Tighten up maf corrections?

mayhemsm · **Posted:** Sat Sep 06, 2008 1:16 am

Ok, here's the run down of what I've got going on, I scaled the stock maf till it was nearly perfect. Installed modded stock injectors, scaled them until they were almost perfect, installed a custom CAI (71mm) and now my corrections are centered around 0% correction but they seem a bit broad, they range from +- 7-8% or so.

Here's the plot of my corrections:

Mods:
2006 WRX
FP TDO6h 20g
Full turbo back 3" exhaust
GM BCS
Custom 71mm CAI
Walbro 255lph
Modded Stock injectors ~900cc
TGV Deletes
Airpump Delete
Big TMIC

FMIC waiting to be installed this weekend.

Here's the log: (it has some junk data in it, so watch for that in the plots)

Attachment:

romraiderlog_20080905_192656.csv [665.64 KB]
Downloaded 6 times

Here's the map:

Attachment:

102_.hex [1 MB]
Downloaded 12 times

So any ideas on how to get those corrections a little tighter around 0%?

Bucket · **Posted:** Sat Sep 06, 2008 8:49 pm

I don't know the answer to your question, but I'm experiencing the same thing after installing an SPT intake. The scatter is much more spread out in the lower voltage range.

I have read that a larger scatter is the result of incorrect latencies.

I don't know if I would mess with what you have. It looks pretty good to me.

Buck

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 1778

Scatter can be due to incorrect latencies.

However, short ram intakes will also increase scatter.

Frankly, that scatter plot looks very good.

eminehart · **Posted:** Sat Sep 06, 2008 11:35 pm

I think things look pretty good but the latency values seem kinda high for modded side feeds. The top feeds need latencies in that area but the side feeds generally need very little changes to latency. But as I look at your maf scaling it seems that without these values the intake would need big scaling at idle. It is about impossible to get things right without doing them one at a time.

mayhemsm · **Posted:** Sun Sep 07, 2008 6:56 am

eminehart wrote:

I think things look pretty good but the latency values seem kinda high for modded side feeds. The top feeds need latencies in that area but the side feeds generally need very little changes to latency. But as I look at your maf scaling it seems that without these values the intake would need big scaling at idle. It is about impossible to get things right without doing them one at a time.

These are modded top feeds, sorry I didn't make that more clear.

But thanks guys, I feel better now knowing others think this scatter is acceptable, I just was used to seeing a bit of a tighter scatter prior to the CAI install.

Guess I better get that FMIC installed now and get to making some more power soon xD

If anyone's interested, I made the cheap-o custom CAI from parts i picked up at Pep boys, 2 extender tubes, a maf housing and a cone filter with a flex tube into the fender well. I saw another guy on another forum use this setup so he could still have a CAI with the SSAutochrome FMIC.

GeneralTJI · **Posted:** Sun Sep 07, 2008 1:52 pm

Good post...

I noticed the same thing when I went to an APS MAF tube that I have in the fender.... bummer having the filter so close to the MAF but I have no choice.

I'm thinking about trying your latencies out (running mod'd top feeds as well) and seeing what I get.... Mine are lower than yours....

BTW, this kicks off a little question, I'm on E85, I wonder how much (if any) I should adjust my latencies compared to gas? Not sure why I overlooked this before... car runs well but I wouldn't mind tightening up the scatter at low maf v's some...

Below I attached a pic of the MAF tube I'm currently building... I machined it out of a piece of cast aluminum (not sure what it was originally).. cast aluminum doesn't machine super nice but it actually came out quite smooth... just need to mount the sensor now, it's about 2.9" ID... and has the 3.5" OD lip on one side for my filter, and the 3" on the other for my intake tubing... just thought I'd share..

mayhemsm · **Posted:** Tue Sep 09, 2008 7:17 pm

Can I get some opinions on the top end of my maf scaling, dose this look acceptable to you guys?

Here's the graph of a 4th gear pull 1800rpm to 5800rpm:

I think from 2000-2750 rpms on that graph has such a large difference (error) because of tip in possibly?

