19 replies to this topic

[Chris Wilson]

Is anyone aware of any tech info on how fast an alternator needs to be driven to give a worthwhile charge rate? I am looking at driving a small modern, lightweight alternator off a Hewland FTR inner tripod joint, by turning and welding on a bell shaped toothed wheel. I see some photos of Group C cars using such a system, but it's far from clear what sort of pulley sizes are involved. Thanks.

[murpia]

Well, modern road car alternators charge OK at 750rpm engine x whatever pulley ratio is fitted.

Average rear axle rpm's are probably about 1000, so maybe gear up slightly from that.

Regards, Ian

[McGuire]

This speed will vary on how the field current is managed in the specific application, among other things... but across the current range of alternators designed for production car use, for the proverbial Rule of Thumb I think you can safely presume that 2300 rpm shaft speed will land you in the meat of the output curve, with the toe of the curve in the neighborhood of 1500 rpm shaft speed.

And naturally you will find that production cars typically drive their alternators at ratios in the range of 3:1. Of course, running the alternator from the final drive is sort of interesting in that charge rate will be roughly proportional to road speed, rather than engine speed. I remember an ancient and charming Voisin race car with a wind-driven generator.

[sblick]

Use to test alternators between 3000 and 20000 rpm on a dyno for noise purposes. Basically seconding McGuire's post about the subject.

[Engineguy]

Originally posted by McGuire
Of course, running the alternator from the final drive is sort of interesting in that charge rate will be roughly proportional to road speed, rather than engine speed.

I'd think a good rule of thumb would be... if you don't have a fan on the radiator, an axle-driven alternator should be fine (i.e. the car cannot/will not be running much while standing still), given a reasonable belt ratio.
Being that a Hewland FTR was mentioned, presumably not a street-driven car.
.

[Greg Locock]

Never tried it, but how long would a car run on battery alone, ie with no charger? I'd guess hours.

[Engineguy]

Originally posted by Greg Locock
Never tried it, but how long would a car run on battery alone, ie with no charger? I'd guess hours.

That was an option in the old days, but with EFI, current ignitions, and electric fuel pumps, it's not now. Clue... even F1 and most drag cars run alternators. An alternator is probably less weight than a bigger battery...

[McGuire]

Originally posted by Engineguy


I'd think a good rule of thumb would be... if you don't have a fan on the radiator, an axle-driven alternator should be fine (i.e. the car cannot/will not be running much while standing still), given a reasonable belt ratio.
Being that a Hewland FTR was mentioned, presumably not a street-driven car.
.

Oh, I agree it would not be a problem except on a road or endurance car. Just interesting to contemplate the issues. If this is the increasingly ubiquitous small-bore sprint class racer/track day car, I wonder if maybe a small PM alternator would be more than sufficient.

[Fat Boy]

Originally posted by Greg Locock
Never tried it, but how long would a car run on battery alone, ie with no charger? I'd guess hours.

It's not a problem on some sprint races. The trick is good mechanics that are religious about putting the car on charge. 1 hour is about the max, though. After that, battery size starts to out-weigh alternator size/drag.

[Fat Boy]

Originally posted by McGuire


Oh, I agree it would not be a problem except on a road or endurance car. Just interesting to contemplate the issues. If this is the increasingly ubiquitous small-bore sprint class racer/track day car, I wonder if maybe a small PM alternator would be more than sufficient.

They are. The belt is not much bigger than a rubber band, but it's enough for a 25 amp draw or so.

[Chris Wilson]

My concern is this car (a Zeus Supersports car, a bit like a radical in appearance, with a Toyota 4-age engine) was run on carbs, with a Red Top pump, and just a very basic Electromotive ignition only managaement. For endurance races someone fitted a small car alternator (Japanese one, from some tiny import car, like Brise sell). They decided to drive it from a longer cam belt, but the mounting of the alternator wasn't brilliantly executed and I lived in fear of it moving and the cam belt failing or coming off. It also suffered failures due to harmonics breaking internal wires. No I am turbo charging the engine and fitting full fuel injection. I am told, and have seen, that modern FI, Motec in my case, doesn't like wild voltage changes, although it does allow for injector pulse width trimming over contained voltage changes. I think I am going to need a charging system. I have recently run the N/A engine on carbs with no alternator and a stock cam belt system, and felt happier. It would run a full track day with just a quick trickle charge into the Varley Red Top 30 battery during lunch, from a 10 amp charger. Onece modified to pressure charging the turbo will be very near the only place to mount an engine driven alternator, so this is why i now look at running one off the driveshaft. It moves it away from heat, allows unadulterated cam belt routing, and should dampen the worst harmonics, indeed, without cam belt tension to worry about, it could be isolated on rubber bushed mounts. I think I'll risk it and maybe aim for a 3 to 1 geared up drive? Thanks

[Bill Sherwood]

Originally posted by Greg Locock
Never tried it, but how long would a car run on battery alone, ie with no charger? I'd guess hours.

