Group Buy: EFI (Fuel Injection) Conversion Kit

[greg737]

Originally Posted by justinswidebody

Hey guys ,

Just so everyone knows. Make sure you get the oxygen sensors inside the exhaust tubing enough. You do not want to be shallow. I double checked my work today and realized I measure wrong. So I cut my bungs off and rewelded them where the o2 sensor hits the middle of the airflow and its a night and day difference in driving. Much better now.

Justin

This post might need some further explaination/clarification. I'm not sure what you meant.

I'm concerned that somebody else working on this conversion might misinterpret this post and make a mistake in their installation.

You say that the O2 sensor needs to hit "the middle of the airflow".

If you're saying that a wideband O2 sensor won't function correctly unless it extends all the way into the middle of the exhaust pipe, you're wrong.

As long as the nose of the sensor is inside the exhaust pipe it will work correctly.

If too much of the O2 sensor protrudes into a small exhaust pipe like the EX-250's has you'll end up restricting the gas flow through the exhaust system which would pretty much kill any performance gain you might hope to realize from the Ecotrons kit.

[ecotrons]

Originally Posted by justinswidebody

Hey guys ,

Just so everyone knows. Make sure you get the oxygen sensors inside the exhaust tubing enough. You do not want to be shallow. I double checked my work today and realized I measure wrong. So I cut my bungs off and rewelded them where the o2 sensor hits the middle of the airflow and its a night and day difference in driving. Much better now.

Justin

Thanks, Justin, I think you clarified one issue on O2 bungs. I was not sure about how long the bung needed to be. We had not tried longer bungs. And theoretically the sensor head could block some exhaust flow, and that is not preferred. Now we know, the O2 sensors have to be fully exposed to the exhaust gas, to be utilized for close loop controls. Too shallow of O2 sensors, could cause mis-reading of AFR, and therefore mis-adjusted fuel. By saying "fully exposed", I mean it does not necessarily have to be "hits the middle of the airflow". As long as the sensor nose (all holes on the nose) are exposed to the exhaust gas flow, it should be fine. There could be some optimizations needed here to find out what exactly the length of the bungs should be.

[flynjay]

Originally Posted by justinswidebody

Hey guys ,

Just so everyone knows. Make sure you get the oxygen sensors inside the exhaust tubing enough. You do not want to be shallow. I double checked my work today and realized I measure wrong. So I cut my bungs off and rewelded them where the o2 sensor hits the middle of the airflow and its a night and day difference in driving. Much better now.

Justin

You make a really good point. I just went out and measured everything.
O2 sensor from crush washer to end ~1.05in
Provided bung length = 0.55 in
Exhaust pipe diameter (pregen) = 1.65in

So with the provided setup the o2 sensor will extend into the pipe by about 0.5 in which is 30% into the pipe.

@greg737 how does that sound.

[ecotrons]

Originally Posted by greg737

If you're saying that a wideband O2 sensor won't function correctly unless it extends all the way into the middle of the exhaust pipe, you're wrong.

As long as the nose of the sensor is inside the exhaust pipe it will work correctly.

Greg, I guess it has never been expicitly clarified: the 2xO2 sensors in the kit are NB O2s.
WBO2 is too expensive to be included in this economic kit.
Certaiinly WBO2 would be better if used for self-tuning. esp. for DIY projects.
NB O2s can do some extent of self-tuning also.

[justinswidebody]

Sorry Greg for what I had said. Just make sure u get enough of the o2 sensor in the exhaust stream. Also on my setup I am only using an accel cable and have no issues with the throttle body not fully closing. As long as you have proper slack in the cable it should be fine. Atleast in my experience it is.

[flynjay]

My 1" bungs from innovative came in today. Definitely too long, the sensor tip does not protrude at all. They either need to be cut down or have the end milled to match the curve for the pipe.

[justinswidebody]

Flyn.

I don't know if you do your own work or not but it is very easy to take a angle grinder and grind the 1" bungs to fit the exhaust tubing. Just hold the bung up to the header and kind of visualize how you need to grind it. Or you can take your other bung and hold the end up to the other bung so u can see the half moon you have to cut.

Justin

[flynjay]

Originally Posted by justinswidebody

Flyn.

