Anyone know the priming, starting, start up, and warm up %'s for a 3.2?

[Tippy]

I've been tuning my MegaSquirt for a couple of years but I've yet to get cold start perfect.

If I could shamelessly plagiarize the factory numbers, that would be great!

Ready for a dead perfect start at ALL temps.

Also, if someone knows the acceleration numbers, that'd be great too.

TIA

[Lapkritis]

What size injectors? What fuel pressures? Are you using MS for ignition control as well? Are you using an idle control valve or a throttle stop?

[Tippy]

55lb/575cc, 2.5-bar-2.8-bar stock fuel pressure, Ford EDIS, stock ICV

[Lapkritis]

MS2 Extra with speed density algorithm example below. Remember, you'll always need to tweak a bit for your own system due to differences. I normally datalog my cold starts and replay in Megalog viewer and repeat until I get it nailed down. Once you're comfortable with the software (strongly recommend Phil Tobin's TunerStudio) then you can start a notebook on tweaks to go with the weather changes in the environment. The air temperature sensors mounted into the aluminum intake manifolds tend to wreak havoc on otherwise properly tuned engines during hot starts. The sensors heat soak and then give a false reading causing the engine to run extremely lean. If your IAT sensor is mounted in the intake manifold then I would suggest a phenolic spacer or better yet moving to a silicone coupling (pre-Throttlebody) where it would not be in contact with any metal.

What brand fuel injectors out of curiousity? I recommend the Siemens Deka as they're the best bang for your buck and have great low PW characteristics. Some of the older injectors that are still in service have irregular spray patterns that can impact running conditions unpredictably.

Any other mods such as higher lift cams or forced induction?

6CYL
Req Fuel 4.9
2 Squirts Per Engine Cycle
Injector Staging: Alternating
Engine Stroke: 4
Number of Cylinders: 6
Injector Port Type: Port
Engine Type: Even Fire

Cranking Settings
Cranking RPM: 300
Flood clear TPS (%): 90
Cranking Fuel Pulse Rate: Every Event

Priming Pulse:
Coolant °F / Prime Pulse-width
10 / 4.1
30 / 3.7
50 / 3.4
70 / 3
90 / 2.7
110 / 2.4
130 / 2.2
150 / 1.9
170 / 1.8
180 / 1.5

Cranking Pulsewidth:
Coolant °F / VE Pulse-width %
10 / 326
30 / 301
50 / 276
70 / 251
90 / 226
110 / 201
130 / 175
150 / 151
170 / 130
180 / 120

Afterstart Enrichment Percentage:
Coolant °F / VE Pulse-width %
10 / 15
30 / 14
50 / 13
70 / 12
90 / 11
110 / 11
130 / 10
150 / 9
170 / 8
180 / 7

ASE Taper Cycles
Coolant °F / Taper Cycles
10 / 350
30 / 330
50 / 310
70 / 290
90 / 270
110 / 250
130 / 230
150 / 210
170 / 180
180 / 150

Warm-up Enrichment (WUE)
Coolant °F / WUE %
-40 / 135
-20 / 132
0 / 129
20 / 125
40 / 122
60 / 116
80 / 109
100 / 106
130 / 105
160 / 100


Just as importantly are Acceleration Enrichment settings... these will have a greater impact on if you can jump in and drive right away instead of waiting and baby sitting the engine for 2mins while it warms.

Accel Enrich Settings
Low RPM Threshold: 1200
High RPM Threshold: 3500
Enhanced Accel Enrichment: Off

Acceleration Enrichment Wizard
Throttle Position vs. Manifold Pressure Accel Enrichment Strategy, Percent TPS Driven (%): 25

Mapdop Threshold (KPA/s): 300
Accel Time: .3 seconds
Accel Taper Time: .3 seconds
End PW (ms): 0.0

TPS dot Threshold (%/s): 65
Decel Fuel Amount: 95
Cold Accel Enrichment (ms) 2.0
Cold Accel Multiplier (%) 175

[Tippy]

Thanks Lapkritis. Running Deka's myself. I'm running 6 squirts/cycle at 6.0 req_fuel.

