Project Execution and Strategy [and NEW QUESTION(S)]

auditude at auditude at
Fri Feb 22 11:37:05 EST 2002

Orin Eman orin at wrote:
> Now, since I actually have some time today...

Thanks for contributing Orin! :-D

> >  I need to understand the output of the hall sensor, relative to
> >  crank location.  Apparently this entails the use of many items that
> >  I don't have possession of or experience with their use.  My
> >  current understanding, from emails and postings, is that the hall
> >  sensor output is normally "high" (at 5 volts?) and "low output"
> >  window is 40 degrees wide, and this window is centered (when the
> >  distributor is indexed where it should be) on the flywheel pin
> >  location of 62 degrees BTDC.  One specific clarification I would
> >  like is whether this is on the 360 scale of the crank, or is it on
> >  the 360 scale of the distributor itself.  I would imagine that it
> >  is the crank, but since I can't measure jack-sh!t currently, I
> >  can't confirm anything.
> On a 4 cyl dwell meter, it measures 5 deg.  That's 5 out of the dwell
> meter's assumed 90 degrees, or 20 out of 360 degrees at the
> distributor, therefore 40 out of 720 crank degrees.  That's all I used
> to get the 40 degree number.  Oh, they are rarely 'centered' on the
> timing pin. As long as the timing pin signal occurs during the hall
> sensor window, the engine will run.

See, stupid me didn't even get the first couple times reading this
that you are referring to measuring a 5 cylinder motor with a 4
cylinder dwell meter.  And never having used a dwell meter makes
me have to think a little more about this to understand it (90
degrees on a 4 cylinder meter, etc.).

(Another example of my misunderstanding, was your reference to
pulling the signals from the MAC11 ecu.  I thought you were
referring to pulling the signals just in order to understand them.
Someone else later pointed out you (may have) meant to
incorporate the actual MAC11 ecu into the new installation, along
with the standalone efi ecu.  I didn't read it that way.)

I understand about the centering thing.  Thanks for the clarification
on the 40 degrees out of what.  I hadn't even realized it was out of
720, I thought it was 360, and wasn't sure if it was distributor or
crank 360.  So wrong I was.  It makes perfect sense now, having
read that.  If it were 40 degrees on the 360 crank scale, it would be
the same as the freaking reference pin.  I feel a bit silly now.  I
think the/a book will help me get into the right frame of mind.

On the centering, if I quote a certain indexing of the hall sensor,
based on math (see below), then I can always rotate the distributor
into position so that it will work with the new ecu.  (meaning I can
center the distributor myself if that's what I tell them I'll be doing)

> >  In order to determine physical TDC accurately, it seems I need a
> >  dial indicator and a suitable mounting base, to make direct contact
> >  with the piston to find where it is located.  I need help here,
> >  both with locating and choosing the correct tool, and with learning
> >  to use it properly.
> >
> >  In order to determine crank position, I also apparently need to
> >  locate and learn to use a degree wheel.  These items apparently get
> >  attached to either a crank pulley or the flywheel, and tell you how
> >  far you are rotating the engine.
> Why?  There is a zero on the flywheel which indicates TDC on cyl 1.
> The timing pin is 62 degrees before TDC on cyl 1.

Well, to answer that question, the reason for that is to be able to
tell the vendor/manufacturer of the ecu when the hall sensor output
will "go low", relative to the position of the crank.  Since it's not
going to "go low" at 62 btdc cyl 1, or at 0 tdc cyl 1, I thought I
needed a way to determined exactly where it does "go low".

That is why I wanted such specific information regarding how it
works.  If I can center my distributor on the crank reference pin,
then I (think I) can just do some math to determine where the hall
sensor window "goes low".  In my mind, if the information I have is
correct, and my understanding of it is sound, then I should be able
to give the manufacture the necessary information to produce the
ecu.  This is without even having to learn about all the extra stuff
that I was thinking I needed to know or measure personally.