I don't know if you do your own work or not but it is very easy to take a angle grinder and grind the 1" bungs to fit the exhaust tubing. Just hold the bung up to the header and kind of visualize how you need to grind it. Or you can take your other bung and hold the end up to the other bung so u can see the half moon you have to cut.

Justin

I decided to go an even easier route and ordered a different set that happened to cost half as much. Amazon returns are easy.

The new set I ordered are a little longer than the ones Matt provided and also are precurved to fit the exhaust pipe.

[n4mwd]

I was wondering how everyone was doing with their conversion. I know some people have already completed theirs. I am still fiddling with the throttle body and trying to make it work with 650 cables. I think my copper tubing approach is going to work well. I just haven't had a lot of time to work on it. I think I have the new pulley arrangement about done.

So how are you guys coming? And did you go with bottom or top feeding?

[Jiggles]

Originally Posted by n4mwd

So how are you guys coming? And did you go with bottom or top feeding?

mine is still in the box

[n4mwd]

Originally Posted by Jiggles

mine is still in the box

Its taken a lot of figuring and plotting, but I have everything mounted as a self contained unit. The fuel pump and regulator are mounted to the throttle body. That way, I avoid drilling holes in the frame and such. Also, I am worried about fatigue of the plastic nipples on the fuel pump so I have to secure the hoses so they don't impose too great of a vibrational load on the nipples.

My new pulley is basically two big washers with a smaller big washer on the inside - like an Oreo cookie.

My design isn't proven yet, but I feel it will be soon.

[justinswidebody]

N4mwd

Do you have any pics of your pulley?

[n4mwd]

Originally Posted by justinswidebody

N4mwd

Do you have any pics of your pulley?

Not right now. I'll try to get something up by tomorrow. Its just a washer sandwich. The two outside thin washers about 1.75" in diameter. The inside is a thicker with a 1.5" diameter.

[n4mwd]

Regarding the fuel tap not having any shutoff valves, I was wondering if something like this would work?



I'm wondering if it just disconnects or if it also shuts the fuel off. The whole reason for having a disconnect for me is so that I can shut off the gas and remove the tank for maintenance. If this thing shuts off the fuel, it looks like a better solution than two inline petcock valves.

Anybody dealt with this issue yet?

[kiler]

Originally Posted by n4mwd

Regarding the fuel tap not having any shutoff valves, I was wondering if something like this would work?



I'm wondering if it just disconnects or if it also shuts the fuel off. The whole reason for having a disconnect for me is so that I can shut off the gas and remove the tank for maintenance. If this thing shuts off the fuel, it looks like a better solution than two inline petcock valves.

Anybody dealt with this issue yet?

Yes, it does shut the fuel off. I have a similar one on my other bike and I got it here http://www.economycycle.com/servlet/...oupling/Detail
I am really enjoying this thread

[n4mwd]

Originally Posted by justinswidebody

N4mwd

Do you have any pics of your pulley?

This is what I came up with:





Note that the center washer is about 0.100" thick whereas the others are 0.050" thick. It was hard finding a 0.100" thick washer that size.

[justinswidebody]

N4mwd

Looks great. Couldn't you just drill two holes through the old and new pulley and bolt it all together with out having to use any glue?

[n4mwd]

Originally Posted by justinswidebody

N4mwd

Looks great. Couldn't you just drill two holes through the old and new pulley and bolt it all together with out having to use any glue?

The glue just makes it easier to put together. Without it, the washers flop around too much. It probably also makes the final assembly a little stronger.

Part of the existing cable holder on the throttle body has to be removed in order to accommodate the larger diameter pulley. I purposely made it bigger to match the Kawasaki OEM Throttle body.

I'm a little concerned about the 1.2" center diameter. It might be too big and I'll have to start over. For the final fasteners, I had originally planned on using pop rivets, but because of the thickness, its starting to look more like little bolts.

[justinswidebody]

Sounds like you have it figured out. I can't wait to see the end result.

[justinswidebody]

I'm waiting on 2 air fuel ratio gauges to arrive for each oxygen sensor so I can see the ratio on each bank while I drive and accelerate. then I'm going to make a cutom dash for the bike with my wanna be replica koso gauge. I'll upload pics when I get it done .

