how to time your engine

[afarfalla]

https://www.flickr.com/photos/37674324@N07/38263238742/in/dateposted-public/

top channel is cam position sensor

bottom is speedsensor

scope horizontal speed is 5ms

scope triggered by cylinder 1 injection

What I see is fuel injected 20ms before tdc and a slight delay till the valves close, car was idling

 

[\/\/0J0]

Interesting... Didn't you have 2 other threads that addressed this same topic?

The question is what does this tell you? 25ms in itself is not relevant data without a rotational speed or rate to be able to correlate to degrees before or after a reference event, ie TDC.
All of this is still not relevant to anything without having measured the sensor trigger notches on the crank and cam to know how accurately they are indexed to each other, how accurately they are indexed to themselves(the TDC condition of the respective shafts they are providing data for) AND how the ecu derives its timing data, whether leading edge, trailing edge, or other.
Another point of interest is... Does this guarantee anything aside from all parts being lined up? Any performance gains to be had or could it be that shifting the valve operation events, advanced or retarded, could have positive effects on performance?
It looks like you have your experiment process laid out but what is your hypothesis? How will you quantify your results? What steps are taken to ensure calibration of devices being tested(this relates to the indexing of components in your engine)?
I'm not trying to drop a deuce in your punch bowl but, instead, figure out what gains there are to be had with this method(if any) and provide some food for thought as you endeavor into the scientific process.


Sent from my mobile look-at device

 

[afarfalla]

yes, I posted on this subject before, this data collection was put up for discussion.

What I see is the ECU injecting fuel 20ms before tdc, I'm sure it takes into account other signals to arrive at 20ms.

There is a slight timing difference (3ms) between speed sensor and cam sensor, I could make this time up by adjusting the cam, but is it prudent? or is OK where its at?

 

[afarfalla]

all I can say is you guys are a bunch of dummy non-believers

I move the cam signal 2ms closer to TDC

Noticeable improvement at idle, nice and smooth, even with delete, it accelerates like a rocket and high speed acceleration, well lets just say much much better than before

Hopefully mileage will improve, also.

I'm happy with my new B5.5

 

[\/\/0J0]

What I see is the ECU injecting fuel 20ms before tdc, I'm sure it takes into account other signals to arrive at 20ms.

That may be so but what does it mean? ...more in a minute

There is a slight timing difference (3ms) between speed sensor and cam sensor, I could make this time up by adjusting the cam, but is it prudent? or is OK where its at?

"Slight" is a relative and highly subjective term. 3ms with regards to a bullet fired is the difference between near impact and post impact where, in that time, it could have ripped through its target. The target would agree that the difference is more than slight. In your engine, it needs to be quantified still. The piece of info needed is the exact rate of rotation that the engine was turning at the time of measurement. I know you said idle, but just for clarification, what is /your/ idle rate?
I have found idle figures between 800 and 900 RPM. Using 819rpm, we find that the engine rotates once every 73.26 milliseconds(60 seconds per minute / 819 revolutions per minute). By dividing this number by 360(degrees per rotation) we arrive at 0.20 ms/degree at 819 rpm. 3 ms of misalignment between crank and cam relationship correlates to 15° of difference, again at 819rpm. This may be significant or it may be slight... I am not qualified to make that assessment. Maybe someone like MrCHill, Franko6, or others that are well versed in the mechanics of engine performance at this level could aid in evaluation of this data.
Following the same math, 20ms of difference between start of injection and crank tdc correlates to 100 degrees at 819 rpm.
These differences in degrees will, of course, increase if the reference rate of rotation was higher. You can see why having accurate relational data is important.


Sent from my mobile look-at device

 

[afarfalla]

well VoDo, what I can tell u is this, my valves, closing, are 2ms closer to TDC than before and that slight change has made a noticeable improvement in my performance? I expect an improvement in MPG's, since my compression loss has been minimized.

After thanksgiving I'll be headed to FL so I'll have a great chance to further test my scientific break thru. 360 that

 

[afarfalla]

when I look at torsion all I see is a number which means nothing to me, I have no clue where TDC and when the valves are closing in relation to TDC, then there is the dude who put my engine together and wrote some number which has long been erased, so I change my torsion from 1.0 to 2, the only adjustment I can make is to the cam to make this change,, so did I move it closer to TDC, how would I know? You can change your torsion value all you want, you will have no clue what the timing is doing.

the timing for the engine in my car is exactly what the scope is showing me, I could care less what my torsion value is. When my number one cylinder is at TDC, I want my cam sensor to tell me my valves are closed, with my setup there is no guessing.

