rusefi.com

With winter coming and the bike resting, its time to plan the next overhaul.

This is my motorbike: A thing that has seen plenty of road time and given much love in the form of regular care and many modifications. But now its suffering from a couple of ailments, mainly a much needed valve adjustment and a bent front wheel.
Much work has been done on restoring the wiring harness, which has been the subject of corrosion.

Some background:
The Ninja 250 is a rather ubiquitous motorcycle that has dominated the small displacement bike market in the USA since its inception. Due to lack of competition, it hasn't changed much from the formula of packaging the best of the 80s/90s technology, old yet reliable, in a low-cost chassis.

The model year is 2010, yet no USDM Ninja 250 has been fuel injected, save for retrofitted aftermarket and EU spec OEM parts. Since 2013, the Ninja 300 was released with modern EFI trappings in basically the same chassis as the 250 (even the engine swaps over comfortably), which has led to many possibilities in parts crossover.

The plan:
Replace the carburetor, external coils and plug wires; and shunting regulator/rectifier with rusEFI, coil over plug, and a MOSFET regulator. This will be managed with (hopefully) reasonable cost by taking advantage of the low cost of Ninja 300 salvage parts. The wiring harness will be replaced entirely with one that I will make myself. A few additional farkles will be added such as a new Koso speedometer, removing the analog speedometer gear, and a new exhaust system paired with intake modifications for more power.

Current implementation as of this thread start:
Acquired a set of good condition Ninja 300 throttle bodies and OEM fuel injectors, an OEM MAP sensor, and a set of Ninja 300 ignition coils are in the mail.
Currently planning the rest of the parts bill. Additional questions will eventually arise in this thread as we get into the details of this conversion.

Possible Hurdles to Discuss:
- The engine speed. I haven't seen many examples of rusEFI projects breaking into high rpm, and the 250's displacement necessitates staying in the 6k-13k rpm range to generate respectable power. I've seen the threads about much faster stm processors being available which can handle the events generated by a high speed engine, but I don't know much about that side of development. Maybe I'm totally wrong here and this is a non-issue, I'm clueless as to whether this is an actual issue, but I would like this point to be clear.

- O2 sensor control/feedback. The plan is to use an LSU 4.9, picking that is easy enough, but I don't know what to do about the actual running of the sensor. Native rusEFI CJ125 implementation didn't seem to be quite there yet since I last checked, and I don't want to use an aftermarket controller. Maybe I'll look into using Prometheus?

- Tuning and idle control strategy. This is much later down the road, but good to keep in mind. The Ninja 300 throttle has a set of secondary butterflies which is used in the OEM strategy, but I will probably be taking those out.

Fuel pump plan? You need one, and those consume many watts. Does your alternator / magneto or what ever you have provide enough amps to power a fuel pump?

Do you know your MAP signal output? Some are frequency, most are voltage.

I think we only have one +10k RPM application. We believe we have Miata's which do 7kRPM, but there has recently been some comments about loss of sync at around 6.5kRPM. We don't currently know if those are a hardware issue, or software issue. We expect any software issues above 6.5kRPM can be resolved. I recall the STM can theoretically handle something like 15k RPM applications. It would be a simple matter of software debugging.

Do you know the pulse pattern your crank wheel will generate? The pulses per rotation effects the max RPM that can be decoded.

Are you considering a smaller package? If so do you have a specific ECU connector you think would work well? I ask because I have a single cyl motorcycle, and some day I plan to convert it to EFI. I've been tempted before to make a smaller package, with a water tight connector, and I would specifically intended for a small package. It would probably be fully potted with something like TechnoMelt which would allow it to be smaller than other options. However Prometheus already has a bunch of testing, and might work well enough for your plans. I think Prometheus in the SECU package can be made to be water tight with a little bit of work. As well Prometheus may be small enough that you can find a place for it. I suspect that Frankenso is to large and has was extra's that you don't need, as well the connector is not water tight.

Great first post

I hope engine speed is not too bad since one high-rpm moped was a success. Worth noting that what matters is not just RPM but RPM multiplied by number of tooth on the trigger wheel. 14000x60 is about twice more CPU usage comparing with 14000rpm X 36 tooth wheel. 8 wheel would be 4 times easier etc.

I would suggest to plan external WBO controller at least for the first phase. I love my innovate where the controller is inside the gauge.

Is the bike 4 stroke? I think it's 2 stroke, and I think you have lots of oil out the pipe. AKA I think O2's don't play nice. Even if it's 4 stroke, I think you still have lots of oil. I'm a skeptic that you'll get a proper operating temperature and that you'll be able to keep the sensor clean. I suspect the O2 sensor tuning will be more problems then helpful. If you have make it to WO2 tuning, then you've moved along a long way from where you are now. I think you'll want to tune more with a thermo couple than a WO2. However that's a long way down the road.

The engine is definitely 4-stroke. Right now I can think of at least 5 other EFI conversions of this motorcycle that use LSU 4.2s and 4.9s with no issues, so I don't think contamination of the sensor is a concern.

The fueling plan took a while to formulate as previous EFI examples using this drivetrain mostly used either an Ecotrons kit or automotive/racing parts.

The most relatable example uses ATV parts, a Suzuki LT-R450 Quadracer fuel pump. However, these units are expensive even when used. After much research I came upon a Honda part that is the same fuel pump for only a third of the Suzuki price new.

The fuel pump will be a new Honda TRX700XX unit, part # 16700-HP5-602. The Ninja 250 equipped with the LT-R450 part reportedly runs well and with room to spare after a few light bulbs are changed to LEDs, so I'm not too worried about power draw at the moment. The fuel pressure regulator will be a new or very lightly used 2004-2006 Yamaha R1 unit which is specified to provide 2.96 bar (42.9 psi) pressure.


As for the signaling side, the Ninja 300 MAP sensor outputs voltage and I haven't thought of the crankwheel pattern yet, though I ought to get to that bit soon.
As for packaging, I haven't set any dimensional constraints, but weatherproofing and NVH-resistance would be nice. I always end up spending hours staring at epoxies/potting, so I'll save that for later. I will definitely look deeper into what Prometheus/SECU have to offer.

Does your pump have an internal regulator, or is the external part going to have a return line? I know some motor cycles have internal regulators. As well is your regulator going to have the line which goes to the manifold? AKA the regulator changes pressure such that the injectors always see 42PSI pressure drop. Keep in mind if you are at high RPM, and snap the throttle shut, you could get an additional 14 PSI drop across your injector. This increased drop will cause more fuel than is expected to be dumped into the cyl. Same goes for idle --> WOT. It would run lean if your pressure regulator doesn't compensate and adjust the fuel rail pressure.

The pump is meant to be used within a fuel return system. I found that the R1 regulator is designed for a fuel rail mounted application, so that's not going to work without an adapter.

I can't find any Bosch stuff that fits my requirements either, so right now I'm looking at other ATV parts. I'm either going to pick up a LTR450 FPR that was originally meant to be used with the fuel pump or find an alternative.

Just pulled the trigger on the LTR450 FPR.


The alternative would have been to buy a late 90s Japanese Automotive OEM part, such as this: https://www.rockauto.com/en/moreinfo.php?pk=482587&jsn=492
A bunch of common mid-size Japanese sedans (accords, camrys, es250/300s) used this particular part, and they all specify 3 bars of fuel pressure. So this regulator plus a fitting or two may have been the way to go. But the price was right on the Suzuki OEM part, and it has been used in a previous EFI conversion, so I'll stick with a better known variable.