Page 1 of 1

Still having trouble tuning - the never ending battle

Posted: Sun Dec 09, 2012 6:43 pm
by sammy
Hi everyone
I hope someone can help with this and I apologise in advance if my questions are a little lame.
I’m still having problems with tuning.
I’m going to try Alpha-n only to see if that helps but in the meanwhile, I’d like to exhaust all avenues with Speed Density and Blended Alpha-N – which I still don’t totally understand.

Speed Density
So far I have Speed Density only & the car is idling reasonably well. I can also drive around the block with relative ease as long as I accelerate smoothly.
However there are a lot of issues, such as flat spots and lean or rich spots all over the place.

I’m hoping the more experience people on the forum could possibly have a look at the attached log and provide some thoughts?
In particular it goes lean and rich when I wouldn’t expect it to. It’s a lot better than it was a few weeks ago but it’s still not crash hot. I want to get this sorted before I try blending Alpha-n.
BTW I’m running 2.86 on V3 of the board and 4 independant Throttle bodies.
2012-12-09_20.10.06.msl
Image 2.png
Alpha-N / SD blend
I have Matt Cramer’s EFI book and I’ve read a lot of stuff on the forums, but I can’t seem to get a straight description of how to blend A-N with SD with 4 Independent Throttle bodies. I know blipping the throttle sends MAP to peak 100kpa immediately, but I need an example of how you fix this with blended on v2.86 of the code. I’ve seen a few samples but they are all done differently and none on v2.86.
Following are the screen and options, I can’t see how these all relate to the VE table
Do I extend the VE above 100kpa to cater for the Aplha-N table above 100Kpa?

Re: Still having trouble tuning - the never ending battle

Posted: Mon Dec 10, 2012 9:47 am
by Matt Cramer
Are you running boost? If not, you won't be going over 100 kPa.

The table generates a "psuedo-MAP" signal that is plugged into the VE table below the "Pure AN to blend threshold" RPM. Above the "Blend to SD threshold" level, it uses the MAP reading straight up. Between the two, it averages them. You'd want to scale it so that the 100% level is the point where, above which, more throttle opening does not affect the way the engine behaves.

Re: Still having trouble tuning - the never ending battle

Posted: Tue Dec 11, 2012 2:43 am
by sammy
Matt ,

I ‘m sorry but I still cant get my head around it.
The Pure AN to Blend & SD thresholds are both based on RPM.

The car idles between 68 and 70 kPa at 750rpm.

If I blip the throttle at idle the map will immediately shoot up to 100Kpa while the RPM bearly moves past 700 to 800 rpm for a second, then the revs pick up but at that stage the MAP has been at 100kpa for a few seconds. Other times I may be crusing then I accelarate hard and the map goes to 100KPa again and is stuck there for a few seconds.

I thought the blending would take place across the entire RPM range and that Alpha N blends upward from a particular kPa
I dont have much head room above idle (i.e. only 30kpa before reaching 100) and there is a very rapid rise from 70 to 100KPa.

I don’t get how its controlled by the RPM

Do you have any real world samples or map files I can look at???
thanks fro you rhelp.

BTW I spoken to a few people about this and most seem to think its a bit of a black art to get ITB running well. There doesnt seem to be much in the manuals either.

Sam

Re: Still having trouble tuning - the never ending battle

Posted: Mon Dec 17, 2012 4:40 am
by whittlebeast
I am developing a web page to help people identify if they really need blended tuning. Here is what I have come up with so far.

www.ncs-stl.com/tuning

The general conciseness is Speed Density works quite well for the vast majority of all motors. Implementing blending is very late in the tuning process. Like years in.....

Hope this helps

Andy

Re: Still having trouble tuning - the never ending battle

Posted: Sat Jan 05, 2013 6:59 am
by sammy
Sorry for the delay in responding, I've been flat out over the Christmas break rebuilding one of my other cars and with the added Christmas and New Year festivities I haven’t had a chance to post a reply.

Andy’s Tuning Website hasn’t been far from my mind since I first looked at it a couple of weeks ago. There is a lot of good information.

I have produced similar scatter graphs as to those you mention on Andy’s Web Site , However despite reading his site over a few times I’m not sure my brain has absorbed it all so I’m hoping you (Andy ? Matt?) can help with a few more questions that I have.

From my observation s I think I’m a candidate for Blended Alpha-N

QuestionsHow do I determine what the 6x6 Alpha-N table should look like ?
How do I determine the best values for the “Pure AN to blend threshold" and the Blend to SD threshold" ?
Can I assume the following
• “Pure AN to blend threshold" denotes the RPM which anything above will be based on TP only
• “Blend to SD threshold" denotes the RPM which anything below will be SD / MAP only
• The RPM range between will be blended.

For example if

• Pure AN to blend threshold = 3000
• Blend to SD threshold = 1000

Then when RPM is between 1000 and 3000 it’s using the Alpha-N Table values and blending them with the VE table values

When RPM is under 1000 than its using the MAP VE tables only

When RPM is over 3000 than its only using the Alpha-N Table values only

Is this correct?

What about any RPM values in the Alpha-N 6x6 tables that are below the Blend to SD threshold? Are they ignored?

Sorry if the questions are lame, but I’m really struggling with this and the Manual is deficient in explaining anything about Alpha-N / SD blending

Sam