zoltan. I'm not sure about sticking a resistor in series with the field winding - I think the regulator will try and fight it. I reckon a resistor in parallel with the field winding will probably have the desired effect - full load is determined by a maximum current value through the field winding - if some of it is being dissipated in a resistor then the field current will drop while full current appears at the regulator.

[TheRecklessEngineer]

Actually, some back of envelope maths would suggest that your engine is probably just about enough to drive the alternator at full current.

I'd be looking at tweaking the engine - perhaps playing with the throttle? I've made some quite substantial differences to lawnmowers in the past by moving arms so the that throttle picks up more with only a minimal drop in rpm. A little unstable at idle, but much better under load.

[zoltan]

Actually, some back of envelope maths would suggest that your engine is probably just about enough to drive the alternator at full current.

It just about manages one 110ah battery but really struggles with two. I get a cycle of the alternator trying to push out about 20 or 30 amps at which point the engine almost stalls, the revs fall such that the centrifugal clutch drops out and the charge light goes out you have to give it a blip on the throttle to get the revs up then the charge light is back on.....then 20-30 amps.....

I think if the resistor could calm this down it would make a huge difference for that first five or ten minutes of running

The yacht forum guy used a PIC micro controller and a servo but that it way out of my area of knowledge. Certainly something I'd like to look at after a while.

I probably need to take the centrifugal clutch out and direct couple with a rubber donut style of coupling. Probably make it quieter

[zoltan]

Hard to tell how noisy it is, this is inside the shed

[Team Idris]

I've been thinking about it, but with a belt. Direct drive never crossed my mind, as I was thinking the speed increase with the pulleys was vital for cooling. But now I'm thinking dynamo, as the current and voltage are easy to tweek in the regulator box.

Great build. Nice bit of machining

[TheRecklessEngineer]

Think I've linked to this before, but thought I would again as it seems appropriate:

http://hackaday.com/2011/09/08/diy-portable-chargergenerator/

[SORNagain]

Lovely bit of engineering! I've been thinking along those lines too, but nothing so neat... I had the same problems with an alternator + lawnmower engine.

Seems to me that although the engine could drive the alternator at full chat, because of the torque response created by the regulator this is not going to be possible.

As you load up the alternator to the point where the engine begins to slow down, the alternator demands more torque in order to keep the output power constant. This of course slows the engine even more - and this continues to the point the clutch disengages or it stalls out.

It would be better to keep the armature current fairly constant and to regulate the output voltage by controlling the throttle. That way the engine revs could be proportional to the power demand, saving noise, fuel & engine wear. I say 'fairly constant' because I'm not sure whether a constant armature current would allow you to alter the revs enough while achieving the required output voltage. Definately microcontroller territory here - lots of experimentation required!

Not sure there is an easy solution to this one

[FridgeFreezer]

Do these motors not have governors?

Have I missed something that means you've got to do different than every other engine-driven generator and throw a microcontroller at the problem? if it's just a question of calming down the spikes in the load that's surely solvable with a small analogue tweak to the regulator circuit to stop it from jumping too fast.

[zoltan]

These little engines are used for water pumps and strimmers so no governor.

I'd really like to avoid micro controllers and servos if at all possible. How would you go about slowing the regulator down? Presumably the internal circuits vary from alternator to alternator according to the way they've been configured for the end users or are these very simple old style regulators??

[SORNagain]

Most engine driven generators don't have a regulator in the alternator. The regulator causes more problems than it solves because it attempts to keep the output power constant. Any drop in speed means the alternator demands more torque, which is probably the reverse of the characteristics of the engine.

I have had an idea though. What the wind turbine guys do is insert permanent magnets into the armature to make a permanent magnet generator. Then you just control the throttle (possibly manually) to control charging current. I think they glue neodinium disc magnets in between the zig-zag bits in the armature, so its not difficult to do. This is the same as driving the armature with a constant current, just more efficient because that current is not being wasted.

Just don't run it without the battery - it will probably kill the rectifier!

[TheRecklessEngineer]

Just wanted to confirm something here, because I think I've made an assumption which is confusing me slightly.

The engine is bogging down with a high load on the alternator and the throttle wide open?

[zoltan]

@SORN The regulator probably needs to stay as it provides the possibility to achieve a two or three stage mode of battery charging. A generator would normally connect to a battery charger which will fulfil the control of the battery charging. That was what I was hoping to avoid by tinkering with the alternator through it's regulator circuit. My charger needs to work with the leisure battery bank so a float charge ability is important

@Reckless, The bogging down does occur at wide open throttle. Basically I don't think in reality this engine is man enough for a 45amp output alternator but I'd hoped to control this output so that the engine could work within its measly power range. Since it needs to be quiet and fuel efficient to run whilst providing the float charge I didn't want a bigger engine both because of it's fuel consumption and for it's physical size

The engine should have enough beans in theory at a steady state, if my maths are right, 30 amps at 14.2 volts is 426w or .42kw, at 45 amps =639w so my 0.75 kw engine should be able to drive the alternator

[Bowie69]

I think I have this bit right...

Alternator efficiency is only around 50-60%, this is possibly where your calculations are amiss, so taking your figures, and the fact that a 45A LR alternator is unlikely to be state of the art, say 50% efficient, it means you are overloading the engine by 41% at 45A, and by 13% or so at 30A.

[Deej]

I think I have this bit right...

Alternator efficiency is only around 50-60%, this is possibly where your calculations are amiss, so taking your figures, and the fact that a 45A LR alternator is unlikely to be state of the art, say 50% efficient, it means you are overloading the engine by 41% at 45A, and by 13% or so at 30A.

You beat me to it. This is why commercial generators tend to have the engine rated for considerably more than the output power of the generator.

For example: http://www.honda.co.uk/energy/generators/EC2000/specifications/ This genny has a rated output power of 1700W but the engine is 5.5hp (4100W)

[Bowie69]

Phew, I don't look like a moron again

[steve b]

Excellent bit of engineering Mr Z

As an apprentice many moons ago I designed and made a load test bed for a 1.5 hp governed 4 stroke engine using a 16ACR Lucas alt , belt driven at around 2:1

alt to engine

The load was banks of 5 x 21W bulbs switched by a mech timer switch with all 6 banks on the engine would stall out within 2 min

...How about a smaller alternator to reduce the max load? there are some nice compact 12v alts on the 2 & 3 cyl Kubota plant engines as used in chiller lorrys/mini diggers/wood

chippers etc and as with larger alts self contained rect/regs

cheers

Steveb

[Deej]

Excellent bit of engineering Mr Z

As an apprentice many moons ago I designed and made a load test bed for a 1.5 hp governed 4 stroke engine using a 16ACR Lucas alt , belt driven at around 2:1

alt to engine

The load was banks of 5 x 21W bulbs switched by a mech timer switch with all 6 banks on the engine would stall out within 2 min

...How about a smaller alternator to reduce the max load? there are some nice compact 12v alts on the 2 & 3 cyl Kubota plant engines as used in chiller lorrys/mini diggers/wood

chippers etc and as with larger alts self contained rect/regs

cheers

Steveb

If you do that the engine shouldn't stall but you may just end up killing the alternator instead. If the smaller alternator only has output voltage regulation and not current (i.e. a bog standard alternator), it will attempt to supply whatever current is necessary to maintain 14.4V at the output. Two dead 110Ah batteries may present too big a load and will cause the windings/regulator to overheat.

Output current control would be the ideal solution but a lightbulb in series is nice and simple if a little inefficient.