PDA

View Full Version : Early Evo rescaling loads - untested as yet


AndyF
08-02-2009, 12:08 AM
Right then people,
Following on from my discussions with Mr Banks on his thread where I asked about linear transformations etc... I now believe I have the understanding of how the scales on the maps are interpreted by the ECU and therefore have a way of rescaling the load limits on the fuelling and ignition maps.

The way in which the ECU determines which cell it will look at is via one of the above mentioned linear transforms. Basically, this takes the form of a list of raw data values of the calculated airflow and which column in the table that value represents.

To try and put it into simpler terms, I'll give an example...

This is the output of the air volume transform that is used as the load scale on the ignition and timing maps

Raw value ---> column number
0 ---> 0
1 ---> 1
2 ---> 2
3 ---> 3
4 ---> 4
5 ---> 5
6 ---> 6
7 ---> 7
8 ---> 8
9 ---> 9
10 --> 10
11 --> 10.5
12 --> 11
13 --> 11.5
14 --> 12
15 --> 12.5
16 --> 13
17 --> 13.5
18 --> 14
19 --> 14.125
20 --> 14.25
21 --> 14.375
22 --> 14.5
23 --> 14.625
24 --> 14.75
25 --> 14.875
26 --> 15
27 --> 15
28 --> 15
29 --> 15
30 --> 15

Now the thing to try and get your head around is that each raw value will be the same size increase and so the increase in values between column numbers differs..

If in the above example, the raw values were psi (which they aren't but it's an example) then we would have the following column definitions in the map

column number ---> psi
0 ---> 0
1 ---> 1
2 ---> 2
3 ---> 3
4 ---> 4
5 ---> 5
6 ---> 6
7 ---> 7
8 ---> 8
9 ---> 9
10 --> 10
11 --> 12
12 --> 14
13 --> 16
14 --> 18
15 --> 26

so you can see that the scale isn't linear like it is on the later evos.

However, because I now understand the linear transforms and how they work (and where they are stored in the code) then we can rescale those values whichever way we like and have them read up to whatever 'load' values we like.

The only problem I have is that my car is off to the bodyshop for a week starting on Monday and so I won't get chance to see if I can alter the load scales (and therefore which cells get 'hit' in the maps) until the week after next but I'll update this thread if all is successful.

Andy

jcsbanks
08-02-2009, 12:20 AM
It isn't linear on the later Evos either, but there is just a lookup table instead of a calculation. Can you log a load variable?

AndyF
08-02-2009, 12:33 AM
Haven't found a specific 'load' variable yet John but I just interpret it from my airflow readings and boost gauge.

soldave
12-02-2009, 06:57 AM
I could give you an idea of load values and their equivalent in boost/vacuum for the Evo I-III timing maps if you want? Will just need to grab the info from my tuning laptop.

merlin
20-07-2009, 09:05 AM
Andy, how did this finish-up, did you get some sensible answers?

AndyF
20-07-2009, 09:11 AM
Hi Merlin,
I've not tested it as I've not maxxed out the std tables yet as they are actually 'mapped' up to approx 1.7 bar.

What I can do though I guess is that now I can log the load column index I could rescale it anyway and then confirm that the column index changes to the appropriate point.

I *might* be able to do this on Wednesday but it depends if I can get hold of an alternator for mine as it died whilst I was mapping it yesterday :(

Andy

merlin
20-07-2009, 09:19 AM
ok, well then can you tabulate what psi your hitting in each of the top 5 load rows?
ie a rough guide as to expected boost/load would be useful.
I an trying to help a mate with an RVR with this.

AndyF
20-07-2009, 11:08 AM
Off the top of my head, row 12 = 1 bar, 13 = 1.2, 14 = 1.5 and 15 = 1.8

All are obviously approximate depending on turbo fitted etc. as it's airflow based rather than boost as such.

Andy

merlin
20-07-2009, 12:12 PM
^ understood and thanks.

zaki
03-08-2009, 04:21 PM
Off the top of my head, row 12 = 1 bar, 13 = 1.2, 14 = 1.5 and 15 = 1.8

All are obviously approximate depending on turbo fitted etc. as it's airflow based rather than boost as such.

Andy

the airflow is frequency based, Hz. Is it possible to assign the rows to frequency values? It will be a lot easier to tune if we refer to the Apexi SAFC as it can show the frequency of the airflow sensor.

AndyF
20-01-2010, 04:17 PM
Just to update this now basically as it's been tested by Redcup and the load scales have been succesfully changes to work with his Evo green turbo so it should be straightforward enough now to change them to work with whatever load we need to be able to map to.

Andy

vr401
20-01-2010, 09:39 PM
ok, well then can you tabulate what psi your hitting in each of the top 5 load rows?
ie a rough guide as to expected boost/load would be useful.
I an trying to help a mate with an RVR with this.
For RVR timing map load scaling is at 7E4D
and for fuel map load scaling is at 7E3D
Cheers, :smile:

merlin
21-01-2010, 02:06 AM
^ Thanks for that Miro.
We keep chipping away at these parameters. I will check it out tonight.

vr401
11-04-2010, 01:08 PM
^ Thanks for that Miro.
We keep chipping away at these parameters. I will check it out tonight.

Mate, I'm getting Moates Ostrich v2.0 next week so I should be able to map trace it now. This way it'll be a lot easier to tune and to use Andy's load rescale