EFI conversion on a Gen-1

[wayne billings]

Mornin' everyone I was looking at this deluxe carburetor air flow tool. Has anyone tried this on the vmax cv carbs and if so which one would be a better tool for the job? I mean stix verses the air flow tool . thanks as I will be attempting to start my '86 after a full overhaul soon. Enjoy the warmer weather.

 

[wayne billings]

Apparently I missed gents vmax post, oops. any way, any reports are welcome. thankyou

 

[joksi]

Hi! A quick update on this topic. After having had some issues which turned out to be a bad starter clutch, I finally have the bike running well again.

Last year, I had the Speeduino controller under the front left air scoop and the O2 sensor in the exhaust. This meant I sometimes lost Wifi signal to the tuning program running on the laptop in my backpack stopping autotune. Also, the O2 signal had too much latency since it was too far away from the cylinders. Now, the Speeduino is positioned under the saddle and the O2 is on the rear right downpipe. This means the Speeduino is closer to the laptop I carry on my back when automatically tuning the fuel table and the O2 response is more direct.

Tonight, I took the bike out with the TunderStudio on my laptop and autotune enabled and just started driving. Since I have the 3d printed manifolds, the amount of air has changed so therefore a tune was needed. Whenever I felt the tune was not right (feeling a dip in power) I held the throttle at that place for a couple of seconds until that cell had the right fuel load and it would run fine. So after a 1 hour drive, this is the result:


View attachment 83422

Blue means more fuel, red means less fuel. The weighting on the right shows how many times that cell was hit, the cell change shows by how much the cell was changes.
For more info on autotune, watch this video:

I have to say it`s quite nice driving with a bike that just keeps running better and better . I`ll keep doing this for a couple of times more until the cell changes are close to zero. Then I need to check some details like quick load changes (going from throttle closed to WOT immediately) to see if I can get those as smooth as possible. Fun stuff!

Will update this topic as always.

Very very interesting, please continue your development!

 

[RempageR1]

A well-written program that automatically updates the fuel mapping as you cruise, what an accomplishment. It sounds like the Holley Sniper system, though I don't know enough about either's operational characteristics, other than hearing they 'learn' and make adjustments as you run the engine. I suppose that once you get the system populated with enough data points to eliminate the red cells, you're closer to where you need to be.

Yes it`s a great tool to have . It`s all about what the O2 sensor registers and the target AFR table you use. Currently I`m using this target AFR table:


The red and blue cells in my previous post show how much the change was in fuel. Blue means more fuel, red means less. The darker the color the greater the change is. So in the end there should be almost no color left as there were no changes.

What will be of interest to Gen. 1 owners is the bottom line: what will the cost be, and how easy will be the change-over? I'm OK with keeping a OEM carburetion system functional. However, learning electronic fueling systems will be a steep learning curve, and will require new equipment.

As said before, I`m sharing all the details of my build so others can take it from here. If I would source parts in the Netherlands and ship to the U.S. it would already be too costly. Also for the throttle bodies some machining is required so someone from the U.S. side needs to stand up to do some fabricating. From there, I do believe a kit could actually work. If someone would go the same route as I have, you would remove the carbs, the manifolds and then start building again. The throttle linkage I have made fits into the original throttle cartridge found on the left side of the steering head.

But to take it one step further from that, I do feel Speeduino is the way to move forward. It is an open source solution that allows others to contribute and it`s low cost compared to the Megasquirt or Microsquirt. Over the last week, I`ve made some changes to my source code and submitted it to Speeduino. To my joy, the code was accepted and is now part of the Speeduino stock software. So if anyone buys a Speeduino and sees this trigger pattern in the drop-down box in Tuner studio, it`s my code .



Further info is now on the standard wikipage belonging with the Speeduino project:
https://wiki.speeduino.com/en/decoders/Yamaha_VMax

 

[RempageR1]

Can we add how the torque, power and fuel consumption comparison to OE?

By feel, I would say the torque and power are the same as stock, although fuel consumption will be better. While the tune is not yet complete, I got 42,34 mpg running on E10 fuel. That was at least 25% better than the other Vmaxes I was riding with (which were on E5). I do run the Venture overdrive 5th gear though which also safes on fuel.

To really get more power, you have to talk to the likes of Sean to find out what`s needed. Perhaps big-bore, high flow heads etc, I do not think that you can really increase the power by much without modifying the other bits.

