WideBand Controlled Stock Type Chip Now Running

[BoostKillsStres]

I've been working on a code patch and doing a lot of assembly learning lately and with help from a couple chip gurus...THANKS GUYS, YOU KNOW WHO YOU ARE... and a lot of coding on my own, I took my car out last night for the first time where WOT fuel is being controlled solely by feedback from the WB sensor and making realtime adjustments.

So far the results look great with a very flat fuel curve target achievement where as I hit WOT, fuel is run off the chip PE tables until about MAF>230 range where the WB adjustments take over and basically correct all fueling on a percentage basis just as a FAST system would to do to maintain a target A/F ratio.

So, I just wanted to let everyone know with the new WB technology out there, I'm sure in the not too distant future you'll see chip vendors selling chips for WB control with the various WB units that are available.

One of the really cool parts of this is I'm running a standard translator so I, like so many others, have the MAF=255 peg problem.....well not any more now no matter how much air I flow over 255 g/s, the new code will continue to adjust for the target A/F I command.....throw heads, turbo, IC whatever on and as long as there is injector pw available and the spool up fuel and initial PE fuel is close in the chip then I'm good to go

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.

 

[Slow91z]

That is AWESOME....Great work, I can't wait to get my hands on something like that.

 

[BJM]

Sounds good. Is the wideband running in place of the stock O2 or in addition to it?

Do you have a fixed target A/F or do you actually have a curve in a new table?

Need a boost controller with that?

 

[BoostKillsStres]

I have it running in addition to the stock o2 currently, on my car I have the closed loop stuff running so well its just easier to leave it for the moment. The patch starts executing when the MAF passes 200 currently but I may raise that a little however on a not quite full throttle run, the MAF will dip to around 210 on the 1-2 shift so I wanted to make sure that range stayed under WB control. Target A/F is currently fixed to right at 11.5-11.6:1

I had the patch also correcting for MAF<60 but was having some issues with it overcorrecting when I blipped the throttle and such. Between 60 and 200 MAF I let the chip retain control for consistant spool fuel, didn't see a need to futz with that right now either although I have been dreaming of a progressive WB target curve through that range. With that and some limits in place for the low speed stuff, along with running the chip in full time open loop WB control, the o2 sensor could be removed.

BJM, I didn't look at your code in super detail so I wanted to ask if the boost limit in your code is a hard limit? In other words if you put extra tension in the WG rod will the patch try and correct for that? Just wondering how adjustable it is.

TIA

 

[BJM]

Correct me if I am wrong but you are not really in open loop, you are just in a different control loop where the wide band is in control. Am I reading you right?

My boost controller right now is purely a derivative loop, it only damps the response of the waste gate. If you raise your boost with the rod, the boost goes up accordingly. If you go from 15-20 psi the damping gain probably would need to change. There is also a threshold boost level below which the loop is held inactive to prevent slowing down spool up. Two parameters to play with.

 

[BJM]

How are you getting the wide band input to the ECM? Through the MAT sensor or through a modified ECM.

 

[BoostKillsStres]

BJM, thanks for the info, might have to try that at some point.

Yes, input through the MAT w/ Ken's boost harness.
When I had the WB controlled low speed stuff in I had the chip forced to open loop all the time so the BLM map was always 128's so the ecm couldn't futz with the changes the WB was making to fueling. So currently back in stock closed loop it runs normal until about 60g/s then the BLM locks to 128 for consistant spoolup and initial PE then past 200 the WB takes over full control but it is using the current PE RPM fuel table as its 'BASIS' to correct from. Then if I want to get real picky I can take the directscan log and see the changes that the WB made and fine tune the PE RPM values so the WB % correction values are less......kinda just like an aftermarket management system where your adjusting the VE tables so that the WB doesn't have to correct as much if your out of tune in a particular range.

HTH

 

[BJM]

It sounds really neat, will any of this be open to the public as it were? I would love to see some.

 

[BoostKillsStres]

I really haven't decided yet. I enjoy giving back to the TR community when I can since I've gotten so much from it but after putting in 3 solid weeks of learning, coding, testing and late nights and a bunch of emails and while I'm thrilled its running I'm feeling a little tight with it at the moment with it if you can understand that. Plus I need to do more driving with it to make sure its doing what its told well enough.

