Jump to content

EFI Conversion Sensor Locations

Recommended Posts

I'm getting down to the end of working on a nearly bolt on EFI conversion for my xr650l (and maybe yours too!)

Our bikes only came with crank sensors so have provisions for mounting them, nbd.

Cam sensors on the other hand...

I believe I can use what appears to be a very convient casting in the valve cover that with a little milling should accomadate a standard sensor. It lines up almost perfectly with the camshaft. I've already found that there is some variance in how Honda cast these bosses, ie: with or without holes in them. Doesn't matter though, just a few more chips left on the table after the machine work.

So what I could use are few quick measurements to make sure this works.

Need inner and outer diameter of the two bosses(red and blue) as well as the center to center distance of the two.

 

sketch1504643412887.png

  • Like 2

Share this post


Link to post
Share on other sites

I don't have an extra cover to measure but if nobody else chimes in I'll pull my tank one evening in the next few days and get those measurements.  I think those are legacy bosses that were used to accept a cam driven tachometer on early RFVC engines.  

I'd love to see what you come up with for an FI system.  

I have to ask though;  on a single cylinder, fixed valve timing, engine why would you need both sensors?  Is it the cam chain and tensioner variable you're concerned with or is it just for redundancy?  I remember the days of early HD fuel injection and the use of both sensors by Magneti Marelli which were eliminated when they swtiched to Delphi.  The cam sensor has never returned.

 

  • Like 2

Share this post


Link to post
Share on other sites

I assume a cam sensor tells the ECM what stroke the motor is on , at least on a lost spark motor like the XR.

  • Like 2

Share this post


Link to post
Share on other sites
On 9/5/2017 at 6:28 PM, Hollerhead said:

I don't have an extra cover to measure but if nobody else chimes in I'll pull my tank one evening in the next few days and get those measurements.  I think those are legacy bosses that were used to accept a cam driven tachometer on early RFVC engines.  

I'd love to see what you come up with for an FI system.  

I have to ask though;  on a single cylinder, fixed valve timing, engine why would you need both sensors?  Is it the cam chain and tensioner variable you're concerned with or is it just for redundancy?  I remember the days of early HD fuel injection and the use of both sensors by Magneti Marelli which were eliminated when they swtiched to Delphi.  The cam sensor has never returned.

 

That makes the most sense now that you mention it. The bosses roughly line up a little off center from mid-line on the cam, which would account for the drive/driven gear interface. Nothing a little crafty machine work and the right sensor can't overcome. 

 

On 9/5/2017 at 6:52 PM, JoeRC51 said:

I assume a cam sensor tells the ECM what stroke the motor is on , at least on a lost spark motor like the XR.

Nailed it, accurate ignition and injection events will get the most power and fuel efficiency out of the system.

Having the injection event timed to intake valve opening = power and/or efficiency, especially when trying to match cam timing/overlap.

Inject it when it's closed = a little extra help cooling the head when things are getting on hot.

I'm also adding kick start to my bike and might have a way to make it way easier to kick these pigs over, especially hi comp ones, and knowing where TDCC is necessary to that end as well.

Edited by IslanderXR

Share this post


Link to post
Share on other sites
19 hours ago, IslanderXR said:

That makes the most sense now that you mention it. The bosses roughly line up a little off center from mid-line on the cam, which would account for the drive/driven gear interface. Nothing a little crafty machine work and the right sensor can't overcome. 

 

Nailed it, accurate ignition and injection events will get the most power and fuel efficiency out of the system.

Having the injection event timed to intake valve opening = power and/or efficiency, especially when trying to match cam timing/overlap.

Inject it when it's closed = a little extra help cooling the head when things are getting on hot.

I'm also adding kick start to my bike and might have a way to make it way easier to kick these pigs over, especially hi comp ones, and knowing where TDCC is necessary to that end as well.

What controller are you using?

Share this post


Link to post
Share on other sites
1 minute ago, IslanderXR said:

Microsquirt V3

That's kinda what I figured and why I asked.  I'm sure you know already but it's programming capabilities easily allow for crank rotation angle input only and either a single or multi-squirt mode to tailor the injection to needs of your goal (max power, max efficiency, or additional cooling as you mentioned).  It would seem to be an easier route that would work great to not modify the rocker cover and cam for the cam sensor.  I honestly didn't know the MS could use both sensors.  

I think it's awesome you're working on this.  It hasn't been very popular on the L (I've only read of a few with it) but something I've been interested in adding since I've gotten back into riding the Honda's on dirt.  I can't wait to see how it turns out.  Keep us posted.  

  • Like 1

Share this post


Link to post
Share on other sites
3 minutes ago, Hollerhead said:

That's kinda what I figured and why I asked.  I'm sure you know already but it's programming capabilities easily allow for crank rotation angle input only and either a single or multi-squirt mode to tailor the injection to needs of your goal (max power, max efficiency, or additional cooling as you mentioned).  It would seem to be an easier route that would work great to not modify the rocker cover and cam for the cam sensor.  I honestly didn't know the MS could use both sensors.  