So your's purrs right after light off in all temps with those settings? <- this what I really want to know!

Noticed I'm quite a bit richer across the board than yours. I had fairly similar settings and it wouldn't even start in the 60's or colder. May have went too rich.

I'm running enhanced acceleration enrichment since it makes the engine rev so quick. I figured out once you go EAE, all those AE numbers mean nothing! Just adjust curves on all the EAE settings and done. I running 100% TPS since the signal is faster than the MAP according to the histogram.

Gonna hotwire WB02 so I can log my startup again. When it's in the warming cycle, obviously no data when you need it most!

[Lapkritis]

I've only tested down to about 0°F as the car is stored for winters.

What algorithm are you using for fuel? Are you configured for sequential fuel or just semi/batch? Mind posting your VE table 1? 6 squirts can cause problems if you're wired for 2 fuel channels as you'll have very short PW. 2 squirts will be better for idle resolution.

[Tippy]

I had a few 20-something degree mornings and it was hit or miss starting and staying running.

Hmmm, been told 6 squirts is better for idle stability, but now you state that, I can see how it could be opposite having large injectors at small PW's.

May try switching. Thanks. Running batch/bank alternating.

Let me post the VE in a minute.

[Lapkritis]

Yep, imagine 1 rpm like a clock. 360° divided into 6 squirts.

You would have squirts at 2 o'clock, 4 o'clock, 6, 8, 10 and 12. The sum of all squirts = to VE.

Now imagine 2 squirts with one at 6 o'clock and 12 o'clock also equal to same sum volume of fuel VE.

The two squirts will need to be larger to equal the same volume of fuel in each port as the 6 smaller squirts.

This can also change if you're wired for sequential control and use a megasquirt that can time and trim your injector per cylinder. I would add a small percentage trim to cylinders 2 & 5 to aid in keeping things cool relative to the other cylinders. Lots of variables to consider... each with a potential to impact how you operate at idle and part throttle. Wide open throttle is far easier to tune than these other areas.

[Tippy]

I'm going to switch to 2 squirts tomorrow morning and see how it does.

Getting a little excited with this nugget of info. The more I think about it, the better it may be for all of my headaches. My PW's are very, very tiny, and the Deka's don't like such PW's after reading up on ID injectors. I have always had wandering AFR's at cruise too that baffled me. Even at cruise, my PW is small having big injectors in a small engine.

This may cure it all. Can't wait to try tomorrow on the way to work.

Thank Lapkritis. I am still baffled how you have such low numbers when the engine is cold and you are not having start up issues. Like I said before, I had numbers similar to yours before, and it wouldn't even start up.

Going to try to follow the MS tuning guide to a "T" and take it from there.

[Lapkritis]

Switching to two squirts will likely require significant changes to VE Table 1 depending on how you're configured.

[Tippy]

Yep! Just found out.

Ugh.....

[Lapkritis]

Save lots of files as you make changes so you can revert if needed.

[scarceller]

Tippy,

The stock setup runs extra degrees ignition advance once the engine catches. So during cold start it has this ignition advance table:

CHT - advance
------------------
-32C 18deg advance (added to base advance)
0c 0deg (no compensation)
65c 0deg

During very cold temps (below 0c) you need to add ignition advance because flame front speeds are significantly reduced.

The above is what the 89 chip does but in my car I found this to work better:
CHT advance
-32c 7deg
0c 2deg
65c 0deg

My cold start AFRs look like this, I recorded these with my WBO2 setup:


This will give you the basic idea of what AFR it's at (red line) vs CHT (orange line)

Also note the RPM based on CHT, I run my cold idle at 1400RPMs and then feather down to 920RPMs when CHT=70c here's my RPM table
CHT RPM
-23c 1400rpm
40C 1400rpm
70C 920rpm

The real trick to cold start is not just fuel, you need extra ignition on those real cold days. Also if engine hunts during cold start it's because it's to rich and/or to much ignition. The 3.2L really does not need to run that rich after the first 20 seconds. Use my data above in the picture as a target goal.

