Hrock

Ok So im getting frustrated with understanding this timing thing.

I got from various threads and am having a hard time in finding what plus this and minus that equals what ..

If you have a Maximum spark _ this should not let the computer calculate higher then this reference point..
Is that right ?

Main spark table
Coolant Vs Spark table
Coolant Bias

Could someone with alot of experience explain when your doing your tables... and other timing related stuff how the math works ?

Base timing ( is this for Grounded ALDL only ?)
So if you get a Coolant offset of 8 degrees, is this primarily for transposing the timing Down 8 degrees ?

My problem is I cant seem to find Perfect when the temperature changes.. I need more Dynamics with the temperatureand timing.

I want to know how to setup the Timing.. Why does 20.04 degrees appear right across the whole timing coolant table ? from a stock chip ? the only modifiers are at Extreme temperature and Low rpm Cold temp.. But mid run it shows 20.04.. So Does that ADD or subtract 20.04 at that particular RPM Vs Coolant ?

Thanks for any help.. Im still searching for clearer descriptions..

RBob

iTrader: (1)

Over on diy-efi (I need to move this stuff to moates.net) on the gmecm area, under papers, or ECMs or such. There is a write up on '747 SA. Maybe even titled "running a SA only '747." Or something of that nature.

That paper will answer your questions.

RBob.

Grumpy

Maybe someone can put this in the stickies.

Ignition Control and the '747 ecm

Robert Rauscher, 5/26/99, Updated 4/9/00

Copyright 1999, 2000 Robert Rauscher, All Rights Reserved.

Areas preceded with a '|' (vertical bar), are updated. The
spk747al.bin file is also updated. It is based on the ACSC
bin.

This write-up is on the spark control only. It also does not
include any egr spark conditions. I have been testing an
ignition only '747 setup, and do not have an egr engine.

|First off, I'd like to thank EcmGuy for providing the referance
materials that made this write-up possible. (originally
credit was given to EcmNut, apologies to both, 4/00, rr).

Second, this material is as accurate as I believe, but may
contain errors. If any, they would most likely be in areas
that use bit mapped status words. It can be difficult to
understand the exact meaning of these bits.


/* ------------------------------------- */


For sensors, I've been using:

a) vss
b) coolant temp
c) map
d) knock



There is also a power brake vacuum line restrictor. This
is a stock GM piece that slows the rate that the p/b's
'get vacuum' from the intake manifold. This reduces the
effects that using the p/b's have on the MAP sensor value.


If your not planning on doing the fuel also, try to
mount the tps to the carb. It would be worth it, it
will allow pe/wot to be determined, along with proper
map test being done. Without it, there was intermittent
error 33's: map sensor high codes.


/* ------------------------------------- */


Ecm error codes in use:

12 drp 14 coolant
15 coolant 24 vss
34 map 42 est
43 esc 51 prom
52 cal pack 55 adu


/* ------------------------------------- */


Some notes and associated files:

1) I recommend using promedit for an editor, I have provided
a .ecu file for use with promedit: spk747.ecu

I'm using the version with the build date of: Aug 05, 1998.

2) spk747al.bin, a bin file with high timing for an alum
headed engine. Use at your own risk! But study it.

/* ------------------------------------- */

(SA => Spark Advance)
(wot => wide open throttle)
(PE => power enrich, same as pe)
(map => manifold absolute pressure)
(esc => electronic spark control)
(est => electronic spark timing)
(vss => vehicle speed sensor)
(tdc => top dead center)
(btdc => before top dead center)
(atdc => after top dead center)
(aldl => Assembly LIne Diagnostic Link)

/* ------------------------------------- */

If you have an aldl scanner, you can monitor some of the spark
advance values with some small modifications to the bin.

SA values can be output at locations normally used for other
functions. Then using a little math, the SA in degrees can
be ascertained.

I use locations that the scanner does no math on, the displayed
value will then be as it is in the ecm. Then for SA values, just
multiply the displayed scanner value by 0.35 (90/256 = 0.3515625).
The result will be degrees of advance. Any initial timing will
also need to be added to this advance (see 0x009).

So, if the aldl INT value is 91:

91 * 0.35 = 31.8 Degrees.

With the initial advance set to 6 degrees:

31.8 + 6 = 37.8 degrees at the crank.

