'94 LT1 EEHack analysis using WB

[babywag]

So I have the WB installed and 99.9% confident it's reporting accurately.
I tuned the MAF table previously using NB data and stoich of 14.3.
I changed stoich to 14.1 in the .bin after installing the WB, because it just seemed to run a little lean to me.
So analyzing some information now I come up with the below results.


https://lh3.googleusercontent.com/-Uu22eS7JVKlZd2wsEwRphzECrB6eeR9iRpAlUnjnsQar7oN9Gy tWTMtMR7y2YUzh5lrsRlA7k2R8xB9B8KX9j8K95dFWAfGhnosV _aSgdMLklIr7KKbFnTIscHTHK_mStG1tTrMg0jhitDyxSYtIjt KrlV1_m_BnofrE1l-Nfm8vxI848JdAEyMJ1f8yUZf__vCPpgAJolpjwzdkVQ3p8WA3j XzYDXDcJyowC0XWFJdQw-boZOi91uTyRIDcM2sKwB9ABYy_dqytKGgj0-Wzswe32UDEEnBIIeW8ppZF8vk91X6eTkol_WOla_MgPWmPuDlg _JqvIDoznwjWffybksKVqRpaLhj4kklUGt2KtwJWPdI7T7mgzm _FD_Z992vJwXEVek4UWX5t6q8nrhRDeiav6etLLnEP5AB4VG9z _3IPiGAOjSJ5PW-wSOzE2TVukk8JQEQY0xHY8lv5KpCeyenAXhfyw6vQaAmFG6nk9 hBltnCX8GTBHYJ5PVhZx2olXhnYGaSVbM-htwLxRjpESnrl7AzIdOzpqRN2MX8tCNi-0yWx_yBzLBMtE28oEPTBM0F_hfRhPwAvb7uFgIxJ0WbhjbiiHJ vM5qyUAXfYOtdC66A=w1024-h600-no

Here is the Display Percentage results.

https://lh3.googleusercontent.com/uCx0h7WKhnG_65oEhuAW-ZGJqY8Ni0GErVxOZsaUdUQOoKPrbmf97TBVgp2wsfzcNjOfbjE eKq6qrBRO9ESVfCy_uiIr3dglsx-nX_IelKyjSyoN3lzQElvC5m3Kwqeih8oqDNOW9hudPd9AE_F4Z fbgmkw2sPLIhgJmBJetjxjGORwW9DMsu6iwiBcAZIzi6fUeTdi 7V6pUUsvg92loTllpTMhmw155WcRp4vMXEh0j1H0oOxh8nkTs6 TN2UTfgbPk022_fLyiQGF9Evxj1Kh5x4fjTBVMsuRuvRd0woCg XEuge3op03sBan2KcNf_OEBlsTG3VnmWQQe36fILh9L4bEaJRd VgOIU4xgEJHMjnQXPBmHPqdfmQFoXcVfcy174vKZgtvDviKdCX NrdP1rp2sWW2fIPUBTg97tUxQVgV47zezhCvyuxs_IeFgqGpb7 d8j8LCY5AW9-dI3th0F1kDxJbLa5vPcn-keBzrG7MyYOkQkZFFtPZ0f0RW9sU3oWFxBlhGUfe3GFiVfbG1j X6yW6iSL1tl4nNY2BPhUGithHbjctR0QoSf8lbhUkHMwQdPG6h pn1sDwmP6vlfdynsEZErtjhZREJvWxN3zltBXl-tNX=w1024-h600-no

Engine really responded nicely to the stoich change, overall seems to run better.
Looks like a touch more to the MAF table, and it'll be good.
The gauge is set to display lambda, and hovers around 1.0+/- most of the time under mild driving/idle.
PE seems good until get to ~3000rpm, then it is pretty rich! 11.8-11.2 and this is not a WOT situation.
Any tips on fixing PE? AFAIK this is a completely stock engine.

https://lh3.googleusercontent.com/JBGH8hCbWTttu0P0BlXwDvWCU6-9UYNmAqzwxhMuKrFsAtDB83xYhye8OXKlQHkwCa0DB8aTTNIn9 5oEGoJDjvfTpiywQ7Y91qYNh_u6D0MZ2H7bK0KZN2Thej15iyz 7Sw2jXkzqXJzb2MMYxQ9nOK1JfRnrvu7ScTwigl9j4TNUfeLEH MBDuuPUlnraYE-E6To1Q5qpsbb22rd7fJBnADSmCfuSHPXs_MTL9y1tOg8x_2iDY z56UUQ07176LyJvnbHPvGdaaibzXwzUtuIWG_TzMfQoT_VJGDU lyH_IZBXihMuvxpjiY8kxpi77jTKxDFffjgbQlY4JhWZluePrd I84GTEujekqXdBzKpn6aq7gZ5-pihlVU4dU8We8SC4acLc_BDPFKxkBcx2X1H0O7O_FqPcYQYvij xFsNJF9C_5Cwc4vEjcZ6Al6FdnzPD560RdTi0TNQ8lZL-WiBj87__z6slN6EJQVFWnah14I7rpDxTeW-Eq5Sdrp8InmD5a7O5Z6EW89tUOo5wf1TYSnNnzNBgkU9xL5LVM uEM7N21Ex4U-h4WyX0Ow4cqa_nia8s1_DUkvz0RWdvpSDduI4cgDU5Tw6FzlMZ A-MPHL2rK4LLCT8=w1024-h600-no

[steveo]

looks like your attachments are broken

i'd start removing 3-4% at a time from the pe/rpm table above 4000rpm and see where that puts you...

personally, i'd push PE AFR as close to 13:1 as you can at high rpms, say above 4000rpm

i'd also reccommend installing the wot blm locker patch if you haven't already.