Very cool. Wondering if this program could be loaded onto small permanent module under seat so it could always autotune. I'm thinking for elevation, humidity, seasonal type of changes. Thanks for sharing.

The program itself takes things like height into account when calculating the mixture. Depending on the hardware type, it takes a map pressure reading when the unit is started or can take continuous readings. Once the tuning is done though, you want to avoid making too much changes. For example, if you quickly open the throttle and run lean for a second, it might overcompensate by adding fuel which makes it run rich the second after. Better to get the perfect tune and then only use the O2 for reference to see if any modifications are required. You could run closed loop though for a while that will correct the O2 if your tune is not perfect yet. But it can sometimes cause the over-correction like mentioned above.

Also, I have seen sourcecode come by that can tune without a laptop but without the graphical representation you might miss out on what is happening.

Humidity I haven`t seen being used in a sensor but you could always program that .

Finally, I did order one of these:

This is a flexfuel sensor. It measures the amount of Ethanol in the fuel and compensates for it. High ethanol levels require a richer mixture so you can correct that with a predefined table:

So then your MPG would really increase if you run low Ethanol since it will use less fuel.

 

[joksi]

As said before, I`m sharing all the details of my build so others can take it from here. If I would source parts in the Netherlands and ship to the U.S. it would already be too costly.

And what about your European fellows like me?? I guess I would not be the only one interested!

 

[RempageR1]

And what about your European fellows like me?? I guess I would not be the only one interested!

For now I`m not able to easily produce anything with the exception of the manifolds. If things get serious, the first thing required would be to find out what people are looking for. So should I focus on standardizing the VFR1200 throttle bodies? Try to create a wiring loom and Speeduino EFI? Should I start an exchange program to weld the bung for the O2 on the rear left downpipe? Another big one is the fuel tank. There needs to be a return line added to it.

To convert to EFI, you need electronics skills, computer skills and fabrication skills (which is my weakest area). So the first pieces of a kit should be going to people that know what they are doing and provide me with feedback for improvement. If someone just comes back with 'it doesn't start', there isn't much I can do remotely. But hopefully someone will stand up that has all three skills and wants to convert a couple of bikes in their local area so we can learn from that.

 

[Fire-medic]

Thank-you for the common-sense replies, patiently describing the plusses and minuses of progressing to a system others can use. Your willingness to open-source the work is a generous contribution. I am glad to see that you're not 'promising the moon' on the behavior of the system in operation. Just meeting the OEM operational parameters with a realistic goal of better gas economy would be enticement enough for many people, where being able to continually tune the bike as you added other modifications, is probably the target group who would most-appreciate the effort.

I suspect the person who has programming experience, fabrication skills, and an electrical engineering background are the skillsets needed to progress to an operational system of functional utility. Once a prototype has been developed, has been tested, and is found to have operational advantages over the constant-velocity OEM induction system, the cost to duplicate the product is what will drive its appeal in the marketplace for the majority of potential customers.

My belief is hot-rodders who wish to add power via larger displacement, forced induction (supercharging, turbocharging, NOS) and associated performance enhancements are going to be able to most-beneficially utilize this developing product. The ease of tuning after significant changes in hardware besides this fuel injection system should be a real benefit to speed-seekers.

 

[one2dmax]

I could possibly make future kits and have access to billet manifolds if people didn't want to print them. I would just need to go through these thread and try to figure out the specifics of what is needed.

 

[joksi]

For now I`m not able to easily produce anything with the exception of the manifolds. If things get serious, the first thing required would be to find out what people are looking for. So should I focus on standardizing the VFR1200 throttle bodies? Try to create a wiring loom and Speeduino EFI? Should I start an exchange program to weld the bung for the O2 on the rear left downpipe? Another big one is the fuel tank. There needs to be a return line added to it.

To convert to EFI, you need electronics skills, computer skills and fabrication skills (which is my weakest area). So the first pieces of a kit should be going to people that know what they are doing and provide me with feedback for improvement. If someone just comes back with 'it doesn't start', there isn't much I can do remotely. But hopefully someone will stand up that has all three skills and wants to convert a couple of bikes in their local area so we can learn from that.

This is probably the most interesting upgrade beside the rims. EFI makes VMax easier to use here, especially because of the mountains with its height differences.

