I am stumped- Shower of Sparks

Chip Sylverne

Chip Sylverne

Final Approach
Joined
Jun 17, 2006
Messages
5,891
Display Name
Display name:
Quit with the negative waves, man.
Engine is not starting properly when cold. Here's what's been done:

"buzz box" has new contacts and capacitor. Coil tests within spec ( but barely) and box buzzes on the bench when 12v is put to it, installed when key is turned, and buzz tone changes indicating load as engine cranks.

Ignition switch has been disassembled, contacts cleaned and continuity tested good.

All wires from switch to box, from box to mag traced and continuity tested good, proper function and no shorts to ground. Buzz box gets full voltage when engine cranks.

Left mag inspected, retard points changed, connectors checked. Condensor changed. Advance points open at 25deg BTDC, retard points open at TDC. A short to ground was found through a paper insulator in the retard points connector housing, and that was fixed, allowing the retard points to function normally. ( I was convinced that would be the gremlin!)

Mag E- gaps are perfect, and engine timing is dead on. Once started mag drop is 50 rpm per side, and smooth as silk. The ends of the p- lead connectors are nice and clean. The copper connector for the switch lead in the mag could be shinier, but it's more than clean enough, and the buzz box indictates by the change in tone that the contact to the primary side of the mag coil is good.

And yet, the engine only kicks off when the key is released from the start to both position...this is driving me nuts. There is no kickback, and the engine spins nicely with the starter engaged.

Any thoughts would be appreciated.
 
Last edited:
"Coil tests within spec ( but barely)"

That may be your answer. It may be doing OK on a bench test but not putting out enough voltage when under the load of cranking the engine. If you can get hold of a spare they're easy to switch for a test.
 
If this is just when cold, have you tried giving it some more prime fuel and a a few extra moments to vaporize?
 
Norman said:
"Coil tests within spec (but barely)"

That may be your answer. It may be doing OK on a bench test but not putting out enough voltage when under the load of cranking the engine. If you can get hold of a spare they're easy to switch for a test.
Click to expand...

I don't know how much voltage it would take to excite the primary side of the mag coil. The output of both my box and a known good is +/- 4v. I used the MK 1 eyeball to compare the "sparkiness" of the points of my buzzbox against that of a known good one on the bench when working with the covers removed, and it seemed pretty close to even. Unfortunately the connectors were different so I couldn't throw the known good one in my bird to see if it worked, and I didn't have jumpers handy.

I had the understanding if the box was buzzing, that it was good. Resistance spec is less than 1 ohm, mine comes in at .8 ohms. TCM doesn't sell replacement coils. The only solution is new box @ $400 or 25 ft of magnetic wire and rewind the old one. That's prolly the next step.
 
Chip Sylverne said:
I don't know how much voltage it would take to excite the primary side of the mag coil. The output of both my box and a known good is +/- 4v. I used the MK 1 eyeball to compare the "sparkiness" of the points of my buzzbox against that of a known good one on the bench when working with the covers removed, and it seemed pretty close to even. Unfortunately the connectors were different so I couldn't throw the known good one in my bird to see if it worked, and I didn't have jumpers handy.

I had the understanding if the box was buzzing, that it was good. Resistance spec is less than 1 ohm, mine comes in at .8 ohms. TCM doesn't sell replacement coils. The only solution is new box @ $400 or 25 ft of magnetic wire and rewind the old one. That's prolly the next step.
Click to expand...

Looking at the spark under atmospheric pressure vs compressed cylinder pressure can be quite misleading, that's why plugs are tested in a pressure vessel.
 
The "P" leads are reversed.
 
Henning said:
If this is just when cold, have you tried giving it some more prime fuel and a a few extra moments to vaporize?
Click to expand...

I can get it started by feeding in some primer while cranking, but it takes quite a few blades. And again, it's not that it won't fire, it's that it fires on key release, when the right mag is ungrounded and the left retard mag is taken out of the circuit.

