I've been looking into converting my vision to EFI. I don't know if I'm going to do it yet. I'd like to be sure I can do it and that it's an improvement over the carbs before I decide. I found this Microsquirt device that will control anything from a single cylinder motor through a V8. From what I've seen this is what I need to do the conversion.
1.) A Microsquirt device.
2.) A set of throttle bodies that will fit the Vision without a lot of alterations. If I can get them with the MAP sensor and throttle position sensor that will save a lot of searching.
3.) An airbox that will fit the throttle bodies. I may be able to change the Visions air box to fit
4.) An electric fuel pump.
5.) An O2 sensor for the exhaust.
I think that it will work best if I can get everything from the same motorcycle. Fortunately these things are not that expensive on eBay.
This is a link I found done by a guy that converted his ZX900.
http://www.kawasakimotorcycle.org/forum/projects-how-write-ups/95445-converting-1985-zx-900-efi.html
And this is a youtube video of him driving it.
http://www.youtube.com/watch?v=QoOFPT-aAck
This is the manufactures intro page.
http://www.microsquirt.info/
The first step would be to find throttle bodies that fit the Visions intake manifolds.
Any thoughts?
Bobby...
A couple of thoughts:
MAP: you could use a ford MAP sensor, since it's a remote, you just need a piece of vacuume line, the pigtail & a place to mount it.
O2 sensor: get a bung welded into the exhaust. most 2 wire sensors should work
fuel pump: any modern FI sportbike pump.
On the throttle body: you really already have one, (stock carbs) all you have to do is either remove the central venturi & fab up a holder for the injector in it's place, or remove it altogether & use the YICS ports in the head for the injectors..
for the TP sensor, fab a bracket on to one of the throttle shafts. doesn't sound all that difficult...
Blake, a member here, installed a megasquirt on an old caddy, & had plans to do the Vision, he even bought a rather expensive MAF meter. he's in, I think, Blacksburg Va. (is or was going to virgina Tech) look him up on the members tab, or i might have his email..
--Lucky
Try Blake at: BStoddar@vt.edu Don't know if this addy is still current...
Burt, did you see this article? http://hildstrom.com/projects/vision/index.html (http://hildstrom.com/projects/vision/index.html) This guy mounted MSD injectors in the Vision's gutted Mikuni's and a TPI on the end of a throttle shaft to make his own TBI system. He ran into some problems and didn't finish the project but interesting reading just the same with pics and fuel consumption charts.
one guy did just what you are saying on an OLD VMax. But you need temperature sensor for engine and possibly air,+ MAF(mass air flow) sensor, possibly barometric(density) sensotr, Crankshaft position sensor. In effect ...Good EFI set up is NOT simple.
Actually, early EFI systems were very simple, but technology in modern systems are much more involved. you can hook up anything to anything, the tough part i think, is going to be the computer setup
This is a subject I have been considering for some time and the fact that there are several fairly late model V twins on the market with EFI systems that can be harvested from salvage shops makes it that much easier. The "system" has been designed already, but now the hitch is to apply it to the XZ and fine tune it. Just having EFI on a bike doesn't mean it is a benefit, it must be adjusted to be responsive and efficient.
My first thought was to use a set up from a '03 or later Suzuki SV650, which is much the same configuration as the XZ with down draft intake, though the brackets that hold the throttle bodies would need some "adjustment" to fit them on the XZ. The Suzuki system uses a cam position sensor, so would have to adapt some such to the V.
Whatever you need in the range of sensors will be determined by the computer you use. For this reason I think transplanting a whole system from another bike is the most likely to work well.
suzuki is 90 degree v-twin. so the intakes would need some rework. The best systems have , MAF , throttle, air temp , coolant(or oil temp), rpm, cam position sensors , are just the beginning. don't forget the hi-pressure fuel pump. Do a search to find Vmax 1200 conversion to EFI. Yes, at times, I wish my Vision was EFI.
That does, I'm sold on the fuel injection conversion. Just looked at al the vmax site. I see now why CV carbs with no slides, and an injector hung in the slide bore is a natural conversion, simple!!!!
I'm thinking that a MAF would eliminate the need for the air cleaner flapper system. Anybody agree/disagree?
That would be nice to not have, don't know how much adjusting the maps allow, good question tho.
There is also a thought I've heard that two EFI units from one of the modern scooters could be used........
WHAT KINDA scooters?
Many of the smaller modern scooters come with fuel injection these days, I have seen them through the shop. Even the venerable Honda 50 now comes with it according to one article I read. Piaggio comes to mind also, I seem to recall that it was a 250.......... We have speculated at some length with my bike being offered up as the sacrificial lamb by others of course........... ;D
Honda's Silverwing scooter is a 582cc DOHC 4 valve head parallel twin with FI.