"vanishing effect" tell my engine that, now that it idles so nicely

 

[afarfalla]

http://forums.tdiclub.com/showthread.php?t=336279 this may help you

my torsion with engine running is 5.9 and injection occurs 1.2 degrees BTDC

 

[afarfalla]

lets just say I agree to disagree, your head is as thick as mine.

I believe if my engine was whacking off I would hear it.

I quote "unless there is a method by which you can determine accurately where the TDC actually is, the setting of the cam's torsion value will be experimental"

 

[afarfalla]

look at the 2 signals on the scope, do you agree that TDC has already occurred before the valves are closed? I see compression escaping because a valve is still open, that's lost power. So why did my engine smooth out at idle? Could it be my belt has stretched? Should I put my torsion back to 1.0? What will that accomplish, I say I leave it at 5.9 and drive to FL. If I whack off my valves then you win, but if I increase my MPG with no whacking off then you lose.

 

[imo000]

I never seen anyone use "wack off" to describe valve timing.

 

[pdq import repair]

whether timing is viewed by a scope or VCDS is really irrelevant in my opinion. Both values are measurements of the same thing just done differently. There are as I see it two great unknowns on this issue.

First unknown, and this is the biggie. As stated above also we have no actual reference to piston or valve position with total accuracy because we can not trust the crank and cam sensors without actual dial indicator off the piston and valve measurements. Sure they are close, but how close?

Second unknown is that we really don't know exactly which setting is best. On my 05 PD I initially set my torsion value at exactly 0 after a timing belt replacement. I noticed on the drive home that it idled rougher and seemed just a touch down on power. Next day I consulted all info I could find and setting varied across the board. I did find on Ross-tech that their recommendation for my Jetta was +1.5. I proceeded to set it there by slipping the sprocket a little. When I checked on VCDS it was 9.0, but it did idle smoother. I figured I wanted to know just what that would do for running so I drove it home that way.

It ran much better than ever before, but the next morning I got a code on startup, and it did start harder. The code was for cam/crank correlation which did not surprise me at all. Over the subsequent days I tried all manner of settings and settled on 6 as a happy medium for torsion setting. When I sold the car I did pop it back to 3 just because.

So we do not know optimum settings, and we can't trust the value of the data presented to be 100% accurate. On customer cars I set them +.5 to 1.5, but on my own I experiment till it feels best.

 

[afarfalla]

[




At a torsion value of 5.6º cam advance, (that's nearly a whole belt tooth of cam adjustment) your intake valves may well be kissing the tops of the pistons between intake and exhaust. If a valve head happens to drop off from whacking on the piston halfway to florida, don't say I didn't warn you...[/QUOTE]

^^^he coined the phrase

 

[afarfalla]

I moved the cam adjust a tooth and a 1/2, the cam signal moved 2ms closer to TDC, torsion value changed to 5.9 and I could watch this all happen with my scope and modified ECU. You can move your torsion values all you want you have no clue what's really happening.

I would think the Germans are capable of putting sensors where they belong

 

[imo000]

[




At a torsion value of 5.6º cam advance, (that's nearly a whole belt tooth of cam adjustment) your intake valves may well be kissing the tops of the pistons between intake and exhaust. If a valve head happens to drop off from whacking on the piston halfway to florida, don't say I didn't warn you...

^^^he coined the phrase[/QUOTE]
No he didn't. On is not off.

 

[afarfalla]

close enough for me, you guys are boring me now...later

 

[\/\/0J0]

close enough for me, you guys are boring me now...later

I'm still waiting on empirical data to be presented, along with the answers to the slew of questions posed in my first reply, many of which quick td repeated but it's us "dummy non believers" that are boring you? You are a piece of work, generating multiple threads on the same topic and not posting any hard data or anything really other than showing off your o scope. I have a few of them, myself, you know. They are useful but without understanding what exactly you're measuring or why, it really is a waste of time, especially when you choose not to listen to even those trying to help you arrive at useful and relevant answers.

Sent from my mobile look-at device

 

[afarfalla]

I'm still waiting on empirical data to be presented, along with the answers to the slew of questions posed in my first reply, many of which quick td repeated but it's us "dummy non believers" that are boring you? You are a piece of work, generating multiple threads on the same topic and not posting any hard data or anything really other than showing off your o scope. I have a few of them, myself, you know. They are useful but without understanding what exactly you're measuring or why, it really is a waste of time, especially when you choose not to listen to even those trying to help you arrive at useful and relevant answers.

