Megasquirt III on CBR600rr ignition advance problem

[spiky1987]

Hi !!!

Thanks all of you for your support, all this time the engine was "running good". I have found a way to make it rev to 12k - 15k rpm . I only had a small VR sensor problem.

After a bunch of tests ... and with the formulaSAE intake restrictor ... we got something around 45HP.
Now, we wanted to try some intakes setup and change the Ignition advance .... our problem ...

I can change the spark table to 45 - 60 - 90° of spark advance and there's ... no big difference, the engine still deliver like 20 lbs of torque...
I can see with the scope that the spark VS injection moves a little bit but not significantly ... anybody have an idea whit the spark advance table isn't working very well ??? Even at the idle I can put 90° and the engine starts ... that's almost impossible

This is our configuration : Toothed wheel ( 12 tooth )
Single Teeth CAM sensor ( on the ms3x )
4 Coil on plugs ( driven by the bosch ignitor chip ) ( ms3x )
4 injectors ( on the ms3x )


Thank you all

Simon

[Matt Cramer]

Could you confirm those results on the scope with a timing light? How far off is the actual timing from commanded timing?

[spiky1987]

It's hard to tell because these's no graduation on the crank.
I did some math to know exactly how much time in miliseconds is 1 deg at a fixed rpm, this way i could get the difference between 2 settings

[elaw]

If you're using a scope you're kind of doing it the hard way!

Here's what I would suggest. First if possible put the engine at TDC and make good visible marks on the flywheel and someplace adjacent with a black marker (I'm assuming the FW and area near it are aluminum). Connect a timing light to the engine on the cylinder #1 spark plug.

Now fire up TS and go to the "ignition options" screen. In the upper right portion of that window, set "fixed advance" to "fixed timing", "use prediction" to "no prediction", and "timing for fixed advance" to 0 degrees. Now start the engine... if it'll start with 0-degree timing. If not, you might have to set the fixed timing to 10 or 20. Use the timing light to see if the advance is approximately right by looking at the marks you made.

Now with the engine idling, try increasing the timing value in increments of 5-10 degrees. In each case, you should be able to see the timing advance.

One other question: have you verified somehow that the Megasquirt is reading the RPM correctly? The timing calculation is based on RPM so if RPM is not reading right, the timing won't be right.

[spiky1987]

It should be right ! We had some problem before ( cant go over 5 - 6k rpm ) but I changed the VR sensor with zener diodes and I now have a nice square wave

[spiky1987]

Does anyone know what settings affect the ignition table ???

Mine does not seems to work over 26 ... From 26 to 90 ... the engine react exactly the same. I wanted to know how the software do it's calculation and which settings may affect this variable ?!

I'm using MS3 with MS3X on a 4 cyl fully seq. with toothed wheels ( 12 ) and 1 cam teeth.

Thanks a lot !!

FsaeCVM
www.cvm.qc.ca/formulesae

[billr]

Post your .msq

[wes kiser]

Is the FSAE restrictor size not still 20mm for gasoline, or 19mm for E85? If not disregard what I am about to post.

You likely have something fundamentally VERY wrong with spark control. Have you actually made marks, and strobed the engine to verify commanded timing matches actual, and that this is true throughout the rev range?

I spent years doing FSAE, and did both F3 and F4's (yours should be "sportier"). The lowest power number we ever recorded was 70 fhp, and this was with 30 degrees retarded spark and primaries hot enough to melt off ceramic in 2 pulls (MOTEC cam sensor right on crank tooth edge, it was either proper or off by tooth/360). If you are less than 50 hp with a 600cc inline 4 and a 20mm restrictor, there is a major fundamental setup problem. Typical numbers for us wer low 70's on a dynojet, and low 80's on the engine dyno.

[spiky1987]

I got the .msq files for those who are interested to help me !!


Oh yeah ? Last year with our power commander ( fuel only ) we got like 44hp because we had no control on the spark, I wanted to get full ocntrol of the engine by putting the ms3 ... i'm disapointed for now

Maybe my mistake is in the "toothe number 1 angle" that's the only parameter I see ... but engine starts so I'm not that wrong.

[spiky1987]

Just got it today Sorry

[wrenchdad]

Does anyone know what settings affect the ignition table ???

Mine does not seems to work over 26 ... From 26 to 90 ... the engine react exactly the same. I wanted to know how the software do it's calculation and which settings may affect this variable ?!

I'm using MS3 with MS3X on a 4 cyl fully seq. with toothed wheels ( 12 ) and 1 cam teeth.

Thanks a lot !!