Anyway, here's the data from the graph:

Code:

RPM   Load   AFR   Map AFR   MAS(V)   AFR Error, %
1835   1.28   14.85   14.01   2.46   6.02
1848   1.30   13.44   13.88   2.46   -3.17
1846   1.30   12.76   13.88   2.46   -8.07
1901   1.36   12.69   13.50   2.48   -6.02
1929   1.35   12.72   13.57   2.48   -6.23
1966   1.40   12.47   13.25   2.56   -5.89
1995   1.41   12.38   13.25   2.52   -6.57
2037   1.43   12.45   13.25   2.60   -6.04
2052   1.43   12.55   13.25   2.62   -5.28
2106   1.42   12.51   13.25   2.62   -5.58
2143   1.43   12.50   13.25   2.70   -5.66
2178   1.45   12.47   13.25   2.72   -5.89
2208   1.44   12.61   13.25   2.70   -4.83
2240   1.46   12.70   13.25   2.72   -4.15
2270   1.49   12.52   13.25   2.68   -5.51
2308   1.52   12.45   13.25   2.80   -6.04
2359   1.51   12.51   13.25   2.82   -5.58
2390   1.55   12.57   13.25   2.80   -5.13
2430   1.52   12.72   13.25   2.90   -4.00
2446   1.59   12.76   13.25   2.90   -3.70
2508   1.62   12.70   13.25   2.92   -4.15
2534   1.70   12.48   13.21   2.94   -5.50
2581   1.71   12.54   13.17   2.98   -4.80
2618   1.76   12.30   13.09   3.04   -6.03
2680   1.78   12.48   13.00   3.10   -3.97
2718   1.79   12.48   12.93   3.14   -3.51
2751   1.87   12.55   12.72   3.16   -1.31
2792   1.98   11.94   12.33   3.16   -3.16
2830   2.12   11.91   11.93   3.26   -0.20
2887   2.12   11.33   11.93   3.32   -5.02
2969   2.15   11.89   11.84   3.38   0.44
3003   2.24   11.79   11.56   3.44   1.95
3046   2.44   11.54   11.10   3.52   3.99
3087   2.54   10.79   10.94   3.60   -1.35
3119   2.62   10.97   10.89   3.70   0.74
3221   2.65   11.13   10.88   3.68   2.33
3286   2.90   11.26   10.82   3.76   4.06
3363   2.99   11.01   10.79   3.84   2.06
3410   2.99   10.79   10.79   3.84   -0.02
3496   2.98   11.23   10.80   3.84   3.96
3552   2.91   11.08   10.81   3.90   2.49
3648   3.01   11.01   10.81   3.98   1.85
3658   3.04   10.95   10.81   3.98   1.32
3755   3.06   10.95   10.80   4.00   1.37
3818   3.05   10.89   10.81   4.00   0.79
3873   3.05   10.98   10.81   4.02   1.62
3928   3.05   10.98   10.81   4.00   1.62
4042   3.04   10.89   10.81   4.06   0.76
4069   3.09   10.82   10.80   4.14   0.23
4140   3.03   10.94   10.81   4.06   1.20
4171   3.05   10.94   10.81   4.10   1.25
4252   3.07   10.86   10.80   4.18   0.56
4296   3.09   10.82   10.80   4.18   0.23
4378   3.08   10.73   10.80   4.20   -0.63
4419   3.08   10.91   10.80   4.20   1.04
4492   3.10   10.89   10.80   4.24   0.87
4540   3.14   10.82   10.79   4.28   0.26
4610   3.13   10.72   10.80   4.30   -0.72
4695   3.22   10.67   10.80   4.34   -1.18
4752   3.13   10.80   10.81   4.32   -0.06
4815   3.11   10.92   10.81   4.32   1.02
4828   3.15   10.89   10.81   4.32   0.74
4913   3.17   11.10   10.81   4.36   2.68
4988   3.13   11.00   10.81   4.38   1.76
5070   3.17   10.94   10.81   4.38   1.20
5067   3.18   10.80   10.81   4.40   -0.09
5137   3.21   10.79   10.81   4.46   -0.19
5169   3.19   10.83   10.81   4.44   0.19
5208   3.17   10.83   10.81   4.46   0.18
5299   3.12   10.92   10.82   4.44   0.96
5354   3.14   10.78   10.82   4.46   -0.39
5374   3.15   10.86   10.83   4.48   0.32
5399   3.13   10.75   10.82   4.46   -0.69
5490   3.13   10.83   10.83   4.52   -0.01
5491   3.11   10.94   10.83   4.54   1.04
5529   3.11   10.94   10.83   4.54   1.02
5574   3.09   10.94   10.83   4.54   1.05
5675   3.07   10.98   10.82   4.56   1.48
5694   3.06   10.89   10.82   4.52   0.68
5753   3.03   10.85   10.81   4.52   0.37
5764   3.02   10.95   10.81   4.52   1.30
5782   3.02   11.01   10.81   4.56   1.85
5825   2.99   10.97   10.81   4.54   1.48
5826   3.03   10.92   10.81   4.56   1.02

mickeyd2005 · **Joined:** Wed Oct 25, 2006 12:32 am **Posts:** 1778

That's pretty good scaling.

RomRaider

Forum rules

Tighten up maf corrections?

Who is online