Sent from my mobile look-at device

In my first post give me your interruption of what your looking at on the scope. Tell me what you see.

 

[Windex]

In my first post give me your interruption of what your looking at on the scope. Tell me what you see.

Doesn't work that way.

"Hey look, I did something, and here are my results. My opinion, based on one car and one car only, is that this is a better way to measure than all of the hundreds and thousands of cars that were measured before me. Prove me wrong"

I'm paraphrasing, but that's essentially what you are saying.

Lunch, you are out to it.

Your scope has no way to account for crankshaft acceleration, something the ECU monitors and adjusts for.

If you really cared for any kind of scientific inquiry, you would make an adjustment then drive the car for a period of time and measure fuel economy etc, maybe go so far as to get the car dyno'd.

Instead, you just spout meaningless observations like "accelerates like a rocket", and then call everyone in this thread "Dumb"

You're a troll, nothing more.

 

[afarfalla]

interpret the waveform, I don't think any of you can understand what your looking at and what it represents...tell me if I'm wrong...the top waveform is telling me that my valves closed 3ms after TDC, I move the cam setting a tooth and a half it moved the top waveform 2ms closer to TDC, torsion was 5.9 car ran and idled perfectly, I moved the cam setting back 3/4 of a tooth, that equaled 2ms from TDC, torsion 3.9, car idles and ran horribly. 3ms from TDC equals torsion of 1 and the car idles bad. So where do I set my torsion?

When I moved the cam setting a tooth and a half, the top waveform was over the trailing edge of crank pulse, car seemed to run the best there

 

[afarfalla]

No, the top waveform tells you the cam sensor was triggered by the tone ring 3ms before TDC. There is no indication of valve position.

The valves have been closed for nearly 180º of crank rotation (they close just past BDC), works out to roughly 37 milliseconds by my quick math.[/QUOTE

so the tone ring does not indicate valves are closed, the valves close at some other point in time so put it back to 0 or 1 torsion and forget it, now I'm totally confused I'll have to find a good explanation of how this engine works thanks

 

[afarfalla]

you guys win, I thru in the towel, set it to +0.5 and thru my experiment in the garbage. Quick TD's post made me a believer

I'll have to find another breakthru

 

[Windex]

The key is that the cam sensor tone ring is not locked to the cam lobes - it is variably positioned based on the three bolts and slots holding the camshaft sprocket to the camshaft hub.

With your scope, you are only measuring the cam sensor, and not the cam itself.

 

[afarfalla]

torsion at 5.9 it thru code P0503

 

[oilhammer]

The key is that the cam sensor tone ring is not locked to the cam lobes - it is variably positioned based on the three bolts and slots holding the camshaft sprocket to the camshaft hub.

With your scope, you are only measuring the cam sensor, and not the cam itself.

This is not true. The tone wheel part is fixed in one position to the camshaft vi a keyway. The sprocket is what adjusts with the three bolts.

I agree that this "experiment" is quite pointless, as the ECU is already doing a good job and this data is displayed already.

 

[Ol'Rattler]

I read that and thought WHA? If the CPS tone wheel was on the sprocket all torsion would be measuring is belt stretch/lenth.

 

[afarfalla]

you may call this experiment "pointless" but consider this...with torsion set to +0.5 (where the experiment started) engine idles rough.

With no scope you adjust the sprocket, read torsion and drive the car, with the scope you could observe the results of sprocket adjustment in time.

Torsion is just a number it has no bearing on how well your engine is running.

The mechanic changes a TB using the prescribe tools, at completion torsion reads +0.5, mechanic is happy he gets paid but is the engine timed?

 

[afarfalla]

Does anybody know when this engine is timed? There has to be a point when fuel delivery, crank and cam position align and make for the best timing for the engine. We can control it but where should it be? What torsion value is correct? A reading of anywhere from -2.0 to + 2.0 just don't cut it for me? Why does my engine idle so nicely at +3.0 thru +5.0?

I see what happens on my scope the cam signal gets closer to the crank signal, the fuel timing changes and the car runs better?

 

[imo000]

close enough for me, you guys are boring me now...later

I thought you are gone!

 

[afarfalla]

you didn't answer my question.