Electronic fuel injection- CV carbs to TBI

[RempageR1]

I don't know the software, but that seems something very-easy to program to work every-other cycle.

The challenge would be to find out what cycle. Because without knowing which is the true ignition stroke, it would start 50% of the time and be wrong the other 50% of the time.

Well I'll add my 2 cents to this really old thread. While the F.I. sounds great as there is no more adjusting the carbs, the FI systems has the same issue. I had the Wild Bro FI on my bike for a while and I always had to travel with a tablet to do adjustments. You need to drill and weld a bung on your exhaust for an O2 sensor. You need to drill and install a fuel pump on your gas tank. You will need to dyno the bike to set up the base configuration. The FI system ran really rich and was hard to get that magic 14.7 -1 ratio. I never was able to get a nice curve but generally it performed pretty good. I'll see if I can find the dyno but I was able to get 120 hp out of with a kerker 4-1 with the small baffle. Unfortunately, unlike your car, this was not a set once and forget system.

I`m wondering how you set it up. From what I read on other forums by birdoprey, this is the firing order of the Vmax:
-------
*Cyl #1 fires, then the crank rotates 70deg
Cyl #2 fires, then the crank rotates 110deg
*Cyl #3 fires, then the crank rotates 70deg
Cyl #4 fires, then the crank rotates 110deg
Cyl #1 fires, then the crank rotates 70deg
*Cyl #2 fires, then the crank rotates 110deg
Cyl #3 fires, then the crank rotates 70deg
*Cyl #4 fires, then the crank rotates 110deg
*Cyl #1 fires(back at the start)

The "*" indicate when that cyl is TDC on its compression stroke(power stroke)
-------
So next to squirting only on the correct stroke, you have to time the fuel injection correctly per cylinder. Since the V4 is configured in 70 degrees V instead of 90 degrees, the injection moments have an offset in timing as well.

Were you able to configure both correctly with the Microsquirt? If not, that might be reason why you ran rich and weren't able to make it run autonomously as it in theory should be able to.

Thanks!

 

[gentsvmax]

A few years back I started conversation with Jack Flemming at roadster cycle, watched every video I could find and bought the plans from him to begin the F.I. build. The planned parts list among alot of other things requires the bunged exhaust for sensor, tip-over fuel shut-off, you decide on internal or external high pressure fuel pump, throttle position sensor GM, inline water temp sensor Ford, Ford 302 injector's and an extensive read about choosing a Megasquirt brain module, either you buy and assemble as a kit, like a Heath kit or a little more expensive module sold ready to program. Both like the fuel pump and water temp sensor are going to be a challenge to find room for to mount. There are quite a few more ingredients can't access the plans as I write this. Jack would NEVER,EVER offer any program advise, he told me "my success as a winner would be based on research, research and more research, dedication and perseverance". He stressed the possibility to quickly burn up a motor with a lean condition programming . LOL every time I spoke with him he would raise his voice saying "Do your home work!". I have a shop here called Quality components thats ready to machine the injector nozzle adapters and plates per the plan drawings for $750,
350 more for the big brain and whatever else Iv already got into the misc. parts doesnt seem like much at all if I could have the performance of the Roadstercycle Vmax( new version) with dual exhausts doing burn outs WITH a 300 wheel and tire!
I just got discouraged about the programming challenge, lack of info and the very real possibility of frying an engine.
Despite the floggings from the mod monkey.
I like the resurrection of the topic and the discussion in hopes of some new info on the possibility of F.I. for the Gen 1 Vmax before fossil fuels are banned.......

 

[02GF74]

So the ignition system is wasted spark?.
How can time the injection without knowing crank position ie having no toothed trigger wheel?

I'm guessing you need to figure out rpm then calculate the delay after an ignition coil trigger when to inject?

And how would you know how much fuel to inject without a throttle position sensor? Use the vacuum sensor?

I need to read up about megasquirt,...

 

[RempageR1]

So the ignition system is wasted spark?.
How can time the injection without knowing crank position ie having no toothed trigger wheel?

I'm guessing you need to figure out rpm then calculate the delay after an ignition coil trigger when to inject?

And how would you know how much fuel to inject without a throttle position sensor? Use the vacuum sensor?

I need to read up about megasquirt,...

A toothed trigger wheel at the crank would give you precise timing, but again it wouldn't know in which cycle the cylinders are. So you probably need to add a cam-sensor. I also read that you might be able to use the MAP (vacuum ) sensor since the manifold pressure drops at the moment fuel is sucked into the cylinder.

But again, I`m looking to see if there is anyone that has actually been able to get it running properly with the correct timings. Since I have the Ignitech, I`m not interested in doing the ignition off of the Injection module, so trigger wheels etc are not required at this time.

