WideBand Controlled Stock Type Chip Now Running

[Reggie West]

Any progress on the project?

Just curious.

 

[BoostKillsStres]

Yeah, some good progress . I burnt another chip tonight with a different approach to the code as my last bit of code was making a bit too course adjustments in the fueling but the new stuff is staying w/in 0.2 of my target A/F ratio of 11.5:1 except for right at the top of the gears where its a little leaner. I believe my new code isn't allowing big enough adjustments in the fuel at those points so a bit more fine tuning is in the works but all in all I'm very pleased. Tomorrow nights chip is going to have WB control for WOT and idle and cruise so we'll see how it does.

 

[BoostKillsStres]

Well, wideband controlled open loop idle and curise is now working but needs a little more tuning, nothing like burning the midnight oil doing chip coding on a 3 day weekend

Thought of another code addition to add which will again be really cool.

How about a chip that watches for knock and will automatically add fuel when its detected? It would then continue to add fuel a little at a time until the knock stops then the code would work its way back towards the air fuel ratio you originally commanded while of course continuing to monitor for any further knock and add fuel again if it happened.

I realize sometimes adding no amount of fuel will stop knock but if your running an A/F ratio near the edge this would seem like a cool feature to have working for you.

Back under my coding rock....
Mark

 

[Ormand]

How did you pick the "target" A/F ratio? With a dyno, you might run at a fixed boost, and adjust timing and fuel for best power, no knock. Without tests of some sort, choosing an A/F ratio is educated guess work, you can get close, but won't be sure if you are right.

 

[BoostKillsStres]

Originally posted by Ormand
How did you pick the "target" A/F ratio? With a dyno, you might run at a fixed boost, and adjust timing and fuel for best power, no knock. Without tests of some sort, choosing an A/F ratio is educated guess work, you can get close, but won't be sure if you are right.

Oh I'm doing tests..... I go about it by picking a generally acceptable target A/F ratio such as 11.5 to 11.7 that I've been advised by people running much faster than me and start w/ a safe boost level and a timing level I want to run and adjust from there using the Directscan logs as a reference until I see the fastest time to MPH with no knock. There's no substitute for real world testing, which is what I do someplace with no traffic, I don't care about putting the car on a dyno, numbers are just numbers...e.t. and MPH are proof. Works for me.

 

[BJM]

Any chance we will get to see how this stuff works soon?

 

[BoostKillsStres]

Its a pretty simle scheme really, way back it all started off w/ a BLM lock patch from Carl Ijames which allowed for adjustment of the WOT BLM lock value and MAF point at which to set the lock. Idea comes up, feed the WB signal into the MAT via either home made electronics or Ken's boost harness which I use as I also like to switch off and log other things such as boost and IAT. Once you have that working and have built a reference table of MAT temp to A/F to WB voltage then your ready to write a routine which will read the MAT and adjust the BLM value at WOT based on the changes in MAT(WB) for whatever starting MAF point you choose and refine to your liking.

I gave up on my full time WB setup since my WB is in the cat area and response is slower than I'd like for normal driving situations and I don't really want to chip the coating off my DP to install the sensor closer to the turbo outlet area. Plus the stock closed loop control I've tweaked in my chip runs just fine and is only used for MAF less than 60, everything above that is either under control of a fixed spool BLM value which I may make user adjustable at some point or WB control once currently past MAF=200.

There are of course many ways to skin the cat, this is just one, may not be perfect but it has the ability to adjust the fuel based on the WB output at about 80 times a second and is working well for me so far. New plans will be to add a CPT-66BB turbo, 60 lb injectors and Art Carr 3000-3200 converter to the mix in the near future so it should continue to be fun.

HTH

 

[at7we2]

Originally posted by BoostKillsStres

I have a distinct feeling there's gonna be a whole new set of record times being run by people with stock type chips w/o the need to pay the big bucks and have the steep learning curve for the aftermarket fuel management systems....should be very interesting to see.
Gotta love high technology.

AWSOME!!
So when can I order an off-the-shelf kit which includes a complete WB setup and a new chip!!

Rob

 

[BJM]

Did you wind up updating the MAT at 80 Hz or is it still at 10Hz? Just wondering if it helped get the control loop response time down.

 

[BoostKillsStres]

I do have that MAT read jump right at the top of my code so certainly thanks for that, as long as I have room I certainly plan to keep it in there, I'm at 66 bytes so far. I was able to measure how fast it was running through the loop when testing my knock fuel addition code in the driveway. It will really sock in the fuel if you don't set some limits running that fast.

 

[Ormand]

How about a chip that watches for knock and will automatically add fuel when its detected

That sounds risky. IF you are lean and getting knock, it would probably work. But you can get knock running rich, too, and in that case it would make things worse, before they got better. As I remember the knock vs mixture curve, it has a little plateau, where knock is least likely, and then it gets more likely as you go either richer or leaner than the optimum mixture.

 

[chris]

Any updates?

