Mircosquirt Fuel Injection Project log.

Everything in-between the gas cap and the carbs. Also how the engine inhales.
User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Mircosquirt Fuel Injection Project log.

Postby Triangles » Thu Jan 14, 2010 1:13 pm

I am investigating the possibility of using the microsquirt controller to fuel inject the Vulcan 500. I know for most people the cost alone or requisite technical knowledge makes this a task that is not worth pursuing. I estimate it will cost nearly $1,000 to do so, not to mention countless hours engineering something that actually works. I want to attempt to do this conversion because it's an engineering challenge and because I can. I'm lazy and don't want to fuss with a choke, or worry about my carbs gumming up. Yes I'm well aware that our carbs are quite advanced and work very well. Also I would expect the power gain if any to be negligible.

I fail to understand why people seem to react so negatively anytime someone mentions fuel injecting a carbureted(is that even a word?) motorcycle. While researching this I've seen comments like;"waste of time", "it's not worth it", "too expensive", "no power gain", etc etc. So please if you don't have anything constructive to add to the conversation, don't waste your time or mine with statements similar to those above.

My current budget doesn't allow me to start this right away, however I can do some research in the mean time. I'll keep my progress updated here.
Last edited by Triangles on Mon Nov 29, 2010 9:10 pm, edited 1 time in total.
Reason: Change thread topic now that I've actually decided to do this project.
ImageImage

hwmccullough
Posts: 10
Joined: Wed Jan 13, 2010 3:28 pm
Location: North Carolina
Contact:

Re: Mircosquirt Fuel Injection.

Postby hwmccullough » Thu Jan 14, 2010 1:41 pm

I don't think I can afford such a change, but I'd really be interested in the difference it makes. I know newer bikes have FI on them.

SandMan

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection.

Postby Triangles » Thu Jan 14, 2010 11:11 pm

I've been doing research on the microsquirt forums and I have it mostly figured out. If I could figure out how to shoehorn a cam position sensor and modify the crank position sensor so that it could be used by the microsquirt controller, you could use the microsquirt controller to control both the fuel injection and the spark advance in a single fire configuration. If I hadn't spent countless hours reading the manual and forums for the microsquirt controller, My previous statement probably wouldn't make much sense. On the EN500A the engine runs at 10 deg advance at 1,200 RPM and then the advance increases with rpm up to 37.5 deg at 10,000 RPM With the microsquirt controller setting the ignition, the load on the engine also influences the amount of advance. The timing is a little different on the EN500C, but I ramble on......I need sleep.

If you want more info on the microsquirt controller go to microsqurit.info
ImageImage

fastpakr
Posts: 68
Joined: Thu Jan 14, 2010 12:30 pm
Location: Roanoke, VA

Re: Mircosquirt Fuel Injection.

Postby fastpakr » Fri Jan 15, 2010 3:08 pm

Good luck with this. Years ago I kept up with the Megasquirt yahoo group and still own one of the early boards I've never used. Can't imagine trying to hide it all (full sized MS, that is) on an EN500, but it would be an excellent modification if you got it rolling. Would love to see more of this as you get started.
1999 Vulcan 500, 23k miles
Image

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Mon Nov 29, 2010 11:02 pm

I have decided to take the plunge re-engineer how this engine operates using the microsquirt controller here after reffered to as "µs." I need somewhere to keep a log of what I've researched, how things work etc etc. I've chosen to do this here since since I'm too lazy to set up a blog and this will server essentially the same function. Feel free to chime in with comments, questions, suggestions, or just be entertained by my progress or lack there of. I'm doing this in my minimal spare time so I expect this project could take upwards of 2 years to complete.

Relevant links(this list will grow as time goes on):
- µs manual - http://microsquirt.info/
- µs forum - http://microsquirt.com/
- electronics suppier (cam sensor) Digikey www.futureelectronics.com
- DIY autotune sells µs controllers and accessories
- EFI parts: Ross Machine Racing Electromotive. - Fuel injector bosses: http://www.sdsefi.com/

Background information on Vulcan 500 engine:
This engine operates like it is 1/2 of a typical inline 4 cylinder engine. This makes sense as it is a descendant of the inline 4 ninja 900. The Vulcan 500 engine is a 180 degree parallel twin. This means cylinder 1 fires, the crank rotates 180 degrees, cylinder 2 fires, then the crank coasts for the next 540 degrees and repeats. The ignition is a wasted spark (dual fire) configuration. This means the spark plug fires on both the power and exhaust strokes. This seems wasteful and counter intuitive but it is actually more simple and easier to implement. The electronic ignition on the EN500A is 10º BTDC @ 1,200 RPM and 37.5º BTDC at 10,000 RPM. Based on the experiments of others on Japanese motorcycles, the ignition curve is probably linear with RPM between these two points or at least that will be my assumption for now.