Addition: This comes directly from the Microsquirt hardware guide:
For fuel-only installs it is possible to obtain a tach in trigger from the negative terminal of a single coil. Note that
this won't work well on a wasted spark setup[...].

So that specifically states the setup`s I`ve seen so far on the Vmax are suboptimal.

 

[02GF74]

As you say, with wasted spark, the fuel will be squirted in on the intake and power strokes.

Potentially the system could squirt on every other cycle and then either engine runs great or like a bag of nails. If the latter, then the system switches to squirt on the other of the wasted spark ignition event.

I'm not sure if the vacuum gauge would react quickly enough, maybe it will.

 

[RempageR1]

Yes it does, watch this one:



Basically, the MAP (vacuum sensor) will only work on low revs, but the megasquirt will only sync for about 10 seconds and will then run without it as it knows the correct phase to squirt.

 

[Big"A"]

A toothed trigger wheel at the crank would give you precise timing, but again it wouldn't know in which cycle the cylinders are. So you probably need to add a cam-sensor. I also read that you might be able to use the MAP (vacuum ) sensor since the manifold pressure drops at the moment fuel is sucked into the cylinder.

But again, I`m looking to see if there is anyone that has actually been able to get it running properly with the correct timings. Since I have the Ignitech, I`m not interested in doing the ignition off of the Injection module, so trigger wheels etc are not required at this time.

Addition: This comes directly from the Microsquirt hardware guide:
For fuel-only installs it is possible to obtain a tach in trigger from the negative terminal of a single coil. Note that
this won't work well on a wasted spark setup[...].

So that specifically states the setup`s I`ve seen so far on the Vmax are suboptimal.

Thought I read somewhere that u need to install 2diodes backward to get it to run right. I'm currently trying to f.i. mine now. If anybody finds something that will work. Please email me at. [email protected]

 

[huw s]

Thought I read somewhere that u need to install 2diodes backward to get it to run right. I'm currently trying to f.i. mine now. If anybody finds something that will work. Please email me at. [email protected]

Fuel injecting this engine is going to be difficult - but possible. Is it worth the effort? The carbs are easy enough to rebuild - how often do they need stripping if the bike isn't left idl?

 

[Fire-medic]

Fuel injecting this engine is going to be difficult - but possible. Is it worth the effort? The carbs are easy enough to rebuild - how often do they need stripping if the bike isn't left idl?

I'm of this expressed-thought. However, if there was a complete ready-to-go system available, where the research was concluded and bugs had been worked-out, for something reasonable in cost, I'd be tempted. It would be good if it was tunable for work being done to the engine, and that there was some-sort of a modest power increase from a stock Mikuni 35 mm set-up. In the meantime, I'm decent at tearing-down the OEM induction system.

 

[one2dmax]

This is not a simple system to add injection to. The biggest hurdle is getting the fuel matched up to the timing events. When you skip a cylinder ever engine rotation set and that the cylinder skipped changes with each set that complicates things. You either batch fire the fuel and just flood the system and get less then ideal fueling, OR, you get a more sophisticated ECU and program it.

 

[RempageR1]

This is not a simple system to add injection to. The biggest hurdle is getting the fuel matched up to the timing events. When you skip a cylinder ever engine rotation set and that the cylinder skipped changes with each set that complicates things. You either batch fire the fuel and just flood the system and get less then ideal fueling, OR, you get a more sophisticated ECU and program it.

That`s why I am trying to figure out how to do the fully sequential setup. If you look at what I wrote before:

-------
*Cyl #1 fires, then the crank rotates 70deg
Cyl #2 fires, then the crank rotates 110deg
*Cyl #3 fires, then the crank rotates 70deg
Cyl #4 fires, then the crank rotates 110deg
Cyl #1 fires, then the crank rotates 70deg
*Cyl #2 fires, then the crank rotates 110deg
Cyl #3 fires, then the crank rotates 70deg
*Cyl #4 fires, then the crank rotates 110deg
*Cyl #1 fires(back at the start)

The "*" indicate when that cyl is TDC on its compression stroke(power stroke)
-------

You could get to this in a fully sequential fuel setup (or non wasted spark if you also run ignition) as such:

-------
*Cyl #1 fires, then the crank rotates 180deg
*Cyl #3 fires, then the crank rotates 250deg
*Cyl #2 fires, then the crank rotates 180deg
*Cyl #4 fires, then the crank rotates 110deg
-------

So from what I understand, you need:
- To get the timing of cylinder 1 (I will get this off the coil)
- To get the correct sequence (ideally using a map-sensor, but CAM might also work)
- To program the timing sequence as above. So you squirt fuel at odd intervals instead of 180-180-180-180.

The Microsquirt cannot do that as far as I learned, the megasquirt I find too big, but the Speeduino (Arduino project) has this cool feature:



In the screenshot above, you can set the oddfire angles! So you could get the fuel timing perfect per cylinder.