I disconnected the starter solenoid and pulled #1 spark plugs. Using a timing buzz box is how I determined the advance and retard points are opening and closing properly. If someone turns the key to start, I get a weak spark from the left mag as I pull through TDC. That's why I rebuilt the SOS box. But even with a new cap and vibrator points, it's not doing much better.

Maybe it is that coil. It's kind of the last thing left.
 
Chip Sylverne said:
I can get it started by feeding in some primer while cranking, but it takes quite a few blades. And again, it's not that it won't fire, it's that it fires on key release, when the right mag is ungrounded and the left retard mag is taken out of the circuit.

I disconnected the starter solenoid and pulled #1 spark plugs. Using a timing buzz box is how I determined the advance and retard points are opening and closing properly. If someone turns the key to start, I get a weak spark from the left mag as I pull through TDC. That's why I rebuilt the SOS box. But even with a new cap and vibrator points, it's not doing much better.

Maybe it is that coil. It's kind of the last thing left.
Click to expand...

P leads are reversed or the key switch is bad (known problem).
 
Henning said:
Looking at the spark under atmospheric pressure vs compressed cylinder pressure can be quite misleading, that's why plugs are tested in a pressure vessel.
Click to expand...

Not talking about spark plugs. Talking about the sparks between the contact points of the vibrator box. I test mag coil output by pulling the plug lead and holding the spring about 3/16 away from engine ground. If it jumps that, it's going to fire the plug.
 
Chip Sylverne said:
N I test mag coil output by pulling the plug lead and holding the spring about 3/16 away from engine ground. If it jumps that, it's going to fire the plug.
Click to expand...

No, not always, What color was the spark? nice and blue or kinda red?
 
Chip Sylverne said:
Not talking about spark plugs. Talking about the sparks between the contact points of the vibrator box. I test mag coil output by pulling the plug lead and holding the spring about 3/16 away from engine ground. If it jumps that, it's going to fire the plug.
Click to expand...

Not necessarily. Under compression the spark needs more snap than that. It should jump a half-inch or more.

What was the fix of the shorted capacitor on those retard points? Did you just take out the capacitor, or replace it? An absent capacitor will kill spark just as surely as a bad one.

Dan
 
Tom-D said:
which one is grounded when the key is in the "L" position?
Click to expand...

The issue isn't which one is grounded in the run positions. The issue is which are grounded in the start position. The right is grounded, the left and left retard points are both grounded until the proper spots in the firing cycle. The advance points open, but the mag does not fire because the retard points are still closed snd in series. The retard points open at TDC, and the BO output from the starting vibrator goes through the start switch "Switch" p-lead to the left mag coil primary. This has been confirmed. If the p-leads were reversed, the vibrator voltage would be going to the right p-lead. The left mag switch and retard collars are different sizes, so that can't be mixed up.

Ignition switch is good. Dissasembled it and the contactors are like new.
 
Last edited:
Tom-D said:
which one is grounded when the key is in the "L" position?
Click to expand...

Dan Thomas said:
Not necessarily. Under compression the spark needs more snap than that. It should jump a half-inch or more.

What was the fix of the shorted capacitor on those retard points? Did you just take out the capacitor, or replace it? An absent capacitor will kill spark just as surely as a bad one.

Dan
Click to expand...

Replaced. Both the mag capacitor and the buzz box cap. Buzz box cap is new, mag cap is known good.
 
Tom-D said:
No, not always, What color was the spark? nice and blue or kinda red?
Click to expand...

Right now its blue, but weak. That's the problem. I don't have a good explanation for it, because by rights it oughta be a lightning bolt!
 
Dan Thomas said:
Not necessarily. Under compression the spark needs more snap than that. It should jump a half-inch or more.


Dan
Click to expand...

Even off the SOS box? 'cause if so, I'm not getting near that.
 