Background information on the Microsquirt controller:
The microsquirt controller has evolved from the Megasquirt open source DIY fuel injection project. It is a compact version in a weather tight case that lends itself to be easily adapted to motorcycle use. The µs controller can control both the ignition timing as well as fuel injection making it a complete engine management system. Why would I care about ignition timing you might think. If the ignition advance is linear, the timing is not ideal while under severe load or very light load such as while cruising at constant speed. I've read that 3 dimensional ignition control can boost fuel economy up to 15%. I intend to find out with my Vulcan 500.

The Plan:
Plan "A" is to have µs controlling the ignition timing and fuel injection using single fire ignition with sequential fuel injection. in simple terms, the spark plug and fuel injector will fire once for every 2 crank rotations. Plan "B" will be dual fire ignition and bank FI (the spark plug and fuel injector will fire on each crank revolution). In order to accomplish Plan "A" I need to shoe horn a cam position sensor in and a crank position sensor. the Crank position sensor will tell the µs what the crank speed and position is. The cam position sensor tells the µs whether the piston is on the combustion or exhaust stroke.

-Requisite hardware:
--Cam position sensor
--Crank position sensor
--Throttle position sensor (TPS)
--Intake air temp. sensor(IAT)
--Coolant temp sensor
--Manifold air pressure(MAP)
--Throttle bodies / intake manifold
--fuel injectors
--Fuel pump/regulator
--Idle air control (maybe?)
--O2 sensor. Wide band?
--Others i've neglected to mention at this point.

Step 1:
- Implement the ignition control first by essentially duplicating the 2d factory ignition and keep the factory ignition intact to make debugging easier. Cam sensor first, crank sensor second.

Step 2:
- Add FI bits and get the FI working with the 2d ignition.

Step 3:
- Tune ignition and FI using the additional sensors.

Step 4:
- Ride happily ever after into the sunset.....
ImageImage

burkbuilds
Posts: 350
Joined: Thu Jan 14, 2010 9:29 pm
Location: Rome, Georgia

Re: Mircosquirt Fuel Injection Project log.

Postby burkbuilds » Mon Nov 29, 2010 11:22 pm

Wow, that's a really intensive project. I hope it goes well. Please keep posting as you progress and let us know how it is going. Good luck! :D

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Mon Nov 29, 2010 11:58 pm

My first baby steps into Step 1:
I'm starting with what I think will be the hardest part. The cam sensor.
Click here to see the original thread on the µs forum
Microsquirt Forum wrote:Slow progress being made. I bought a Ninja 500 valve cover off fleabay for my prototyping before I cut any holes in my valve cover.
Here's what the top of my engine looks like before removing the valve cover.
Image
Under my valve cover this is what I have to work with.
ImageImage
Here's the ninja 500 valve cover. Kawasaki's "clean air system" injects air into the exhaust for emissions. If I'm going to use an O2 sensor I can't have this so I've tapped, plugged and filled in with JB-weld the small ports that lead to just outside the exhaust valves. I also made some caps to cover these cavities.
ImageImage
My initially thoughts were to simply mount a sensor directly above the cam lobe and use the cam lobe as a trigger, but after reading around on here I'm inclined to think that this wont work.

Second thought was to copy mxrob by putting a bolt in my cam chain sprocket. This doesn't look very feasible as there are many bolts etc in the way that would make it difficult if not impossible to get close enuff to where I could possible put a bolt thru the cam sprocket.

As suggested above I thought about having a trigger welded next to the cam lobe but, that would likely hit the valve adjustment screw/locknut.

I then noticed that the camshafts have a roughly 11-12mm hole all the way thru and about a 5-6mm gap at the ends. I was thinking of fabbing a trigger wheel to stick on the end of the cam shaft. I could put a bolt thru the entire cam shaft and screw it into the threads of a trigger wheel. This doesn't seem like a very good solution though since I would think finding a bolt that exactly fits the internal dimension of the hole unlikely and this would probably cause vibration issues throwing the camshaft way out of balance.

I discovered a small issue with the ninja 500 valve cover. it is about 3mm shorter in height. I'll have to verify this doesn't cause any interference issues. Plus I'll have to either get ninja 500 valve cover bolts or just shim the vulcan 500 bolts with some washers.

