still not starting

[rhyatt]

Not much left in the way of it bursting into life and you hurtling towards the sunset!!!!!

[Barnsey]

I've given up for today. I need some rest time before work tomorrow. But I'll tweek the timing next weekend and give it another kick. See what happens.

 I set the points timing on the crown of the cam, which isn't quite right. It should be on the rise.  So that needs adjustment first. 

[rhyatt]

If you do it on the crown it's wrong.
I thought you'd sorted the timing with your meter on beep telling you when the points just opened.

Obviously you've already got the piston in the right place on the right stroke

[Barnsey]

Yes, my mistake, but it doesn't need adjusting by very much. 

However, I have a query about the ignition coil.  I wired positive to breaker and negative to switch, being as the bike is positive earth. OR, is this misleading and makes the assumption of a negative earth machine? 
Should I connect the negative terminal to the breaker?  Although this is contra-intuitive, the two coils are commoned to the negative terminal, so it seems to make sense that this side should go to ground via the breaker. 

 Earlier coils were labelled as contact breaker and switch. 

All the sites I look at seem to say to connect negative to negative and positive to positive, but I am not convinced that this is right. 

[rhyatt]

If your bike is pos earth , then the coil terminal +( or the cb) should go to the points.
And the negative of the coil( old sw) back to the ignition switch.
If you've converted your bike to neg earth then reverse these of course.

[iansoady]

That is correct in theory but in practice I don't think it makes much difference.

[Steve.S]

Both ryatt and iansoady are, of course absolutely correct. However, although this will be of no use to you, one way of telling if the coil connections are reversed, is to use an oscilloscope where the trace spike on points opening will be upside down.
Moving on, a spark jumping only 4mm is no good at all. I would remove the coil, and connect it to a good fully charged battery showing 6.4V i.e Positive to positive and negative to negative, with the HT lead fitted, and as you remove the feed wire there should be a big fat spark of about 8mm. But don't leave it connected for any length of time.
Can you confirm that your battery was showing 6.4V on your return from holiday?
If both your old and new coils were only jumping 3 or 4mm, it indicates there is nothing wrong with your old coil.
Since you are now feeding the coil direct from your fully charged battery, assuming the earths are good, then the only reason I can think of for a weak spark, is a problem with the points or condenser.
The points gap should be 0.015" at their widest opening. If you are not sure, you could establish this using feeler gauges.
It's possible, when using a meter for ignition timing, to go a little beyond points opening. I use a fag paper, where the slightest slackening of grip indicates points opening. These are available at your local Tobacconist.

[Steve.S]

P.S You will have to earth the piece of metal you are sparking to.

[Barnsey]

I've set up a simple improvised loom on the bench with the original coil and a variable power supply.  I achieved a 10mm spark. 

[Barnsey]

The mottobat battery has been holding 6.4V. With the Mottobat battery I achieved around 8mm.   I didn't try the Cyclones.  They are not holing voltage.

[Derek996]

The mottobat battery has been holding 6.4V. With the Mottobat battery I achieved around 8mm.  I didn't try the Cyclones.  They are not holing voltage.

Just a thought. I'm wondering if you may have a high resistance somewhere in the ignition circuit. With the battery disconnected and the contact breaker points fully open try measuring the resistance between the wire that connects to the -ve side of the battery and the supply side of the coil (I'm assuming your bike is +ve earth). You should see a resistance of less than 1Ω. Sometimes the resistance build up across the ignition switch and multiple bullet connectors can drop the voltage to the coil considerably.
It's also worth measuring the voltage across the battery and then between the battery positive terminal and the -ve connection to the coil when the contact points are fully open. The coil won't be drawing any current with the points open and the voltage at the coil should be much the same as measured across the battery, certainly no more than about a 0.25V drop.

[Steve.S]

Excellent. Now we're getting somewhere. So your new coil and battery are working well. You could test the old coil. It's probably OK.
I completely agree with Derek, although I hope you're bypassing the ignition switch and feeding directly from the battery? With this direct feed, I hope you're NOT switching the ignition on out of habit? Bear in mind that the bike started well before you rewired it. Don't forget to disconnect the feed before you pop indoors to bathe your swollen knee in Epsom Salts. I would also check the resistance between the points base plate and the 'distributor' body and the body to the crankcase. While you're at it, check the resistance between the battery positive and frame.
Are you sure the points are assembled correctly with 0.015" gap? Is there much sideways play in the 'distributor' camshaft bushes?
Are the advance/retard springs fitted and in good condition? I ask this because it could be that when you kick the engine over, the ignition goes to full advance, which may be why it kicks back?
Keep at it.

[Gromit C15]

Hi Barnsey and contributors to this thread, can I congratulate you all for your tenacity. I have ridden motorcycles for many years however stayed away from the mechanics and starting to learn, this thread has been so helpful to me with wrestle to understand the complexities creating efficient spark/fuel combustion... Keep at it Barnsey... I'm looking forward to reading the words.... 'It's working'

[Barnsey]

Okay, I have been measuring resistances.  1.5 ohm battery negative to coil.  0.5 ohm coil to points and another 0.5 ohm points back to battery positive. 

  Closing the points and measuring battery positive to battery negative gave 2.8 ohms including the 0.8 ohms of the coil itself.  So I have 2.0 ohms to lose.

 Both paths go through the ignition switch, so maybe I will need a new switch, but I will try some more experiments first.   

[Derek996]

You will never be able measure low ohms readings of less than 1Ω accurately as the meter leads have some resistance themselves but you can get a better idea by touching both meter leads together and seeing what ohms reading you get. Subtracting this value from the readings you are getting will give you a more accurate result.
Assuming the coil has a resistance of around 3Ω and there is a 1.5Ω resistance between the battery -ve and coil and that the battery voltage is 6.4V. Then total resistance, ignoring the 0.5Ω to earth, is 4.5Ω which will draw a current of 1.5A. This will mean a voltage drop of 2.13V at the coil. So your 6.4V at the battery is going to drop to only 4.27V at the coil at best. No wonder the spark is weak.
With the wires disconnected measure across the ignition switch terminals. When the switch is at the ON position you should see almost zero ohms. If there is anything more than about 0.5Ω the switch would be replaced.