tuning strategy ?

[Yves]

I have been doing some tuning based on what I see in the logfiles.
I put a window with only the AFR, AFR target and tps on it, so in essence you should be able to see where it runs lean, rich and so on.

However I sometimes get different readings for the same cell.

What tells you when the actual AFR reading is correct ? I have been looking at both the map and tps and looking when the exhibit kind of stable values, using the middle of an extended periode with the same tps reading, but unsure if that is correct.

Also : how close does the AFR need to be versus the AFR target ? Seems that's nearly impossible to get it spot on.

[elutionsdesign]

You'll never get it spot on, that's what the EGO control is for. Too many variables in a running engine let alone the weather and fuel quality. Most tuners are happy with 5% or less variance from target.

[lagos]

I enable EGO correction and look at the amount of correction thats going on to see how close I am. 5% correction is pretty good in my book.

[Yves]

Do you use the standard settings in EGO correction ?

[elutionsdesign]

No, I don't let it correct my idle tune so it starts around 1200 rpm. Of course you want the motor warm. I generally give it 25% authority and never more than 1% step size. Something around 20 events per step gives good response.

[Yves]

I found out that EGO control was not disabled when I tried to do my fuel tables. Does this screw up results ? Does the VE analyser in mlv take this into account ?

Current settings are :
- events per step : 40
- step size : 0.1%
- controller auth : 0
- active above 160°F
- above 1300 rpm
- below 70% tps
- below map : 90
- above map 20

It would look as if it doesn't work with 0% authority. Not sure if i need to enable it.

[elutionsdesign]

Not if you're using tuner studio, it takes both the accel enrichment and EGO into account when calculating VE so you can leave them on.

Give it 20 events per step, set controller authority to 25% and step size to 0.5% for now.

[Yves]

Question about lag factors. I have them set at 50% now. Should they be set higher for better response ? The signals will jump around some when I do.

Do I also need to set the PWM current limit to 100% ? Currently it's at 91

[dontz125]

The higher the lag factor, the faster the response, but the greater the effect of noise or erratic response. You need to set each factor to give the best combination of responsiveness and smoothness.

PWM current limiting is only used with lo-z injectors. If you're using hi-z injectors, then yes - it should be 100%.

[Yves]

Ok, I tried the following, I set the pwm current limit to 100%. I also set the ego correction and I worked on the IAC some to get it to work. When warmed up I blipped the throttle some and took a log. Usually I was getting some backfiring/popping, the ego usually went really lean when blipping the throttle.

I ran that through the VE analyser in MLV just as a test. It set the VE values in those high kPa(>90 kPa)/low rpm zones (1300-2300 rpm) immediatly to very high VE's (118 and such). For a test I ran the table the VE analyse gave me an it seemed not to backfire/pop as much. When running that log through the VE analyse it made the VE's even higher with some bins in 146...Can I assume that VE analyse is off here ?

PS this is an ITB setup on a short stroke 5.7 liter V8 with long rods (4.125" bore, 3.250 stroke, 6.250" rods)

[elutionsdesign]

Are you running SD or ITB mode or other? I wouldn't immediately assume the VE is off but if you're in SD mode you won't have any fueling resolution in the 90KPa range. NA or turbo?

[Yves]

Are you running SD or ITB mode or other? I wouldn't immediately assume the VE is off but if you're in SD mode you won't have any fueling resolution in the 90KPa range. NA or turbo?

SD and NA

FWIW I have 42 lbs injectors, which are based of an estimate on horsepower gotten from several sources for this combination as there is no real comparison to an engine with similar characteristics outside maybe a Nascar engine. If the VE's are that high I fear the injectors stay open too long.

If this is in any way correct I'm looking at a really bumpy VE table with a very high VE's at the high kPa zones. The engine is not even in it's power band in that rpm zone.

This is the log of the last run with the modified table.

PS : manifold is a crossram ITB, so the path from stack to valve is almost straight. ITB's are 50 mm.

[elutionsdesign]

I wouldn't go any further with SD, you're just not going to get the drivability it deserves. Switch to ITB mode.

[kaeman]

Hello Yves, we are working on the same engine combo. your injector times are Not to long. I am working on a rich misfire issue when I stab the throttle at idle. I found that the auto tuners and the ve analyzers don't work in some sections... if you have a rich misfire as I had, the auto tunes and ve analyzers kept adding fuel because the afr is real lean, but as I have found rich misfires show up as lean spots and the tuning software keeps adding fuel to a place where the engine cant use the fuel its already getting. I fought that problem for a long time the software had some of my bins in the 240's... and the problem just kept getting worse. so I looked at an old fuel table that had been tuned by hand through out and copied the ve table values into the new ve table.. and the engine ran way better. now I am just fixing the off idle misfire during hard launches. look at my msq. my ve table needs some small tweeks but the engine runs good and the vehicle pulls hard during acceleration. if I bring the engine to about 2000 to launch, the misfire is gone and the stumble is also.

2014-07-29_18.03.44.msq

[Yves]

Hi,

Thanks, I have something to compare to now.