I guess I'm still looking for ideas on how I could do a cam position sensor and/or how I can secure a trigger wheel to the end of the cam shaft. I suppose I could weld one on but I'd prefer an option that could be repositioned if necessary.


There are two common types of crank/cam sensors. Variable Reluctance (VR) and Hall Effect. VR have an analog sine wave output where Hall Effect sensors have a digital output. From what I've read VR sensors are good for high speed signals but, don't work so well at low speeds (ie when cranking to start) Hall effect sensors are good practically down to zero. For these reasons I'm going to use a Hall type sensor for the cam and a VR type sensor for the crank. One of the problems with both of these sensors is they must be extremely close to sense their metal target. Some are less than 1mm!

The first thing as you can see from my copied post above is where the hell am I going to put a cam position sensor and how am I going to make it work?!?!? I am going to try to get a signal off the cam lobe. If that fails I'll fashion some sort of trigger wheel to the end of the cam shaft.

After researching some sensors at Digikey I've settled on Cherry's GS100102 gear tooth sensor. Unfortunately Digikey has min quantity of 1,000 so after some googling I found one at future electronics for about $45 shipped

It is important to note the max temperature for this is 125C(257F) and according to Cherry it has a sensing range of up to 3mm. So cross your fingers that this just might work. I hope my assumption that the top of the engine is normally not hotter than 100C(212F) is correct.

read about engine timing methods here, your brain will hurt I guarentee!
This also has some useful timing info
ImageImage

burkbuilds
Posts: 350
Joined: Thu Jan 14, 2010 9:29 pm
Location: Rome, Georgia

Re: Mircosquirt Fuel Injection Project log.

Postby burkbuilds » Tue Nov 30, 2010 6:55 pm

I like your new "avatar" but I found it difficult to concentrate on your post. ;)

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Fri Dec 03, 2010 2:13 pm

burkbuilds wrote:I like your new "avatar" but I found it difficult to concentrate on your post. ;)

ROFL. I saw it on the "darkside" forum and borrowed it from there. Oddly I suffered the same concentration problem as you but on that forum. Thankfully this forum doesn't show the "avatar" while posting. If it did I wouldn't be able to post anything intelligible :D. I suppose I should change it to something more palatable to the female members.... Although I'm not sure how much better my mug would be....

For your viewing enjoyment just return to this post:
Image :lol:

Anyway... I've got the hall sensor in. In my lack of foresight I didn't realize a 12mm x 1 tap would be so difficult to find locally. Amazon.com to the rescue. I'm JB welding a chunk of 1/4" aluminum to the flat spot above the cam lobe. This will give me a little more meat to cut threads in. As soon as the tap is delivered I'll figure out if the cam lobe provides a usable signal or not.

Then I move on to the fun of trying to figure out how to implement a crank position sensor. Unfortunately the factory trigger pattern on the flywheel is incompatible with the µs controller. I purchased an EX500 flywheel and a spare flywheel cover from fleabay to experiment on. I believe the flywheel is the same but the factory timing trigger is different from the vulcan :? . That's ok because I'll either be machining teeth into the wheel or pressing on a trigger ring.

Here is a secondary idea that a guy used on a thumper:
Image
Image
I'm first going to try what mxrob from Thumper talk did. I'm actually modeling a lot of what I want to do after what he did. Here's pictures of his trigger wheel from page 24 of his thread.

Image
Image
Image
Image
Image

I think I'm going to try using one of the stock VR sensors if I can figure out how to make it work.
ImageImage

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Tue Jan 04, 2011 11:50 pm

First baby steps toward success. I have a functional cam position sensor. It will be a while before I know if the µs controller can use the input signal as I want to have the crank position sensor and cam sensor worked out before splurging on the µs controller.

Here's the Cherry's gear tooth sensor:
Image

I stripped the paint off (well mostly off) the ninja valve cover and used JB Weld to attach a 1/4" thick piece of aluminum so that I have more "meat" to cut threads into.
Image

I positioned a piece of foam the same height as the sensor would be to verify it would clear the gas tank:
ImageImage

I installed the sensor with an o-ring and was slightly concerned by the proximity of the ignition wire as I feared it may cause interference.
ImageImage

wiring up a test LED:
I used the +12V power source for the turn signals at the flasher.
A convenient bolt was the ground.
I used two 4.7K Ohm resistors in parallel to get a 2.35K ohm equivilent "pull-up" resistor (Cherry documents recommended a 2.4K Ohm with a 12V source). This probably means nothing to you unless you've read the sensor's documentation. The sensor's output signal is the source voltage (12V in this case) continuously and goes to 0V when it detects a "gear tooth"
I used a 470 Ohm current limiting resistor for the LED.
ImageImageImage

