Miata Turbo Forum - Boost cars, acquire cats.

Miata Turbo Forum - Boost cars, acquire cats. (https://www.miataturbo.net/)
-   MEGAsquirt (https://www.miataturbo.net/megasquirt-18/)
-   -   Table switching to increase VE and spark resolution for MS2 (https://www.miataturbo.net/megasquirt-18/table-switching-increase-ve-spark-resolution-ms2-56241/)

miatauser884 03-13-2011 07:16 PM

Table switching to increase VE and spark resolution for MS2
 
Anyone out there try using fuel/spark table 1 as an idle only table, then use table 3 as the main fuel/spark table.

Set the table switch to something like TPS > .5%

It should allow a couple more rows of resolution for the normal tune for table 3, and a huge idle area for table1

dynokiller90 03-14-2011 07:02 PM

Is that possible? I'm going to have to look into that if it is. not for idle, but make it switch based on map. use one table for boost one for vac that way there is a lot more resolution to tune in cruize and such.

miatauser884 03-14-2011 07:06 PM

Quote:

Originally Posted by dynokiller90 (Post 701399)
Is that possible? I'm going to have to look into that if it is. not for idle, but make it switch based on map. use one table for boost one for vac that way there is a lot more resolution to tune in cruize and such.

Exactly, I like your thinking. I haven't tried it, but I don't see why it can't be done.

As long as the latency for table switching is short. You should be able to have a high res cruise map and a high res boost map that you could switch at ~ 100kpa.

I'm surprised this post didn't get more responses. I just figured I'm retarded.

y8s 03-15-2011 09:59 AM

the MS3 has a feature like this called "idle advance" (and another one for idle VE).

it waits until closed loop idle parameters are met plus delays and then uses the idle advance (or VE) table.

so yeah, I'd say you could do it.

the parameters for idle advance are (my settings):

TPS Below value (1%)
RPM below value (1500)
load above value (15%/kpa)
CLT above value (50C)
after delay (1sec)

Braineack 03-15-2011 10:18 AM

use dual maps for this. make a 16x30 table. or it that a ms3 thing again?

Ben 03-15-2011 10:30 AM

The table switching is instant. It's recommended that you overlap your break point from table1 to table3, though I know some people do not.

However if you can demonstrate (with datalogs to back it up!) that your car needs more resolution than 16x16, we would like to see.

I think that you'll find that more "resolution" in cruise not only isn't necessary, but adds significant complexity to getting a smooth driving tune.

miatauser884 03-15-2011 10:38 AM

Quote:

Originally Posted by y8s (Post 701654)
the MS3 has a feature like this called "idle advance" (and another one for idle VE).

it waits until closed loop idle parameters are met plus delays and then uses the idle advance (or VE) table.

so yeah, I'd say you could do it.

the parameters for idle advance are (my settings):

TPS Below value (1%)
RPM below value (1500)
load above value (15%/kpa)
CLT above value (50C)
after delay (1sec)

Unfortunately, it appears that MS2 only uses one of the parameters (TPS,RPM,KPA) at a time.

Quote:

The table switching is instant. It's recommended that you overlap your break point from table1 to table3, though I know some people do not.

However if you can demonstrate (with datalogs to back it up!) that your car needs more resolution than 16x16, we would like to see.

I think that you'll find that more "resolution" in cruise not only isn't necessary, but adds significant complexity to getting a smooth driving tune.
You are probably right. It is also going to be a pain to tune if VE analyze doesn't switch tables as well. It might be nice to have the boost VE and spark kpa cells seperataed by only 10kpa. Right now mine are separated by 20kpa from 180-260.

That being said, I really want it for an idle table.

Quote:

use dual maps for this. make a 16x30 table. or it that a ms3 thing again?
For VE you could do a 30x16 or a 16x30 Depends if it is better to have higher MAP resolution or RPM resolution
For spark you could do a 22x12 or a 12x22 depending on the same thing MAP vs RPM

Ben 03-15-2011 10:48 AM

Quote:

Originally Posted by djp0623 (Post 701671)
It might be nice to have the boost VE and spark kpa cells seperataed by only 10kpa. Right now mine are separated by 20kpa from 180-260.

Why? What do you gain? Your VE is going to be pretty linear, so adding "resolution" doesn't improve your tune any. It just makes for more cells to tune.

When you're working with a management system with very large tables, what most tuners will do is skip to every 3rd or 4th row, and interpolate between the tuned rows. Why? Because the added table size does nothing for performance. Also systems with very large tables generally do not allow for custom break points, where we can set them wherever we need/want them.

I've heard people complain that MS2/MS3 16x16 tables are too big. That's why in MS2 you can bump your VE table back down to 12x12 if you want.

