[Diy_efi] ignition timing variation

Arnoud van der Wel A.P.vanderWel at utwente.nl
Thu Aug 1 15:09:27 GMT 2002


This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C238B7.EC247CA0
Content-Type: text/plain;
	charset="iso-8859-1"

Hi,
 
I built a microcontroller based ignition system for my motorbike. The system
is up and running; I use one pulse every two crankshaft revolutions (the
sensor is mounted on the camshaft) and I don't get more 'jitter' than approx
+/- 1 degree, and that only at very low engine RPM. (approx 1200). The bike
does not have a particularly heavy flywheel. Above 1500 rpm, timing is
immaculate (fluctuation << 1 degree).
 
I base my timing lookup on rev period like you do and I use the previous
revolution period for the lookup. I do no averaging.
 
Beware, your sensor may have erroneous inputs. In my first prototype, my
sensor would trigger on the ignition firing pulse, then it would trigger on
the ignition firing pulse of the other cylinder, and then it would trigger
on the generator which is mounted close to the ignition pickup. Now it
doesnt trigger on either of those three sources and timing is much better...
:) 
 
For diagnosis, you could program an 'advance curve' that is constant for all
rpm and all throttle position.
 
FWIW...
 
Regards,
 
Arnoud.

-----Original Message-----
From: Stewart Prince [mailto:sprince at csun.edu]
Sent: Wednesday, July 31, 2002 9:39 PM
To: diy_efi at diy-efi.org
Subject: Re: [Diy_efi] ignition timing variation


After speaking with the student writing the software, I now have more ammo
to respond with....we're using a Motorola 6811 for both timing and fuel.
With 8 teeth on the crank, there is 45 degrees between teeth.  The time
between teeth is inversely proportional to engine speed, however only engine
period and throttle position are used in the ignition lookup table.  They
have a bench version working with a hall effect sensor and wheel mounted on
an electric motor, and I observed little or no fluctuation of timing here.
The engine is mounted on a water brake dyno with a large flywheel.  I looked
at the timing with a standard timing light at 3000 rpm, and the timing
signal jumps around, both advancing and retarding about 2 degrees.  I asked
about programming errors, but the student ensures me that the timing is top
priority on the interrupt list.  A quick calculation shows me that 100
microseconds at 3000 rpm is 15 rpm, thus if the engine changes 15 rpm in 45
degrees and the timing is based on the previous tooth information, a two
degree error could easily occur.  I now have to find out if we're using
speed averaging (as per Eric's email) as I think this may smooth out the
pulsations.

Bruce:  I'll look into a better timing light, and no, it is not just a
timing delay.  It really advances and retards

Joel:  More teeth mearn more accuracy, but when I was a student I designed a
system based on one tooth and don't remember this being a problem.

Many thanks for all responses

STEWART PRINCE



PROFESSOR, MECHANICAL ENGINEERING



CALIFORNIA STATE UNIVERSITY, NORTHRIDGE



Bruce wrote:


How are you detecting this jitter?.
If you're using a conventional *home grade* timing light, that might be a
characteristic of the timing light.  Some just don't trigger accurately.
Might consider mounting a second sensor done differently, and then compare
the two output signals, and see if the timing is really wandering around.
  If the timing error is constant, then why even worry?.   Just include that
in your timing table.    After all does it matter if the timing is a cal'c
34 d or actual 34d as long as the engine is making max HP?.
Bruce

----- Original Message -----
From: "Stewart Prince"   <mailto:sprince at csun.edu> <sprince at csun.edu>
To:   <mailto:diy_efi at diy-efi.org> <diy_efi at diy-efi.org>
Sent: Tuesday, July 30, 2002 3:57 PM
Subject: [Diy_efi] ignition timing variation



Does anyone have any information on "acceptable" variation of ignition
timing?  We're getting about +/- 2 degrees of jitter on our home-built
EFI system (Formula SAE).  By my calcs, 100 microseconds of error
corresponds to 1.8 degrees of rotation at 3000 rpm, and 7.2 at 12,000
rpm (assume constant angular velocity). I believe the students are using
the time between the last two teeth on an 8-tooth wheel (crank mounted)
to calculate engine speed.  Is it possible that some sort of averaging
scheme should be used to account for the speed variation due to

combustion?

--
STEWART PRINCE




_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org <mailto:Diy_efi at diy-efi.org> 
http://www.diy-efi.org/mailman/listinfo/diy_efi
<http://www.diy-efi.org/mailman/listinfo/diy_efi> 