The dial indicator/degree wheel/o-scope path was suggested to me
by the ecu vendor, hence my thinking I needed to go all that schtuff.

Seems to me if the math/logic question below isn't
sound/applicable, then I need to know where the hall sensor output
edge is relative to TDC, and if there isn't a mark on the flywheel
corresponding to that exact position, then I either need a degree
wheel, or need to use some other method to know how many
degrees are between one of the two reference points (0 on the
flywheel, or the pin) and the edge(s) of the hall sensor.

> If you are going to be using the distributor for timing, dial
> gauges etc aren't going to help... you need to be thinking
> timing light once the engine is running.  If you are worried about
> timing sequential injection, it really doesn't matter that much -
> totally irrelevant at high load and only really affects emissions at
> low load.

Well, I have to admit that I am concerned about timing sequential
injection.  But to put it in perspective, having a running motor would
be a pretty good start.  At this point, I only want to be able to tell
the efi ecu when to start the counting off of cylinders.  Which
cylinder is number 1, where is it, and when should it fire the spark.
all the other cylinders are calculated off the rpm sensor.

I would like to say that people seem to discount the value of
sequential injection as only being beneficial at low rpms, and only
for emissions.  There are benefits, and "if it's all the same in the
setup", then I'd like to "switch" that feature on.  If I can incorporate
that it into the system, especially by correctly providing them the
sync trigger location (is there really an argument "not" to do this, or
is it just the extra "trouble" to do so?), then I want to do that.  I've
already gone through the trouble to locate and buy a system that is
capable of running it in the course of a normal installation.

> >  I'd love to hear that I'm wrong about how complicated this is, and
> >  that there is either some other way to get the information I need.
> >  For example, it was said that SJM's site had the o-scope waveforms
> >  of the hall sensor.  I didn't find that, I only found the rpm and
> >  timing reference sensors output waveforms.  Does anyone know where
> >  I can find a waveform of a hall sensor output?
> Just a rectangular wave.

Okay, that's perfect.  "Now," I just need to know if I can go ahead
and subtract half of the 40 degree window from the reference pin
location of 62 btdc cyl 1, to get 82 btdc cyl 1 as the position of the
falling edge of the hall sensor.  Then I can do the opposite and say
that the rising edge of the hall sensor output will occur at 42 btdc
cyl 1.

Then I can call the vendor, give them the specs they've been
waiting for, and get my stupid ecu made, get it, and install the thing!
Right?  Or, wrong?

It's the fact that it could be that simple, and I just needed
clarification to that effect in order for me to proceed, that "had"
gotten me a bit frustrated.

It is my obligation to be humble and appreciative of any and all
information someone decides to give me on the matter.  So, I've
intentionally "not" sent more than one email to individuals who
didn't respond to questions.  I know there is no obligation to
respond, I was just in the dark, not knowing if I'm a big jerk-off that
comes across like he's insisting on answers, or if it is simply a
matter people not having time to give me a detailed response.

> >  If anyone has the ability to measure the output of the hall sensor,
> >  and is willing to do so, will you please contact me or post to the
> >  list?  Maybe that's what I should ask?
> It's all above.  Low for 40 out of 720 crank degrees.  Pulled up
> to 5V by the ECU otherwise.

Thank you SO MUCH, Orin!  That is absolutely fantastic!

My understanding about the "pulling up" by the ecu is limited.  If
interested, see my questions above as to what I wonder

A question that comes up (I'm sorry for being this way
(overanalytical), but I can only be myself.  Believe me, it sucks
often.), is:

Does the ecu do anything that affects the output of the hall sensor,
"besides" providing a constant 5v power source?

Writing that it is pulled up by the ecu when the shutter is not
blocking the signal (input or output?), makes me think that the ecu
does something different when the shutter is in position.

This is the meaning of ambivalence.  I have renewed excitement
and energy for the turbo/efi project in general, but the uncertainty
pains me.

Thank you all for all your help!


More information about the quattro mailing list