Nice work!
Can I ask what software you used to generate the Alfa-N map?
I've not managed to get one out of MSTweak3000 or VexMe.
Paul.
Anyone care to comment on state of tune of my Zetec?
Forum rules
Forum rules
Read the manual to see if your question is answered there before posting. If you have questions about MS1/Extra or MS2/Extra or other non-B&G code configuration or tuning, please post them at http://www.msextra.com The full forum rules are here: Forum Rules, be sure to read them all regularly.
Forum rules
Read the manual to see if your question is answered there before posting. If you have questions about MS1/Extra or MS2/Extra or other non-B&G code configuration or tuning, please post them at http://www.msextra.com The full forum rules are here: Forum Rules, be sure to read them all regularly.
It's an application I wrote when I was investigating the relative merits between SD & Hybrid Alpha-n with my ITB setup. I uses the vex files for the current VE and AFR tables and outputs the files I posted and updated vex files. It is still in a crude form, I've not had chance to make it suitable for general use yet. It's written in C++ so is significantly quicker thanVexMe and performs extensive filtering to only use relevant data.ptownsin wrote:Colin,
Nice work!
Can I ask what software you used to generate the Alfa-N map?
I've not managed to get one out of MSTweak3000 or VexMe.
Paul.
Cheers,
Colin
Hi Colin,Cairb wrote:Chris,
I've had a quick look and have the following observations:
From the O2 readings it looks like your useing wideband, is that correct?
If running wideband with 0v = 10/1 and 5v = 20/1 then it looks like your experiencing the usual steep rise in VE with MAP above 90kpa. as it looks like your running rich at 90 and about stoic at 97 where I think you should be closer to 12.5.
If you've just used autotune then it may be that you have too few bins around this non linear region causing this.
With this configuration some consider placing more bins in the 90-100kpa range but I have found that Alpha-n gives better control at full bore which would imagine your wanting.
Search some of my previous posts for details if relevant.
I have run your datalog through my analysis software and attached some xls files based on SD and hybrid Alpha-n file based on a variable AFR table running about 16/1 at light load and 12.5 at full load. The zero's are missing elements are where there were no data points so would need.
I'm running a 2.3 Duratec on ITB's and noticed a significant improvement on Alpha-n as have a few others.
PS which version of megatune are you using as I couldn't load the msq?
Cheers,
Colin
Many thanks for your responses so far.
It's taking a while for your points to all sink in and I guess it's hard for you to really see what's going on if you can't see my MSQ!
I am using Megatune 2.25 built on Feb 5th 2006 which I thought was the most recent?
I have not been aware of the issues you mention at high MAP, could you point me to some areas of the datalog where they occur for me to fully understand. Also is it possible for anyone to run Hybrid Alpha-N or are you using a customised code of your own?
I am attaching images of my AFR targets and my VE table to help you to understand what is going on here and yes you are correct in asuming a wideband sensor, it is an LC1.
Thanks again,
Chris
Chris,
I'm assuming that you've set your LC1 for 10/1 at 0v & 20/1 at 5v.
If you look at the MAPBins file I posted you will notice that the MAP values are closer together above 86kpa. This is because the slope of the VE v MAP changes in this area so that the linear interpolation that megaquirt does has less errors from the desired value. If you look at the 4k line you will see that the VE that I have estimated goes from 53 at 86kpa to 107 at 98kpa. Whereas your table has 80 at 85kpa & 100 at 100kpa.
This increase in slope of VE wrt MAP seems to be a characteristic of ITB's. An additional problem is that with ITB's you either have a pulsating MAP signal that results in wide fluctuations in fueling or a very slugged system that doesn't respond well to throttle control. This is why I changed to hybrid Alpha-n as there is a much better correlation between VE and TPS (see the TPSBin file as a comparison). I still use the MAP signal as part of the fueling calcs, I believe that there were some changes since the code that I am running in this area but mine still proportions the fuel based on MAP. This is not quite ideal and I think that some sort of blending between TPS based and MAP based would be better but mine runs well as is. The main are affected is at high engine loads which is probably our main region of interest (;-{).
I you look at your datalog in excel, filter for the MAP values you will see that you are running lean at high MAP and rich as the MAP come down. This shows if you compare your VE table against the one I posted. The AFR table I used was similar to yours except I go to 12.5/1 at high MAP.
If you want to stick with SD I suggest that you increase the resolution around 85 - 100 kpa region and re - autotune, but I reckon that you would be better off with hybrid alpha-n. The maps I posted may be of some use for either method, don't forget to fill in the blanks though.
The mods that I have done to the code don't affect fuelling. I made a mod to keep the spark table MAP based, other mods were to suit my installation.
Are you also using spark control only some of the values look a bit retarded? This can also influence fuelling.
I think that the problem I have in opening your MSQ is that I need to set up the config files as I tried using the latest Megatune.
Cheers,
Colin
I'm assuming that you've set your LC1 for 10/1 at 0v & 20/1 at 5v.
If you look at the MAPBins file I posted you will notice that the MAP values are closer together above 86kpa. This is because the slope of the VE v MAP changes in this area so that the linear interpolation that megaquirt does has less errors from the desired value. If you look at the 4k line you will see that the VE that I have estimated goes from 53 at 86kpa to 107 at 98kpa. Whereas your table has 80 at 85kpa & 100 at 100kpa.
This increase in slope of VE wrt MAP seems to be a characteristic of ITB's. An additional problem is that with ITB's you either have a pulsating MAP signal that results in wide fluctuations in fueling or a very slugged system that doesn't respond well to throttle control. This is why I changed to hybrid Alpha-n as there is a much better correlation between VE and TPS (see the TPSBin file as a comparison). I still use the MAP signal as part of the fueling calcs, I believe that there were some changes since the code that I am running in this area but mine still proportions the fuel based on MAP. This is not quite ideal and I think that some sort of blending between TPS based and MAP based would be better but mine runs well as is. The main are affected is at high engine loads which is probably our main region of interest (;-{).
I you look at your datalog in excel, filter for the MAP values you will see that you are running lean at high MAP and rich as the MAP come down. This shows if you compare your VE table against the one I posted. The AFR table I used was similar to yours except I go to 12.5/1 at high MAP.
If you want to stick with SD I suggest that you increase the resolution around 85 - 100 kpa region and re - autotune, but I reckon that you would be better off with hybrid alpha-n. The maps I posted may be of some use for either method, don't forget to fill in the blanks though.
The mods that I have done to the code don't affect fuelling. I made a mod to keep the spark table MAP based, other mods were to suit my installation.
Are you also using spark control only some of the values look a bit retarded? This can also influence fuelling.
I think that the problem I have in opening your MSQ is that I need to set up the config files as I tried using the latest Megatune.
Cheers,
Colin