[Jiggles]

Originally Posted by justinswidebody

I'm waiting on 2 air fuel ratio gauges to arrive for each oxygen sensor so I can see the ratio on each bank while I drive and accelerate. then I'm going to make a cutom dash for the bike with my wanna be replica koso gauge. I'll upload pics when I get it done .

Sounds awesome! Links?

[n4mwd]

Originally Posted by justinswidebody

I'm waiting on 2 air fuel ratio gauges to arrive for each oxygen sensor so I can see the ratio on each bank while I drive and accelerate. then I'm going to make a cutom dash for the bike with my wanna be replica koso gauge. I'll upload pics when I get it done .

Matt @ Ecotrons stated earlier that the sensors were narrow band so they don't give an air-fuel ratio reading. As I understand it, the NB sensors only give an Ok/Not-Ok reading.

Did you get yours installed with the injectors upside down?

[justinswidebody]

Yes I installed the injectors upside down. I wanted to get it installed quickly for I try to use my bike as everyday transportation. However one issue I still have with mine being that it seems in the 2012 ninjetts they changed the throttle cables. So my throttle body won't open all the way cause the pulley hits the end of the threaded end of the cable. So if I adjust it where it doesn't hit then the throttle body won't close all the way. So when I get time I'm either going to cut the throttle screw shorter or make a new mount tab for the cable.

[justinswidebody]

Can the ecu bennefit from swapping out the two narrowbands. with wideband o2 sensors. Being that nb o2 sensors were made only to tell the ecu if it had a 14.7 ratio around idle and cruise conditions.

[n4mwd]

The same problem exists for the 2009 650 cables. If you use the existing mount plates, the pulley hits the cables. I'm mounting the injectors on top, but I am also making a new bracket to hold the cables. After Greg737 pointed out how kinked they were with the stock mount plates, I decided to make sure they went in straight on mine.

[justinswidebody]

Turns out if you wanted to replace the o2 sensors with widebands I believe the ecu would still only ready them as narrow bands with a 0-1v signal. You have to have a wideband controller to make use of having wideband o2 sensors to be able to read between a 0-5v signal.

[n4mwd]

Since 0-1v is in the range of 0-5v I don't see that as a problem assuming he is reading it in with an ADC. The more important question is whether or not the firmware can make sense of the wideband data. If he is just reading it in as a logic level, then all bets are off. Maybe Matt can step in and say for sure.

I hope to get some more work done on my throttle body today.

[atomicrm]

any chance of doing another group buy on this in the future?

[greg737]

Narrowband sensing and the ECU programming for it are a lot simpler and cheaper than dealing with a wideband O2 sensor (and its controller, which a wideband sensor has to have).

This added programming complexity, plus the cost of wideband O2 sensors and the cost of the controllers would change the nature of the Ecotrons kit. Matt has pointed out several times that the Ecotrons kit is meant to fill the niche of a "low-cost" FI kit.

Believe me, there are plenty of more expensive ways to configure a do-it-yourself FI kit. In my own version of a do-it-yourself FI just the Microsquirt ECU alone cost me $370 and the Innovate LC-1 wideband controller/with Bosch O2 sensor was a little over $200. So I had over $570 into it before I even got around to purchasing the European EX-250 throttlebody, the Suzuki Quadracer fuel pump and fuel regulator, the various sensors, fuel filters, tubing, wiring and the list goes on from there.

So, as Matt has said, the Ecotrons kit is a low-cost option.

That said, I'm pretty sure that the programming and such necessary to utilize wideband O2 sensors is probably it well within Matt@Ecotron's ability. But the new, much higher price of the kit would probably kill off much of his sales volume.

But I do have to say that being able to program the Microsquirt to "auto-tune" for any specific air/fuel I want across the whole engine operating envelope is SWEET!

[n4mwd]

Originally Posted by greg737

Narrowband sensing and the ECU programming for it are a lot simpler and cheaper than dealing with a wideband O2 sensor (and its controller, which a wideband sensor has to have).

This added programming complexity, plus the cost of wideband O2 sensors and the cost of the controllers would change the nature of the Ecotrons kit. Matt has pointed out several times that the Ecotrons kit is meant to fill the niche of a "low-cost" FI kit.