A kit would be a dream, and if the price isnt too high, I would go for it immediately. Not just because of fuel savings.

The smallest problem are the manifolds, Im running a print farm and I would MJF print them, since of heat resistance, precision and finally cost per piece. Plus maybe I can assist with board designa nd production since a close friend of mine was employed by KTM motor development division.
However, I am following with highest interest.

 

[RempageR1]

Hi!

Another update. Autotune is a fine tool to correct the fuel load table, however I found out it has trouble modifying the right cell under varying load. Since the O2 signal always has a little lag, it`s hard for it to modify the correct cell. To take the TPS out of the equation, I`ve created this rather ugly but very effective stopper:



I can set that so the throttle stops at a certain percentage. So you could stop it at 25% and have autotune only correct the 25% TPS horizontal row. Then you get it down to 1500 RPM, open the throttle to the 25% stop and let it run until redline (or where the RPM maxes out in a higher gear).

The first run I made had one very badly configured cell so the O2 goes up to 19.7 (way too little fuel) and the bike stayed at that RPM for a while. Then autotune corrected the cell and it started accelerating again. Fun stuff .

The graph below shows this, top is the RPM, middle is O2 and Bottom is the TPS.


15 mins later, the cells were corrected and it looked like this. A lot better. It still goes a bit lean at the beginning, which is likely caused by the cells below being too lean. This will be corrected when I get to those lower TPS% cells. However, from 2K to redline it`s looking very nice.



This evening, I`ve done two rows, 30% and 26% and need to continue with the others until they all are within limits. After that, I can have a look at acceleration enrichment to add the correct amount of extra fuel if you open the throttle rapidly.

The 3D printed manifolds are still holding up very well.

Over the next days, I`ll burn some more fuel to get all the cells right. To be continued!

 

[RempageR1]

I`m done with all the runs and this is what the fuel table looks like:



And the 3d model of it:


The top left quadrant is mostly the same, at those low RPM`s the butterfly valves in the throttle body open so far that there is no decrease in pressure anymore. From 2500 RPM and up it does start to make a difference which becomes more evident in the higher revs. Even though the Vboost is not present anymore, the engine still likes the extra fuel from 5500 RPM up on open throttle as you can see. It still gives you the kick in the back around those RPM`s.

Hopefully this table can serve as a starting point for others that want to have fun with EFI .

 

[one2dmax]

This entire thread was awesome! Would be great to add back in the vboost and use it with the rest of your work.

 

[Jeff Jarquin]

Would there be any possibility of adding a tps to a gutted set of stock carbs to maintain stock looks?

 

[user 77084]

and then there's this:

 

[V4max]

Be nice if he wasn't scared of it and really showed what it has he was clearly scared of it.

 

[RempageR1]

This entire thread was awesome! Would be great to add back in the vboost and use it with the rest of your work.

Thanks Sean! Really appreciate the compliment . I might have a look at it in the future if people are really interested in preserving the Vboost. As mentioned it`s kind of hard to time everything but perhaps the Vboost itself could be equipped with a TPS to monitor the valve angle and make sure everyting is calculated correctly. Or perhaps I could write some code to open the Vboost using the Speeduino so it knows the angle. That could work.... Perhaps a cool winter project .

Would there be any possibility of adding a tps to a gutted set of stock carbs to maintain stock looks?

The injectors need to be below the butterfly valve so you would have to do some heavy modification of the stock carbs but it might be done. I`ve run TPS on the carbs for 8 years or so using the Ignitech for ignition. I did that by drilling and adding a nut to the end of the butterfly valve at the position shown below. That M5 nut fitted exactly into the TPS unit I bought on ebay. But again thats the easy part.



You could also mount the caps covering the diaphragm sliders onto the blank aluminium end plates I use at the moment so it at first sight looks like a stock setup. Anything is possible

 

[Parminio]

You could also mount the caps covering the diaphragm sliders onto the blank aluminium end plates I use at the moment so it at first sight looks like a stock setup

I wouldn't go through the trouble of doing something as involved and elaborate as this and then try to hide it. I'd show it off. Big time.

 

[Jeff Jarquin]

That looks great. Definitely sparking my interest.

 

[one2dmax]

For a stockish carb set you could possibly use a later Royal Star rack. I believe they had a TPS on them.