Air lock in fuel line?

Will-do I think I'll try a swap of both coils, which means cutting them off the harness down near the gland where the leads head inside the crankcase, then fit the XJ coils with enough spare wire to connect via a weathertight 3 terminal sumitomo connector pair; plenty of room down there to locate it.

 

I did an interesting experiment last night; put one XS400 coil and one XJ coil on the milling machine table. I mounted a short length of steel on a length of threaded which I chucked so the machine would spin the steel bar across both coils, similar to the ignition trigger bar which bolts to the crank. I put the same value load resistor across both coils and hooked the two oscilloscope channel probes across the resistors.

The idea is to compare the XS vs XJ coils, see if the XJ's is close enough to use, without risking damage to the CDI inputs ie what if the XJ's signal is double the voltage etc. I don't have matching probes so the readings aren't properly comparable yet but so far it looks like the XJ's are effectively close to the XS's as far as signal profile and amplitude.

I have a pair of cheapie import probes on the way so I can better compare the two signals. If it all works out I think that means trigger coils from at least some other bikes will be direct replacements. The wire colors are the same though the harnesses will be different so using them will be a cut and splice sort of deal... but XS400 stuff is so often unobtanium/$$$ alternate parts are most certainly welcome.

 

I finally got every sorted and up on the milling machine table, with the steel bar prepped to swing past both the Seca coil (left) and the XJ coil (right)




I put a 4k or so resistor across the leads of both coils, and connected a scope probe to each. With the mill spinning the bar a 200rpm or so, I got the following traces. Seca's coil in yellow, XJ's in blue;



So it looks like the XJ trigger coils will drop right in. The voltage swing is about the same- I was concerned about double the voltage etc leading to the possibility of damaging the CDI. From previous tests with the Seca coils in the bike and operating, I found the trigger signal would substantially drop in amplitude with increasing engine temps- will be interesting to see how the XJ's react.

The XJ coil is about 600 ohms vs the Seca's 120ohms or so. The XJ coils are similarly about 5 to 6 times the inductance. So the next step is to cut the Seca coils off and wire in the XJ's via a waterproof connector; I have a 4 terminal sumitomo pair which should fit OK. Probably the win is the next warm day to put the Seca harness back in, then cut into it and fit the XJ coils & connector with everything in place.