Single Phase brushless motor controller?

[davegsm82]

I was wondering if any of you chaps have experience in the aformentioned area?

I'm looking to have a little bit of an experiment to create a starter system for my motorbike engine using the permanent magnet alternator on the crankshaft.

Ideally I would have liked it to be 3-phase but in my case its only a single phase unit, I know some of the theory required for 3-phase as the company I used to work at made 3-phase motor drives.

The alternator output is 15A, so at a guess I'd say you could quite easily dump 100A or more through the stator for the short amount of time that would be required to start the engine. The Winding resistance is very low, meaning that the high current required won't need any form of voltage boost.

Initial thoughts are of using a normal H-bridge configuration and installing a Hall sensor between the coils in order to detect the direction/rotation. At a guess I'd say that in order to start rotation, it would probably be necessary to 'swing' the rotor back and forth until enough inertia is achieved to start rotation.

Any thoughts on this are welcomed.


Dave.

P.S. I have the facilities and the parts to make the large H-bridge and the usual PIC's and programming facilities so these aren't a problem.

The problem with this is getting enough magnetic field from the permanent magnets to give the torque. Associated with this is very real risk of demagnetizing said magnets. This is why starter motors use a wound field. If it was possible you can bet the makers would have done it already to save the cost of a starter motor. There were dynastart systems where a large dynamo was used as a starter motor, usually on small two strokes.

[Leon Crampin]

It won't work - for several reasons - but demagnetisation of the rotor (field) is not one of them. Most ordinary cars now have starter motors which are a conventional PM field DC motor with an epicyclic reduction gear and of course, a further reduction at the pinion/ring gear interface. The overall ratio is therefore about 30:1.

There is no hope whatever of developing sufficient cranking torque from a motorcycle alternator geared 1:1 to the crankshaft. Increasing the stator current will merely saturate the iron - just enough of it was put there originally to provide the maximum alternator output. Auto component makers are not charities...

Don't mess with it - grief awaits.

Leon (ex Lucas employee).

If you want to see how an imbricated pole synchronous alternator can be used as a starter motor, Google "Valeo STARS".

[davegsm82]

Thanks for the input chaps, I should probably say a little bit more about the application that I have planned for it.

The engine in question is a 300cc single cylinder 2-stroke, relatively low compression at 10:1 which only begins half way up the bore (ported). This is the only reason I was considering this venture, as I guess (rightly or wrongly) that the heavy crank could be persuaded to lump over with the assistance of the alternator alone. The crank can be turned over TDC by hand quite easily.

I guess its one of those things that you'd have to try before coming up with a real-world answer.

I've looked at that 'STARS' system, looks quite interesting although massive overkill for this application, this is a Low-Budget 'Lets see what we can do with what we have' kind of job. I would consider a normal PM DC starter but the location of the alternator doesn't lend itself to any modification without serious cutting and welding of the crankcase/cover.

For those in the know, the bike is an MZ ETZ301.

Dave.

[Leon Crampin]

The Japanese motorcycle starters are PM DC motors with an epicyclic reduction box in the nose. They are chain coupled to the crank via a sprag clutch (usually at the motor end) to prevent overspeeding.

Energy = force x distance x time.

Anyone can crank an engine at 5 rpm. Try doing it at 250.

The Japanese of course, got it right. If you want an electric start, use one of their motors. I don't envy you the task of fitting it - a kickstart has some advantages...

Leon.

[DangerMan]

If you want serious torque.... and you do, as well as choice of direction... you'll need a rotating magnetic field to be generated by the stator, as it is in a three phase machine.
You just can't do this from a single winding, it can only produce an alternating field.. think of N and S poles swapping over between 12 o'clock and 6 o'clock rather than them rotating steadily around the clock face.
If you do create such a single phase field, the motor can go backwards equally well (or rather, badly) depending on which way you start it rotating. Just like a synchronous clock motor.
The torque that could be developed is at best proportional to the strength of the existing permanent magnet (ignoring air gaps and losses in pole pieces etc), which is optimised for a lower torque task and not likely to be enough for engine starting.

Pete

Most ordinary cars now have starter motors which are a conventional PM field DC motor

I didn't know that, I stand corrected.

[davegsm82]

Leon, do you know if the japanese motors have the clutch built into the planetary bit at the front? or is it usually further down the line, 'post chain' as it were.

The alternator on the bike looks like the one in the picture attached, I could attach the sprag clutch to the end of the rotor, although balance may be a minor issue.

Dave.

[MichaelR]

Leon is absolutely right

My professional experience of electric starter motors on bikes is that at the very minimum you will need to have a battery that can deliver 200+ amps to get a quality start.Albeit it will only have to provide this for a few seconds but it will need to be able to hold its terminal voltage up to about 8 volts under these loads. These are the sort of cold cranking amps that you get on even small bikes with a purpose starter motor suitably geared. The initial torque to get the motor to crank is the important bit and it will be very high. In cold conditions of course even higher.

At the other end of the scale the very large Bikes can go much higher certainly upto 800 amps+.

Mike

[Brian R Pateman]

Please keep this to the original subject chaps (borderline for this forum anyway).

I'm quite sure that there are forums specifically for the discussion of motorcycles!

[val33vo]

I have some experiance of a dynastart ( marine two stroke Dolphin ) and to be honest I would avoid them like a plague, even the Dolphin which was custom designed has a serious lack of torque, but at least on the Dolphin the generator and starter have their coils on the circumference of the flywheel so a wee bit of mechanical advantage, I would imagine a small alternator on a motorbike would be much less suitable, if I was you I would just stick with the kickstart

You can be sure if the idea was practical the manufacturer would have done it

regards

val33vo

NB I have some sympathy for you, I was once hopping about like criple for a couple of days after my Honda 125 backfired whilst kickstarting it