It's wintertime here as you all know and almost all tuners stopped any mapping activities saying that in temperatures below +10 degrees it's impossible to do it properly. My problem is if it's true or is it just lack of ability to write proper temperature correction maps?

Also question connected to this one. Is very cold temperature safe for cars tuned lets say in UK that never been tested in such temperatures? Let's say my STG1 Escos in -30 schould the fueling be spot on or schould it be checked?

 

Well I've not heard that one before. Maybe cars aren't meant to be driven in the cold...............

 

proper calibration work should definitely include sub zero temperatures.

depends on the individual maps. i would guess that most maps have the same correction factors as factory, which you would hope were done properly by ford in the first place. most chips are just tuning of original maps for increased boost, different cams etc, so do not necessarily have to change any modifiers or correction factors for temperature

 

So you think it's not needed to compensate for much higher intercooler efficency and things like that ? I heard that sometimes highly tuned engine give problems in very low ambient temperatures.

 

that's not what i was trying to say jakub. i was trying to say that yes of course you need to compensate for higher intercooler efficiency etc (although the ecu does not know why the engine is receiving cold air), but that it should already be compensated for in the original complete mapping process done by ford. when it has been tuned, all that they have probably done is to adjust the main maps that are applied at a higher temperature, so that the original compensation should still apply (roughly).

 

The key question is how roughly


I mean I'm not sure if it makes a difference if you push it (intercooler, turbo system) to the limit or not.

 

Here's the standard ST Escort Cosworth borderline spark compensation for ECT. The table is the spark advance/retard for ECT and is based upon rpm and load (I haven't worked out the conversion factor for the tables yet). The multiplier is then used to calculate the amount of advance/retard to be used according to temperature.

SPARK_BDLN_RETARD_ECT_TABLE (rpm vs load)
1.375 4.000 4.750 5.750 2.500 2.500
2.125 4.625 5.500 6.375 3.875 3.875
3.125 5.625 6.375 7.250 8.250 9.500
4.625 6.875 7.625 8.625 9.500 10.875
7.125 10.250 11.125 12.000 13.000 14.000

SPARK_RETARD_FOR_ECT_MULTPLIER (degrees C vs multiplier)
123.33 -0.21
117.78 -0.21
87.78 0.00
-40.00 0.94
-160.00 0.94
-160.00 0.94
-160.00 0.94

The same process is done for ACT.

SPARK_BDLN_RETARD_ACT_TABLE (rpm vs load)
5.750 5.750 10.000 9.000 9.000 5.750
5.750 5.750 10.000 9.000 9.000 5.750
5.750 5.750 10.000 6.250 5.750 5.750
5.750 5.750 10.000 6.250 5.750 5.750
2.500 2.500 10.500 6.250 5.750 5.750

SPARK_RETARD_FOR_ACT_MULTPLIER (degrees C vs multiplier)
123.33 -0.59
121.11 -0.59
37.78 0.00
-40.00 0.27
-160.00 0.27
-160.00 0.27
-160.00 0.27

As you can see the temperature adjustment is very much 'broad brush' and doesn't seem to be needing too much fine control on an ecu that calculates fuelling and spark timing from load vs rpm based maps.

Incidentally there is no idle speed adjustment for ACT. Also, IIRC correctly, the optimum boost setting is around 40 degrees C ACT on the ST.

Edited following a bit more investigation.

 

WOW, do I see 120 to -160 degrees celsius???


Pretty wide and not many points in the correction curve..hmmm... I though more compensation was needed...

 

I've edited my post after a bit more investigation, this increases the number of correction points but it still remains a very coarse correction. The extremes of the temperature are probably more symptomatic of lack of need for fine control rather than you ever seeing –160 The thing to bear in mind with eec-iv’s is that the fuel and spark tables are load based (in this case load = volumetric efficiency) and because all the latter ones use mass air sensing the fuelling and spark are shifted automatically. e.g. high ACT’s mean a less dense air charge and therefore calculated load is less and different parts of the tables are used.

 

Yes but apart from VE temperatures affect such things as knock treshold. I presume that this correction is build in to compensate for that. Still I'm surprised it's that rough. Interesting....

 

There is a SPARK_BORDERLINE_TABLE which is used as part of the calculation to find the most spark advance permissable before knock occurs, the tables shown above are used for part of this calculation (amongst others). See http://eectuning.org/forums/viewtopic.php?t=5791 for a bit more detail if you're interested.