Bicycle generator on a Seagull - example shown

[Rex NZ]

Phil

That's a great idea

Converting an ignition coil would be very neat & compact.

Rex

[Charles uk]

If you bought a Seagull in 1979 & any time afterwards a CDI was all you'd get!

And you could buy the optional charging system.

[Rex NZ]

Charles

All very interesting. NZ saw bugger all CDI or lighting coils, (if any). Resultantly I know nothing about them.

So what's the story with them?
* could the lighting coil be retro fitted to a mid 1970's classic?
* was the magneto plate the same as a mid 1970's classic or does this need to change?
* are the lighting coils readily available, or are they a rare item?
* what's the lighting output in volts & amps?

Rex

[Charles uk]

Rex,

1 could the lighting coil be retro fitted to a mid 1970's classic?

No, different crankshaft, ignition, base plate & needs flywheel cover.

2 was the magneto plate the same as a mid 1970's classic or does this need to change?

Yes

3 are the lighting coils readily available, or are they a rare item?

Yes, but very few new ones but a few second user, I have 3 or 4, as I always remove them, I use my motors to push boats not to power a christmas tree.

4 what's the lighting output in volts & amps?

Circa 12 volts & by rotating connector plug in socket 40 or 80 watts (I think)

But note 80 Watts is more than 1/10 of 1 Hp, ignoring any losses in it's generation.

[40TPI]

Seagull magneto ignitions do not run on 6 or 12V. The primary resistance is around 0.2 to 0.3 Ohms. A little mental arithmetic with Ohms law quickly convinces that it is highly improbable that a peak DC current of 60A is flowing during dwell time. And that is certainly the case! My earlier post at :

http://www.saving-old-seagulls.co.uk/fo ... php?id=183

showed the dwell period on a Villiers MK1 magneto points waveform. The trace is 10 ms/div and 2v/div. O volt is grid centre line. This confirms that the voltage across the coil under load, ie when the points are closed, is zero to all intents and purposes. (A slight downwards IR droop is just visible as the primary current builds to a peak before the points open and it is more helpful to think of the magneto being current driven rather than voltage driven esp. when calculating the stored energy per spark.)

The first waveform below is the primary voltage across the points with a paper wedge holding them open. This is 4 volts/div, 10ms /div. . The peak amplitude is highly dependant on RPM. Be aware that this is open circuit; no load is being drawn.

The second is also 4v/div and 10ms/div and the points jammed open but with just under half the RPM (about 935 RPM) . Note the far smaller peak amplitudes.

(I am aware that this post now contains more than five sentences, all using more than five words, with most words having greater than five letters. There will no doubt be a challenge from someone that it is a linguistic rant …………..)

This asymmetric waveform is highly RPM amplitude dependant and is clearly not the most attractive proposition as an efficient starting point for rectification to provide a simple regulated generator output. For anyone wanting to start a design from here the primary of most early Seagull magnetos is around 22 SWG DCC ( close but not the same as AWG) and about 140 turns.

A better arrangement for an alternator would be multi pole /multi coil ……..like the Far Eastern motorcycles. A Marston Villiers magneto however already has two empty laminated core arms which would have had ac lighting coils fitted when it was supplied for motorcycles.

The Seagull alternator that Hugo and Charles mention also suffers badly from the output being unregulated for variations in RPM; note the half hearted attempt to get round this by having “High and “Low” charge positions on the cable loom. This is a recipe for wrecking the battery being charged. Older wet lead acid batteries were much more forgiving of such brutal treatment than modern gel types. All in all this really calls for some sort of buck /boost DC to DC converter to make a reliable and usable charging system especially if modern electronic equipment is to be safely powered. I don’t think it is worth the candle.

Peter

>

[Keith.P]

I would have thought that a small charging coil could be fitted, but the dynamo would look in keeping with the seagull.
DC output from a coil?

[40TPI]

DC output from a coil?

The diode rectification depends on the waveform(s) shape and to some extent on the design voltage you choose. Bridge arrangements are not attractive on low ( < 20ish) volt designs because of the power loss in the second diode.

Peter

[Keith.P]

DC output from a coil?

The diode rectification depends on the waveform(s) shape and to some extent on the design voltage you choose. Bridge arrangements are not attractive on low ( < 20ish) volt designs because of the power loss in the second diode.

All I was saying was, that a magnetic field alternately passing a coil, as per setup on the seagull ignition is not going to give you a DC output.

[40TPI]

Exactly so, it will need rectification to get DC. Which is when the choice of diode arrangement arises. A single coil will give the waveform above. Most efficient use of that waveform would be to half wave rectify pos and neg pulses to separate capacitors arranged symetrically about chassis potential and then feed the combined dc voltage across the two caps to either a buck or buck/boost dc to dc converter. That takes care of the voltage regulation problem with RPM and also neatly allows the final output voltage to be unipolar and referenced to chassis. (Takes a lot more words to describe it than a circuit/ picture !) But as I said before not worth the candle! Just a paper design!

Peter

[Rex NZ]

40TPI

This thread is very interesting. I think I'm going to learn a few things here.

The scope traces are not quite what I expected. However, they make sense when I think carefully about the dynamics. The manufacturers common description of AC lighting coil, (as opposed to charging system) this seems appropriate.

My small brain is having difficulty understanding your worded circuit description. I'm not familiar with buck/boost config. Maybe can you post/photo a circuit diagram sketch. Your input is appreciated

Rex

[40TPI]

Google frequently saves wear on my typing finger ....!

And once again it comes to the rescue:

http://www.eetasia.com/ARTICLES/2007MAR ... RLD_TA.pdf

First hit was from Texas; many others too numerous to mention. This one looks like it was a paper for a conference.

Peter

[atoyot]

Wouldn't it be a lot neater and perhaps easier to place a charging coil inside the flywheel on the stator plate and run the output through a rectifier?

Something like this perhaps?

http://cgi.ebay.com/Volvo-Penta-outboar ... 3a5ff12e01

It may take a lot of experimentation. I tried one something like this and got almost 8v at top revs out of it. I think the trick's going to be to find something the laminations of which pretty closely match the inner circumference of the flywheel, for maximum pickup.

Who can provide me with the inner radius of a Seagull flywheel? Or the diameter across the inside, even better. Both of mine are in storage at another fellow's place (and I'd rather not take off a flywheel for no better reason). Surely, someone here has one disassembled already; knowing the diameter or radius will allow me to ask questions of Ebay sellers regarding their lighting coils. In the interest of science of course....