[Tippy]

Thanks Sal for the info. Going to evaluate again. Problem is, is I think MS has additional fueling the Motronic doesn't do adding complexity.

[scarceller]

Motronic does have a cold start squirt just as cranking begins. This squirt is delivered in 2 shots of fuel over 2 crank revs, where each shot is about 2ms pulse, it really does not deliver much prime fuel at all till cranking really gets going.

This may also help, this is the fuel compensation in % based on CHT. It's how much extra fuel is delivered % wise for a given CHT. The table looks like this:
CHT - % extra fuel
-----------------------
-32c 50% (very cold start)
-16 42%
0 23% (at freezing)
19 12%
51 6%
60 0% (Considered warm enough, no more extra fuel needed)

Notice how significantly more fuel is required at CHT below freezing! The relationship between CHT and % fuel is NOT linear. If you graph this you'll see it.

Originally Posted by Tippy

Thanks Sal for the info. Going to evaluate again. Problem is, is I think MS has additional fueling the Motronic doesn't do adding complexity.

[grendiers]

Great info! Also, as you make changes, and save tune(s), check your VE table. Your idle/cruise hunting might be associated with too big of an increase/decrease between numbers in the table. Also, for fun, once your basic engine parameters are set, try the Autotune feature. I use this same setup (MSII, EDIS) on my 924 Turbo track car.

[scarceller]

What I've learned about the 3.2L is that if you make any changes from stock and open up air flow these engines surge at light loads. If you alter exhaust, intake or cams they tend to be very difficult to tune at very light street loads.

My engine surges when you have very little load on it. For example slight down hill with foot still slightly on the gas. No injector shutoff, I'm not in decel cutoff. Also when doing parking lot crawl in 1st gear I get some bucking from time to time.

What has helped is to lean the fuel out significantly at these very small injection pulse widths. For example have a row for very light load in the fuel PT table and try leaning out. Even as much as AFR 17:1 helped. Also adding ignition at these very light loads helped, as much as 50deg advance!

I'm not 100% certain by I think these motors tend to do this because of feedback pulses in the intake, just the way these engines ingest air. Several other tuners I've spoke with have also seen this. I have not been able to totally eliminate in my setup either. It seems to happen at RPM range 1000-3000RPMs at very light loads.

Engine fully warm idle surge I fixed by simply running slightly richer mixture around 13.8 to 14.0 AFR, and ignition at 15 degrees. Give the engine what it likes and don't focus to much on target AFR. You really don't need to run at 14.7AFR at idle if it does not like it.

Originally Posted by grendiers

Great info! Also, as you make changes, and save tune(s), check your VE table. Your idle/cruise hunting might be associated with too big of an increase/decrease between numbers in the table. Also, for fun, once your basic engine parameters are set, try the Autotune feature. I use this same setup (MSII, EDIS) on my 924 Turbo track car.

[Tippy]

Sal, I'm glad you mentioned the surging at light loads. I've tuned that area a decent amount and have a ton of timing there and while it's far better, it still happens.

I think what happens is pulse width gets so low, the injector get out of their efficiency range (less than 2ms) delivering inconsistent injections. Or, I wonder if Motronic can't generate the right values or literally shuts the injector off due to not being to produce such small pulse widths? Just theories.

Only way to prove out on mine is running ID injectors (big $$) that can inject at tiny pulse widths and see.

[Lapkritis]

A lot of surging is due to lambda correction thresh-hold being too low. The surge is very often the o2 correction %.

[Tippy]

Are you speaking of Motronic or MS? I'm running no EGO control with MS if you meant that.

BTW Lapkritis, I pulled the dead time from 1.0ms to 0.950ms and holy crap, the AFR's calmed down, accel enrichment is way more consistent, and the light load surging is almost null.

That with 6 squirts, not 2 using 55lb'ers seeing sub-1ms PW's. Thanks for the tip.

Cannot wait to run 2 squirts and retune everything. I believe the light load surging will completely disappear and the idle will be dead steady.