But there are some drawbacks as once you go to this type of control, fuel pressure adjustments aren't going do anything so if you wanted to try a little leaner or richer you'd have to have the chip adjusted unless one of the chip guru's comes up with an external input of some type to somwhow adjust the target A/F value...that would be cool...hmmmm. So, you'd really only be able to adjust boost and maybe timing if you had a xlator+ and I'm not sure if it would even work w/ a xlator+. Plus I imagine the code may run a little different depending on which WB unit is being run and everyones car seems to like a little different A/F ratio so its not going to be a cookie cutter application from what I've seen and heard talking with another chip maker working on this type of project.

Plus you can't run alky with it per say as w/ alky the A/F will drop off into the very low 10's usually and sometimes the WB unit will just report RICH so its either not going to have control or you don't want it to have control. But with Directscan you could use it to work with your chip vendor and get your chip dead on then run the alky w/o the WB control and you should be about dead on w/ additional timing in the alky chip at that point.

Another thing is w/ this I can't switch to logging boost or intake air temp thru the mat harness unless I change to another chip sure wish we had 2 more mat type inputs we could plug stuff into then we could log WB, boost and move the mat sensor to the upper intake like I have and log charged air temp too, that would be too cool....oh well.
We'll see.

 

[ijames]

Mark, the reason I switched to a maf threshold for locking the blm in the first place was so that the spoolup would be more consistent. Let the ecm learn the fuel pressure change and it does a good job reproducing the afr during spoolup, then lock the blm so fuel pressure only changes the wot-already-spooled-up fueling. Otherwise you wind up trying to tune both the spoolup and wot fueling simultaneously with just one thing to change, the fuel pressure. (That's what I didn't like on the Thrashers, the open loop fueling at idle, highway cruise, and at wot - he had to have the "steps" correct in the chip and you had to run the boost he expected so the wot fueling matched.)

 

[BoostKillsStres]

Thanks Carl, I really appreciate your point, its just that I'm currently stuck with the mindset that locking to 128 while spooling will give me the most consistant fueling and as long as my spool fuel is correct I should be fine unless I do something like turn the boost up a bunch at which point I may need a little more fuel. I've seen cell 15 swing from level ground 128 to 138 and to 118 when driving hills on the highway and I don't like the thoughts of it possibly being that far off the next time I stand in the gas for spool fuel. And if your cell 15 likes to learn a decent amount away from 128 and you decide to change chips and reset the ecm and it gets reset to 128 then there goes your spool fuel calcs again until it relearns which at the track wouldn't be nice. But those are just my thoughts, you know more about this stuff than I do.

 

[bruce]

WOW.

And I don't say that often.

CONGRATS ! !!

That's really impressive.

I for one would like to see it Pulbic Domained. It's thru Public Domaing that the effort moves forward. If you don't that's totally understandable, and I can really see your point. But, if everyone just sits on what they have, then nothing pushs the system foreward. Since everyone is always having to start at square one rather then building on what others have done. I know the grief of seeing a project pirated, and some clown making a few bucks off the work of others, but, it was worth it in the long term.
It was truely worth while, dispite the piracy.

 

[3.8TransAM]

Sounds fantastic !

I certainly hope this sees the lite off day . I am currently having my TTA done by ANS and hoping for bottom 11's on pump gas when all is said and done.. (too bad not in time for the NAts, Im gonna cry now). DIY Wideband assembly in progress.

Im just beginning to manipulate actual code to test/change issues that arent available in your common editor programs and going thru the whole hex/decimal curve when u get inot the actual source code for changes... Definitely frustrating yet rewarding once u figure somehting out and a way to make it work!

Hopefully one day I'll be the next guy going hey check out this patch i made to run your Mr. Fusion closed loop at WOT and open loop at idle

Thru my programming/tuning friends I've met thru the net, Ive seen some truly neat stuff and been privy to some fo it before anyone else . (Hows the 749 $58stuff going Bruce? ) So much of this wouldnt get done if the folks who are versed in assembly and programming didnt contribute or help others like myself who struggle to change the already existing source code ..

