the plx is a logging solution like many others that i mentioned but the thing is timing has to be read from the ecu's facilities, i don't know of any timing sensors.

and that being the case timing is usually read one of two ways. way one is the normal odb2 port giving you about 15 events per second which is quite slow.

the other way is to bypass the port nad use some oem diagnostic function which bumps the data rate upwards of 200 events per second, reliably logging at least 100 while logging other things.

 

I haven't looked into it yet, but it would seem to me that I should be able to look at the cam angle sensor and the signal to one of the coil packs and log both. From that it should be possible to derive the timing advance. I'd have to pull out the manual to be sure (and it's out in the car), but I think I can log at least two channels at a 200 Hz. rate.

Am I overlooking something?

 

mmm how are you conncting to the cam angle sensor? ecu? if so then it'll go through the odb2 stuff i think...

actually, to make some corrections to this statement. if you were to do it generically by tapping a wire or two i think you'd be going through odb2 systems but there are ways to bypass it through software or maybe some firmware as that's how ecu+ does it.

i'm not sure of the details at all, since i'm not a programmer i never plugged for the details.

 

I haven't actually looked into the details of connecting to the cam angle sensor, but it should be possible to connect the signal (possibly after some conditioning or scaling) directly to one of the two digital channels, and they both operate at a rate from 0 to 10kHz., which seems as if it should be fast enough. When we hit the fuel cut at 9100 RPM that's still only 9100 pulses per second. Nyquist may apply if the signal can't be handled as a pure digital pulse (that is, if it has to be sampled), but on the surface at least I don't see any reason it can't be treated like a true digital source and trigger on either a rising or falling edge.

If anyone is aware of any reason this want work I'd appreciate a heads up. No need to go through the hassle of trying to make it work if the effort is doomed to failure.

 

Oh, BTW, I do also (sometimes) log OBD2 data to the ICPC, but I've never actually tried to use it for anything. Maybe I should look into it some more just to see how useful it might be, but at this point I really don't know. I'm pretty sure the data rate is too low for many data logging tasks.

 

Just did some more checking, and according to PLX the R500 can be used to log both the ignition pulse and shaft angle sensors (or anything else that produces or can produce a pulse). Since they specifically call out ignition pulse logging and shaft position logging in the owners manual (and a max pulse rate of 10kHz.) I can find no reason to believe that I can't log this stuff.

Guys, what am I overlooking?

 

How many teeth are on the timing wheel? What's coming out of most of those sensors is an analog signal that the ECU reads, amplifies and turns into a digital signal. Many of these signals are riding shielded cables back to the ECU (surprised me the first time I cut into a harness).


9100 RPM/60 = 182 RPS.
182 RPS * 12 CKP teeth = 2184 hertz.

So, 10K is overkill.

Looks like Honda depends on an accuracy of only 30 degrees for timing and injection events. They must be interpilating in the ECU. I don't believe the CKP would give a real leading and trailing edge though you could probably fake one with enough amplification (run an amp into clipping and analog becomes square)

From my reading the two cam angle sensors are just used as backups to the CKP sensor.

 

The cam position sensors are used to determine where cylinder #1 is to start operating the injectors.

CKP is also used to determine cylinder misfire .

 

Yeah, looks like 3 teeth on the cam position plate. Cam, runs at 1/2 the speed of the crank right? Wonder why honda uses 2 TDC sensors? Failsafe or increased precision?

 

The 3 teeth are not in the same position on the back of the cams. It's looking the correct part of the pattern to locate #1

If the cam sensor plates are bent slightly or the CMP A or CMP B signals are interupted, the PCM also goes into back-up and takes away most of the rev range and VTEC.