auto advance unit

[RDfella]

Bit of info that may be useful to anyone with / contemplating an electric starter.

Now I know at least one supplier of a starter kit recommends using an ATD unit as opposed to manual. I suspect this is because the spark will be of better quality when retarded, and therefore the engine more likely to start.
However, my A10 used to kick back occasionally on starting - something that would likely destroy an electric starter. Mindful of that, and as I'm in the process of fitting an electric starter, I thought I'd better take a close look at the auto adv unit. Calculation showed that the range from retard to full advance was 11* - ie 22* at the crank. This in turn meant I was running around 8 - 10 deg of static advance - no wonder the odd kick-back.
As the unit on my bike showed evidence of substantial wear, I obtained another - in largely better condition apart from looseness in the spindle. I duly stripped it, sleeved the central pin, and ground / filed the stops to achieve 15* of advance (30 at crank). This will enable little or no static advance and a substantial reduction of the likelihood of kick-back. And hopefully save my starter from destruction.

Addendum conclusion: I'd recommend anyone with a classic wearing a starter kit to check their static (retarded) timing and, if 5* or more, to modify that before a kick-back damages something. I have a classic car that sometimes backs onto the starter (7* static) but, being electric start is standard fitment, that at least has a built-in shock absorber. The kits for our bikes don't.
 

[KiwiGF]

Bit of info that may be useful to anyone with / contemplating an electric starter.

Now I know at least one supplier of a starter kit recommends using an ATD unit as opposed to manual. I suspect this is because the spark will be of better quality when retarded, and therefore the engine more likely to start.
However, my A10 used to kick back occasionally on starting - something that would likely destroy an electric starter. Mindful of that, and as I'm in the process of fitting an electric starter, I thought I'd better take a close look at the auto adv unit. Calculation showed that the range from retard to full advance was 11* - ie 22* at the crank. This in turn meant I was running around 8 - 10 deg of static advance - no wonder the odd kick-back.
As the unit on my bike showed evidence of substantial wear, I obtained another - in largely better condition apart from looseness in the spindle. I duly stripped it, sleeved the central pin, and ground / filed the stops to achieve 15* of advance (30 at crank). This will enable little or no static advance and a substantial reduction of the likelihood of kick-back. And hopefully save my starter from destruction.

Good research, my B44 bought not long ago kick-backed terribly and the PO only sold it (reluctantly) due not being able to start it, it had weak ATD springs and not enough advance retard range to get anywhere near TDC when static, so I modified the stops like you did. Bent the spring stops to make sure it had full retard when static (springs were loose when static before this). It still has std points of course.

The ATD’s come in various ranges, mine had one with less range than std (probably). It was marked 6 degrees from memory, and should be 11 degrees (I think). Modified It’s now about 15 deg.

It doesn’t kick back now and starts as easily as my B31 (manual retard), usually first kick, but it does have the odd hissy fit and sometimes takes half a dozen kicks when hot but left a while. Not sure why.

The pinking at low revs is also almost gone. A strobe shows the ATD hits full advance a bit over 2000 rpm.

I wonder how much of the B44’s reputation for poor starting is due to the ATD not working correctly, many owners fit electronic ignition to help with that, when maybe they just needed to fix the ATD.

[Colsbeeza]

I am starting to get my mind around the Steve McFarlane electric start, soon to be installed.
Could I get an opinion on why Lucas provided a range of advance.?
I have attached the Lucas specs for ATD's and you will see that they all have a range. The BSA unit 45703 for my bog standard Flash with 7.25:1 pistons and 356 cam states a range of 11-13 degrees, others are different.
My best guess is that it is either the factory tolerance, ie. some came out as low as 11 deg advance and others as high as 13 deg advance, or alternatively, it may be for wear tolerance.
It may not matter much normally as you set the ignition with ATD wedged fully open, but I would like to ensure that when my electric start is fitted, that the maximum advance at start is low enough for me not to have any kickbacks on starting, thus risking the sprag bearings as some have experienced.
I have been running at 31 Deg BTDC which has been running great, and assuming my ATD is 11 deg, then advance at start is about 31 - (2 x 11) = 9 degrees BTDC. I may decide to do some filing on the stops to reduce this to maybe 6-7 deg???
If it is 13 deg, then ignition advance is 31 - (2x13) = 5 deg BTDC and no filing needed.
 I should first establish what advance the ATD has now, but I haven't got it off the bike yet. Is there an easy way to measure it on the bench?
If 13 deg, it is already there. What ignition advance should I aim at for starting? Opinions from the Forum seem to suggest that below 8 deg will prevent kickback, providing carb timing etc are spot-on.
Good question for Bill Groily? The ones you tested Bill were all 11 deg.
Col
The knee is improving - best low pain night last night.

