New MS project - KZ400

[BLSully]

Hi folks,

I've been wanting to MS something for a few years now, and I think I'm finally going to tackle it this winter. I have a couple of old Kawasaki KZ400s, which are vertical twins with 360 degree firing. Makes spark pretty easy. Not rocketships, but pretty bulletproof little bikes and very simple machines, so I figured it would make a good intro to a FI conversion.

Air and fuel will be handled by a pair of Suzuki Burgman 650 throttle bodies i picked up on ebay for a song. They're a bit narrow compared to the KZ's intake spacing, but a small manifold with angled runners should do the trick. 2x32mm butterflies, and has 12ohm 230cc injectors (part # cross references to the same injector as an 03 GSXR 1000 which i know people commonly use), fuel rail, TPS, and it looks like a fast idle valve with balancers. Still investigating that part. Only 2 wire so I don't think it could be a stepper (but my electronics knowledge is lacking).

Fuel pump duty looks like it'll be handled by a pump assembly from a Honda TRX420 ATV. Pump, 43psi reg, and swirl tank all contained in one tidy package a bit bigger than a soda can. This should make the lines easy as there's no need to run a return line back to the tank. Also, reportedly, the pump has a very low 1-1.5amp draw which will come in handy with the universally bad charging systems Japanese bikes from that time had.

Ignition I'm very undecided on. I don't really mind sticking with wasted spark (OEM is single points, single coil, wasted spark), but i'm not sure if I should switch to a trigger wheel/sensor or just lock the advance on the points. May leave ignition alone to start and try to run fuel only for now since this is my first conversion. If(when) I do ignition, though, something like an LS2 coil w/ 2 outputs would be pretty sweet if someone knows of such a coil. I like the built in driver (ignitor?). Reading the manual that seems like it's good for reducing electrical noise issues.

I'm sure I'll have plenty of questions in the future, and will attempt to keep up with this thread as the project progresses. So far I have the TB's in hand, and made an offer on ebay for the fuel pump. Next step is analyzing the 3 motors I have and deciding which one is going to be the lucky candidate.

[Matt Cramer]

It often works pretty well to start with fuel only and then add ignition control on a conversion like this.

[old guy]

Just use two LS1 coils and fire them off one output. People do it all the time. You might also look at some of the Japanese coils which look like a coil on plug with another plug wire coming out of them which hooks to the other plug.

[BLSully]

Thanks for the input and suggestions

I have the fuel pump on its way, and found a local engine shop that is going to check out my motors. Want to start with a known good top end, and I have roughly 3 complete engines to pick from. Once that's sorted out, I have a sketch of my intake manifold to marry the throttle bodies up to the KZ head. Need to render that in modeling software and figure out how I'm going to make it. Two plates with angled runners... Shouldn't be too tricky.

Microsquirt will probably get ordered in a month or two, giving me some time to get it all sorted before we hit early spring riding

Discussion question for the evening... How do you best read engine temp on an air cooled bike? Seems like mounting to the fins is likely inaccurate, let alone how to physically secure it, but maybe "good enough"? What about substituting oil temp? The kz has an external line running to the head, I think I could fit a temp sender at the joint where it exits the case. Temp and MAP are the only two sensors I haven't accounted for just yet.

[ArttuH]

For engine temp you want to measure head temperature on intake side, in theory. I have found out that outer edges of the head tend to cool down too much in cool weather or rain. So for example edge of the valve cover wasn't good location for the sensor. The cylinder block seems to follow head temperature well enough and it's usually less affected by ambient conditions. Now I'm using the cam chain tensioner bolt in the middle of back side of the block as mounting point for the sensor. I'm using a small NTC thermistor glued on a ring cable terminal as sensor. That can be mounted under any suitable engine bolt. Oil temperature often rises much slower than head temperature so it isn't perfect source for engine temp.

I would investigate a bit more for better suitable throttle bodies before making an offset manifold. There are good changes that you can find something that fits in right away or with minor modification. But this is personal preference of course. If you have tools and skills to make a manifold easily then it might be viable option as well.

[BLSully]

ArttuH, Makes perfect sense about the temps.

The misalignment of the TB's isn't much really. I probably over emphasized how much the angle will be. The center to center difference between the head and the TB's is about 10mm. I measured it the other day and have it written down, but don't remember exact numbers off the top of my head. It's definitely not bolt on, but it will not take much work to make them fit. The TB's i have are such a good fit otherwise (compact, well matched bore size, etc) I have a hard time believing I could find something more appropriate. There's a polaris 500 snowmobile set I am curious about, and looks like could be a closer fit, but the bores are larger and still not guaranteed correct spacing.

[ewflyer]

Here's a possible solution for your throttle body needs: the Kawasaki Ninja EX300 fuel injection throttle body.

I believe this throttle body would match perfectly with your Kawasaki KZ400. The EX300 has its cam drive chain in between the cylinders (Kawasaki standard engineering) so the throttle body spacing will be wider. The bore-size on the EX300's throttle body will be large enough to provide for good breathing at the KZ400's (much lower than the EX300's) engine redline speed while also being small enough to be very tune-able for idle engine speeds and it should allow good off-idle acceleration characteristics. You could even experiment with controlling the secondary butterfly assembly.

http://www.ebay.com/sch/i.html?_odkw=&_ ... acat=10063