there is NO battery fitted

I assumed there was a battery.
In that case the voltage tolerance is less of an issue.

TL431 IC type circuits will not work

They will work fine without a battery, but their accuracy is probably not needed if there is no battery, although bulb life is improved if they always have the correct voltage.

 

RE: Crustchov ""The 2N3055 has a 10A maximum rating with the circuit operating limit being about 5A.""

Oh I have many hundreds of them without of any proper use. What to construct by it honourable, best if soldable. By the way, their old-fashioned body demands a radiator what is not anymore fashionable nowadays, however they are cheap and very much of.

 

I assumed there was a battery.
In that case the voltage tolerance is less of an issue.

They will work fine without a battery, but their accuracy is probably not needed if there is no battery, although bulb life is improved if they always have the correct voltage.

Yes sorry I Really should have made that clear ;(
I was focused on describing the alternator so as to avoid any suggestion of a series type regulator
I will have another look at the TL431 but im worried how it will cope with the very choppy dc waveform
thanks for taking the time to reply and help

 

I will have another look at the TL431 but im worried how it will cope with the very choppy dc waveform

The response time of the TL431 should be fast enough to respond to the frequency of the chop.
Or a capacitor could be added to the TL431 input terminal, which will filter the chop, thus it will respond to the average voltage level.

Do you know what the waveform looks like?

 

The response time of the TL431 should be fast enough to respond to the frequency of the chop.
Or a capacitor could be added to the TL431 input terminal, which will filter the chop, thus it will respond to the average voltage level.

Do you know what the waveform looks like?

No because it varies lots with the engine revs
had a look with a friends handheld scop and it was all over the place
you can even see it on a an analogue volt meter, it drives digital meters nuts

 

No because it varies lots with the engine revs
had a look with a friends handheld scop and it was all over the place
you can even see it on a an analogue volt meter, it drives digital meters nuts

The TL431 circuits should still work.
Adding the capacitor might be the best because it then regulates to the average value of the voltage, which is what you want to regulate (actually you want the RMS value, but the average should be close enough).
And you can add a pot to easily adjust the voltage to whatever you want.

 

found a 5amp 14v zener and it worked ok with lights on
but 2n3055 failed after about 30 seconds with lights off
it was on a fair sized heat sink bigger then I normally use for the original lucas zener
is there any other higher output device I can use

 

Joseph Lucas- Prince of darkness!

 

found a 5amp 14v zener and it worked ok with lights on
but 2n3055 failed after about 30 seconds with lights off

Is that the circuit in post #13?
Did you use heat-sink grease when you mounted the transistor on the sink?
Without that the thermal resistance between the transistor and sink can be very high.

You could put more than one 2N3055 in parallel.
You would need to add a 0.1Ω or so resistor in series with each emitter to equalize the current.

Edit: See next post

 

It just occurred to me, that if the output is indeed AC, then you need back-to-back Zeners to clip both the positive and negative peaks (which is likely what the original Zener was), thus the transistor circuits won't work properly as shown.
That is probably why your circuit failed.

To use any of the transistor circuits they could be put inside a rectifier bridge, as shown below:
Note that you'd have to reduce the zener voltage to allow for the forward drop of the two rectifiers in series.



Edit: Alternately you could connect two of the transistor circuits back-to-back with a rectifier across each to carry the reverse current, which is a more complex circuit.
But that would cut the dissipation in half for each transistor.

 

It just occurred to me, that if the output is indeed AC, then you need back-to-back Zeners to clip both the positive and negative peaks (which is likely what the original Zener was), thus the transistor circuits won't work properly as shown.
That is probably why your circuit failed.

To use any of the transistor circuits they could be put inside a rectifier bridge, as shown below:
Note that you'd have to reduce the zener voltage to allow for the forward drop of the two rectifiers in series.

View attachment 152729

Edit: Alternately you could connect two of the transistor circuits back-to-back with a rectifier across each to carry the reverse current, which is a more complex circuit.
But that would cut the dissipation in half for each transistor.

Thats quite an interesting idea
but no I use a 35amp bridge rectifier and 6300uF cap in place of the battery
dosent have much effect on the flickering lights but essential for starting the bike

just for interest this is how Lusas and Norton coped with the higher output alternator in the 70s
http://imgbox.com/cyP5gCfh
https://thumbs2.imgbox.com/39/19/cyP5gCfh_t.jpg

if the above image does not work, i give up just how do you like to a photo

 

if the above image does not work, i give up just how do you like to a photo

The link doesn't work.
I just do a screen grab of the photo (with the Snipping Tool in Windows) and paste it into the post.