I've noticed that if I hover around -5 psi with light acceleration during a cold start the car feels like it is fighting something to make the expected power at that throttle position. Hovering around -10 psi or moving to Zero and on into boost feels fine but hovering around -5 psi it's only giving me 11* timing around 2750 rpm and slowly climbs to 14* by 3500 where it begins to smooth out. Now that I can datalog exactly in the range this is happening I compared it to a warmed up run:

Timing Advance at 15% TPS and ~ -5 psi with
Coolant at 135 F vs 176 F
2500 13* vs 38*
2750 11* vs 41*
3000 12* vs 42*
3250 13* vs 43*
3500 14* vs 43*
3750 24* vs 43*
4000 33* vs 44*

Both runs are around 14.8-14.9 AFR during the test and showing zero knock sums. Load averages 46% when cold and 43% when warm, with the main difference being at the lower RPM's where the load is forced higher due to less timing.

So what exactly is taking place here that is yanking so much timing advance and not allowing the car to smoothly accelerate? I can drive around it by avoiding the 15% TPS range and go faster or slower but I shouldn't have to do that!

Tables that I'm looking at:

Ignition Warmup Retard - I only see 3's and 5's on this table which is much more than 11* to as much as 30* being pulled.

EGR Timing Advance - This table is very uneven but as far as I can tell the effected load % cells would be advancing timing between 5*- 7* in the 2500-3000 range, not retarding it!

Evap Map #1 - This table is leaning things out to a target 18-19.8 AFR from 2500-3500 during warmup presumably. Gave me a nice scare when I was testing lean burn in open loop but I'm not sure this is the timing retard culprit.

Anyone getting similar behaviour or can suggest where to adjust first? What are the functions of these 3 maps since they are not always in play?

 

I've never done any logging on a cold start but I have always had hesitation on light acceleration during cold start.

 

Same here, especially at about 1800-2100 rpm's...it feels like the car won't go, and then all of a sudden it will built boost again as the rpm's get a little higher. Feels like the car is being intentionally held back.

 

Yeah it's always been there even with the stock turbo but now with the larger turbo it's more pronounced since boost hits later. Granted the car is warmed up inside of 2-3 minutes and I can drive around this but there should be a reason it's happening.

Perhaps some hidden table that is altering the timing that we don't know about yet. I also want to learn more about the operations of the Ignition Warmup Retard, EGR Timing Advance and Evap Map #1 tables. Maybe this is something in play to heat the cat?

 

Throttle tables?

#5 is the cold one - #3 is the HOT one.

So just as a check copy #3 to #5 and see if that helps any...

 

Hmm I could try that but it doesn't seem TPS related. Comparing the two tables they are almost identical in the same columns for the 10-20% TPS range at the 2500-4000 RPM's.

As far as I can tell the hesitation feeling is from the car running 30* less timing then when warmed up. The numbers also bounce around a few degrees while it's doing it so the car picks up power on it's own without any increased throttle input which is very unsettling.

I'm going to log a -10 psi run and a zero psi run tomorrow and compare them. I don't really mind if it's retarded until it warms up, it just needs to be more consistant.

 

Yea the current warmup map is incorrect or defined incorrectly. I messed around with it a bit last year and it didn't do much.

On the 8 and 9 the ignition warm up tables have -20 to -30* in same places, from 10 to around 90 load. The 8 and 9 can have the same issue you are describing, even more so with E85, turning off the warmup tables helps a ton.

Mr. Tephra?

 

Evo 9 warmup tables:

 

ill see what I can dig up

 

??????

 

this might be it for 53040010:
evo10base.xml:
<scaling name="Percent255" units="%" toexpr="100*x/255" frexpr="255*x/100" format="%.0f" min="0" max="100" inc="1" storagetype="uint8" endian="big"/>

<table name="Percent Low Temp Timing Trim vs Coolant Temp" category="Timing" type="2D" level="2" scaling="Percent255">
<table name="Engine Temp" type="Y Axis" elements="8" scaling="Temp"/>
</table>


53040010.xml:
<table name="Percent Low Temp Timing Trim vs Coolant Temp" address="56174">
<table name="Engine Temp" address="6111c"/>
</table>

 

Ahhhh!

That makes a ton more sense in what I am seeing then, we just need to turn up the appropriate Evo X table then. I would imagine that leaving some retard in there is fine (and safer?), it just needs to be a little less intrusive though. The way it is now it almost doesn't encourage light throttle, you just want to get into low boost moderate throttle to avoid that area. I suspect it's an emmisions thing for heating up the cat earlier, which may or may not make a difference depending on how high flow your cat is. Cough... cough...

HB Speed:

I think that table you posted is just the EGR Timing Advance table. I tried smoothing it out a bit so there weren't so many huge timing swing but since I can't tell when it kicks in it's going to be hard to test any difference.

 

That looks very promising, if anyone can figure out the coding for a 52680015 US ROM I can test it out!

 

yeah I am not 100% sure..
52680015:

<table name="Percent Low Temp Timing Trim vs Coolant Temp" address="56170">
<table name="Engine Temp" address="610e0"/>
</table>

 

Ok that seems to bring up a table:

Percent Low Temp Timing Trim vs Coolant Temp

It has -40 C in all the engine temp column entries (which is odd) and 100% in the 2nd - 5th fields in the next column over. I'll try it tomorrow morning with 50% in there and see if only half the timing is pulled out on a cold start and report back. If it works well to just reduce their effect then finding the actual Low Coolant Temp Timing Trim maps wouldn't be needed as much.

Thanks!