I put in the key turned it on and nothing happened. This was good since I set the sensor position by having the cam lobe pointing up, screwed in the sensor until it hit the cam lobe, backed it out about 1/3 to 1/2 turn and locked it in place with the lock nut. I hand rotated the engine and the light went out. Several crank rotations verified it was working. I finished putting the bike back together and fired it up. Never thought a blinking LED would make me so happy :D As I revved up to about 3K RPM the blinking got so fast that my eyes could no longer tell it was blinking. Doing some math 3K RPM = the LED blinking at 25Hz... 10K RPM = 83.3Hz. I have a friend with an oscilloscope who unfortunately lives a few hrs away. I'll have to brave the cold winter air some weekend and go to his place to hook it up to the oscilloscope to ensure the sensor is indeed functioning normally between 3K and 10K RPM. I was going to post a youtube video but I doubt the low video quality would show the LED strobing accurately.

Installed sensor clearance from gas tank. Note I had to use washers as spacers on the valve cover as the ninja cover is shorter than the Vulcan. Frankly I expected oil to leak pretty bad around the washers but after 30 miles it hasn't leaked yet.
ImageImage

Next step is the crank sensor:
I need to pull things apart to figure out where to mount a sensor and how/where to get the trigger teeth cut into the flywheel. I picked up an '08 ninja 500 flywheel and a spare stator cover for the sacrifice.
ImageImageImage

The OEM pickups go inbetween the two pairs of screw holes at about 2 and 8 o'clock in the picture above. The EN500C only has one pickup at the 8 o'clock position in the picture. I would like to use an OEM pickup, using spacers to move it toward the starter clutch where I could have teeth cut into the flywheel. I won't have an idea if this is feasible or not until I tear things apart. I also wanted to keep the OEM system intact so I could switch between µs and OEM ignition to aid in trouble shooting. However I doubt the trigger teeth on the '08 ninja flywheel match the '94 vulcan 500 making this a less likely option unless I am willing to carve up my '94 flywheel instead!

Until next time......
ImageImage

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Thu Jan 06, 2011 11:31 pm

After brainstorming playing around with the flywheel and cover above, there doesn't appear to be enough extra "meat" on the flywheel to machine teeth into. Then examining how the starter clutch bolts to the fly wheel it looks like In between the starter clutch and flywheel would be a good place to put a trigger wheel. I'm thinking 1/8" thick or maybe 3/16" if there's enough room. I'll have take measurements draw up the trigger wheel in AutoCAD and send it off to be waterjet cut out. This will require a bit more experimenting and brainstorming to figure out where the missing tooth or teeth need to be and where / how I'm going to mount the crank sensor. Plus I need to disassemble the starter clutch to see if there is any reason I can't throw it on the lathe to machine it down to make up for the thickness of the trigger wheel.

WOW! I'm beginning to think there is a slight possibility this just might work!!!
ImageImage

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Sun Jan 09, 2011 5:54 pm

Due to my current employment situation this project is on hold indefinitely. :(
ImageImage

User avatar
Triangles
Site Admin
Posts: 819
Joined: Sun Dec 27, 2009 10:35 pm
I ride: '94 Black Cherry Vulcan 500, '06 Candyfire Red Vulcan 500 LTD
Location: Toledo Ohio
Contact:

Re: Mircosquirt Fuel Injection Project log.

Postby Triangles » Wed Mar 02, 2011 12:10 am

Employment situation is good! Expect a progress update within a week or two.
ImageImage

burkbuilds
Posts: 350
Joined: Thu Jan 14, 2010 9:29 pm
Location: Rome, Georgia

Re: Mircosquirt Fuel Injection Project log.

Postby burkbuilds » Wed Mar 02, 2011 8:45 am

Hey Triangles, I'm glad to hear your employment situation has changed to GOOD. Looking forward to your update in a few weeks. Congratulations.

JWR
Posts: 18
Joined: Fri Jan 07, 2011 12:06 pm
I ride: 2005 Vulcan 1989 Pacific Coast

Re: Mircosquirt Fuel Injection Project log.

Postby JWR » Wed Mar 02, 2011 10:01 am

Good work... keep it up.


Return to “Carbs & Fuel System and Intake”

Who is online

Users browsing this forum: No registered users and 2 guests