I think it's awesome you're working on this.  It hasn't been very popular on the L (I've only read of a few with it) but something I've been interested in adding since I've gotten back into riding the Honda's on dirt.  I can't wait to see how it turns out.  Keep us posted.  

It can run with just the crank sensor, absolutely. The additional input from the cam is definitely in the double bonus area for the FI conversion. I'm sure I could set it up to run with or without and in turn offer it as a delete/add option. Not making someone get their valve cover machined would make the install easier and cheaper. I'm currently basing it around the components of a TRX700XX, kinda have a thing for keeping the Honda as much Honda as possible. The thought had crossed my mind to see if it would fire up on just the TRX ECM. But it's not easily programmable, doesn't have gauge outputs and won't give me the data I want. 

  • Like 1

Share this post


Link to post
Share on other sites
1 hour ago, IslanderXR said:

...I'm currently basing it around the components of a TRX700XX, kinda have a thing for keeping the Honda as much Honda as possible... 

Yep, I have the same thought.  You probably could make the 700 stuff work by adding a Dobeck (or similar) tuner to handle the fueling changes needed.  You'd still need to handle the ignition timing.  I'm not sure there is anything, other than Ecotrons, that can take total control and use a wide band O2, which is my goal.  I don't have any experience with the Ecotrons stuff but I've come close to just dropping the $800 and trying it.  I just haven't been able to pull that trigger.  May be a winter project if the real job slows down. 

Edited by Hollerhead

Share this post


Link to post
Share on other sites
On 9/7/2017 at 6:58 PM, IslanderXR said:

It can run with just the crank sensor, absolutely. The additional input from the cam is definitely in the double bonus area for the FI conversion. I'm sure I could set it up to run with or without and in turn offer it as a delete/add option. Not making someone get their valve cover machined would make the install easier and cheaper. I'm currently basing it around the components of a TRX700XX, kinda have a thing for keeping the Honda as much Honda as possible. The thought had crossed my mind to see if it would fire up on just the TRX ECM. But it's not easily programmable, doesn't have gauge outputs and won't give me the data I want. 

Aside from facilitating "sequential" fuel and spark, the cam sensor would also allow you to sample MAP in any specific window of the engine cycle. It can be quite useful as single-cylinder engines with tiny plenum volumes like ours have very wide MAP fluctuations at lower RPM.

Does the TRX brain generate any sort of data stream that can be tapped? It would be interesting to see the inside of one.

  • Like 1

Share this post


Link to post
Share on other sites

Latest MS code allows using the MAP data to indicate which part of the cycle the engine in on, so no need for a cam sensor. 
How many teeth on the crank ring?

There are several threads on TT and Advrider about adding EFI to singles, and they have discovered, and solved, most of the issues.

 

  • Like 1

Share this post


Link to post
Share on other sites

Interesting.
I wonder what purpose was intended for those two castings? Any other similar models that might have sensors mounted there?
And what is immediately inside the cover that could be used to trigger a sensor?
Most sensors detect the sharp edge of a passing ferrous material like a tooth wheel or gear tooth, but could  sense the proximity of a ferrous material like a rocker. Some also sense changes in a magnetic field, like the ones a XR200 uses on the cam to trigger ignition.

Since a 4T cycle involves two revolution of the crank,   and for sequential injection the ECU needs a way to determine what part of the engine's cycle the crank is on;  a cam sensor can do that but so can the latest uS code via the MAP sensor, or sensing the ignition spark if it is not wasted spark.   And for a lot of applications batch injection works just fine so it doesn't matter. One TT member used wasted spark for batch injection on what I remember as a DR350 or 400 single, he setup the uS as a 2 cylinders engine, then  only used one injector; and it worked good, so lots of latitude.
On a single cylinder engine crank speed varies a lot during an engine cycle and  uS really needs good crank position information, so a multi tooth sensor wheel on the crank will be needed if you want sequential injection. 

Share this post


Link to post
Share on other sites

The non-US XLR series had a cable driven tachometers (driven off the cam) as well as decompression levers that I believe went in those locations. The RM/LM versions though had electronic tachometers.

Share this post


Link to post
Share on other sites
On 9/27/2017 at 8:15 PM, Tmosely1 said:

I couldn't tell if you ever got these or not... this is from a 1989 XR600 cover. 

 

IMG_2831.JPG

IMG_2832.JPG

IMG_2834.JPG

IMG_2835.JPG

Thank You! That's almost exactly what I have measured out on mine.

  • Like 1

Share this post


Link to post
Share on other sites
On 9/28/2017 at 6:46 AM, Fingers said:

The non-US XLR series had a cable driven tachometers (driven off the cam) as well as decompression levers that I believe went in those locations. The RM/LM versions though had electronic tachometers.

That is also pretty useful. Get ahold of the tach drive housing/cable end, little reverse engineering, whammo, sensor adapter.

Thank you!

Share this post


Link to post
Share on other sites

It would be easier to put a magnet on the cam sprocket and an inductive sensor through the rocker box.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Reply with:


×