How cool is that?

 

[one2dmax]

Haltech I think can also do the job

 

[Parminio]

Fuel injecting this engine is going to be difficult - but possible. Is it worth the effort?

Exactly.

In my humble opinion, it isn't. At all.

By the time you sank the volumes of time it would take to custom tune a fuel injection system with new ECU and all else that goes with it you could have simply run out and bought a Gen 2 Vmax and made that your baseline.

As picky and sometimes finicky as carbs can be, they belong on the Gen 1 forever.

I was at a car show a few years back and saw a beautiful 1970 Chevelle SS just like I used to have 30 years ago. I fast walked over to it coming up from behind it, walking down the driver's side nearly in tears. Then I got to the motor compartment anxious to see the glorious LS6 once again.

They fuel injected it and painted the block blue. I was disgusted and walked away.

 

[Fire-medic]

The fuel injection product for a Gen 1 probably has such a small market, it's not going to be worth-it in the marketplace to place it there. But, for those who want a premium product, there it is, assuming someone does the work.

Once a viable product is developed and a price is assigned, we'll see what the market says. There are so-many components out there being used on other bikes, it seems that the programming is the largest obstacle. Adapting the injectors from say a Hyabusa, or a CBR1000 would seem to be the easy part. A sheet metal or extruded metal weld-up for the plenum I assume would work. Doesn't the current product use stripped CV carburetor bodies fitted with injectors?

 

[MaxMidnight]

Just thinking aloud but why do we have fuel injection?

Is it because we want to have more power, better fuel consumption...or were manufacturers given no choice because of ever stringent emission requirements? (Place you vote NOW. Results will be given after several days of counting, many tweets and possibly several law suits.)

I suspect it was the latter as the majority tend to be conservative in their outlook on new technology (regardless of what they say) and sit back to see which new innovations may give them a competitive advantage.

Whilst anything is possible with enough determination and money I think we have to recognise that the Max motor was never designed with PI in mind.
I can understand that someone may wish to do it 'cos they can' but I have read quite a few threads over the years from those that have attempted it (who I applaud).
I can't recall anyone who has succeeded and also shown that it give a tangible benefit over a well set up carburetor.

 

[huw s]

This bit about firing the injectors to co - incide with valve opening is a nonsense - most 1st and second generation fi systems were batch fire and fuel control was many times better than carburretors. Also, just about all modern systems have injectors open way beyond valve opening periods - they have to in order to get sufficient fuel in.
Years ago I made a system for my electronics bEng based on a 16f877 pic microcontroller. I wrote the code for two batch fired injectors firing fuel into an A series mini engine. Everyone tells you this can't work due to the shared inlet ports and charge robbing. Well it worked, as well as needed, and was better at fuel contro and smoother than the SU carb. If anybody wants pics I still have the report and the inlet manifold I used. So a vmax, where ports are individual up until the ports open up would be fine imho.
As for implementing the hardwarem making an insert to fit the mikunis would seem a good start. On my mini I just removed the dashpot.
Oh, code wise, I had control of the fuel pump, warm up mixture, a throttle potentiometer and temp sensor. I also had a driver controlling ignition timing - this was by far the most difficult bit to do - you have to convert speed signals into time to get it smack on. I never wrote the gui for it, it was to be phase 2..
Oh, perhaps I should say, I'm not suggesting anyone should write their own code, but adapting an industry standard bit of kit could be doable - batch fired, or as mentioned previously some do allow staged injection start. Getting ignition to work might need something clever like two seperate pickups to fir the coils, but I can't see the point of going to an ecu unless the ignition timing is done as well. Still a lot of work...

 

[MaxMidnight]

I'm not suggesting it isn't doable but question if it is worth the effort unless this sort of thing is your passion.

Not sure if it is still available but there is/ was a PI system that mounted the injector in the OE carb bodies but don't remember anyone reporting any success with it.

 

[Parminio]

Not sure if it is still available but there is/ was a PI system that mounted the injector in the OE carb bodies but don't remember anyone reporting any success with it.

Isn't that essentially what Triumph did with the Bonneville to keep the twin carb look but with fuel injection?

 

[MaxMidnight]

As far as I know it was a pukka PI system that was designed to look like carbs rather than carbs that had been adapted to run PI.

 

[huw s]

The possibilities are to fit them in the airbox (at least anywhere before the throttle plates), with the pistons removed, or cut into the inlet manifolds and weld in bosses.
I had some thoughts on this today - you might also want to wedge the vboost open, since the change in manifold vacuum, flow and ultimately fuel demand would be difficult to map. Having a single big throttle body would do away with this issue I expect, but it would still be a transition you would need to fuel exactly...