I do have a caution on the use of the aldl on the '747 ecm,
it is Slow. Hence, the entire data output during one frame, may
not be relevant to that frame. IE: If your capturing a frame,
while accellerating, things are changing, and the data is not a
'snapshot' of one point in time. The data output is gotten
from memory as it is being output.


Ok, locations I use: BLM, INT, IAC, (0x509, 0x4f7, 0x4eb).

I output RAM locations:

0x0067 (total spark advance)
0x006E (PE/WOT spark advance)
0x0077 (Retard spark amount, only half this
value is sub'd from the total SA)

The value of location 0x0067 will include the SA of the
PE/WOT and Retard spark also. This is nearly the final
value used to program the spark control counters. Only
a spark latency correction is made after this.


/* ------------------------------------- */


Here are the more important spark locations in the bin:

The left hand column are the eprom offset location for each.


0x009: Initial spark advance. This value should be what
the physical distributor timing is set to. This
value gets subtracted from the Total Spark Advance.
(but is physically added back in by the position
of the distributor relative to piston tdc).

This is also called 'static' timing.

0x00C: Max RPM used in the spark slope calculation. This is
maximum value at which additional slope spark will
be calculated to. Once this value is reached (you
have it floored, right)?, the slope increase will
be held at this value.

0x00D: Slope in degrees per 1K rpm. This is the rate that
additional SA will be added.

--- These last two work together to add additional spark
at a linear rate as the rpm increases. This additional
spark starts at 3600 rpm, and increases until the
max rpm value is reached. 3600 rpm is where the main
SA table ends. Above the 'max rpm' the total slope will
be held at the value calc'd for 'max' rpm.

0x014: Warm spark bias. This value is subtracted from the
Total SA. It's purpose is to bias the main SA table,
so that negative advance values can be in the table
as a positive number.

0x015: Cold spark bias. This value is subtracted from the
Total SA. It's purpose is to bias the Coolant
compensation spark table. Again, to allow for a
negative advance in this table.

0x019: Maximum spark (2 bytes). This value is the highest
total advance that can be programmed into the
distributor. Any Calculated Total SA above this
value gets clipped to this value.

--- It's purpose is due to the physical limitations of
the distributor. You can only advance the timing
so far until you are sparking on the previous
| cylinder. With eight spark firings per distributor
| revolution, there are only 45 degrees between each
| set of plug wire terminals on the cap. GM uses the
| value of 42 degrees as the maximum advance. This
| always a safety margin to direct the spark to the
| proper terminal.
|
| However, this is not the limit that can be run at
| the crank. If you set the static timing to 10 degrees
| BTDC, a maximum of 52 degrees can be had at the crank.
|
| The 42 degree limit is only the distributor limit.
| Crank timing is the total of the static setting, and
| the amount programmed by the ECM into the distributor.
| It just becomes a balance of how much total crank advance
| is required, versa, how much static (inital), timing can
| be used (and still hot-start).

(if it becomes a limitation, you need a DIS system).

0x01B: Maximum Retard (2 bytes). Maximum total amount of retard
allowed. For the same reasons as max SA above.

0x023: aldl/closed loop added spark. This spark is added when
in aldl diagnostic closed loop mode.

0x024: Hi-way spark mph threshold. Minimum MPH to start timer.

0x025: Hi-way spark time threshold: Amount of time required before
hi-way spark is added. Must be above MPH threshold for this
amount of time, in order to get hi-way spark.

--- Note: The ecm does an ESC test once each time after an
engine start. This will typically cause two (2) knock
counts to occur (see loc 0x555). Until this test occurs,
and passes, hi-way spark won't be added.

Sooner or later, these counts should show up on the
aldl scanner, under 'knock counts'.

Note: Another parameter (hard coded), is that no err43
(est) must have existed since power up(?). Or, is that
the test for err43 hasn't yet been run(?).

0x026: Startup added SA is reduced (re-calc'd), on this time period.
Every 'n' seconds, any added startup spark is decayed.

0x02B: Knock retard mph cutoff threshold.

0x02C: Knock retard rpm cutoff threshold.

0x02D: Knock retard temperature cutoff threshold.

--- For knock retard to take place, the engine temp needs to be
greater then the temperature threshold and the vehicle voltage
must be greater than 9.0 volts, then, either the VSS or the
RPM threshold must be exceeded.

