Compatible Crank Angle Sensors (CAS, CKP)

FoundSoul · Post by **FoundSoul** » Wed Jan 18, 2006 7:56 am

Guys-- I'm constantly helping people determine which MegaSquirt systems will be compatible with their stock ignition systems, or if that's possible at all. And in most cases this comes down to if the stock CAS can be decoded. Currently though (at least that I can find searching around) we don't really have this information together for people to be able to help themselves, and I've sometimes got questions myself.

For instance-- the 'given' configuration that will definitely work: A single single coil, distributor based, even fire engined vehicle with an Optical, Hall or VR sensor (sensor doesn't really matter with v3 PCB anymore). As long as the CAS has an equal number of 'positions' to the number of cylinders that the vehicle has, and they are evenly spaced, then presto! No problems. For VR sensors this would be teeth on the wheel. For Opto or Hall sensors this would be slits or Windows.

So what about a CAS with 360 1 degree slits around it? That sounds like it should be easy enough to read... but does the MS-II code do so? What about MSnS_E? MS-I? (Most standard MS-I B&G fuel only code people would just trigger from the coil I suppose)

Here's a specific situation that motivated me to post this. The early 90's Nissan 240sx. Two optical sensors. Two rings of slits on the CAS wheel. The outer ring is 360 slits 1 degree apart. The inner ring has 4 slits 90 degrees apart (or 180 degrees of crank rotation apart since this wheel spins at cam speed which is half crank speed). I was just looking this over last night as I want to buy a project car for my new employee where he can use the stock ignition system and I feel pretty good about this but I have one minor hesitation…. The spacing of the inner 4 slits is right, but the No1 cyl slit is much wider than the rest, and still centered at 90 degrees it appears, meaning that the square wave will start sooner and end later than the other three. If it started at the same time as the other three and just ended later that may be fine as we could trigger from the rising edge and ignore when the pulse ended but I’m not sure about this arrangement as it appears the rising edge will start sooner.

What do you guys think? I'd love to see us turn out a resource of examples of supported configurations, illustrated. I can pulls pictures like this one of different CAS configurations to use as examples if needed.

allen22 · Post by **allen22** » Wed Jan 18, 2006 9:20 am

FYI, 95-02 Nissan Maxima's use a 360 tooth crank wheel on the flywheel, any compatibility information would be helpful for us Maxima users too.

allen

woh · Post by **woh** » Wed Jan 18, 2006 9:32 am

Jerry

To be sure of what the waveform is we need to look at this on a scope. The diagram may or may not reflect the actual wheel.

FoundSoul · Post by **FoundSoul** » Wed Jan 18, 2006 9:47 am

Can't you tell though from looking at the disc that the waveform would look like this:

Or am I wrong to assume that? I don't have the car here to test unfortunately... I know that wider window is there to indicate no1 cyl TDC so that info must get to the stock ECU. So I'm thinking it's got to be reflected in the waveform. If I'm right on that the question is what will the MS do with this?

woh · Post by **woh** » Wed Jan 18, 2006 10:04 am

The point of interest is the alignment of the edges from the four slits with one large slit. As you said if the edges aligns to TDC of each cylinder then that would make things easier.

I don't think we should assume what comes out of the module. If we know what actually comes out we may have several options.

Do you know if there is one output (ie combining both tracks) or two outputs, one for each track?

FoundSoul · Post by **FoundSoul** » Wed Jan 18, 2006 11:35 am

There are two outputs, one for each track.

newtyres · Post by **newtyres** » Thu Jan 19, 2006 2:50 am

Jerry,
I am playing with a very similar wheel and sensor from a Nissan SR20 engine, the inner track has 4 different sized windows, small to large, that are equally spaced 90 degrees apart on the trailing or leading edge depending on what direction the wheel is turning (or what side of the wheel you are looking at). It also has the 360 outer ring. This one can be set up to deliver 4 equally spaced triggers. With a pullup resistor to 5V, the windows give a high signal. I've run this up on the bench with the twin sensors driving the MS input pins directly (bypassing the optocoupler) and it looks good, i.e. MS is decoding and rpm is steady, and spark outputs OK. Because I am using 2nd trigger I shaped a jeweller's screwdriver into a miniature chisel and punched out a window in the right place in the outer ring with it's 360 slits to make one window (2nd trigger) and painted in all the rest of the 360 slits so I have 4 inner windows and one outer window. Note that the 2 optical sensors are staggered - not in line - and they will output a correct signal at zero rpm which makes it easy to set the timing. I think this type of sensor could be the universal sensor of choice for those that don't have an engine with direct code support. I have the experimental odd-fire MSnS-E code running on the bench with this setup.

Ian.