FsaeCVM
http://www.cvm.qc.ca/formulesae

Well, are you reving above 7000rpm? Because you timing table doesn't go above about 25 degrees until you get above 7K rpm. You may want to lower you rpm range on your table to better match what rpms your engine will do. If your engine is able to do those rpms, I would suggest that you blend your timing table more than you have now.

just my two cents
later wd

[jsmcortina]

Very important:

You likely have something fundamentally VERY wrong with spark control. Have you actually made marks, and strobed the engine to verify commanded timing matches actual, and that this is true throughout the rev range?

This is a critical setup step.

I'm concerned that your zener diodes on the tach input could be causing problems and throwing the timing off.

James

[zappaMan]

We had all sorts of trouble with the stock Honda VR sensors on a CBR600 F4i (for Formula SAE) awhile back. Timing would drift all over the place and no amount of swapping pins and noise filtering would help. Finally switched to hall sensors and everything works waaaaaay better. We didn't figure out what was going on until we set the timing to locked 13degrees and confirmed it with a timing light and the stock timing mark across the entire rev range (1500rpm-14000rpm). A timing light is your friend!

PS See you in Michigan... I'm car 33 (University of Victoria)

[taylorhoff]

We had all sorts of trouble with the stock Honda VR sensors on a CBR600 F4i (for Formula SAE) awhile back. Timing would drift all over the place and no amount of swapping pins and noise filtering would help. Finally switched to hall sensors and everything works waaaaaay better. We didn't figure out what was going on until we set the timing to locked 13degrees and confirmed it with a timing light and the stock timing mark across the entire rev range (1500rpm-14000rpm). A timing light is your friend!

PS See you in Michigan... I'm car 33 (University of Victoria)

Out of curiosity what type of hall sensors did you use for the cam and crank, and where did you find them?

We are running a CBR600RR for our FSAE car and have also found that the stock VR sensors are a pain. We are going to Lincoln though, so we won't see you in Michigan.

[zappaMan]

I ordered the hall sensors from digikey (they are really quite nice honeywell sensors. They are a bit pricey, but not too bad: 480-2811-ND)

[spiky1987]

Hi !

Thanks a lot for your tips, with the vr sensor on the crank, we had some problem ( like at 5k rpm ) some .... lag or whatever. Putting 5v zener made us have a nice little square wave, seems to be good.

Can you tell me what it your tooth #1 angle ? So i can compare with ours .... at least it's gonna be 180° offset if you have cut the opposite tooth on the cam.


My main point here is that I can Start the car with 90° ignition advance which is almost Impossible

[taylorhoff]

I ordered the hall sensors from digikey (they are really quite nice honeywell sensors. They are a bit pricey, but not too bad: 480-2811-ND)

Thank you!

Hi !

Thanks a lot for your tips, with the vr sensor on the crank, we had some problem ( like at 5k rpm ) some .... lag or whatever. Putting 5v zener made us have a nice little square wave, seems to be good.

Can you tell me what it your tooth #1 angle ? So i can compare with ours .... at least it's gonna be 180° offset if you have cut the opposite tooth on the cam.


My main point here is that I can Start the car with 90° ignition advance which is almost Impossible

Our crank wheel is modified with one tooth removed, making it a 12-1 wheel, which changes the definition of what tooth #1 would be. But in any case, our tooth #1 angle is approximately 0 degrees.

[spiky1987]

Oh s*** ! I didn't want to remove a teeth , cause it leaves you with only 11 tooth.... this is working well ?
But the tooth are not exactly in front of the sensot when the engine arrives at TDC no ?

[spiky1987]

Our crank wheel is modified with one tooth removed, making it a 12-1 wheel, which changes the definition of what tooth #1 would be. But in any case, our tooth #1 angle is approximately 0 degrees.

But do you still use the cam sensor ?

[taylorhoff]

Oh s*** ! I didn't want to remove a teeth , cause it leaves you with only 11 tooth.... this is working well ?
But the tooth are not exactly in front of the sensot when the engine arrives at TDC no ?

Our 12-1 wheel has the tooth removed so that tooth #1 (the first tooth following the gap) arrives at the crank VR sensor just before TDC #1 (approx 10 degrees BTDC). We are still actually experimenting with tooth #1 angles in TunerStudio, but it runs "ok" at "0" degrees (although we've just started tuning so there are many things to iron out).

The MS has no problem distingiushing teeth/gaps though if that is what you're asking (after some dialing in of the hysteresis/zero crossing potentiometers for both the cam and crank VR sensors).

Our crank wheel is modified with one tooth removed, making it a 12-1 wheel, which changes the definition of what tooth #1 would be. But in any case, our tooth #1 angle is approximately 0 degrees.

But do you still use the cam sensor ?

Yes, we still use the cam sensor as well, with two teeth shaved off of the original 3-tooth wheel.