Burt from all the reading I've done so far on EFI on the XZ the latest V3 of the microsquirt is the way to go for the Yamaha Vision. The biggest problem with V-twins is they are an odd fire engine, meaning their cylinders fire at odd timing and can't be triggered off one point for both cylinders. V3 Mircosquirt solves all this in one system for ignition and fuel as it's "new options allow independent dual outputs for odd fires engines". The problem with using an OEM management system is being able to set the maps up and tune the computer, the microsquirt is more expensive to start with but I reckon setting it up and getting it running correctly will be easier.
I posted about looking at using the megasquirt 4 years ago but life got in the way and bikes have been in storage since I got back from working overseas. I'm kind of pleased it was put off in a way because I could see lots of problems with the XZ and EFI due to the odd fire issues. The MicroSquirt developers where working on solving these problems for V-twin enthusiasts and it's nice to know the waiting has led to an easier solution with the latest version now available.
It's still on my list of things to try out on my bike, hope to have them out of storage and in the garage to work on next year. One of the reasons I'm back on ROV, I was looking to see if anyone had converted to EFI yet:-). I know a couple have been seriously thinking about it.
Hi Steve,nice to see you back. Are you back in Wellywood now?
As far as I know there is not yet anyone who has done a MS implementation on the XZ. It's on my list too, but about 10 things down.
Quote from: roro on October 21, 2012, 01:03:31 AM
Hi Steve,nice to see you back. Are you back in Wellywood now?
Hi Rohan
I'm happy in Rotorua now, have updated my location thanks for reminding me. Have settled down and should be here for a couple more years. Need to be here a while to enjoy all the time that's been put into renovating the house and get some time on the bike:-).
I'm in the USA at Christmas and very tempted to buy the MS V3 while I'm there to encourage me to get working on it.
I have been looking at Suzuki SV650 throttle bodies as a near fit without too much fiddling. There are some fairly good photos on eBay so you can get an idea how they might be adapted.
Quote from: fret nut on October 21, 2012, 11:52:54 PM
I have been looking at Suzuki SV650 throttle bodies as a near fit without too much fiddling. There are some fairly good photos on eBay so you can get an idea how they might be adapted.
Interesting I'll take a look. I know some have mentioned the R6 or maybe it was the R1 carbs.
I've started ordering parts to start attempting a conversation myself, I'll be sure to post my findings on here for others to use. The ECU I got seems very similar to magasquirt, it's speeduino. my biggest concern at the moment is figuring out how the timing sensor works on the vision and how to configure for it in tuner studio.
does anyone definitively know how the timing is measured on the vision?
I could be mistaken I'm not an expert, but I saw 18 coils on the stater, and I'm under the impression that a coil of a stater may require 2 magnets of opposing polarities to pass over at the same time, so it should have 36 magnets in the flywheel maybe? I know it's common to use a 36 tooth trigger wheel for timing, so is it possible the timing coils work off the same magnets as the stater for position somehow?
If anyone knows anything, that'd be helpful, i'll update you on what I find myself.
Two VR sensors positioned 70 degrees apart, and a single crank tooth on the outside of the flywheel.
If it supports single tooth you should be able to drive speeduino off these but they don't produce a square wave so you'll need to digitise them with the speeduino dual vr conditioner. For the Vision, I think you want to set it up for leading edge triggering.
POD is correct. For each cylinder individually, the leading edge of the timing rectangle on the flywheel passes the sensor 36-38 degrees BTDC. The trailing edge passes at 10 degrees BTDC. This generates an AC voltage with a crude sine wave. Leading edge positive, trailing edge negative. This goes into the TCI where it is half wave rectified. The negative pulses "disappear" and the positive pulse triggers the TCI. As the revs increase the voltage from the sensor increases and at about 3100 RPM this increased voltage causes components in the TCI to switch to full wave rectification. This means the "missing" positive pulses now appear at 38 degrees... 28 degrees before the negative pulse. The 38 degree pulse triggers the TCI and the 10 degree pulse does nothing as it occurs too quickly.
This is all analog, nothing digital about it. There are a lot of temperature and voltage compensation components and a multi vibrator circuit to control the dwell time but it is really a clever and simple solution for it's day.
I learned about this by researching the Hitachi and Yamaha patents.
Hey, right up my alley. I was finally was going to get back to my EFI conversion on the vision this winter, as we moved for the last time, and dont plan on moving again- i can finally unpack all my stuff from storage after 5+ years.
You can technically use the stock crank triggers but its going to be a headache if you really want to modify the timing, injection, etc. and it's going to be very touchy.. Megasquirt/ speeduino really needs some sort of constant reference of crank position otherwise its going to be guessing after it receives the signal. In other words- doable yes, but not ideal. If you're going to make the effort to convert, i'd recommend doing a big more work in that area before hand vs trying to fix issues with the single tooth. I learned the hard way with my car that getting that crank signal solid on megasquirt is #1 priority.
