High-Resolution VE Table (option for 10ths)

This is a forum for discussing the development and testing of alpha MS2/Extra code. Documentation
(Runs on MS2 and Microsquirt)

Moderators: jsmcortina, muythaibxr

gslender
Super MS/Extra'er
Posts: 1072
Joined: Fri Sep 16, 2011 5:29 am
Location: Brisbane, Australia
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by gslender »

dontz125 wrote:I don't know how the tables are stored relative to each other, whether VE1 and VE2 are contiguous or well separated. The thought occurs to wonder how hard it would be - rather than have cell 08-11 of VE-HR consist of 2 consecutive bytes - to have the data byte stored at what would be 08-11 of VE1 be the 0-255 value, and what would be cell 08-11 of VE2 contains the tenths value. This means that whether you are using HR or standard tables, the values and handling of VE1 remains the same. VE2 will either be the swap table or the tenths table. Did that make any sense? Would there be any benefit to this scheme, or would it make it harder?
Both VE tables are a contiguous section of flash memory that could simultaneous exists as either the existing two tables, or as a single array of 16x16 ints (two bytes) so that it represents 1000 for 100.0%, 505 for 50.5%, 101 for 10.1% etc.

TS is easily than modified through the ini to be told that a contiguous 16x16 table is int with a scale factor or 0.1 that essentially provides a similar HR VE table to what is presented in the TS MS3 tunes.

All doable on the surface, but won't really know what other issues present until more code review is done.

G
Mazda MX5 + MS3 Pro
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: High-Resolution VE Table (option for 10ths)

Post by muythaibxr »

What makes you think you *need* the VE in 0.1%?

The current MS2 code interpolates in 0.1% units. What do you hope to gain here?

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
gslender
Super MS/Extra'er
Posts: 1072
Joined: Fri Sep 16, 2011 5:29 am
Location: Brisbane, Australia
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by gslender »

muythaibxr wrote:What makes you think you *need* the VE in 0.1%?

The current MS2 code interpolates in 0.1% units. What do you hope to gain here?

Ken
Ahh, perhaps I'm confused then.... if I have a cell at 55 and its neighbors are 56, but really I need 55.5 to get to the right AFR, how do I do that?

G
Mazda MX5 + MS3 Pro
Black99rt
Experienced MS/Extra'er
Posts: 364
Joined: Tue Mar 14, 2006 10:38 am

Re: High-Resolution VE Table (option for 10ths)

Post by Black99rt »

That's a bad example because one can answer- cut your req fuel in half and make sure you have map multiple on so you don't go much over 200.

I think the the req fuel should be left at the calculated value, other sections of the MSQ are more intuitive this way. We were told there would be features that relied on a "correct" req fuel and VE number in the future as well.

Now if you had said you needed 55.2, you might have an arguement for it :-p
1999 Dakota 5.9L R/T-
MSIIe 3.1.2 Batch Fuel/MSD single coil dizzy/Stepper idle
LC-1
gslender
Super MS/Extra'er
Posts: 1072
Joined: Fri Sep 16, 2011 5:29 am
Location: Brisbane, Australia
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by gslender »

Black99rt wrote:Now if you had said you needed 55.2, you might have an arguement for it :-p
Actually, I don't know if 55.5 is needed - I only know that 55 is too lean and 56 is too rich... so somewhere between those two is the best value... and it could very well be 55.3.

G
Mazda MX5 + MS3 Pro
93white3400z
Master MS/Extra'er
Posts: 685
Joined: Mon Aug 25, 2008 11:37 pm
Location: Montreal, Quebec

Re: High-Resolution VE Table (option for 10ths)

Post by 93white3400z »

But is 55.3 really gonna help anything over 55, I think that's Ken want to know. I mean in the end its probably so little fuel less, especially with EGO correction enabled. I honestly haven't noticed any improvement with my MS3 by using VE table in 0.1% increment but of course in MS3 its nice to have them since the flash is large enough anyway, might as well have that!

