Mercury 500 Spark for 15-20 seconds at start-up. then nothing.

Mercury 500 1974 SN 4072091 Type II Thunderbolt. I have good spark for 15 seconds and the motor dies. Will stay dead for an hour and re-fire but again dies after 15 -20 seconds. As soon as it died, I tested for spark off of the coil wire and had no spark, I tested ohms of the 3 wires going to distributor according to the manual which were fine. The mercury switch is disconnected. In the manual, I cannot locate a test for the ignition coil which I am suspicious of but cannot verify. When testing the ignition coil I have continuity between the - & + which does not seem right to me. Does anyone have a method to test the ignition coil? Thank you

The two sides of a coil are indeed connected, it's when the magnetic field that's created by that winding collapses that the spark is produced. What is easier to test than what you have already done is to check the voltage at the negative terminal when turning the engine over, the voltage should pulse on and off. It would be interesting to see what would happen if you just turned the key on for the same amount of time and see if there would be any spark. If there isn't then I would be looking at the control unit to see if there is a part that is failing from heat. If you find no spark try getting a can of "compressed air" that's used for cleaning computer parts and turn it upside down so you get liquid instead of gas and spray the control unit, that will cool it, and see if that will allow it to spark for a few seconds.

Just to show you how much fun it can be to diagnose an intermittent problem like that, it could be defective pickup in the distributor if there is one, defective cdi unit or just a poor wire connection that opens just enough with heat to loose contact. If you verify that the issue comes with heat from use then at least you have somewhere to start. Along tne way, test to see how the voltage acts when the engine runs so you can compare that with when it doesn't.

Take the pink/salmon wire off switchbox middle terminal ..I think its the blue one and make sure someone is not applying 12V to red terminal on switchbox. This bypasses switch so make sure it in barrel as it will start and will have to ground the box or choke to kill...

Swagonmaster, I was unsure of how to perform the negative side test you had mentioned. I used a test light and clipped to the + battery cable and touched the - of the coil with the test light probe while it was still connected to the switchbox, 12 volts is there, when I turned the key the test light dimmed out and I did not see a pulse. My test light is more like a test candle with the very old style bulb so I wasn't convinced that I had or hadn't seen a pulse. I then tried it without battery and pull started it, there was nothing there. So instead I disconnected the - wire of the coil from the switchbox and used my ohm meter as a jumper between while it was set up for amp drawtest. I reconnected the power and turned it over, it showed nothing happening for amp draw so that tells me that nothing is going to the coil.

And Faztbullet had a good idea, I actually disconnected the kill wire early on to eliminate the keyswitch possibility. Thank you to both of you guys for responding to my problem.

I have since bought 2 different manuals for this motor, a Seloc and a Clymer. The ignition troubleshooting section of both outline the ohm test for the coil, but when it gets to the trigger and switchbox, both manuals say that both of these components need to be tested by a certified Mercury dealer and then that's all. I am thinking that the AC charge coming out of the ignition driver could be tested while it's spinning with the starter to see if it's sending current out, and there must be a way to trigger the switchbox to make the coil spark but nothing I have found in my research over the past few days has helped. Does anyone have any ideas to pinpoint the problem? I just found an early 500 with a stand alone straight magneto ignition and I'm getting tempted to try switching the ignition out to the earlier style so I can get back to catching fish.

The test of the output of the coil needs to be done when the engine will run and then compared to what you see when the ignition is kaput. Seeing no pulse at the coil likely just means that the engine doesn't run right then. Testing a coil can be hard because the resistance is so low and resistance changes with temperature so that normal variations will make you think that the coil is out of spec. If it works then you can say it's ok.

I see a problem with what I said about testing the power to the coil. On a cdi system the positive is pulsed to create the spark rather than the ground however the net result is the same, pulsed power causes a spark to be created. There should be a timing signal to the cdi unit and a voltage output to the coil resulting in a spark, you just need to see what is missing.

If anyone arrives here with the same problem on a Type II Thunderbolt ignition, I have finally figured this problem out. I found a boat salvage yard selling used parts and called them. I said that I needed ignition parts and ran the problem past the owner. He said that it sounded like a bad coil in my ignition driver and that he would just send me the entire distributor shaft assembly if they had one, I told him the number of my switchbox and told him to send one of those also if he had it. The parts arrived and I tested the old ignition driver and the new one. Here is how I tested them: white lead and red lead of the ignition driver measured 48-50+ volts AC while turning the key as a build-up reading on a digital volt meter (this is the wrong way to test it but gets you in the ballpark), blue and white read 10-20+volts, this was the same with both units. I ran the new ignition driver in my motor and I had the same problem as before, it ran for a few seconds and died with no spark detected immediately afterwards. I had a different switchbox also so I connected that in to the party. This is called swapnostics and it's the lowest form of mechanical problem solving there is but in this case it proved that the switchbox was the culprit. The motor was now making hellacious spark, the switchbox was loose from the frame and hanging by the wiring harness, I had one sparkplug out with a plugwire on it from an earlier spark test. The motor was running on 3 cylinders, I hit the kill switch/key, the switchbox was loose so the ground was intermittent, it wouldn't die, sparks were were jumping everywhere and I was happy again. I pulled a couple other plugwire boots and she died out.
Most of the older outboard mechanics I've talked to all have said that the symptoms I have explained did not sound like a switchbox issue, they said that switchboxes usually quit and stay quit. My problem was that I had 20 or so seconds of ignition and then it would be dead for awhile. I would get spark back after a few hours and then loose it again. As time went on the amount of time having spark would be less and less.
I hope this helps somebody out there someday.