MicroSquirt.com read only forum (Jan.15,2018)

efahl wrote:

flyguyeddy wrote:along the same lines, can you set a specific afr target for ego correction at a certain map/prm?

Yes, if your wideband and firmware support it. Usiing my example above, you really only need a wideband, that would work with any firmware, so long as you can get the WB data signal into MT, you are all set.

Eric

I'm sorry if I seem dense on this, but does that mean that even the B&G v2.98 firmware and the above code snippet in http://www.msefi.com/viewtopic.php?p=97678#97678
inserted into the custom.ini file will give an autotune like feature?

I assume that the afrtable.vex file would go into the same directory as the custom.ini file? (i.e., my ...\car1\mtcfg\ directory)?

I'd then be able to use the v2.98 .s19 in MS and any of the latest Megatune versions (i.e., 225b757) with the above code in the custom.ini file and a corresponding afrtable.vex to have an autotune feature that went after the targets in afrtable.vex? Of course, I'd be using a wideband sensor and have that input being read by the MS.

does that mean that even the B&G v2.98 firmware and the above code snippet in http://www.msefi.com/viewtopic.php?p=97678#97678
inserted into the custom.ini file will give an autotune like feature?

pparaska,

The key to autotune is that it uses the EGO correction to change the VE table. So while it can work with the v2.98 code, it only works at one EGO switch point at a time (since this is all there is in the V2.98 code).

You can tune different parts of the MAP to different switch points by changing the EGO and autotune parameters appropriately, but it isn't as simple (or automatic) as MS-II or MSnS-E which have AFR tables.

Lance.

Thanks, Lance.

I will go ahead and move to MSnS-E.

I'm still having trouble getting my head around the autotune feature how or if the "AFR Targets for VE Table 1" under "Fuel Tables 1&3" in MT225b717 using the msns-extra.ini file from the 024s13c MSnS-E firmware package are even used as EGO correction set points.

What I'm failing to grasp is how do MT (in the above configuration) and MSnS-E (024s13c) interact to tune to an AFR target table.

Eric has stated:

Well... Actually both simultaneously. Your MS will compute EGO correction and apply that internally. MT's auto-tune algorithm is watching that EGO correction and minimizing it by messing with the VE table. Obviously the MS does its EGO correction all the time (assuming closed loop is enabled), but MT can only do its bit when you have MT connected to a running engine.

By that, I guess that MT225b717 (or similar) will look at the "AFR Targets for VE Table 1" table and use the ratio of the AFR value that the "EGO switch point" under "Exhaust Gas Settings" in MT corresponds to to the AFR(rpm,map) target in the table to scale the EGO correction value that the MS sees?

So now I'm wondering if the above code snippet is even needed to use autotune with MSnS-E.

I am confused!
Pete

Am I getting close?

pparaska,

Yes, I think so. but I'm more familiar with MS-II, so let me rephrase it in MS-II terms.

MS-II has an AFR table, and a calibration table that tells MS-II which voltage corresponds to which AFR.

When EGO correction is active, MS-II looks at the AFR table to see what the AFR should be for the current conditions. It then looks up the EGO voltage that corresponds to that AFR.

- If the actual measured EGO voltage (from the wideband) is above that voltage, then the engine is lean, and the EGO correction is increased (within the EGO limits).

- If the actual measured EGO voltage (from the wideband) is below that voltage, then the engine is rich, and the EGO correction is decreased (within the EGO limits).

The EGO correction is applied to the pulse width, and the fuelling is changed appropriately.

None of this has involved MegaTune at all (other than to set the limits - but there's no 'active contribution' from MegaTune).

What MegaTune does is look at the EGO correction:

- If it is positive, then the engine must be lean at the current conditions (relative to the AFR table and corresponding switch point), and the VE should be increased.
- If the EGO correction is negative, then the engine must be rich at the current conditions, and the VE should be increased.

So MegaTune increases/decreases the VE a bit, depending on the EGO feedback, just like you would if you were watching the screen.

MegaTune's autotune doesn't know anything about the AFR table or switch points, it only knows if the EGO correction is above or below 100% for the current conditions, and uses that to adjust the VE table. These VE table adjustments bring the EGO output closer to the desired AFR, and EGO correction drops to 100%, at which point no further adjustments are made.

Lance.

Pete,

To reiterate, Auto-Tune will work with anything, any version of code, any hardware, any fuel algorithm (MAF, SD, A-N). It only requires two things: an EGO correction number and a VE table. I could enumerate all the combinations that this includes, but it's easier to say "it works with every combination."

Now you have a lot of choices as to how you generate that EGO correction value, and this is typically where people freak out. If you have a narrow band sensor, then only have one choice: set a switch voltage, set the Auto-Tune parameters such that it only tries to tune the safe/stoich part of the VE table and let 'er rip.

If you have a WB sensor and controller, then you can tune the whole VE table. If you are using an embedded code like MSnS-Extra or MS-II that supports an AFR table, you can use that MS-internal mechanism to generate the EGO correction value; this is pretty simple.

Only if you are using a WB with a code that does not support AFR table, like B&G 1.x, 2.x or 3.000, and you want the full AFR tuning capability, do you need to use my code snippet from above. The example with the AFR table in a VEX file uses MegaTune-specific functionality to compute the EGO correction that is normally done inside the MS. This is a bit more difficult than using the built-in mechanism supplied by Extra/MS-II, but it should work just as well.

Eric

is it wise to use the AutoTune for tuning a supercharged application?

arnold wrote:is it wise to use the AutoTune for tuning a supercharged application?

Arnold,

Maybe, depends on how you use it. If you've only got a narrow band O2 sensor, then no. If you have a wideband and have set all the proper parameters, then yes.

Eric

Ok, could you give me a bi of advise on this? I use the LC Wideband

Eric, would it be possible to add a colour change to a cell of a ve table which got affected?

red=richened
green= made leaner


This would be cool...in case all the cells dont get affected and ...well....i think it would be useful.

Hi! I think I saw a couple other guys with a similar problem, but never really found anything that seemed to help me. My autotune stays "ok" for about 30 seconds and then says "autotune outside area" - it thinks its outside its bounding box.

Heres the deal though, in the car specific custom.ini I set:

x 0-->10k
y 0-->300
z 0-->300

Trying to give myself a box that would cover ANY possible operating conditions, yet I still have this problem. It has the same problem with the default values and it seems like any value I put in there.

Oh, and i'm using a 1 day old downloaded copy of MT2.25, and MS&S-E 29q2... Everything is brand new / downloaded from official sources.

TIA

Pete