I think that also Ken said the interpolation is in 0.1% increment which mean that since your engine will not always be directly on the specific MAP/RPM of that cell, it will be interpolated with other cells around and will end up with a value in 0.1% increment, which makes the 0.1% increments for the VE table itself even less important like he said.
1993 Chevrolet Cavalier Z24
3400 Turbocharged Inter-cooled
MS3 v3.57+MS3X
gslender
Super MS/Extra'er
Posts: 1072
Joined: Fri Sep 16, 2011 5:29 am
Location: Brisbane, Australia
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by gslender »

93white3400z wrote:But is 55.3 really gonna help anything over 55, I think that's Ken want to know. I mean in the end its probably so little fuel less, especially with EGO correction enabled. I honestly haven't noticed any improvement with my MS3 by using VE table in 0.1% increment but of course in MS3 its nice to have them since the flash is large enough anyway, might as well have that!

I think that also Ken said the interpolation is in 0.1% increment which mean that since your engine will not always be directly on the specific MAP/RPM of that cell, it will be interpolated with other cells around and will end up with a value in 0.1% increment, which makes the 0.1% increments for the VE table itself even less important like he said.
Interesting point - perhaps I'd best look at if this VE value with this 0.1% interpolation occuring is producing the needed resolution anyway. Not sure if this is available by logs, so perhaps a simple addition to the logging to include this might be a more reaonsable way to reduce the concerns, and if what you are saying is correct, then maybe it isn't needed.

I need to look at the code then... :RTFM:
Mazda MX5 + MS3 Pro
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by racingmini_mtl »

I also think the effort would be better spent on something more useful. If you really need to tune to that level, you'll be using sequential injection. And then you already have the trim tables available to do the fine tuning.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: High-Resolution VE Table (option for 10ths)

Post by muythaibxr »

gslender wrote:
muythaibxr wrote:What makes you think you *need* the VE in 0.1%?

The current MS2 code interpolates in 0.1% units. What do you hope to gain here?

Ken
Ahh, perhaps I'm confused then.... if I have a cell at 55 and its neighbors are 56, but really I need 55.5 to get to the right AFR, how do I do that?

G
You will have 55.5 right between the bins. What I am trying to say is that the wideband doesn't have the ability (and most likely neither does the human mind, aside from placebo) to discern the difference between 55 and 55.5. The engine will most likely run the same in that situation.

Heck, the factories start bragging when they can model an engine to 5%.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
Martin Young
Helpful MS/Extra'er
Posts: 99
Joined: Fri Feb 25, 2011 6:31 am

Re: High-Resolution VE Table (option for 10ths)

Post by Martin Young »

Speaking as a beginner I have to say I was surprised to find the VE table could only accept whole number values. If a cell is at, say, 40 then the smallest adjustment possible (to 41 or 39) changes the amount of fuel injected by 2½%. That seemed surprisingly coarse to me.

What's the largest value the table accepts? 255?
dontz125
Super MS/Extra'er
Posts: 4200
Joined: Mon May 11, 2009 7:14 pm
Location: York, ON
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by dontz125 »

Martin Young wrote:What's the largest value the table accepts? 255?
Correct.
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module
TTR Ignition Systems
Greg G
Experienced MS/Extra'er
Posts: 311
Joined: Thu Mar 10, 2011 11:33 pm

Re: High-Resolution VE Table (option for 10ths)

Post by Greg G »

The workaround to getting the afr just right, with the current resolution of the ve table is by playing with the values of the bins (X and Y axis). A little cumbersome to get it just right for idle...

(short reply, am on vacation :D )
1996 Mazda MX-5 1.6L NA6/ Mazdaspeed M45 SC/ BSP AW Intercooler/ Maruha F-cams/ 425cc RX-8 injectors/ DIYPNP

MS2/Extra test mule :)
kjones6039
Super MS/Extra'er
Posts: 1986
Joined: Sat Jul 22, 2006 8:02 pm
Location: Eureka, NV USA
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by kjones6039 »

Martin Young wrote:Speaking as a beginner I have to say I was surprised to find the VE table could only accept whole number values. If a cell is at, say, 40 then the smallest adjustment possible (to 41 or 39) changes the amount of fuel injected by 2½%. That seemed surprisingly coarse to me.