Believe me, there are plenty of more expensive ways to configure a do-it-yourself FI kit. In my own version of a do-it-yourself FI just the Microsquirt ECU alone cost me $370 and the Innovate LC-1 wideband controller/with Bosch O2 sensor was a little over $200. So I had over $570 into it before I even got around to purchasing the European EX-250 throttlebody, the Suzuki Quadracer fuel pump and fuel regulator, the various sensors, fuel filters, tubing, wiring and the list goes on from there.

So, as Matt has said, the Ecotrons kit is a low-cost option.

That said, I'm pretty sure that the programming and such necessary to utilize wideband O2 sensors is probably it well within Matt@Ecotron's ability. But the new, much higher price of the kit would probably kill off much of his sales volume.

But I do have to say that being able to program the Microsquirt to "auto-tune" for any specific air/fuel I want across the whole engine operating envelope is SWEET!

I'm curious why the WB sensor needs a separate controller. I know that yours has a fancy digital display, but that's not absolutely necessary. If the WB sensor puts out 0-5v, then why is a separate O2 sensor controller necessary? The ECU should be able to handle that even though it will probably take a little more firmware to actually do anything with it.

[flynjay]

Originally Posted by justinswidebody

I just went to a auto parts place and bought 2' of 1/4 fuel line. It was like 4 dollars. I had better luck with that and no leaks. Mine right now has issues w popping when shifting gears and decelerating. Turns out the 2012 ninjas use different cables which caused tiny complications with my install. However I think I'm gonna order the 650 cables and put the injectors on top . Matts working with me trying to figure out the popping when shifting gears and decel. Sometimes decel is really smooth cause it cuts the fuel.

Flynjay,

My bike still runs a little ruff under 2500-3000 rpms. I'm gonna hook up my laptop today and see what's going on.

I was having a similar issue where the bike started fine and ran fairly decent but would get progressively worse as the engine warmed up. I logged a ride from cold to warm-up to see what was going on.

Looking at the data it seemed the bike was starting to run worse while the warm up factor dependent on engine temperature was being fazed out (basically the engine was beginning to run to lean and it was popping and stalling). I adjusted the factor to be constant from 30 deg to 70 deg and the running has become much more stable open loop.

[greg737]

I'm curious why the WB sensor needs a separate controller. I know that yours has a fancy digital display, but that's not absolutely necessary. If the WB sensor puts out 0-5v, then why is a separate O2 sensor controller necessary? The ECU should be able to handle that even though it will probably take a little more firmware to actually do anything with it.

The digital air/fuel ratio display gauge (that I have up front on my instrument panel) is just an "add-on" that I bought along with the controller and O2 sensor package. It's simply hooked up to a spare output line from the Innovate LC-1 controller (it has two output lines). So the gauge really doesn't have anything to do with the need for a wideband controller.

I think you're correct in thinking that you could actually incorporate all the functions of a wideband O2 controller into a FI ECU if you really wanted to do that. I know that my Innovate LC-1 is actually designed to operate by itself, not even connected to an ECU (don't ask me what the application would look like, because I don't know).

But fuel injection system designers aren't O2 sensor designers so it's easier to just add one to the other rather than trying to incorperate the two together in one product.

[Daves2JZ]

Originally Posted by justinswidebody

Yes I installed the injectors upside down. I wanted to get it installed quickly for I try to use my bike as everyday transportation. However one issue I still have with mine being that it seems in the 2012 ninjetts they changed the throttle cables. So my throttle body won't open all the way cause the pulley hits the end of the threaded end of the cable. So if I adjust it where it doesn't hit then the throttle body won't close all the way. So when I get time I'm either going to cut the throttle screw shorter or make a new mount tab for the cable.

I have the same problem with my 08 throttle cable so it's not you it's the placement of the bracket. I will be welding up a tab to place the cable in the right location. I will also be porting the injecter bung hole in the TB so they match. The fact there is a huge lip for the fuel to hit rathern then spray into the TB is simply stupid in my opinion.

[ecotrons]

Thanks! Greg,
you helped a lot to answer the questions. And you are right on WBO2.

