Power Arc: runs well BUT kickstarter kickback can be severe

Status
Not open for further replies.
Joined
Jan 1, 2017
Messages
349
Country flag
I thought I'd post this seperately from the previous thread about getting the norton to run smoothly on idle. That got sorted out.

A question for Power Arc users:

So here is another "one last thing". And this one nearly wrecked my right shin yesterday. It's a question for Power Arc users. Sometimes the '74 850 with one Mikuni VM34 and #40 idle jet actually starts on first kick. And sometimes on first kick it stops my swing near the top and then on second swing it might kick back really hard. Might I have it a touch too advanced? I was very careful with making sure the light just next to the optical sensor (in my case, hard to see without bending way down to look just below the sensor module) came on at top dead center. Method I used was to first lightly tighten down the timing module in the middle of it's limited range of rotation. Then fully tighten the timing disk bolt, getting the notch as close to directly under the optical sensor housing I could without worrying about getting it exactly correct. Then, making sure of tdc, loosening and rotating the module to get it just right -- light coming on -- and tightening the module down again. Is this kick back due to not giving the sensor wheel a full rotation in the correct direction to get the correct count between firings? Could that be caused by lack of sufficient swing in the kick? I don't remember a large backward rotation on the first kick that would have increased the count and caused it to fire very advanced though that certainly would cause a kick-back. Also noticed often on first kick there is an "almost" kick-back where, after the slightest downward movement in the kick, my kick is stopped dead as if it has fired very early in the downward swing rotation. It often fires up next kick, but just as often kicks back violently at the bottom of that second swing. By the way, that's when my foot sliped off the kick-starter yesterday and the kick-starter banged back up. I got a good whack in the shin. That, I really want to prevent happening again. Is there, perhaps, a 'slight' backwards rotation where I percieve a dead stop in rotation that increases the count and is enough to cause kick-back? I have thought to kick it through with the key off, but if it's all turned off, what good would that do to the count? Also wondered if there is still power to the timing module if the red ignition cut-off button is pushed? Must say that once running it is now running very well. And would it do any good to install a cutout swich for power to the coil only so I could get a full rotation? Anyone else experiencing this -- or am I doing something wrong?
 
Time it with a stobe light while running after static timing it. it sounds like you are too far advanced.
 
I have this ignition and suffered the same until I worked out that the ignition does not arm until it sees TDC using the slot that brings the light on when setting the timing. If there is any backwards movement of the engine after it has armed, ie rocking it over compression to get a good kick, then all movement of the sensor wheel is counted as a forward turn and therefore added to the advance.

My technique is to turn the engine over until it has just gone over compression, then turn the ignition on, then give it a real good kick so it goes over TDC twice. It needs the first rotation to arm, and then the second to fire. If for any reason the engine does not fire up then I turn off the ignition and start again, just in case the engine has rocked backwards.

I have also found a couple of kicks with the ignition off helps prime the engine.
 
Last edited:
I've been massaged by my 850 kicking over with too much advance. It doesn't feel good. I've read a little on your system and its supposedly tops when you get it set in. Good luck.
 
The power arc ignition is prone to this.
Carefully positioning the kickstarter with the ignition off and then use a serious kick is the only way I know to help. [Don't move the engine with the key on until your ready to kick.]

I wouldn't use this ignition on anything. Running it on the dyno with the gas analyzer showed lots of problems with combustion efficiency. Plus it's a shake and bake ignition with electronic parts under the cover -it's not if it will fail, but when.
 
I was trying to keep on the upside, but.... cat is outta the bag. My right knee is bad anyway.
 
Actually, on a low compression [8.2 measured] 850 engine, the PowerArc performed on par with most ignitions as long as the plug gap was set to their spec. [.015]
Just takes a little care when starting.

On a 10.5-1 engine is where I had real problems.

One other thing that may help with starting is to make sure the timing chain is near it's minimum play. A little extra slack will cause mis-counted teeth on the reluctor.
 
You shouldn't be getting kick back from a Power Arc ignition, if you are starting it correctly.

With the ignition powered on, you need to turn the engine over slowly two complete revolutions.
The first revolution is so the ignition can learn where top dead centre is.
The second revolution is so the ignition can apply the correct timing/map/advance curve (the unit can hold multiple maps)

On our bikes, this equates to just under 2 swings on the kick starter.

On your third swing, the engine should fire up, and you shouldn't get any kickback.
 
You shouldn't be getting kick back from a Power Arc ignition, if you are starting it correctly.

With the ignition powered on, you need to turn the engine over slowly two complete revolutions.
The first revolution is so the ignition can learn where top dead centre is.
The second revolution is so the ignition can apply the correct timing/map/advance curve (the unit can hold multiple maps)

On our bikes, this equates to just under 2 swings on the kick starter.

On your third swing, the engine should fire up, and you shouldn't get any kickback.

Sorry, the ignition unit looses track of TDC the moment the engine stops turning. But pushing the engine though may give you a better mixture for easier starting.
 
It's crazy to put up with a starting drill like that on a Norton. I can't see any advantages of this ignition. I'm in the Pazon camp with an Altair on the Norton and a Surefire on the 350 ~M~. I don't have to turn on my key at just the right time. Just move past TDC and kick.
 