What's the largest value the table accepts? 255?
Perhaps I'm missing something here. My experience (however limited) has been that I tend to use higher numbers in my VE bins. Instead of say, 40 in your example, I might use 60. (with an attendant decrease in Req_Fuel of course) Now, going from 60 to 61 is a smaller percentage change than going from 40 to 41 isn't it?

Or perhaps I'm wrong in my thinking! (which wouldn't suprise me!)

Feel free to slap me around and get me straightened out! :lol:

Ken
1979 Corvette - 383 CID SBC w/ Holley Pro-Jection 900 CFM TBI, 4-85 lb lo-z injectors & Walbro 255 pump
MS2 v3 w/extra 3.4.2 Release
36-1, Delphi LS2/7 coils in wasted spark, driven by v2.0 logic board from JBPerformance
Spartan Lambda Sensor from 14point7
TinyIOX from JBPerformance
piledriver
Super MS/Extra'er
Posts: 1679
Joined: Tue Oct 27, 2009 6:24 am
Location: Van Alstyne, Texas

Re: High-Resolution VE Table (option for 10ths)

Post by piledriver »

OK, just to make certain...

You can go to 255 as max VE, so as an example, reduce reqfuel 50% we can double our VE# and have 2X the resolution, right?

I guess my "mentally" equating and limiting VE% to "sane" VE numbers is limiting the potential resolution artificially.
Where are the gotchas?

I just did some simple math and I can scale my VE table by 2.5X and still have much headroom, dividing reqfuel by same, now idle is at ~105 and all the scaling and smoothing will be much finer % steps.

At idle/low PW this could help.

On an effort vs. benefit basis, I'd rather have a per-cylinder ignition trim table to play with vs. .1 VE%, my motor doesn't seem to care that much about AFR,as long as I feed it enough fuel.
(If that would be allowed)
Last edited by piledriver on Mon Feb 13, 2012 3:05 am, edited 1 time in total.
Always doing things the hard way, MS2 sequential w/ v1.01 mainboard, LS2 coils. 80 mile/day commuter status.
gslender
Super MS/Extra'er
Posts: 1072
Joined: Fri Sep 16, 2011 5:29 am
Location: Brisbane, Australia
Contact:

Re: High-Resolution VE Table (option for 10ths)

Post by gslender »

piledriver wrote:reduce reqfuel 50% we can double our VE# and have 2X the resolution, right?
I guess my "mentally" equating and limiting VE% to "sane" numbers is limiting the potential dynamic range artificially.
Where are the gotchas?
At idle/low PW this could help.
That's a good point. I could actually halve the applied calculated resolution in code, and then change the ini file to show .5 values (by scaling by half). That way you'd have an effective range from 0.5 through 127 in half step (0.5) increments (but yet the values would still be from 0-254) and no need to change the reqfuel.
piledriver wrote:I'd rather have a per-cylinder ignition trim table to play
Can you explain further why this would be required - how does ignition vary on a per-cylinder basis such that you'd benefit from that?
Mazda MX5 + MS3 Pro
piledriver
Super MS/Extra'er
Posts: 1679
Joined: Tue Oct 27, 2009 6:24 am
Location: Van Alstyne, Texas

Re: High-Resolution VE Table (option for 10ths)

Post by piledriver »

No two cylinders are ever identical.

This is far more of a problem on engines with "interesting" intake manifolds or siamese ports than for me personally.

As an example, the end cylinders on many (OK, most) inline 4s, I6 or V6 often run a tad hotter than the middle etc.
This will change the "best" timing relative to the inner pair, all else being equal.

(This was added to MS3 for good reason, it concerns me that it would be "allowed" for MS2, even though it has been a requested feature since dirt was new)

If you have sequential fuel, you can compensate some with fuel tweaks, but that's the wrong knob, even though it works.
A lot more folks are running sequential spark, so this could be of wider use.

The devs seeing it as a time sink for them when they'd rather be working on MS3 is rational, but...
...the thought of MS2 as "competition" for MS3 is pretty ridiculous, a MS2 with all the features implemented is a hack job at best, and still isn't close.
Always doing things the hard way, MS2 sequential w/ v1.01 mainboard, LS2 coils. 80 mile/day commuter status.
Post Reply