0x02E: Burst knock MAP threshold, in Kpa.

0x02F: Burst knock TPS threshold, in percent.

0x030: Burst knock differential TPS threshold, in percent.

0x031: Burst knock, duration of retard, in seconds.

--- The last four, plus Knock Temp Threshold (0x02D),
determine whether burst knock retard takes place.

There is also an enable bit at location 0x006, bit
zero, that must be set, if burst knock is desired.

Right off, the engine temp is compared to the knock
temp threshold (0x02D). If below this temp, burst
knock can take place, if above this temp, only regular
knock retard can take place.

The burst knock qualification is as follows:

Option byte @ 0x006, bit 0 set (= 1)
Engine temp below knock retard threshold
Not in closed loop
Map value above threshold
TPS % above threshold
TPS differential above threshold
TPS differential is for more throttle opening

If all of the above qualifies, then a lookup is done
based on current engine temperature. This value is
subtracted from the total timing. It is not available
for output to the aldl link, but is reflected in the
value at location 0x0067.

Burst Knock is an all or nothing deal, no attack rate
or decay rate. Once started, it exists for the entire
duration set at 0x031. The actual retard value is
adjusted each pass through (80 times/second), relative
to the engine temperature (from the lookup table).

Once the Duration is met, the retard instantly ceases.


0x035: Main SA table. This is the beginning for what the Total
Calc'd SA will be. This is a 3-dimensional table with
rpm and map in Kpa as the look up parameters. The value
at that location is the third dimension, which is the
base or starting SA. The warm spark bias is reflected
in these table values. (ie: actual SA will be the lookup
value minus the warm spark bias).

0x10A: Coolant Compensation SA table. This table is used to add
or subtract SA, based on engine temperature, and less
so on manifold vacuum. Typically, more SA at low temp's,
less SA at high temp's, and no change at normal operating
temperature.

This is a 3 dimensional table, with coolant temp and
map VAC (inverse of map Kpa), as the look up parameters.
The cold spark bias is reflected in these table values.

0x157: Hi-way Spark Added. This spark is added while in a steady
state cruise. See 0x024 & 0x025 for qualifiers. If qualified,
a new value is calulated each pass through the code. It is
based on engine vacuum.

0x15E: PE Spark. This spark is added when at WOT. It is based
on engine rpm. I believe that a TPS is required for
the code to recognize a WOT condition.

0x16E: Startup Spark Added. This SA is added after an engine
startup. It then decays at a rate defined in two
other parameters (shown next). The amount of spark
added is based on coolant temp at startup.

If the engine is startup while in clear flood mode, no
statup SA is added.

0x174: Startup spark decay. This is the initial amount of time,
in seconds that the added spark will decay over. The decay
is set once at each start time.

0x17a: Startup spark decay multiplier. This value is a multiplier
of the Spark Decay parameter. Since this value is updated
on a regular basis, as the engine warms up, this increases
the initial decay rate. So, on hot day, the added SA will
decay faster than on a cold day.

0x17f: PE Knock Retard. In degrees, the amount of retard to
use based on engine rpm, during WOT knock detection.

0x188: Knock Retard. In degrees, the amount of retard to use
during normal engine operation. Based on map VAC.

0x18d: Retard attack rate. Amount of degrees per millisecond
that any required retard will occur at, based on rpm.

--- When knock is detected, a count is available to the
ecm code. This is read from the ecm hardware. This
count is the number of milliseconds that knock has
been occurring. This is the 'knock count' displayed
on a scanner.

The total amount of retard is calculated by:

1). Multiplying the attack rate by the 'count'.

2) Lookup on either PE retard table if WOT, if not in
PE, then it uses the normal table.

The amount of added retard will be the lessor of the
above two values; the attack rate, or the table value.
It is then divided by two. Yes, it appears that the code
divides the retard value in half before subtracting it
out from the total spark advance.

Of course, a maximum retard check is then done using the
value in location 0x01b. The ecm will not retard the
total timing any more than this value.

0x196: Retard recovery rate. Percent per millisecond that any
retard recovers back to normal at, based on rpm.

The decay routine runs 10 times a second. The decay starts
immediately upon the addition of retard.

Again, required for a smooth transition?