Have you tried to see spark at the plug during start? sounds to me like there is a high resistance connection in the wiring some place. do a voltage drop test of every connection in the entire circuit.
 
Chip Sylverne said:
Even off the SOS box? 'cause if so, I'm not getting near that.
Click to expand...

Yep, I just went through a no start on a freshly overhauled engine (fuel injection system came in set waaaaaaaaaaaaay to rich) so I checked out the vibrator, pulled the wire off the starter, positioned the engine with the retard points open and hit the key, it jumped nearly an inch with a bright blue spark.

You say you can hear it change pitch when the starter hits compression, possible that a fault in the starter or its circuit is drawing the voltage down too low?
 
Skylane81E said:
You say you can hear it change pitch when the starter hits compression, possible that a fault in the starter or its circuit is drawing the voltage down too low?
Click to expand...

According to TCM, the vibrator pitch changes should happen when the retard points open because the vibrator signal is no longer grounded through one or both sets of points, but instead is grounded through the primary side of the mag coil which causes some amp draw. If the sound remains constant, it's supposed to indicate the retard points aren't opening. Makes sense.

The starter spins the engine just fine. It's a new Skytech. The cables have been upgraded to copper. A voltage drop test showed the voltage at the buzz box is around 9-10v with the engine cranking, which is still greater than the 8v min required. With the starter out of the circuit, the voltage is 12.1 at the in terminal.

One of the Bendix service pubs says in a pinch, you can hand prop if you can put two d- cells together and get the buzz box buzzing, so it must not take much voltage to be operational.

I agree it sounds like a high-resistance connection, but every connection in the starting circuit has been taken apart and cleaned. Could possibly be the spring plate p-lead connector inside the mag. But if you turn the key to "off" momentarily at idle, both mags ground. Is there a good way to measure the resistance across that terminal with the p-lead connected? Remove the left ground connection at the ign switch and measure that to an engine ground?
 
Any of you know what the voltage output of a healthy buzz box should be? I can't find a spec in any of the manuals. At least that would tell me where I stand up to this side mag connector. Or failing that what the minimum primary coil voltage input required to excite a good spark out of the secondary?

Maybe my p-lead has continuity, but some broken strands that are causing a voltage drop.
 
Chip Sylverne said:
Any of you know what the voltage output of a healthy buzz box should be? I can't find a spec in any of the manuals. At least that would tell me where I stand up to this side mag connector. Or failing that what the minimum primary coil voltage input required to excite a good spark out of the secondary?

Maybe my p-lead has continuity, but some broken strands that are causing a voltage drop.
Click to expand...
"P" leads only make continuity to ground. there is no voltage involved. they have exactly the same electrical potential as the points.
 
When it runs great after start and give good mag drops it is not a mag issue, it is a start issue, that is why you need to pull a spark plug lead and hold it off a good ground and then crank the engine see the SOS is sparking the plugs during start.

If it not, the ignition switch is at fault, because you know that it has good power.

If the spark from the SOS is weak during start, but sparking just the same, it is the coil in the SOS. because the mags are working.
 
Tom-D said:
"P" leads only make continuity to ground. there is no voltage involved. they have exactly the same electrical potential as the points.
Click to expand...

Not in SOS systems. The voltage output to excite the primary side of the mag coil from the SOS vibrator goes into the the left mag through the "switch" p-lead when the ign. switch is in the start position. That is the same lead used to ground the mag when the switch is off.
 
Tom-D said:
When it runs great after start and give good mag drops it is not a mag issue, it is a start issue, that is why you need to pull a spark plug lead and hold it off a good ground and then crank the engine see the SOS is sparking the plugs during start.

If it not, the ignition switch is at fault, because you know that it has good power.

If the spark from the SOS is weak during start, but sparking just the same, it is the coil in the SOS. because the mags are working.
Click to expand...

