1982 DOHC not running on right cylinder

[sipma02]

Fuel. Air. Spark. Timing. Its simple right? This is the XS400 set up with the Arduino ignition, and its actually going decently well. The left cylinder is working great. The engine starts more or less right up on just that cylinder! But for some reason, when I hook the other coil up, it WILL NOT start. Or even pop. This engine/carbs ran great 3 years ago. I currently dont have an airbox or exhaust on it. Ive re-cleaned the carbs multiple times. Assuming these are 180º engines, i only have one pickup sensor triggering both coils. (wasted spark) Ive swapped coils, spark plugs, Ive taken both spark plugs out and made sure both are sparking. They are. Put my thumb over the spark plug hole. Plentyyyy of compression. When i squirt a bit of gas in the spark plug hole, I miiight get a backfire through the carbs. So something is going on in there.Then, I disconnect the right cylinder and it fires right up (as well as an engine can with one cylinder.

I seem to remember reading that these bikes have an asymetrical airbox, and different sized jets in the carbs, is that true? Is that causing my issue? The thing that confuses me is that when i have both coils connected, it WILL NO do anything. As soon as I disconnect the right one, it fires right up. Which makes me think its a spark issue.

Any ideas? Or anything I can try?

 

[markmakeitso]

Hi. A 180 crank can't run both cylinders directly off one pickup, so your other cylinder is probably sparking at a completely wrong time. That seems likely to be your problem. Popping up through the carb also points to this, since it might be igniting fuel when the intake valve is open, rather than when both valves have been closed for a little while to compress the mixture.

Another possible issue when one cylinder fires but not the other is the vacuum petcock possibly leaking into one runner, which can wet the plug enough to prohibit starting. Running both coils on one pickup could be tough on the ignition system also, where plugs might fire when open to air, but struggle to create adequate spark in the compressed mixture.

 

[sipma02]

Wow, so embarrassing! I don't know why I didn't think of that. Just a majorrrr brain fart. Thanks! I will definitely give that a go and report back.

Sweet to see you're around the cities! Where from?

 

[markmakeitso]

That's still my guess on your problem. I didn't know about the differences between the SOHC engines like on the two I've worked on and your DOHC one though.
If I were experimenting with the ignition system I'd be tempted to go for some modern coil on plug or coil near plug parts. Lots cheaper than the 70's tech that Yamaha was churning out. Ford Escape COP modules are under $10 each if you don't mind trying eBay specials, and GM truck coil near plugs are just a little more. Some intelligent coils allow just a TTL level control input, so the bulk primary current is handled internally. Not sure how you're switching the coils, but that might be a nice improvement too. I've done some dabbling in Arduino and engine management in my travels, and controls and data acquisition is kind of my professional niche.

I'm actually on the edge of downtown Minneapolis currently, where you at?

 

[sipma02]

Funny you should mention those COP modules. A friend of mine just gave me one to experiment with. I definitely plan on converting most/all electric bits on the motorcycle to some modern alternative.

Thats awesome you've worked with an Arduino + engine management stuff! I'm really excited about this project. I'm just stoked that this crazy idea is even working. Its been such a blast to experiment with! I'm near the west end of Uptown, Hwy 100 & Mtka Blvd. Probably not too far away from ya!

 

[markmakeitso]

Until recently I was working on a fuel injection setup for an old 4 cylinder pickup truck using an Arduino. Had to sell the truck when I moved, so not sure if/when I'll keep going with that. Also a school side project where we replaced a broken ignition system on a small turbine engine with a PIC (another family of microcontrollers) and a COP. That was similar to your current programming, just a "while powered: turn on, short delay, turn off, longer delay, repeat."

If the DOHC bikes use the separate voltage regulator you could also measure system voltage with an AI pin and use a PWM output to control the field coil power. Can even keep the field coil completely unpowered until above a certain RPM to save battery power and make starting slightly easier. And use LEDs to flash with pickup inputs for troubleshooting. And a serial output of measured values. Tons of cool improvements you could add to the bike.

 

[sipma02]

Wow. Thats actually a very interesting idea. I love the voltage regulator idea. I have been using led's flashing as a timing light! it worked well. Agreed though. Theres so much potential!

Currently, I'm using hall effect sensors cuz theyre cheap and easy. Ideally, I'd like to use a variable reluctance sensor. Those are basically a mini AC generator, which would create a voltage, and a larger voltage if the engine is spinning faster. Then I could use the frequency (when the sensor is triggered) and the varying voltage to calculate advance (voltage) and timing (frequency). Unfortunately theres a bit of a learning curve and special circuitry to use a VR sensor, and Im getting along alright with a hall effect. Super fun to work on though!

See my build thread on the ignition: http://www.xs400.com/threads/82-maxim-dohc-400-electronics-retrofit.15941/page-4

 

[markmakeitso]

I've gone the other way. The engine I was working on for in-cylinder pressure testing has an AC output for the crank pickup, and I just wanted a 0 to +V signal rather than the +/- signal. That's fairly common, and there are dedicated chips and boards out there for getting to a square wave from an AC. Both your analog and digital inputs on the Arduino will be happier with positive only voltages.

You do you, but calculating RPM based on the peak of the AC seems very difficult. It may vary quite a bit even with an engine running at steady state. Next time I'm testing it might be possible to make a quick application to calculate the max and min peak values. I'd guess a 25% difference is quite possible. You could instead calculate RPMs from the time from one pickup event to the next by using a timer of adequate duration.

I'll check out your build thread, interesting stuff. Do you have a scope? I have an older benchtop one in the garage that you could have. Could get you going with a PC scope for pretty cheap also. Those would be incredibly useful for your work.

Have fun,
Mark

 

[sipma02]

Sorry for the slow replies, I dont have notifications setup for this thread. I think I do now! Anyway, I'd definitely be interested in the scope, and maybe to meetup and discuss strategy. I'll PM you. In any case, its interesting stuff and fun to tinker with!

Josh

 

[sipma02]

Back on thread topic, I cannot figure out this engine! I had it sort of running on one cylinder last week, but now I cant seem to get it to even pop! What am I missing? The carbs are cleaned, 3 times. Ive air compressor blown them out very carefully. Ive confirmed that I'm getting spark (by holding the spark plug on the engine block–its a strong and consistent spark). Ive put my finger over the spark plug hole (obviously w/o a spark plug in) and the compression easily blows past my finger. I tried the other cylinder Ive even dribbled some gas in the cylinder via the spark plug hole and NOTHING. Not even a pop/combustion! Its so frustrating!!!

One thing to note is that my airbox/filtes are off, and the exhaust is off. Obviously this causes a very, very lean condition. I also am only attempting with one cylinder. But shouldnt I at least get a bit of a reaction from the engine?! However, like I said, last week I have it "running" (if you can call it that) on one cylinder:

Any ideas of what I can try next? My goal is to simply get the engine to pop or combust just to see if there is life. I dont even know what to try next...

 

[sipma02]

Also probably worth noting is that the engine has little to no oil in it. I have the side cover off and over the past few days of testing and cranking it has evacuated a fair amount of oil. Could that be it?

I also can easily smell gas in the first 5-10 second of cranking it from sitting overnight. Stands to reason that gas is getting to the cylinder.