Weighing Fueling Options

[Syscrush]

Rob also provided some really valuable insight about fitting the fuel pump, filter, etc. into the tank on a bike with a saddle-shaped tank like the CBX. He used the setup from the early EFI Ducati Monsters before they went to a plastic tank. Hunting around a bit, I found that (perhaps unsurprisingly), the classic-look EFI Triumph bikes are similar. My plan is to go with the Ducati or Triumph parts and modify the bottom of the tank enough to include a large removable panel (like that Triumph setup) so that everything can be fitting from below so I can retain the twin-shock's distinctive gas cap, and keep the fuel pressure regulator and filter inside the tank.

[Syscrush]

And just because I think it looks cool, here's the scan of one bank of carbs:

[Syscrush]

I'm pretty intent on an internal pump and regulator (which will require significant tank modifications), but for others considering an EFI conversion and wanting to leave their tank stock, there's an OEM Honda option for an external pump with an integral regulator and sock-type filter - it works like a surge tank or swirl pot. I came across it today. There were units with the same external dimensions used for both the TRX ATVs and supposedly some larger cruisers. Here's a bit of info about the ATV one for those who are interested:

http://forums.banditalley.net/suzuki-ba ... #msg107979

I wasn't able to find concrete info or part number for the version meant for bigger bikes, but maybe this lead is valuable for others who want to go that way.

[Syscrush]

There's been some discussion here about ways of getting a cam phase signal for running sequential ignition. Fitting some kind of trigger wheel or other sensor to the cam presents numerous problems. I looked into adapting the technique from the Saturn Compression Sense Ignition which infers cam phase from the timing of signals on the two separate channels of a waste spark system - that would be good as it does not require any additional sensors under the cam cover, but the electronics of it are at least a bit tricky.

The builder of that crazy V12 308 Ferrari suggested I look at making a cam phase signal from the MAP#1 sensor, and that lead me eventually to this product:

http://efi.ttrignition.com/quadramap.html

It's a single module that takes the inputs from multiple MAP sensors (one from each ITB) and outputs the current strongest signal so that the ECU sees the output as a single MAP sensor like you'd have with a traditional plenum setup. I've seen other such solutions before and was already planning to use one (as documented in this thread).

HOWEVER, this one also sends a cam phase signal when MAP#1 is the strongest signal - so I no longer need to worry about fitting a degree wheel or hall effect sensor somewhere on my cam, and I also don't need to worry about that tricky compression sense ignition stuff.

I've contacted the seller and they say that the 6 and 8 cylinder versions are designed but not yet in production. Hopefully they're able to come through with a single off-the-shelf solution for these 2 problems.

[Rick Pope]

Picking up your timing signal from any source other than the crank induces a bit of variability because of the inherent slack/wear in the chains. Does it matter? Probably not.

[Syscrush]

Timing signal still comes from the crank, but with only crank position, you can't tell if a given cyl is on its intake stroke or its power stroke.

By combining a "cam phase" signal with the crank position signal, you can tell not just where the crank is, but also what stroke each cylinder is currently in. That cam phase signal does not have to be as accurate as the crank position. In fact, once the ECU has had a lock on the cam phase signal for a while, it doesn't even care if that signal goes away - as long as the crank position signal remains consistent.

[EMS]

The controller requires the exact engine rpm and the crank position to determine the appropriate fuel amount and the correct ignition point. To provide this, a gear was installed into the housing of the ignition pulser. This gear has a notch which sends the data to the controller via an inductive sensor

[Syscrush]

The controller requires the exact engine rpm and the crank position to determine the appropriate fuel amount and the correct ignition point. To provide this, a gear was installed into the housing of the ignition pulser. This gear has a notch which sends the data to the controller via an inductive sensor

Thanks for this. There are a few good examples of nice custom trigger wheels fitted to the CBX in place of the stock ignition signal generator. My plan is to go with something like the nice 36-1 wheel made by Mike Lynch. That should give the ECU excellent crank position information.

That would be enough to run waste spark ignition and batch or semi-sequential fuel. By adding a cam phase signal from the MAP sensor on cyl #1, the ECU will have enough info to run full sequential fuel injection. I will keep the waste spark system because I don't want to or see any need to run 6 separate coils. If I liked the looks of the coil-on-plug systems I would go that way, but I don't think it's right for this bike.

[Syscrush]

So, I've ordered the fuel pump assembly from a Triumph Rocket III. That bike makes a claimed 148 crank hp, so I'm pretty confident that it will have no problems feeding my engine.

The assembly as it's used in the Triumph can't be expected to fit the CBX tank:



But I chose it because it includes a pickup filter, pump with a rubber bumper to keep it clanking around in the tank, fuel level sender, filter, and internal pressure regulator - and it's modular. It should be possible to modify the mounts and plumbing to make it fit the CBX tank properly.