[KiwiGF]

Not an answer but related, it's important the ATD gives full retard at starter speed eg springs are good. My B44 kicked back really badly when it had an 8 deg ATD (should have 11deg, I think) with slack springs, but some filing and bending of tabs fixed that. My guess is the timing is still advanced 8 degrees during kick starting.

You could prob set up a strobe and disc arrangement to test things out on the bench, given the consequences of the ATD not behaving as expected this might be worth doing regardless.

[groily]

Cheers Col - and glad pain levels dropping by stages.

As you say, there were differences between the units of a few degrees here and there. RDFella went into this too, in connection with his own electric start design. I am sure he will recall where he ended up with his, having gone through the same thought processes.

The easiest way to check what one actually does is 'on the bike' with a strobe I'd say, but that does assume the thing operates properly across its range. Off the bike,  a circular protractor (degree disc eg) can be your friend,  absent a dynamic rig for running the thing up on an actual mag.
I don't know for certain how many ° BTDC is optimal for 'no kick back' and long life for the clutch - but RD will I am sure have his figures to hand, and 'sub-8°' sounds pretty sensible.

No problem to increase the movement a bit by attacking the stops, as you say - but best to know what the start and end points are/will be, I do agree!
The Triumph ATDs, both Lucas and the BTH type with clockspring and rollers, had more movement than most others of the anti-clockwise variety,  but then their full advance setting was always pretty big. Bloke I had round here for a play-date the other day with a much-tuned T110 motor in an historic Moto Cross Rickman-framed machine was running his at 40° BTDC, no less.

Good luck!

Edit: If I'd thought to scroll up I'd have seen what RD reckons!

[Colsbeeza]

Thanks RD, KGF & Bill,
That is great experience. It seems that as long as I can approach about 6 Deg with springs taught I should avoid significant kickbacks. Other than that, it seems well-tuned. Check out the plugs recently. Champion L82C at 20''' gap, Unleaded 95 no ethanol.
I will update in a few weeks after I have bench-tested the ATD.
Cheers Col

[Joolstacho]

Phil Irving 'Tuning for speed' etc stressed that (contrary to some people's belief ), almost all well tuned OHV / OHC motors operate within a fairly limited ignition advance figure, which I think is in the range 35-38 BTDC. (Not when retarded for starting of course). -The theory being that it's to do with the speed of the flame-front in the combustion chamber. A more advanced figure denotes a less than optimal combustion chamber form because the flame-front travels slower. I'm no expert (as you know) but... Interesting. I wonder if modern fuels have changed things. I believe if you have 2 Velo motors with the same cam and same compression ratio, a squish converted one uses less advance and gives more power because the combustion happens quicker.

[groily]

I'm sure Phil Irv was right on that, as so much else!
I was also always brought up to believe that it was sub-optimal head design / lack of squish that - principally - demanded large advance figures.
Of the bikes of the day, Notruns pass the 'How little advance needed?' test better than most - I've seen them running at 28-30°, depending on model. 
Trihards seem to require the most  . . .

[Joolstacho]

Which all reminds me of my old 450R/T Desmo Ducati I bought in the early seventies. Fabulous machine, if not perfect (what is!)
The owner's manual got the Ignition timing figures a little wrong, in that they gave a flywheel mark that the book said was the static advance (retarded) when in fact it was the advanced mark. Of course as any sane human (perhaps not an Italian) would do, I set the points to what the the manual advised, I added the ? degrees advance. Unfortunately... and my left foot limp even now attests to the error, the book should have read 'advanced' not retarded (static). It took me some time to cotton on to the fact that it was horribly over-advanced (mind you once - if- you started it sure went well).

One day a snide wise-guy copper stopped me on the Esplanade after I had embarassed a joker on a 500/four off the lights at Middle Park. The unsuspecting rozzer decided he'd like to try to start the R/T. -He'd never seen anything like it, and fancied himself. (Of course I invited him magnanimously, -having some clue as to what might ensue, and with hopeful fingers crossed behind my back).
The poor bugger ended up on his back in the gutter with the Duc on top of him, -it had kicked him back with a vengeance. He was not happy!
The bloke gave me a canary (non-roadworthy notice), guess what for... not having a kickstart rubber! Hahah.