You don't need to crank to test the SOS system. All you need to do is disconnect the starter solenoid, position the #1cyl so the retard points are open, pull the lead from the left mag off the plug and turn the key to start and hold it there. When I do that I get spark, it's just weak, even after replacing the buzz box cap and points.

Which means either there is high resistance getting the vibrator signal into the mag primary resulting in weak spark (connector or wiring, because switch is good) or the output of the vibrator is poor, even though it gets sufficient voltage all the components meet spec, though not by wide margin.

I'll see if I can borrow a known good buzzbox and jumper it in. If I get good spark, I'll rewind that coil.
 
Umm, the buzz box shouldn't be hooked to the P lead, it should be connected to the retard post.
 
Chip Sylverne said:
You don't need to crank to test the SOS system..
Click to expand...

YES YOU DO, because the ignition switch is in a different position. It may not be powering the SOS.

OR

the starter circuit is drawing all the power and the SOS is not getting the power it needs.
 
Last edited:
Skylane81E said:
Umm, the buzz box shouldn't be hooked to the P lead, it should be connected to the retard post.
Click to expand...

Ummm, no. The bo lead from the box goes to the switch, the switch sends the signal to the mag via the switch/ground lead contantly with the switch in the star position. The retard points ground the signal in parallel with the advance points until tdc, when they open and the bo signal goes through the primary.
 
Last edited:
Tom-D said:
YES YOU DO, because the ignition switch is in a different position. It may not be powering the SOS.

OR

the starter circuit is drawing all the power and the SOS is not getting the power it needs.
Click to expand...

No you don't. You'll hear if the buzz box is getting juice from the switch, because it'll be buzzing. Or you can put a VOM on the " in" terminal and see the voltage with starter connected and/ or disconnected. If the starter is disconnected, it draws nothing away from the buzz box.

Either way, the mag sparks from the buzzbox when the retard points are open, regardless of whether the engine is actually turning.
 
Chip Sylverne said:
No you don't. You'll hear if the buzz box is getting juice from the switch, because it'll be buzzing. Or you can put a VOM on the " in" terminal and see the voltage with starter connected and/ or disconnected. If the starter is disconnected, it draws nothing away from the buzz box.

Either way, the mag sparks from the buzzbox when the retard points are open, regardless of whether the engine is actually turning.
Click to expand...

I'm confused, Who is stumped, who is trying to help, and who is arguing?

I'm gone.
 
Tom-D said:
I'm confused, Who is stumped, who is trying to help, and who is arguing?

I'm gone.
Click to expand...

Sorry Tom, not trying to p*** ya off, just telling you I've already been down the road your suggesting.
 
Just for future reference for anyone interested, I borrowed a brand-new TCM Vibrator and looked at BO output with my o-scope. Output is a pulsed 2v DC with a full 12v input. My box outputs about 1.5v pulsed, so it's a little shy.
 
I've got it!

It seems that you can increase the amount of voltage the vibrator is sending to the left mag in starting position by bending the tab upon which the lower vibrator contact assembly in the vibrator is mounted. If you bend it slightly so there is more pressure on the point faces while the vibrator is off, you can almost double the voltage output. There is a point of diminishing return, bend too far and the unit vibrates, but the signal is very dirty with a lot of transient voltage. You want a nice clean pulse. You can do it by ear, listening for good, strong even tone, but an oscillosope takes out the guesswork.

So, long story short is changing the contacts and capacitor isn't sufficient. You need to "tune" the voltage output to compensate for any change in coil resistance. Spark easily jumps 1/2" or so now!

Or you can part with $400 for a new box.
 
Last edited:
Shower of sparks makes a real hot spark your plug wires have to be in good shape to take it might use a plug wire tester to confirm the wires can take the hot spark going through them. When you release the start switch putting the mags in the run position (normal lower spark level) the plug wires work and the engine fires. The Mag cap and rotor also need to be in good condition more so than impulse type, shower of sparks are great systems but need a little more care.
 
You must log in or register to reply here.
Back
Top