Starter woes (yet again)

... a new pin out of a 4mm HSS drill which should, I think be tough enough. I accept plain carbon steel may be less suitable. I do wonder whether the pins I already had - installed by a previous owner - were strong enough but they have stood up to considerable pounding!

I think I'll just clean up the pawls with a small stone in the Dremel and see how I go with that.

... but will postpone any testing till tomorrow giving the loctite on the crank bolt time to set.

I am convinced the reluctance to start is down to the BB box not getting enough voltage and am starting to suspect the battery. I'm charging an old car battery in the shed and if it takes charge will try that with jumper leads.

My current proposed solution - which may be a non-starter - is to fit a small 12V battery dedicated to the BB box. This will be activated by a relay switched by the ignition switch and charged via a diode (if I can remember which way round to put it!) so that it doesn't try to contribute to the starter motor load.

All good fun eh? At least I have the B'Zuki which is turning out to be a nice bike to ride.

... and the mechanical side is working fine. The old car battery was a non-starter(!) as it wouldn't hold charge so would be worse than useless.

It still wouldn't start off the main battery although turning over briskly. It did start using the auxiliary one. So I'll buy a little sealed item and connect it up. This is what I'm thinking. The diode (which I may have drawn in the wrong direction!) is to allow the auxiliary battery to be charged but preventing it being drained on starting.

... across the battery terminals which is where I get the 10+ volts. I've also measured it at the ignition wire (white) going into the BB box and this is the same as the voltage across the battery give 0.1v or so. Static volts with acharged battery are 12.7 in both places. I'm convinced that whatever Boyer say it's not working effectively at this voltage.

I've ordered a battery from Tayna and will wire it all up tomorrow, probably omitting the charging connection till I've proved (or not) that it works.

for continued interest. Re where I'm measuring voltage: it's the same at battery terminal and engine. I've used 10 sqmm heavy duty cable for both earth connection and live (negative) starter connection.

Many thanks for offer Dan but I'm in Redditch. The bike came with a Mk3 Boyer system which is known to give problems at low voltages so I've replaced that with a new Mk 4 box direct from BB. The PCB and rotor are as original but BB assure me they should work OK with the Mk 4 box. Resistance across both coils on the PCB is 138 which is in line with BB recommendations. I've replaced all those horrible pre-insulated connectors with properly soldered ones.

As you say Al, electricity is good stuff. I have a wild theory at the back of my mind that because I have used the heavy duty cable for starter and earth connectors that this allows the starter to grab all the available power leaving little for ignition. Totally daft probably.

What do you think about my idea in the sketch? I've ordered a 4.5AH battery from Tayna to try it out. I'm not sure if the diode will achieve what I want - it should probably be a split charge relay.

Duncan: I've found a split charge relay for a tenner that should work better than the diode, however as Al says the current(!) problem is verifying that the ignition side works. https://www.carsparesltd.com/product/towing-trailers-b5350/dual-charge-rely-COMMP288/COMMP28

That device is negative earth but I should be able to use it to drive (yet another) relay to bring in the auxiliary battery when the system is charging ie the voltage across the main battery is above 13.5 or so. It may need its threshold tweaking a little.

Al: I have measured the voltage both at battery and BB feed (in the original configuration) and both are the same at just over 10 volts. My configuration is intended to completely isolate the ignition system - apart from the earth - from the main bike system. I haven't shown all the connections in my diagram but the BB white wire (negative) will be fed from the auxiliary battery via the relay so will be permanently live when the relay is closed. Other connections will stay as standard. I measured the coil resistance just for elimination purposes.

The auxiliary battery is due to arrive this afternoon......

Of course I may be barking up entirely the wrong tree and find the problem is carburation.......

... the battery arrived (great service from Tayna as always), I gave it a quick refresher charge and fitted it. Reslt: The bike started easily after a few revolutions and seemed to run well. I'm reluctant to keep trying it at the moment but feel as if I'm getting somewhere.

I've bought the split charge device and will fit that when I've recovered from the heat.

As Duncan says, these devices are designed to connect a secondary battery when the input voltage exceeds a predetermined level - preset to 13.5 volts in this case - and to disconnect when it drops below another level - 13.2 volts. This is precisely what I need. This layout is designed for motorhomes etc and is used for the same reason - to charge the auxiliary (leisure) battery but preventing it from contributing to the starter. Obviously this relies on the charging system (full wave rectifier plus Zener diode in my case) providing that level. There is an adjusting screw so I may have to tweak the cut-in value a bit.