With the ignition powered on, you need to turn the engine over slowly two complete revolutions.
The first revolution is so the ignition can learn where top dead centre is.
The second revolution is so the ignition can apply the correct timing/map/advance curve (the unit can hold multiple maps)
----------------------------

Is this correct, this ridiculous involved process in order to start a Commando with a PowerArc ignition?
Seriously, kicking the motor over slowly two times, finding TDC, etc etc?????
Why would anyone put up with such a starting process in the first place?
What am I missing, this just sounds so horrible to be true?
 
It’s not complicated, you just kick it over twice, and your good to go - I am quite sure most people can manage that.

This was a feature I believe Fred at Old Britts designed in to the software in conjunction with the guys at Power Arc. I think he’s got an exclusive on the Commando for this reason.
It can be disabled if you don’t want it.
You can also disable the multiple sparks per revolution (three I think) if you don’t want that either.

Here are the Old Britts notes:

Recommended Starting Sequence

If you experience an occasional kick back when starting your bike it is because the ignition thought it was at a different position of the crank rotation than it actually was.
The ignition will fire the first time after it is turned on when it sees Top Dead Center (TDC) then start counting teeth on the encoder disc to start the advance curve firing sequence.
If you turn on the ignition and kick the engine through to find the position where you like to start your bike, the crank may slip backwards under compression a bit and the ignition does not know the engine is running in reverse and start to count teeth on the encoder disc.
Then when you start kicking the engine the ignition will fire before the desired position and can cause kick back.

The recommended starting sequence is:

  1. With the ignition off, kick the bike through until you have the crank in the position that you normally like to start your kick.
  2. Turn on the ignition.
  3. Kick the bike over and always kick through the complete throw of the kick start arm. Do Not do a half kick since the kick start pawl does not disengage from the engine until the kick start arm has completed its rotation.
  4. If you turn off the ignition, the ignition will reset and look for TDC when it is turned on again.
 
My Wassell is key on kick and go also. I've no complaints after a year. All I can say is it reminds me of myself with lacquer paint & TT100's. We make our choices and go with it.
 
  • Like
Reactions: baz
With the ignition powered on, you need to turn the engine over slowly two complete revolutions.
The first revolution is so the ignition can learn where top dead centre is.
The second revolution is so the ignition can apply the correct timing/map/advance curve (the unit can hold multiple maps)
----------------------------

Is this correct, this ridiculous involved process in order to start a Commando with a PowerArc ignition?
Seriously, kicking the motor over slowly two times, finding TDC, etc etc?????
Why would anyone put up with such a starting process in the first place?
What am I missing, this just sounds so horrible to be true?

It is not true. Slowly pushing the motor over does give a better charge that is easier to start. That's all and it will help with any ignition.

I am familiar with the processor used in this ignition. It does not remember it's position past a steady pulse stream. Moments after the pulses stop, the location is lost. The only truth is that it requires a complete kick.

Also it can not choose it's own map. It can store an extra map but it must be chosen by the operator. There is no other sensor [load,altitude,temp,tp, ect] to base a map choice on.

Power Arc also wanted me to sell the ignitions. I believe a free steak dinner was part of the deal. I declined. Jim
 
After a Honda 90 in 1968 and then a 350 Kawasaki in 1969 -- both easy starters -- I got a used 1966 Norton in a Matchless frame. GSR 15 I recall. It was in ok condition. And it had a magneto. I guess a previous owner worried about clutch slippage because he'd installed extra springs inside the standard clutch springs. It was a bear to pull the clutch lever and soon the large nut in the center of the clutch plate started to push past its threads. I had the plate re tapped and a larger nut put in and it worked. But I was too young and dumb to just take out the extra inner springs. I thought they were supposed to be there, so I suffered with that till I sold the bike and moved on. But one thing I nevered suffered was starting it. I would tickle the old monoblocks and not even a full kick and it would fire. Now, when I get a little extra cash, I'm buying a magneto again. I hear one thing about one electronic ignition and another thing about the next electronic ignition. They're all wonderful -- until you write in about your problem figuring you must have done something wrong. Then one after another others begin describing the same thing. So until I switch to a mag, maybe I'll just wear a metal shinguard! Yeah, ok, I'll memorize the safe starting proceedure. I can do that -- and chew gum at the same time -- and I won't forget, no I'll never forget, and end up having to wear an ace bandage for a week
 
Hey, don’t shoot the messenger, I’m only going by the documentation and what I have been told by the sole reseller.

I never said it chooses it’s own map, but it does apply a map at each power on, counting the slots in the encoder wheel from the TDC slot.
Yes - they are user selectable (that’s what switching the brown and blue wires does)
You can implement a vacuum switch on the blue wire if you wanted to automatically advance and retard timing based on engine revs.
It can store two different maps but four different advance curves.

If it’s all so wrong - particularly the startup procedure, maybe someone needs to tell Old Britts.

I give up.
 
Nothing wrong with Fred's startup procedure.

Some of the reasons are a bit "imaginative" but not surprising without an electronics background.

A vacuum switch can sense the load and switch between pre-programmed maps. A TP sensor could also be used.

I don't mean to sound condescending but it has been "one of those days." Sorry if I came off that way. Jim
 
I also am NOT a fan of the power arc or for that mater any IGN with the processor in the timing cover. been there done that with tri spark. i am also with jim as to trouble with a constant timing issues with the optical wheel system counting slots as the harmonics of the cam drive and a slack cam chain could cause count issues and through off the timing. I also remember the mallory optical distributor (unilight) and its high failure rate, it got the name of uniwalk and like my experience with trispark you better carry a spare.
 
Status
Not open for further replies.
Back
Top