Here's a fun video of the Huber Mustang. Nearly 1000 whp 40 psi turbo 4-banger. It runs a fancy MS1 based system. 12x12 tables. This is a 9 sec run, but he's well in the 8's now.


miatauser884 03-15-2011 10:53 AM

Quote:

Originally Posted by Ben (Post 701680)
Why? What do you gain? Your VE is going to be pretty linear, so adding "resolution" doesn't improve your tune any. It just makes for more cells to tune.

When you're working with a management system with very large tables, what most tuners will do is skip to every 3rd or 4th row, and interpolate between the tuned rows. Why? Because the added table size does nothing for performance.

Instead of asking why, ask why not??? Who cares if it takes extra tuning. I can easily tune VE with VE analyze. I can even work around the table switching issue. I could fine tune my spark in boost with a few more cells. Especially since I know right where my car gets finicky with spark adjustment in boost. Why does MS3 use higher resolution tables if it is unnecessary? I like tinkering, so more table gives me something to do.

I'm merely pondering ways to get a little extra from the MS2.

JasonC SBB 03-15-2011 12:02 PM

Can you set your own breakpoints in the MS3?

In my AEM I have closer breakpoints for both RPM and MAP around idle.

Ben 03-15-2011 12:22 PM

Quote:

Originally Posted by djp0623 (Post 701684)
Instead of asking why, ask why not???

I just answered the why not, or at least hit the highlights. But if you can demonstrate a need, with datalogs to back it up, then I will present a case to Ken.

Quote:

Originally Posted by JasonC SBB (Post 701726)
Can you set your own breakpoints in the MS3?

In my AEM I have closer breakpoints for both RPM and MAP around idle.

Yes. You can set all breakpoints and the MegaSquirt uses a weighted interpolation between cells.

y8s 03-15-2011 02:00 PM

Quote:

Originally Posted by JasonC SBB (Post 701726)
Can you set your own breakpoints in the MS3?

In my AEM I have closer breakpoints for both RPM and MAP around idle.

with the separate idle VE map, I can do both/either and have different AFR for light cruise at low rpm (say for when I am doing 25mph in 5th gear) :loser:

yeah all MS3 maps have fully parametric breakpoints.

Reverant 03-15-2011 02:55 PM

Same goes for the MS2 as well.

miatauser884 03-15-2011 02:59 PM

Ok, so does anyone see any benefit to using table-switching to increase resolution or to create a seperate idle map on the MS2?

Braineack 03-15-2011 03:05 PM

you shouldn't really need it to be honest; only when A/C is active if anything.

Jeff_Ciesielski 03-15-2011 03:09 PM

Quote:

Originally Posted by djp0623 (Post 701819)
Ok, so does anyone see any benefit to using table-switching to increase resolution or to create a seperate idle map on the MS2?

Aside from the neat-o factor....No.

If you cannot get a stable AFR in boost, good sub 100kpa response, good cruise mileage, and a stable idle with a 16x16 map, your tune sucks and needs to be reviewed.

JasonC SBB 03-15-2011 03:20 PM

Quote:

Originally Posted by y8s (Post 701790)
with the separate idle VE map, I can do both/either and have different AFR for light cruise at low rpm (say for when I am doing 25mph in 5th gear)

So if you can set up fuel, timing, and afr target maps for more breakpoints around idle, what's the need for a separate idle VE map?

JasonC SBB 03-15-2011 03:21 PM

Quote:

Originally Posted by Ben (Post 701666)
However if you can demonstrate (with datalogs to back it up!) that your car needs more resolution than 16x16, we would like to see.

Another way of putting this: How many map points are you willing to tune, on the road or on the dyno?

miatauser884 03-15-2011 03:32 PM

Quote:

Originally Posted by JasonC SBB (Post 701837)
So if you can set up fuel, timing, and afr target maps for more breakpoints around idle, what's the need for a separate idle VE map?

So what rpm break points are you guys using from idle to 2k rpm if you don't have MS3 resolution for VE and spark?

I've noticed that having a large area around idle (up to 1400rpm) that maintains my idle advance (14 degrees) My idle settles very quickly at various humidity, temp, etc conditions. However, notice a significant off idle power reduction by removing this much timing.

y8s 03-15-2011 03:39 PM

I thought I was clear above with the 25mph in 5th gear...

idle load and rpm is a pretty small region of operation, right?

normal (or abnormal) driving can occur in the cells also used for tuning idle in theory.

you may also want to tune the initial fuel delivery for idle cells to be slightly richer to prevent stalling but once things settle down, you want to go back to closer to 14.7 AFR.

I use it mainly to avoid autotuning those cells incorrectly. the idle ve table provides a "dont touch me" table.

would it blow your mind if I said you could have a second idle VE table that is used for certain circumstances?


All times are GMT -4. The time now is 06:46 PM.