You can (I thought about it) use the stock crank triggers along with a secondar sensor on a cam sprocket. This is sort of how i have my car wired up for megasquirt. On my car i have a 36-1 wheel on the crank, and a half moon on what used to be the distributor, which gives it crank position and cycle information. That's where I got before having to pack it all up and move. I do remember clearly that I wasn't happy with that solution since there is no "good" place to read the camshaft teeth. In the end, my plan is to just attach an actual 36-1 crank trigger wheel to the flywheel. It's relatively easy to do on a mill/lathe if you have access or someone who can do it for you. If you're not in a rush, i have both and can mod both our flywheels when i get to it later this year after finishing up some other stuff before it gets cold. Doing it that way, you only need one sensor that you'd mount through the crank case and deal with the rest (cylinder offsets, etc) via tuner studio. You still wouldn't be able to do sequential fuel or ignition since there is no measure of which cycle it's on, so it'll be batch fire/wasted spark- but you're no worse off than stock.
I'm not sure how far out you've thought with parts- LS coils are easy, cheap, and reliable to wire up. I'm using microsquirt, the LS coils, throttle bodies from a triumph, and some aftermarket injectors that i spec'd for the engine. (They were relatively cheap on summit racing). maybe 18lbs or so if i remember right. Also, thankfully the price of wideband 02 units have come down considerably so tuning will be a snap. I dont remember if i bought one for the conversion yet or not. I had an old one back in the early 2000's rigged up to the bike to tune my 82 carbs with k&n pod filters. Worked real well but i had to lug a huge laptop around and tune by logs.
Thanks for the quick reply! I also just found this link, pages 15-17 have some relevant info:
http://www.jetav8r.com/Vision/IgnitionFaq.pdf
That with the information you guys provided is Very helpful!
Now I should just have to figure out which setting to use for such a setup. This is my full list of my options in Tuner studio:
Missing Tooth
Basic Distributer
Dual Wheel
GM 7X
4G63/Miata/3000GT
GM 24x
Jeep 2000
Audi 135
Honda D17
Miata 99-05
Mazda AU
Non-360 Dual
Nissan 360
Subaru 6/7
Daihatsu +1
Harley EVO
36-2-2-2
36-2-1
DSA 420a
I've crossed out all the car names as they shouldn't be relevant. I cant find much, or any, documentation to describe some of these settings. It seams Harley EVO might be what I should use. I think the Evo may have used a points system so I'll have to investigate that to see if it would work/how I'd configure it.
Again if someone has some solid information that'd be helpful.
While I was typing this Blake actually replied. Regarding triggers wheels, this link seems to provide a way to order custom ones, but I'm not sure how I'd mount one to the vision. https://speeduino.com/wiki/index.php/Trigger_Wheel_Generator
My current plan is just to get my Speeduino to read timing and adjust with speed the same as the original module did (if possible). Not looking to go crazy with timing adjustments. I realize a 36 tooth wheel is far more accurate and would have "less problems", but if the original can work similarly as well as it does with the stock ignition module, I'd be happy with that. So long as we can configure for it, it definitely makes the conversion more approachable. Also I do plan to get coil overs, mainly because I think I have a bad wire now. The carbs I ordered were from a 2012 ninja ex650; They already have "all" the sensors I need, injectors and the injector fuel rail for like $36 on ebay. wouldn't be heart broke if I have to go with another option, that's just what I figured I'd try. I got a small 12mm 02 sensor for a Harley, was like $28. Its from an Electra Gide 2010-2014 if you want to look them up. For my setup I actually have a steeper driver module for the stepermotor on the ex650 throttle body, I can use it to help control Idle; its configurable in Tuner studio. All the parts for the Speeduino itself cost about $200, I had looked at mega/micro squirt a while back and they all just seemed to be a little steep for me ~$400 if I'm not mistaken. For many, its probably a better option as it can come with a harness. Both are pretty similar. I do still have to get a fuel pump and it needs to provide 45psi for the Kawasaki injectors, I have a few in mind but have yet to order.
So for what you say you want to do, the "harley" route would indeed be best.
http://www.useasydocs.com/details/dualspark.htm
Take a look at picture 1b. that's what we're working with with the stock ignition on these things.
I'm actually working through troubleshooting my car again at the moment- i think my sensors are bad (old generation of diyautune hall sensors) so i just ordered a couple new ones that should be here tomorrow/ i'll be rewiring it this weekend and see if that fixes it. Also check them out for 36-1 trigger wheels if you ever need some in the future. Relatively cheap and they have a few different sizes.
I am looking at a Speeduino for my XZ.
Also consider the idea of how complex you want the system to be. An open loop system requires much less (sensors, etc) because of the assumptions it makes in its fuel and ignition maps. Many of the OEM systems on dirt bikes, scooters, etc are open loop. Fuel delivery is a function of RPM and throttle position. Timing is similar. These often require only cam, crank and TPS sensors to run. Some have BAP to compensate for altitude.
Closed loop adds air (IAT/MAF/MAP/BAP/idle) and feedback (O2/lambda) to the mix.