 

[BoostKillsStres]

All work and no play makes Mark a dull boy and I've been very busy with other stuff than the car.....

However that being said I have done some coding and I plan to get back to some testing and code tweaking this weekend. Thanks to BJM's help and code, I've added a user adjustable A/F setting for the WB control and am in the process of adding user adjustable boost timing for when the alky is running or race gas is added.

The final plan being, throw the alky activation switch on and maybe one other switch and:

1. The alky will have power waiting for boost switch activation.

2. A pressure switch in the engine bay will activate switching over from normal stock WG solenoid control of low boost to a preset high alky boost bleed setup.

3. ECM will now "know" the alky will be running and automatically add in the extra user adjustable timing when in boost.


Should make for a VERY fun setup once its all working....

 

[chrisj]

I am very interested in this topic of stock ecm modification. the problem is that i am a complete newbie but can program pretty good. can you folks share any links to get me started down this path? thanks ,chris

 

[BoostKillsStres]

Chris,
Goto the gnttype.org site if you haven't already, in the ecm and chip area, there are a bunch of chip links there along with sample chip images, decoded and commented stock chip text and a spreadsheet listing all the table information and more. From there you can purchase a fast scantool such as Directscan which gets you all the info you'll ever need and purchase a programmer, chips and programming software which you can get all from BM computers on this board then you can have at it.

HTH

So far the chip is working really well. Haven't had much drive time in the car lately but did bolt on a new turbo in my sig and have other parts to go on one by one(converter, heads, injectors and cam and intake at some point. The initial boost setting of the new turbo was a little more than I expected and the chip compensated automatically with 11% additional fuel and only had 3.2 knock when it hit 21 psi in 1st...oops...so its working nicely and I've since soed up the fuel loop a little so it can add fuel even faster if running extra boost or if the PE vs. RPM tables are way off target

I've added 4 user programmable functions in the chip: Wideband Target Air Fuel Ratio, Extra Timing for Alky, Spool Fuel and Shift Light along with 2 extra degrees of timing in 1st and 2nd gear. I've got more testing to do to make sure the timing is working correctly but I'm about ready to call it done and just add more parts, drive it like I stole it and tune.

 

[at7we2]

keep posting!

that is truly awsome!

I really don't know diddly-squat about chip programming and all that, but I do my best to follow some of the threads about it and what you're doing sounds really cool! Keep posting your progress and results.

thanks!

 

[BoostKillsStres]

New Idea for an upgrade to my wideband controlled chip.

How about this:
When running a chip which is controlling WOT fuel with a wideband it will "learn" so that you can then flip a switch in the car while driving which will then cause the chip to run off of what the wideband has "learned" for your fuel requirements, which is based on whatever air fuel ratio you have programmed into the user settings of the chip which you want the car to run.

This would allow a user to tune the car to street boost levels or race fuel and race boost levels and if you feel you've reached the performance level for the octane, boost and timing your running you can then run at very close to those same performance levels and remove the wideband completly to use on another car or maybe just remove the sensor to prevent it from lead fouling if running race gas.

The "learned" fuel value would of course also be user adjustable so you could continue to tune the WOT fuel for the car from inside the car. This way you wouldn't run any risk of a wideband mishap or failure causing a fuel "issue" with the car.

Thinking out loud.

 

[BJM]

I thought about making my boost controller learn for quite a while. The stumbling block was not enough RAM and then not enough space in the chip. Ultimately I got a version of the controller working so that it became way less sensitive to weather changes. So I went away from a learning routine. My controller is working pretty well if I do so myself. I turn it off sometimes when I forget how bad my spike used to be and scare myself.

For your WideBand idea, would your learning routine require multiple RAM locations or are you just going to use the PE-RPM table and offset it by a learned amount. I guess you need a target and a learned value to store. Am I close?

 

[BoostKillsStres]

Hey Man, glad to hear that boost control is doing so well for you, thats cool stuff.

I wish we had move VMRAM to work with, just not enough there to play with unless you go full bore like a ME chip. I have one cell for the user target A/F value and am thinking on using another cell for storage of the learned value whether it be calculated or a peak type fuel value or something. That's why I said "run at very close to those same performance levels" as the full time WOT WB correction in my above post because when the wideband isn't under control your again at the mercy of the accuracy of the PE RPM tables and even when there perfect, which mine are very close, the wideband is still making minor corrections since you have 3 or 4 gears to run thru depending on how fast your running or plan to run and we know the cars will utimately run best with individual fuel needs corrected for in each gear which is what running on the WB at WOT gives me.

Running off a single learned value would be similar to running off a locked BLM value but this value would be a true learned A/F value and would be abjustable so you could fine tune from there. There would be a requirement, I'm thinking, of the code that you would have to run the car well into 3rd gear for this value to "learn" a valid fuel correction since fuel amounts change between gear shifts and a good 3rd gear pull would give a very good "learned" value to use thru all WOT driving as a base.

 

[turbo2nr]

I'm not buying any thing else, untill this awsome project is released to the public! You must continue, test, produce, and sell!