Looks that your cam is a little tamer in the sense that the timing is shorter, lca is wider, but you have more lift. I have 247/249 @ .050 with an lca of 106 so this will definately play a role. I run a Mike Jones cam and from what I'm told these are rowdy. If I cannot get it to work, I might have to consider going to a different cam. At least I was able to get a reasonable idle out of it.

I have been thinking that the ecu was being misguided by somethings among others misfires and most probably the accel shot that is not good. I have been readjusting the AE some (put in more fuel and have it come on faster). Still have to see what this is providing me.

I also think that VE analyser/autotune is no good below a certain rpm, which in my case is about 2000 rpm, where I have lots of reversion and misfiring possibly taking place, so like you said it might be necessary to tune this section by hand. Still a lot of work to do that's for sure.

The other thing I have been thinking of were the mesh filters I'm currently using. I'm of the impression that they screw up flow into the stacks and restrict things severly. I found out that after balancing the throttle and setting the idle speed screw that the rpm would drop ca. 150 rpms when putting on the filter.
The co. that sold me the manifold sold these as a filter solution and said they were not restrictive, but ...I hope it's only a temporary solution as I set to build myself an airbox around the stacks. Some of the alu plates are being machines as we speak, i'm just waiting on them to arrive.

I do have a question though : I see you have VE values with decimals. Mine are all values with no decimals. If I can use this, it would be better since I feel that at idle I fall in between values.

[dontz125]

A rich misfire reads as dead lean because of all the unconsumed oxygen going past the sensor.

[kaeman]

I am currently using ms3 ecu, it has the decimal place in the ve tables, my ms1 didn't. what ecu are you using. I had to make a box for the top of my stacks because my manifold is old enough that hilborn didn't make the individual foam filters that fit the manifold. so I fabricated my own box and used a readily available flat square filter from a mid 90's to 2008 pick up with 454. its flat and about 8 x12 iches.. plenty of airflow. but when I went from no filters to this one it completely changed my tune. I found changes when I used screen in the bottom of my stacks to keep bolts etc out to the intake. it had 1/4 mesh with .o28 wire and it affected flow.... DONT be surprised when you need to change the tune after putting a filter on. don't open the idle screw to adjust for the filter.. fix the ve tables, the futher closed your butterflys can be at idle the better.
Good luck.

PS I ran my engine with the no filters for 3 years, my stacks fit under the stock hood and I had it tuned for that, when I went to the airbox I had to cut a hole in the hood to make the air cleaner assy fit. my tune changed because of the hole in the hood.. tuned for that because I hadn't finished the box and was heading for a cruise.. then finished the airbox tuned for that change then put on the cowl on the hood and it changed the ve tables again... my point is make all the changes to the filter and the hood if those mods are needed before tuning again.. sure the more you practice the quicker you get the hang of it but it sure becomes a pain quickly...LOL

[Yves]

I am currently using ms3 ecu, it has the decimal place in the ve tables, my ms1 didn't. what ecu are you using. I had to make a box for the top of my stacks because my manifold is old enough that hilborn didn't make the individual foam filters that fit the manifold. so I fabricated my own box and used a readily available flat square filter from a mid 90's to 2008 pick up with 454. its flat and about 8 x12 iches.. plenty of airflow. but when I went from no filters to this one it completely changed my tune. I found changes when I used screen in the bottom of my stacks to keep bolts etc out to the intake. it had 1/4 mesh with .o28 wire and it affected flow.... DONT be surprised when you need to change the tune after putting a filter on. don't open the idle screw to adjust for the filter.. fix the ve tables, the futher closed your butterflys can be at idle the better.
Good luck.

PS I ran my engine with the no filters for 3 years, my stacks fit under the stock hood and I had it tuned for that, when I went to the airbox I had to cut a hole in the hood to make the air cleaner assy fit. my tune changed because of the hole in the hood.. tuned for that because I hadn't finished the box and was heading for a cruise.. then finished the airbox tuned for that change then put on the cowl on the hood and it changed the ve tables again... my point is make all the changes to the filter and the hood if those mods are needed before tuning again.. sure the more you practice the quicker you get the hang of it but it sure becomes a pain quickly...LOL

I know what you mean. The problem here is that this is a crossram. The bellmouths are really close to my tall valve covers and the fuel rail runs underneath them. I have some plates being made up for them so that I can attach an airbox. Then I still need to make a filter box that will take air from the cowl. I already guessed that it would upset the tune if you change these things. Right now I guess the airflow is about 15% down from running no filters.

[suberimakuri]

Changed to ITB mode or AN yet?
ITB's on SD are recipe for disaster... especially if you do any racing at different baro's.

A SD table on ITB looks logarithmic with map.
A SD table on single throttle looks linear with map.
An AN table on ITB looks linear with throttle position.
An AN table on single throttle? No idea, no need for me yet.
An ITB map? No idea haven't done one personally...should be linear due to switch point between SD/AN.

[Yves]

I would like to know why it's such a disaster with itb's ? Not that I might not try but I need to know where to set my switchpoints, so that cannot be done with guessing.

I'm not sure to what this SD disaster is related to, maybe irratic map signal ? I have a 50% lag and a pretty large vacuum chamber underneath the runners. The vacuum seems to react like when I was running on carb and single plane manifold.