FoundSoul · Post by **FoundSoul** » Thu Jan 19, 2006 6:30 am

newtyres wrote:Jerry,
I am playing with a very similar wheel and sensor from a Nissan SR20 engine, the inner track has 4 different sized windows, small to large, that are equally spaced 90 degrees apart on the trailing or leading edge depending on what direction the wheel is turning (or what side of the wheel you are looking at). It also has the 360 outer ring. This one can be set up to deliver 4 equally spaced triggers. With a pullup resistor to 5V, the windows give a high signal. I've run this up on the bench with the twin sensors driving the MS input pins directly (bypassing the optocoupler) and it looks good, i.e. MS is decoding and rpm is steady, and spark outputs OK. Because I am using 2nd trigger I shaped a jeweller's screwdriver into a miniature chisel and punched out a window in the right place in the outer ring with it's 360 slits to make one window (2nd trigger) and painted in all the rest of the 360 slits so I have 4 inner windows and one outer window. Note that the 2 optical sensors are staggered - not in line - and they will output a correct signal at zero rpm which makes it easy to set the timing. I think this type of sensor could be the universal sensor of choice for those that don't have an engine with direct code support. I have the experimental odd-fire MSnS-E code running on the bench with this setup.

Ian.

Interesting... What year/make/model? Maybe I can look up an image of that CAS as well... Were you getting a stable RPM signal with just the inner track of 4 windows or did you have to do the 2nd trigger? Did you put the 2nd trigger right at cyl1 TDC?

78Spit1500Fed · Post by **78Spit1500Fed** » Thu Jan 19, 2006 8:49 am

FoundSoul wrote:Can't you tell though from looking at the disc that the waveform would look like this: ...
Or am I wrong to assume that? I don't have the car here to test unfortunately... I know that wider window is there to indicate no1 cyl TDC so that info must get to the stock ECU. So I'm thinking it's got to be reflected in the waveform. If I'm right on that the question is what will the MS do with this?

I think you're wondering if the waveforms line up like the ones below, right? Either the rising edge or falling edge are aligned, or neither?

FoundSoul · Post by **FoundSoul** » Thu Jan 19, 2006 9:16 am

Correct--- Am I right that this is what's important? Or will the MS's coding 'figure it out' based on the rest of the pulses and timing of it all?

510rob · Post by **510rob** » Thu Jan 19, 2006 12:50 pm

http://www.autronic.com/page_files/trigger_disks.htm

?

78Spit1500Fed · Post by **78Spit1500Fed** » Thu Jan 19, 2006 2:17 pm

510rob wrote:http://www.autronic.com/page_files/trigger_disks.htm

?

Way cool.

There's your solution if you find that the wheel doesn't have an evenly spaced rising or falling edge!

-Brian

78Spit1500Fed · Post by **78Spit1500Fed** » Thu Jan 19, 2006 2:20 pm

FoundSoul wrote:Correct--- Am I right that this is what's important? Or will the MS's coding 'figure it out' based on the rest of the pulses and timing of it all?

That is what is important... either the rising or falling edge should be the same as the rising or falling edge of the other windows.

I would still scope it to verify your suspicions. The drawings in books may or may not reflect the actual production part... it may be for concept only.

-Brian

PSIG · Post by **PSIG** » Thu Jan 19, 2006 2:52 pm

My view is that this could be a very desirable wheel in that you could block the combination of holes to suit your engine and distributor alignment or use EDIS/DIS coil packs. You could block all but 8 slits and just run the dizzy to MS and back through a VB921 to a single coil like points. You can block all but 7 slits on the perimeter and run missing-tooth code to dizzy or coil packs. You could just block 3 slits on the inner section and run dual wheel code. You can leave every 5th slit open (except 2) and run EDIS. You could leave every 3rd (except 4) open and run Bosch-style 60-2 code. You could....

FWIW,
David

newtyres · Post by **newtyres** » Fri Jan 20, 2006 1:09 am

Jerry,
I am a newbie at spark implementation, but it appears on the scope that the MS spark input is triggered on the falling edge of the input signal, and with this sensor that is the trailing edge of the window. MS doesn't care (within limits I guess) how long the signal is high for before it falls. I know this because I've run it up, if I put my wheel in the wrong way around the rpm is all over the place, since then the trailing edges are not evenly spaced.

The rpm is stable on my setup on the bench, this setup should be OK for nearly any "2nd trigger" or "wheel decoder" MSnS-E setup I would think but I really have only researched the project I am doing and don't know much or anything about any other setup. The sensor I am using came from a Nissan SR20 engine which is apparently used in the 200SX and Sylvia from 93/94 on. I'm sure it's the same as the one pictured here:

http://megasquirt.sourceforge.net/extra ... wheel.html

Ian.