We absolutely have the technical capabilities to integrate the WBO2 (Bosch LSU 4.9) into EFI system, but then it becomes very expensive. A LSU 4.9 sensor costs easily more than $60 each and times 2 ; you have $120 just for the sensors, and then the control circuits for Bosch LSU is way more complicated than a NBO2. That's why there are so many WBO2 controllers in the market. The lowest price could be $170 each.

We do have our own wideband controller: ALM with LSU 4.9. $259 each. Not cheap compared to others, because of premium Bosch genuine parts.
ALM is good for tuning, especially for those that want to do more tuning with more performance parts than what we provide.

You can certainly use other wideband controllers to do the tuning. It's up to you to find out what is best option for you. Wideband controller market is another big topic and has a long history. Google it.

Do not try to connect your WB O2 sensor(s) directly to our ECU. It will not work. Mis-connection could even damage the ECU. Because a WBO2 sensor pinout is definitely different than a NBO2. Our ECU works with NBO2 by default.
It can also work with a WBO2 via a separate Controller (which has a linearized analog output), but via different ECU pin (check out ALM-ECU integration manual at our website):
http://www.ecotrons.com/support.html

It can also work with a Wideband controller if it has a simulated NBO2 output.
Actually this is a good way to use a WBO2 if you want to run both close-loop fuel control and at the same time display your real AFR on the gauge. (again check our ALM-ECU manual on how to do that).

WBO2s for this Kit are only recommended for advanced tuners, or at least do some homework before you buy those wideband controllers.

Matt

[ecotrons]

More on NBO2 vs WBO2:

For gasoline engine controls, the engine runs at stoic AFR (14.7) most of time. And that is what a gasoline engine is designed to run at most conditions. It only runs rich at WOT (wide open throttle) to get more power and at the same time to have some cooling effects. This is how all the cars on the road work so far. One main reason is for emission controls, of course. So when we talk about close-loop controls, it really means stoic AFR (14.7) controls. There is not much of close-loop control for AFR other than 14.7, at least not practical in automotive industries. Even with a WBO2, which has been widely used for today's cars already, the close-loop controls pretty much means stoic AFR controls. WBO2 provide faster and more accurate controls than NBO2, that is main reason it is replacing NBO2 in gasoline engine controls.

In short, your engine is supposed to run 14.7 AFR most of time, except WOT conditions. For this reason, you do not really have to use WBO2 to control the AFR, at least not cost-effective for that amount money you gonna spend. NBO2 can do the job for stoic AFR controls.

It is certainly nice to have a WBO2/controller to tune the engine and also show the actual AFR on a guage.

[Scattcatt]

Kit just came in today! Tearing my baby apart and starting tonight! Hopefully have it done by Sunday! With the exception of perhaps the welding job.

Excellent packing job by the way, everything was wrapped up super tight in bubble wrap, looked like bricks of cocaine lol THAT'S when you know care was put into the packing

[choneofakind]

just hope that it wasn't shipped in some of the same methods as cocaine

[Scattcatt]

Back to the whole throttle cable thing. I've stuck with the stock throttle cable and an upside down TB and I've discovered first hand all the problems. As stated previously, if the accel cable is installed into the actual hole on the throttle body, it does not allow the throttle body to completely close. Would it be safe to install the acceleration cable to the deceleration tab instead and leave the deceleration cable off entirely? It seems stable enough to me through my basic tests and allows full twist and return.

[justinswidebody]

Scatt I'm a little confused. If you use the decel tab then your cable is on the other side of the pulley. I don't see how this would work as for the decel tab is to bring the throttle back to close.

[n4mwd]

Originally Posted by ecotrons

I don't think copper is a problem here. Injectors are fine with copper fuel rail.

I spent two weeks and about $50 working on this and its a no go. The problem with using copper pipe and fittings is that the tolerance needed for a leak proof system just isn't there. So if anybody was thinking about doing the same thing, forget it. It was just too tempting not to try it because the injector connection is exactly the same as a 1/4" copper Tee. You would think that would be close enough, but it isn't.

Unfortunately, my pulley bracket system is not compatible with the stock injector holders. I'm going to work on making a machined aluminum fuel rail and see if that works any better than the previous design.