Awesome work and hope to see some more
thanks
Jeremy

 

[87GN_70GS]

try a little leaner or richer you'd have to have the chip adjusted unless one of the chip guru's comes up with an external input of some type to somwhow adjust the target A/F value...that would be cool...hmmmm

At first blush it seems like simple hardware. Intercept the WB02 sensor voltage, incorporate a high-impedance op-amp summer that adds or subtracts a driver-adjustable bias (very stable reference voltage, using a pot - one o' them lock down vernier dial types). Unless I'm not considering something important....

 

[bruce]

Originally posted by BoostKillsStres

Another thing is w/ this I can't switch to logging boost or intake air temp thru the mat harness unless I change to another chip sure wish we had 2 more mat type inputs we could plug stuff into then we could log WB, boost and move the mat sensor to the upper intake like I have and log charged air temp too, that would be too cool....oh well.
We'll see.

Actually you can.
By switching to a Lockers Dataloging system. It's DIY, and is an edge card system like DS, but it gives you all 256 RAM locations, and room for 5 additional inputs.

If you were to get an ecm modified by Bailey for the MAFless ME system then you gain an extra A/D channel. Then use that for the WB. Actually there's also at least one more A/D channel availale even using that one.

FWIW..

 

[dennisL]

Bruce, where can I find out more about Lockers Dataloging system ?

Also, do you happen to have the data sheets / specs for the ADU chip used in the 148 ecm? Your mention of an unused A/D channel reminded me that not all the input channels are being used. I would like to research this because, like many others, I have a good use for another input.

 

[BJM]

Can I assume you are now calling the MAT input within your patch to bring the speed up to 80 updates per second as opposed to the normal 10 times? It would be scary running your engine with infrequent updates.

 

[bruce]

Originally posted by dennisL
Bruce, where can I find out more about Lockers Dataloging system ?

Also, do you happen to have the data sheets / specs for the ADU chip used in the 148 ecm? Your mention of an unused A/D channel reminded me that not all the input channels are being used. I would like to research this because, like many others, I have a good use for another input.

I just found out there aren't any more lockers boards, but searching at DIY-EFI should get you the info about it. Maybe some more boards could be made.

Searching thru here, will give you the schematics of the 148, if I'm not mistaken.

http://www.cruzers.com/~ludis/

 

[Reggie West]

Congratulations on bridging the gap between a " factory " ECM and the aftermarket units.

If all can agree on a plan here I think it would be really neat to see an ECM that can be tuned using the WB patch as you describe here.

I am not up to speed on what the Extenders and Translators can do but if you can have the ability to add or subtract fuel at lower boost settings for launches on the strip and utilize the WB sensor to monitor or make changes to the map on the fly you will really have something and people will buy it.

If you can make a box that can give me feedback on AF ratios, a data logger and access to the fuel map to change values you will have a fire sale on in the Buick world no one has seen in years.

Good luck

 

[bruce]

Originally posted by Reggie West
Congratulations on bridging the gap between a " factory " ECM and the aftermarket units.

It's actually the other way around.

The aftermarkets are still trying to catch up with the oem ecms. The oems have millions in R+D, and documented reliability reports, and the aftermarkets with each generation get more and more complex to get the manners, and strategies of the oems. The WB option on the aftermarkets are more a marketing ploy then anything, IMO. You still have to pick what AFR they command, and that's where the talent of tuning comes in. Yes, there is an ocassion when something goes wrong and the ecm can make a correction to try and deal with it, but in the same sense the WB can fail and generate a problem.

While designed so that if they fail a WB should indicate lean and that would force a rcih correction which is realitively safe, I have one here that failed rich, and that would force a lean correction.

Yes, this is a great tuning tool, but like all serious tuning you use as MUCH info as possible to develope a tune, is read the plugs, performance measurements, ETG, etc..

The biggest advantage here, IMO, is that with the stock knock sensor, you have the added security of somewhat good fail safe mode. ie In a failure, the K/S will trigger some retard to the timing. And the GM K/S tuning is very good..