ModToad · Post by **ModToad** » Thu Feb 09, 2006 12:31 pm

Just some hearsay for you ...

I'm attempting the same setup on an FJ20ET (old Skyline 4 cylinder from '84) and it also uses the dual wheel CAS setup as mentioned above.

A few people on this board have come forward to tell me that they have used the inner wheel only (the 4 slotted one, since they are separate outputs) with pullup to 12v and triggered the factory ignition module/transistor with a pullup to 5v.

They all have stable RPM (from their seat-of-the-pants testing) and get full advance as well.

It seems this wheel is common on all Nissan engines.

I've had a few apart (I run a Nissan tuning shop) and can verify the SR20DET, SR20DE (RWD) KA24DE (USDM S13/14 Silvia 240sx) CA18DET, CA18ET all use the same CAS setup.

canadiandrifter · Post by **canadiandrifter** » Sun Feb 19, 2006 1:52 pm

I am a newb to the MS and I am looking to run the MSnS-E on my NIssan 240sx with the KA24de engine. I am very interested in seeing the resultsof every ones findings and I will add this, If you can use this CAS tor trigger the ms then you should be able to use the MS to trigger the COP's right?

Just to let you know the SR20det engine found in Japan use this very sensor as mentioned but it also has coil on plug. Would you need the ignitor to run such a set up?

Chris

Oh by the way I have a bunch of the nissan dizzy's with the opto sensors laying around so if any one would like one let me know.

FoundSoul · Post by **FoundSoul** » Mon Feb 20, 2006 9:38 am

The problem with triggering COPs is two-fold, A) you need a sync signal to tell the ECU when piston one is a TDC (this is why one hole is bigger, but the MS won't decode that), and B) you need multiple spark outputs.

Both can be addressed using the MS-1 with MSnS-E firmware combined with extra coil drivers and a missing tooth wheel. Though this doesn't use the stock CAS at all. This page covers more of this info: http://megasquirt.sourceforge.net/extra ... wheel.html

Drop me an email about one of those 240 dizzy's, I would mind testing one out if you've got one to offer up. websales AT diyautotune DOT com.

NA_KA24DE · Post by **NA_KA24DE** » Tue Apr 04, 2006 5:24 am

newtyres wrote:Jerry,
I am a newbie at spark implementation, but it appears on the scope that the MS spark input is triggered on the falling edge of the input signal, and with this sensor that is the trailing edge of the window. MS doesn't care (within limits I guess) how long the signal is high for before it falls. I know this because I've run it up, if I put my wheel in the wrong way around the rpm is all over the place, since then the trailing edges are not evenly spaced.

The rpm is stable on my setup on the bench, this setup should be OK for nearly any "2nd trigger" or "wheel decoder" MSnS-E setup I would think but I really have only researched the project I am doing and don't know much or anything about any other setup. The sensor I am using came from a Nissan SR20 engine which is apparently used in the 200SX and Sylvia from 93/94 on. I'm sure it's the same as the one pictured here:

http://megasquirt.sourceforge.net/extra ... wheel.html

Ian.

Ok, the question still isn't clear to me. Did I miss something? Did you have to use the second (i.e. 360 slit) wheel? I'm considering setting up my 95 240sx to run ITB's managed with MS, and I'm wondering if I'll have to go EDIS or not.

FoundSoul · Post by **FoundSoul** » Thu May 04, 2006 11:52 am

Here's an update and great news for guys running a 240sx or 300zx with the KA24DE or VG30E engine--

I’ve been wanting to test this for a while now to make sure the 240sx and 300zx distributors would trigger the MS properly. I went to a local junkyard and picked up a 300zx distributor to test it with since the trigger wheels are very similar and I built my test rig this afternoon and good news—the 300zx dizzy triggers the MS-II just fine, which means it should also trigger the MS-I with MSnS-E without a problem if you want to go that route. I did use a 12v pullup on the ignition input through a 1k resistor, and you have to trigger from the falling edge as they are all in alignment, the leading edge is off alignment on the bigger hole which causing instability in the tach signal. Falling edge is rock solid—I was able to put in on my hand drill and test it at 700 and 1600 RPMs, and then on my drill press and I cranked it up to 6100 RPMs and it was rock solid. This should work great.

Here’s what the KA24DE trigger wheel looks like:

The 300ZX VG30DE Distributor’s wheel looks just like this, but with six 120 degree slits (crank rotation, = 60 degrees on this cam-speed wheel) instead of 4 180 degree slits. And it triggers the MS just great. I would have tested a 240 distributor but I couldn’t dig one up at the local pull-a-part.