Yep.
I run my carby-fed racer on battery alone and although the races are quite short - less than 20 minutes or so - there's no problem at all with battery power only.

I've also run one of my road cars on battery only (alternator **** itself) for over an hour with no problems - EFI engine and some headlight time in that as well. The battery was just a regular size & type for an AE-86.

[ben38]

Hey guys,

I've been working on a big GT car with all the EFI and full spec PI datalogger and a big v8 in it with an alternator like you suggest. Never had a single problem.
The thing is to be carefull to charge your battery and during long systems check up plug in an extarnal one on top of that and everything will be fine.

The only thing is just try not to spin/stall 4 times per lap or won't do

[murpia]

Is there an easy way to turn off an alternator whenever an engine's on full throttle, to increase torque?

Same as an aircon clutch which disengages at WOT in a road car. Battery would still charge at less than WOT and take up the load at WOT.

Regards, Ian

[McGuire]

Sure, and it has been done forever. Just cut off the current to the rotor winding and the alternator will freewheel. Now the only drag from the unit is due to the friction of the bearings and the aero drag of the rotor in the case... which often includes a cooling fan. Many alternators have two or even three fans, inside and outside the case.

On modern alternators with solid-state voltage regulators, often employing pulse-width modulation from inside the CPU itself, this feature is written in. At somewhere near WOT the field excitation is cut back to nothing or very close to it. With some recent alternators this is necessary due to their excess of high-speed output. Engineered to produce high current at low speeds, at prolonged high speeds they can overrrun the battery.

[murpia]

Thanks for the info.

So to interface with the rotor winding what would work: lowside driver, highside driver, solidstate relay, good old mechanical relay?

PWMing the cut-off to blend charge current on and off sounds better than just switching it off, but would that be OK on an alternator not designed for it?

Thanks, Ian

[J. Edlund]

Go to http://www.mclarenel...ternators&id=70 they offer a few current output vs. alternator speed graphs aswell as some additional info about alternators that can be good to know.

[McGuire]

Originally posted by murpia
Thanks for the info.

So to interface with the rotor winding what would work: lowside driver, highside driver, solidstate relay, good old mechanical relay?

PWMing the cut-off to blend charge current on and off sounds better than just switching it off, but would that be OK on an alternator not designed for it?

Thanks, Ian

All the above but depending on the alternator's output, field current can be up to 8A so plan accordingly. PWM is very elegant but since you are not load dumping to the B+ you are not going to hurt anything in any event.

Back in the Day it was typically done with a kickdown switch for a GM THM 400 automatic transmission (bolts right to the throttle linkage) in series with the coil circuit of a SPST mechanical relay like the ubiquitous Bosch or whatever else is in the junk drawer; field circuit in series with the switching side A pin. The switch closes on throttle application (adjustable) so when the relay's coil is energized the rotor circuit opens. Recalling this now makes me laugh. As simple and dumb as it comes, but at one time this was trick stuff.

[murpia]

Originally posted by J. Edlund
Go to http://www.mclarenel...ternators&id=70 they offer a few current output vs. alternator speed graphs aswell as some additional info about alternators that can be good to know.

Thanks, very helpful. To quote this

The excitation coils in the rotor draw current from the vehicle battery via the Ignition (IG) connection. IG should be disconnected when the engine is not running, otherwise it will slowly drain the battery.

When IG is disconnected, the excitation coils are not magnetised and the alternator does not deliver any current. IG may be disconnected while the alternator is being driven: the alternator will not be damaged and the load on the engine is decreased.

A little more Googling suggests an IG pin is present on most modern alternators - so job done!

Regards, Ian

[Powersteer]

The transaxle input shaft should not be far off the drive shaft. If you are running the clutch at the back as well, that would work having the alternator running just before the clutch assembly. A rear mounted alternator and battery set up is really neat.