John: I'm quite happy with the performance of the starter motor - it turns the engine over readily. As you say there are well known weak points in the sprag / transmission but not insuperable. I think it would be a major engineering job to fit any other sprag system. I think Peter Holland has fitted an alternative motor.

I remain convinced that the problem I have is that the BB system is not receiving enough power when the starter is running. I've been talking to BB who have been quite helpful. However, the system was designed for bikes with kickstarts. In their own words: "Although the MK4 box can operate at lower voltages, power will still remain very low". It may well be the case that there are other contributory factors - I haven't checked compression for instance, but it seems quite low on kickstarting. I've rebuilt the carb and once running carburation is clean and the engine runs well.

This is a rough schematic of what I propose. Note I've added a 10 amp fuse in the auxiliary battery output.. On the VSR, terminal 12 is positive input, 6 is positive switched output and 0 is negative. Fortunately the whole lot is in a plastic box so no worries about the chassis being earthed etc.

... all connected up as in diagrams and tested. It works just as I hoped it would. Starts readily - albeit after a bit of churning over - and runs well although at the moment still a static test. The charging system is also working - I temporarily included an LED to show when the auxiliary battery was getting charge and that showed it was at everything over a fast idle. Of course it needs a road test and some time to show that it's a permanent solution.

A thought has just struck me - the same charging circuit could substitute for the warning light assimilator with a couple of minor adjustments - particularly replacing the N/O relay with a N/C one.

... of the installation. Note, if it all failed then it's a simple matter to just reconnect the white ignition wire to the white on the BB box which will omit all the stuff I've added.

Relay and BB box:

Voltage sensitive relay and charging relay:

Auxiliary battery installation:

... but the drive has failed again. The silver steel pin has bent! So I have now tried 4mm drill shank and silver steel. I've asked on a model engineering forum and there are 2 suggestions: hardened and ground dowel pins and piano wire. I'm quite attracted by the former as they arrive in exactly the right size.

I've also noticed that there seems to be a huge amount of play on the pawl pivot pins. I was hoping not to have to do anything with them but this can't be helping anything. Plus the pawls are now showing significant wear so I'll try the weld repair suggested by either Al or Andy.

I have to say it's getting me down, and the hot humid weather doesn't help.

I've now dismantled the  whole thing, grinding the peened part of the pivot pins and driving them out. One of the pawls (actually not the one with the offending drive pin) is very badly worn. It makes me wonder whether that pawl has not been engaging but has been riding on the ratchet.

The pivot pins are worn giving a lot of play. Interestingly, even the unworn portions are significantly smaller diameter than the pivot holes in the pawls. This seems to indicate that a certain amount of "slop" is built in although I can't see why that would be a good thing as amongst other things it would let the pawls cant over. Maybe it's to allow them to find their own position?. The holes in the pawls allow a 5mm drill shaft to fit easily. Logic says that I should open up the holes in the main component to 5mm but I have been wrong many times with this bike.

I plan to have the pawls built up with weld - hopefully I can find someone locally to do this - then file back to the correct profile.

What is ironic is that I had it starting and running well earlier in the year but seem to be going backwards...... But I'm sure I will get there eventually.

I have a copy of the drawing and will check dimensions. I've already looked at the hole sizes which correspond to those in the drawing. I've just tried drawing a file across a pawl and it appears they are at least case hardened although when I try on the worn portion the file marks it. There are no apparent wear marks on the ratchet attached to the crankshaft.

I'm still puzzled that the large hole in the pawls (for the pivot pin) is so much larger than the pin size - 0.202" as opposed to 0.184" (on an unworn section). Seems a huge discrepancy and can only add to misalignments.

This is proving something of a trial but I'm sure it will be sorted out eventually. I agree the pawls are worn out however new ones aren't available and my fitting skills are inadequate to make new ones. I'm off to a local engineering firm (Nametab- hope they're still in operation) this morning to seek advice. I am considering whether metal spraying may be a route to recondition the pawls. Alternatively they may be able to make them although I suspect the cost would be prohibitive especially as they're such an odd shape.

Interestingly the badly worn pawl is not the one that has been causing me trouble - I've only been changing the pin in the one that's broken / bent. I can see that when that happens that pawl is not going to engage with the ratchet correctly and may ride over it thus increasing load on the others. There are no signs of wear / damage on the ratchet itself.

I still can't fathom why the pivot pins are so much smaller than the holes in the pawls with about .015" clearance. This can't help in getting them all to share the load unless I'm missing something. Not unusual......