/* ------------------------------------- */

See Part II

Grumpy

Part II
Ignition Control and the '747 ecm
Robert Rauscher, 5/26/99, Updated 4/9/00
Copyright 1999, 2000 Robert Rauscher, All Rights Reserved.


OK, so what do I do with the above, and how does it all fit
together?

Note: These calc's are completed 80 times a second, asynchronous
to the actual timing signal from the distributor.


First, a quick run down of what the code does. Remember, just
because a value is calc'd, doesn't mean that it has to be
greater than zero in value.


1) A look up is done on the Main SA table, the value retrieved
is based on the current rpm and map values. Call this Total
Spark Advance, or TSA for now.

2) The Slope, if any, is calculated and added to the TSA.

3) A look up is done on the Coolant Compensation table using
the current engine temp and manifold vacuum. Add this to TSA.

|4) (adjust for any egr spark, SA is added if egr is active).
| (a lag filter calculation is done here also, but appears
| to be egr related only. The lag filter allows for the smooth
| change of SA whenever the EGR is enabled/disabled).

5) Add in any PE spark to TSA. Calc'd in a separate routine.

6) Add in any aldl/closed loop spark, to TSA.

7) Add in any Start-up Spark, to TSA.

8) Add in any Hi-way Spark, to TSA.

9) Subtract the Warm Bias spark value, from TSA.

10) Subtract the Cold Bias spark value, from TSA.

11) Subtract the Initial spark value, from TSA.

12) At this point, the TSA value is tested against
the Max Allowed Spark Value. If it is greater, it
will be set to the maximum allowed SA value.

13) Subtract any Burst Knock, from TSA.

13) Subtract any Knock Retard, from TSA.

14) The TSA value is tested against the Minimum
Allowed Spark, and adjusted to this minimum value
if required.

15) A Spark Latency correction is made and the end
value is sent to counters in the ecm. An output from
the counters control when the distributor will spark,
relative to the incoming distributor pulse.

/* ------------------------------------- */

As I explain this, I'll describe how I set up this engine. I
can't say whether the same setup will work for you. The engine
is as follows:

327 ci SBC
'91 alum corvette heads, no heat riser
flat top forged pistons
~ 9.75:1 cr
Isky 280 Mega cam, .465 lift, 224 deg @ .050, 108 deg LDA
Edelbrock Performer RPM dual plane intake
Carter 625cfm afb
headers & duels
4-spd w/3.55 rear

This engine idles at 850 to 900 rpm. Keep this in mind
as I discuss the changes I made to the bin.

Note: I don't recommend this cam for an efi engine, it
idles rough, along with a highly fluctuating map. It
works OK with the ignition system, but have my doubts
once I get a TBI on it.


/* ------------------------------------- */


1). I set the Initial Timing to 6 deg. This is both in the bin
and at the distributor. (disconnect the bypass and with the engine
running, time it with a light to 6 BTDC).

2). Set the Warm Spark Bias to 0 degrees.

Between these two, the values you see in the Main SA table
|will be what you have at the crank. The initial 6 deg will
be subtracted from the main SA look-up value, but will be
physically added back in by the distributor timing.

This makes it both easy to start the engine, and easy to
read the main SA table.


3). Set the cold spark bias to 10 degrees.

4) Set each value in the coolant compensation spark table
10 degrees higher than the actual desired value. This is
due to the cold spark bias being subtracted from each of
the values of this table.

If you set a value to 15 degrees, the spark will be advanced
only an additional 5 degrees. If set to 7 degrees, this will
in effect subtract 3 degrees from the total SA, (7 - 10 = -3).


5). Spark Slope values. I've extended the Max RPM to 6375,
along with reducing the slope to 1 degree per 1k rpm. This
will add a maximum of 2.7 deg over 3600 rpm through 6375 rpm.


|6). Set the Maximum Spark value to 42 degrees. The initial SA
|will be subtracted from this; this allows a max of 48 degrees
|of timing at the crank (with the 6 degrees inital).


7). aldl spark. Just set it to zero and forget about it.


8). Startup Spark Added. At the lowest temp, I added about
three additional degrees. Changed from 5 to 8 degrees.


9). Retard Recovery Rate. Note that at 400 rpm, the recovery
rate is high (on some bin's). With an idle speed of 900 rpm,
I brought the recovery rate higher at 600 and 800 RPM also.
This is an anti-stall tactic.


