Audi 5 cylinder turbo triggering for LinkPlus standalone engine management
Wed, 30 Jan 2002 11:41:20 -0700 (MST)
I put in an order for a standalone engine management system. The system ca=
n control all the important engine parameters, and even controls the carbon=
canister purge valve for emissions. The idea here is to use it on my MC-1=
based turbo conversion into an '85 4ksq. Initially, I plan on using the i=
gnition distributor, later switching to direct ignition if it makes sense.
I was queried by the LinkPlus vendor for information regarding how the stoc=
k MAC11 engine management gets it's sensor information. I got some feedbac=
k from the q-list members and archives regarding hall sensor types and oper=
ation, number of teeth on the flywheel and ignition timing reference pin lo=
cation on the flywheel. I forwarded that info to the LinkPlus vendor, who =
forwarded it to the manufacturer and was going to let me know what they sai=
I received this (quoted below) from my LinkPlus vendor, who got it back fro=
m the folks at http://www.link-electro.co.nz. It summarizes the built-in c=
apability of the LinkPlus, and asks for more information regarding the hall=
sensor signal on the single-window distributor on a MAC11 or similar.
I have to look at this and think about it. It looks like I will need to ca=
lculate the position of the distributor window relative to TDC #1.
Which I guess, if that window is "centered" on the 62 degree BTDC timing re=
ference pin and the window itself is 40 degrees "wide", then the window sta=
rts at 22 degrees BTDC and ends at 82 degrees BTDC? Then I have to convert=
that to a "720 degree scale" somehow?
Any comments, questions, or advice would be greatly appreciated. Please co=
py me directly as well if you post to the list. I will pass along updates =
as I receive them.
Audi 5 cylinder ignition system.
The Link ignition timing software requires one pulse/edge for each cylind=
at about TDC, or preferably a little before (say 10 BTDC). The actual ang=
is not hugely important so long as it is stable and consistent since timi=
can then be added to or subtracted from this "base" timing as required.
Note that this "base" timing is what the engine will crank (start) on sin=
programmed advance is not reliable due to the irregular cranking RPM all
engines exhibit. The software measures the time interval between every one
of these pulses and mathematically divides it into 144 (for a 5 cyl. engi=
time slots. Each of these "time-slots" represents 1 degree of rotation and
the advance (or even retard) can be built on this data.
Note that the Link ECU generates an advance signal by predicting the time
slot for the NEXT cylinder based on what has occured for the present/past
reference signal. Some systems start with an advanced reference pulse and
delay the output signal to generate the advance curve. Should that really
be a retard curve ????
(Higher resolution is possible by introducing more "time-slots"
as required. For all PRACTICAL applications 1 degree resolution is plenty.
If an engine requires sub-one-degree accuracy to run THAT close to
detonation then you had better be VERY careful with every other variable
in the fueling/timing/control system.)
In the case of the Audi, the flywheel trigger consisting of 135 teeth is
used to derive the 5 x 144 degree pulses by running a divide-by 54 equati=
This division needs to be INDEXED so that the /54 starts at the correct
(same) place for each engine revolution. The co-mounted, single, 62 degree
pulse may used to index (pre-set) the count sequence so that the "54th"
tooth/pulse occurs at some convenient point for base timing purposes.
Alternatively, the Hall-Effect, cam speed trigger signal can also be used
for indexing. The actual location (in degrees) of this signal is arbitrary
but must be consistent at all times. This "720" degree index is prefered
to the crank "62" degree index since it defines a unique angle within the
engine's 720 degree cycle which may also be used to control injection
phasing for true sequential injection, and also allows coil-per-cylinder
direct fire ignition to be implemented.
(A "crank only" trigger system cannot run sequential injector timing since
it cannot differentiate between the two crank revolutions per 4 stroke
engine cycle. Likewise ignition coils must be fired using "wasted spark"
method. This may prove very difficult with an odd cylinder number engine
such as this. This probably precludes the use of primitive engine control
systems which require the addition of a 58 tooth crank trigger wheel.)
One further piece of information is required regarding the cycle degrees
at which the Hall Effect (cam) sensor's edges occur. This must be express=
on a 720 degree scale starting at TDC #1 cyl (call this 0 degrees and cou=
upwards). Please record the angle at which the magnetic beam is broken by
the shutter wheel and also the angle at which the beam is restored I.E.
the leading and trailing edges of the shutter.