The big challenge will be the interface between this assembly and the tank bottom - which will have to be custom fabricated. That will mean cutting an irregular shape from a non-flat piece, fabricating a flange for reinforcement, and fabricating a matching base for the pump assembly, and coming up with a custom gasket. The oddball shape will probably require some CAD/CAM techniques including more hi-res 3D scanning and some 3D printing, custom casting, or CNC machining. For the gasket, I'm hoping I can make something work with off-the-shelf O-rings. Time will tell.

EDIT to add: The more I think about this, the more I think it will be relatively straightforward to do with fiberglass. I can make the base plate using the bottom of the tank itself as a mold before cutting it open. Once cut open, I can lay up fiberglass reinforcement with threaded inserts inside the tank to serve as a flange. This avoids having to use complex fabrication techniques and guarantees a perfect fit. I can even use an off-the-shelf O-ring as part of the mold for this process to ensure that the resulting groove is precisely the correct shape & size.

[Syscrush]

I've started work on learning CAD and modeling the new throttle bodies. With the removal of the choke and slide, I'm going to end up with much better flow and reduced pumping losses. My goal with this project isn't to increase power, but to improve reliability, tunability, and fuel consumption, but if I end up picking up some power due to the more efficient TB's, I certainly won't complain.

To make room for the fuel injector, my plan includes moving the throttle assembly slightly upstream (toward the airbox, away from the head). This gives me the option of going with a significantly larger throttle butterfly if I choose to, and I'm considering going with 34 or even 36 instead of the OEM 32 mm.

I think I have 4 broad choices here:

1. Reuse the OEM throttle plates, shafts, and linkage - AKA the smart choice.
2. Design for OEM throttle plates & shafts from a more current bike. These aren't really a wear item, but they could be easier to find for longer than the CBX pieces.
3. Design for aftermarket throttle plates & shafts from someone like Weber, as replacements are widely available at low cost.
4. Design custom plates in CAD and have them fabricated.

Approach #4 takes 2 minutes in CAD, but adds significantly to the cost for no benefit.
Approach #2 only makes sense until you realize that you can't order OEM throttle plates on their own.

So, that leaves reusing the OEM throttle plates, shafts, and linkage, or building for readily-available and standard butterflies like the Weber ones. In addition to having options for larger sizes, going with the Weber butterflies would let me use more parts that are fresh and new, and having fresh new parts is very appealing.

But since I want to retain near-stock looks and keep the stock intake manifolds and airbox, I think it makes the most sense to stick with Approach #1. It has the added benefit that anyone who knows how to work on the OEM carb rack for assembly and sync would also be able to manage the EFI ITB rack - so that's the way I'm going to go - even though it means reusing (and being reliant on) 40+ year old parts.

[Syscrush]

So, I've ordered the fuel pump assembly from a Triumph Rocket III. That bike makes a claimed 148 crank hp, so I'm pretty confident that it will have no problems feeding my engine.

This arrived yesterday, and looks exactly as in the picture, but it doesn't include the fuel level sender. I'm going to concentrate on a solution to fit the necessary parts of the assembly to the CBX tank first, and then decide whether to fit a simple reserve indicator, the level sensor, or nothing.

[Syscrush]

I don't know if these intermediate updates are of interest to anyone else, but I do like being able to go back and review the process afterwards, so I might as well keep updating.

I just ordered a set of sealed ball bearings which I intend to use for the throttle shafts. I know that's probably overkill, but they're inexpensive off-the-shelf parts and if they do give me lighter/smoother throttle action, I'll be very happy to have it. Trying to salvage the bushings & seals from the OEM carb bodies seems excessively difficult. I should be able to make the new bearings & TBs work with the OEM throttle shafts and linkages, but I'll know better once I have the bearings in hand and can model them up along with everything else.

Having decided to stick with the stock 32mm throttle plates and the OEM parts, I'm feeling good about the decision and the design I'm finalizing now for the TBs. With the elimination of the choke, the sliders & associated vacuum passages, the flow through the new TBs should be a massive improvement over the carbs - not to mention eliminating the pumping losses associated with keeping the slides lifted against their weight & spring pressure. Merging the injector nozzle with the airflow downstream of the throttle plate is going to add back some flow issues, but I'm still confident that it'll be a significant improvement overall.

[Syscrush]

While we're at it, what is this screw for? It looks like some kind of air bleeder - is this the air fuel mixture screw?

[heli_madken]

While we're at it, what is this screw for? It looks like some kind of air bleeder - is this the air fuel mixture screw?

It looks like the vacuum take off plug for synchronising to me?

Please keep on coming with the updates what you are doing is fascinating stuff

[heli_madken]

Double post sorry