10). For the following I have not changed anything:

0x157: Hi-way Spark Added. Typically disabled.
0x15E: PE Spark. (Need tps to hit PE).
0x174: Startup Spark Decay.
0x17a: Startup Spark Decay Multiplier.
0x17f: PE Knock Retard (Again, tps needed).
0x188: Knock Retard vs VAC.
0x18d: Retard Attack Rate.


11). This leaves the main SA table and the coolant compensation
table.

Set the main spark table values to what you want to see at
the crank during normal engine operation. At the low rpm idle
area, I level out the SA so that fluctuations in rpm and map
don't cause the timing to change. This gives a more stable idle.

Along the low map values (foot off the pedal, slowing down),
you can put less timing. This will allow the car to slow
down better, as you are pulling power from the engine.
Change the timing up and down until you like the way it
drives. Pulling a lot out can put your nose into the
windshield, and then light the tires when you re-apply
the gas.

For the cruise areas, map between 35 and 50 Kpa, along with
the rpm range of 1800 to 3000 rpm, adding timing helps.

As the map rises to 100Kpa (wot), reduce the timing to
prevent detonation. Keep an eye on the knock counts while
tuning.

With aluminum heads, you can run more timing throughout.
This is reflected in the current bin.


12). For the coolant compensation table, enter the Cold Spark
Bias value throughout the normal operation areas. This will be
around the temperature range of the thermostat. You don't want
to use this table to change the spark in this area. So by using
the cold bias value, will in effect, cause no change to the TSA
during normal engine operation.

With a cold engine you want more SA, as the engine starts
to run a hot (overheat), pull some timing out.

Then blend in from the edges to the normal range for a
smooth transition.


/* ------------------------------------- */
/* ------------------------------------- */


All in all, this should get you started, examine these bin's,
and some stock ones with a promedit'r. You will soon get a feel
for how to do spark timing.


/* ------------------------------------- */
/* ------------------------------------- */

| New section describing how loookups are done into 3D tables:

It's a 3-way, 16-point interpolation between the table values.

Think of the point of 53kpa map & 1700 rpm as being someplace inside
of a box. The four corners of that box, are the four table values
that surround the defined point. The distance from that point, to each
of the four walls is estimated, with the final result based on those
four values.

Works something like this...

Part of main SA table:

map> 50 60
rpm
1600 20 24
1800 22 26

If the map is 53 and the rpm is 1700, first lookup gets the 20 & 24
degree values @ 1600 rpm, and interpolates between them. The difference
between these two are divided into 16 points:

24deg - 20deg = 4deg, 4deg / 16points = .25 deg per point

60map - 50map = 10map, 10map / 16points = .625 kpa per point.

53 map @ 1600rpm => 20deg + (3map / .625) * .25 = 21.2 deg.

So, 21.2 deg is the 16 point interpolation of 53 map @ 1600 rpm.

(Note: On a 2D table, we would be done, and this value returned).

The 3D lookup, however, calls the 2D routine a total of three
times. In this example, the first call we already completed.

The routine then advances a full row to interpolate between the
22 deg and 26 deg values found at 1800rpm.

So, using the same math as above, 23.2 deg is the 16 point
interpolation of 53 map @ 1800 rpm.


The third and final interp. is done between the answers of the
first two lookups: 21.2 deg & 23.2 degrees (@ 1600 & 1800 rpm).

The rpm is divided into 16 points the same. For this example,
it's right in the middle of table values 1600rpm and 1800rpm,
with 1700 actual.

So, 21.2 deg & 23.2 deg => 22.2 deg as the actual returned,
looked up finalized interpolated value to be used.

/* ------------------------------------- */
/* ------------------------------------- */

--

Hrock

Thats the stuff.. Im assuming Gm wouldn't go out of there way to change much to the 7165... Thanks Rbob and Grumpy-

liquidh8

RBob and Grumpy, you's guys are great!! Glad I found this post.

So with the warmSA @ 0, your main SA table is the entered value, + your static timing + your coolant comp spk adv? I think that is right.

Sorry for the noob questions.

Hrock

Yes thats the sequence. Now its just to find the missing Locations in the Binary

I have a $6E definition in the TunerCat.. and I run a 7165 ECM.
What are the locations for the options.

The $6E is alot different then the $42 definitions.

I can not seem to find the WARM Spark bias...

drb930

Subscribed!