-- 

STEWART PRINCE



PROFESSOR, MECHANICAL ENGINEERING



CALIFORNIA STATE UNIVERSITY, NORTHRIDGE





------_=_NextPart_001_01C238B7.EC247CA0
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 6.00.2716.2200" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2>Hi,</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff size=2>I 
built a microcontroller based ignition system for my motorbike. The system is up 
and running; I use one pulse every two crankshaft revolutions (the sensor is 
mounted on the camshaft) and I don't get more 'jitter' than approx +/- 1 degree, 
and that only at very low engine RPM. (approx 1200). The bike does not have a 
particularly heavy flywheel. <SPAN class=809002217-31072002><FONT face=Arial 
color=#0000ff size=2>Above 1500 rpm, timing is immaculate (fluctuation &lt;&lt; 
1 degree).</FONT></SPAN></FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff size=2>I base 
my timing lookup on rev period like you do and I use the previous revolution 
period for the lookup. I do no averaging.</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002></SPAN><SPAN class=809002217-31072002><FONT 
face=Arial color=#0000ff size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2>Beware, your sensor may have erroneous inputs. In my first prototype, my 
sensor would trigger on the ignition firing pulse, then it would trigger on the 
ignition firing pulse of the other cylinder, and then it would trigger on the 
generator which is mounted close to the ignition pickup. Now&nbsp;it doesnt 
trigger on either of those three sources and timing is much better... :) 
</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff size=2>For 
diagnosis, you could&nbsp;program an 'advance curve' that is constant for all 
rpm and all throttle position.</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2>FWIW...</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2>Regards,</FONT></SPAN></DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=809002217-31072002><FONT face=Arial color=#0000ff 
size=2>Arnoud.</FONT></SPAN></DIV>
<BLOCKQUOTE 
style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid">
  <DIV class=OutlookMessageHeader dir=ltr align=left><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Stewart Prince 
  [mailto:sprince at csun.edu]<BR><B>Sent:</B> Wednesday, July 31, 2002 9:39 
  PM<BR><B>To:</B> diy_efi at diy-efi.org<BR><B>Subject:</B> Re: [Diy_efi] ignition 
  timing variation<BR><BR></FONT></DIV>After speaking with the student writing 
  the software, I now have more ammo to respond with....we're using a Motorola 
  6811 for both timing and fuel. &nbsp;With 8 teeth on the crank, there is 45 
  degrees between teeth. &nbsp;The time between teeth is inversely proportional 
  to engine speed, however only engine period and throttle position are used in 
  the ignition lookup table. &nbsp;They have a bench version working with a hall 
  effect sensor and wheel mounted on an electric motor, and I observed little or 
  no fluctuation of timing here. &nbsp;The engine is mounted on a water brake 
  dyno with a large flywheel. &nbsp;I looked at the timing with a standard 
  timing light at 3000 rpm, and the timing signal jumps around, both advancing 
  and retarding about 2 degrees. &nbsp;I asked about programming errors, but the 
  student ensures me that the timing is top priority on the interrupt list. 
  &nbsp;A quick calculation shows me that 100 microseconds at 3000 rpm is 15 
  rpm, thus if the engine changes 15 rpm in 45 degrees and the timing is based 
  on the previous tooth information, a two degree error could easily occur. 
  &nbsp;I now have to find out if we're using speed averaging (as per Eric's 
  email) as I think this may smooth out the pulsations.<BR><PRE class=moz-signature cols="$mailwrapcol">Bruce:  I'll look into a better timing light, and no, it is not just a timing delay.  It really advances and retards<BR><BR>Joel:  More teeth mearn more accuracy, but when I was a student I designed a system based on one tooth and don't remember this being a problem.<BR><BR>Many thanks for all responses<BR><BR>STEWART PRINCE

PROFESSOR, MECHANICAL ENGINEERING

CALIFORNIA STATE UNIVERSITY, NORTHRIDGE

</PRE><BR>Bruce wrote:<BR>
  <BLOCKQUOTE cite=mid:02a201ba5eaa$bc3bd900$13258fd1 at nacelp type="cite"><PRE wrap="">How are you detecting this jitter?.<BR>If you're using a conventional *home grade* timing light, that might be a<BR>characteristic of the timing light.  Some just don't trigger accurately.<BR>Might consider mounting a second sensor done differently, and then compare<BR>the two output signals, and see if the timing is really wandering around.<BR>  If the timing error is constant, then why even worry?.   Just include that<BR>in your timing table.    After all does it matter if the timing is a cal'c<BR>34 d or actual 34d as long as the engine is making max HP?.<BR>Bruce<BR><BR>----- Original Message -----<BR>From: "Stewart Prince" <A class=moz-txt-link-rfc2396E href="mailto:sprince at csun.edu">&lt;sprince at csun.edu&gt;</A><BR>To: <A class=moz-txt-link-rfc2396E href="mailto:diy_efi at diy-efi.org">&lt;diy_efi at diy-efi.org&gt;</A><BR>Sent: Tuesday, July 30, 2002 3:57 PM<BR>Subject: [Diy_efi] ignition timing variation<BR><BR><BR></PRE>
    <BLOCKQUOTE type="cite"><PRE wrap="">Does anyone have any information on "acceptable" variation of ignition<BR>timing?  We're getting about +/- 2 degrees of jitter on our home-built<BR>EFI system (Formula SAE).  By my calcs, 100 microseconds of error<BR>corresponds to 1.8 degrees of rotation at 3000 rpm, and 7.2 at 12,000<BR>rpm (assume constant angular velocity). I believe the students are using<BR>the time between the last two teeth on an 8-tooth wheel (crank mounted)<BR>to calculate engine speed.  Is it possible that some sort of averaging<BR>scheme should be used to account for the speed variation due to<BR></PRE></BLOCKQUOTE><PRE wrap=""><!---->combustion?<BR></PRE>
    <BLOCKQUOTE type="cite"><PRE wrap="">--<BR>STEWART PRINCE<BR></PRE></BLOCKQUOTE><PRE wrap=""><!----><BR><BR><BR>_______________________________________________<BR>Diy_efi mailing list<BR><A class=moz-txt-link-abbreviated href="mailto:Diy_efi at diy-efi.org">Diy_efi at diy-efi.org</A><BR><A class=moz-txt-link-freetext href="http://www.diy-efi.org/mailman/listinfo/diy_efi">http://www.diy-efi.org/mailman/listinfo/diy_efi</A><BR><BR></PRE></BLOCKQUOTE><BR><PRE class=moz-signature cols="$mailwrapcol">-- 
STEWART PRINCE

PROFESSOR, MECHANICAL ENGINEERING

CALIFORNIA STATE UNIVERSITY, NORTHRIDGE

</PRE><BR></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01C238B7.EC247CA0--

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi



More information about the Diy_efi mailing list