[Colsbeeza]

Bill Here in Australia, in 1960 when my Flash was built, we ran such bikes on Super Leaded fuel with an Octane rating (or RON) of about 98 or a little more. I think it got to about 100 in mid sixties. The old Standard leaded ( about 92 octane) used to knock a fair bit up hills with WOT. Standard ran OK in the old 6.5:1 Holden.
I find Unleaded 95 runs very nicely with 7.25 : 1 CR (compression ratio).
Rolls Royce for their 1960's lower compression S3's and Shadows with about the same CR, recommended a retarding of ignition of "1 degree per octane number reduction" when changing to unleaded, as unleaded was lower RON.
For my bike, with BSA's recommended 13/32" BTDC (cig paper setting?) after 1959 presumably on available British fuel, this means a setting of about 36 Deg BTDC, about where Phil Irving said.
With the 95 unleaded, this calculates to a retardation of about 33 Deg BTDC (plus or minus a fair bit of uncertainty). Although many Flash owners set theirs lower than 13/32". I find 31 Deg seems best, and is close enough to what Orabanda's dyno testing showed.
Although fairly theoretical, general opinions suggest our lower RON fuels need the ignition retarded below book settings by a few degrees, which is what Orabanda found out. If you run best on 98 octane for say a Super Rocket, then more advance is beneficial performance wise.
Col

[Colsbeeza]

I am having a go at increasing the ATD Advance Range from 11-13 (standard) to 14.5 in readiness for fitting the electric starter.
Stephen McFarlane requires that the ignition at full retard must be 2 degrees maximum to avoid risk of kickback on starting and damaging the sprag clutch.
I want to set my full advance at 31 Deg BTDC, so the ATD Range must be increased to (31-2) /2 = 14.5 as the magneto rotates at half engine speed.
Rather than butcher my pristine and well-proven ATD, I got hold of another for me to cut around. I'll try it on the bike and if OK I'll leave it on. I was able to calculate how much metal to remove by simple trigonometry (Calculations attached)- The diameter across the stops is 59mm, so radius is 29.5mm. The gap over which the advance moves is 5.83mm. This is the gap you jam a clothes peg in when setting the ignition timing. The range angle is calculated from Tan X = 5.83mm/29.5mm =0.1976. So X = 11.18 Degrees, This suggests that my ATD had little wear.
To get 14.5 deg, The new gap works out at 7.63mm, an increase of 1.8mm.
The Stops are 9.2mm width, so if I chop metal off the stops, the stops width will be 9.2 - 7.63 +5.83 = 7.4mm, which is getting a bit thin. Several ATD's I have seen usually have the stops bent, as they are soft mild steel and they get a hammering.
Another alternative is to remove metal from the bolt housing, as per my sketch.
The only effect I can envisage is the lighter housing may mean full advance is reached at slightly lower Revs.
My question is whether any members can see any other issues, ie should I remove 1.8mm of metal from the STOP ( or Advance Limit Post on sketch attached) or from the Bolt Housing?
Cheers
Col

[RDfella]

Col, your figures closely match mine and follow the same reasoning / calculation method. The 2* max static advance S MF suggests aligns with my thinking as well.
I removed metal from the stops, but my stops were thicker than in your diagram, and my carrier [housing] thinner. Remember, however (as I'm sure you're aware) it needs to be removed from the advance stop (the timing is then set appropriately) because removing from the retard side will achieve very little due to the fact the bobweights can't really move any closer in.

[Colsbeeza]

Thanks RD for the method confirmation. Yes either remove metal from the advance STOP or from the adjacent housing/carrier.
My preference is to remove it from the housing/carrier where I have marked it on the diagram.
BTW, anyone with a standard ATD can use this method, just measure the advance gap and look up trig tables for tan.
Cheers
Colin

[Joolstacho]

Col, you arrive at a fully advanced figure of 31deg? That's a fairly retarded fully advanced position isn't it? Phil Irving's 38 to 36 degrees has been like a benchmark for an efficient engine for many years, and yes I know modern fuels may have changed things a little, but I can't see that 'modern' fuels would require retarding roughly 7 degrees. A Super Rocket surely would run more advance than 31. (357 cam, approx 8.5:1cr)
I'm no expert, but I do know that running retarded is not good.

There's been this myth that a more advanced ignition produces more power... mainly I think from *ossers trying to tune old japanese bikes. The fact is that the opposite is more likely. A somewhat more retarded figure indicates a more efficient combustion space, - a faster moving flame-front that requires less advance. (Though I'm talking sports/racing engines running at 'our' sort of RPM here).

Just interested in opinions...

[RDfella]

Usually the most appropriate advance for a particular engine is just before pinking occurs. Relatively easy to find that spot with a distributor insomuch as one just rotates the unit until appropriate point is reached - unfortunately not really possible with a magneto with ATD. With a manual advance, one could set the engine over-advanced and then play with the retard lever for best results - and then measure what that advance is and re-set the position with manual lever at full advance.
As others have stated, amount of advance required varies wildly between different engines, so currently I go with data recently published on this forum, where 31* seemed to be the best place.