[babywag]

I dunno wth is up with pics? google photos has been an issue lately.
all 3 show on my laptop or android phone.
on my iphone they do not show. Seeing same on other forums as well.
I'll definitely play with PE table, cuz right now it is stupid rich.
Also like to say thanks again for EEHack, it's a great tool.
Even tuning with NB it got it in the ballpark. Now that I have the WB it's an even better tool.
Car runs so much better than when I bought it.

[steveo]

~12:1 isn't stupid rich by any means. ive seen stock lt1s about 11:1 wot.

you could aim for 12.5:1 and be very close to peak power while still playing it safe for unknown fuels...

glad eehack is working for you, your results look very good and you are definitely using the analyzer as intended.

consider though that you have very little high rpm data being analyzed, if you disable pe and do a bunch of high rpm playing around, you might find your pe and non-pe both gain consistency, and your pe targets will function as intended..

[babywag]

Well...looking closer @ log, picking snapshot(s) so to speak.
RPM 4489, TPS 79.2%, TARGET AFR 11.40, WB AFR 11.63, PE active Lambda = .79
RPM 4637, TPS 69.0%, TARGET AFR 11.40, WB AFR 11.06, PE active Lambda = .75
IMHO too much fuel.
I adjusted the MAF table, and I'd like to get some more data again before adjusting the PE table.
It's getting there though.

One small oddity that has been present ever since I bought it, haven't figured out yet.
Sometimes on a warm restart after sitting for a few minutes it goes real lean initially on restart.
Only momentarily less than 60 seconds, and not every restart.
The WB has verified again what I thought/felt was happening. Not sure what/where this problem actually is.

[steveo]

Not sure what/where this problem actually is.

if you get a log of that vs a normal restart, maybe you'll spot something?

[babywag]

Haven't been able to capture hot restart idle oddity yet. It's about 1 out of 50 starts...elusive bugger.
My gut is telling me it may be a wiring issue? I've already had to fix several PO repairs, and some other faulty wiring,

I am curious though, how does EEHack calculates the MAF AFR percentage trims with wideband data?

If getting 14.6 AFR, and BLM's lower than 128, shouldn't I be seeing some 0% or -% trims?

[steveo]

I am curious though, how does EEHack calculates the MAF AFR percentage trims with wideband data?

it uses a patch to insert target AFR into the datstream, then compares that target data to the wideband to arrive at a percentage difference.

this is mostly intended for (and works very well with) open loop tuning.

(it doesn't work with WOT, as there's no good way to put a real WOT AFR target in the datastream)

[babywag]

Ah, so then I need to ignore the %, and adjust manually.
Not sure how report 14.1 output accurately with MTX-L.
Works well using lambda, not so much as anything other than 14.7 AFR
Last time I tried when tuning my 16197427 equipped Jeep, only way I could get any accuracy was changing the .adx and using lambda.

[91ss]

Lot of people get themselves wrapped around the axle with widebands, stoich, etc. And there are too many things to set right that can go wrong if you want something other than 14.7 The general recommendation is to move to Lambda and work soley in that. For me, being old school, I leave everything in straight gasoline at 14.7. Then my brain recognizes stoich, WOT values, etc as they use to be. EVEN if the actual combustion values are different because of the actual fuel. So, you're 99% sure your wideband is correct. 1) Did you reprogram the display for something other than straight gas? 2) What output from it are you using, analog, digital? 3) Does it need to be programmed separately from the display. (only familiar with the LC-1) 4) How does EEHack interpret that input? If you did not change the WB to display correctly for a different fuel, eg. still 14.7 and you 'set' your bin to 14.3 and think it's lean, it would in fact be spot on at 14.7. You wouldn't see 14.3 as that is rich for gasoline.

[babywag]

Yeah the WB is reconfigured. The 2 outputs are analog and also reconfigured.
I have always preferred lambda, it is just easier to work with IMHO.
When software uses AFR it bugs me, but it's just an extra step to deal with to make corrections.

EEHACK confused me is all, because I was seeing 14.6-14.9 AFR
The EEHACK % trims were off from what I thought they should be. Steve confirmed this with the info that the target AFR was used by EEHACK.
Since my WB analog output was configured to report 14.7 as stoich to EEHACK.
However my stoich/target AFR in .bin was set @ 14.1, so the % trims should simply be ignored/not used when I do my adjustments.

Maybe he could add a target AFR override on the analyze window, or maybe it's possible to manually edit the config file? I dunno...

[91ss]

Cool. Curious, how does EEHack accept the analog input? In mustang world, we tend to use the formally used egr input into the EEC for analog or the separate digital output straight into the laptop.

[babywag]

The MTX-L seems to work very well.
I have both outputs in use, stock o2 sensor was removed, and pcm is being fed an analog signal from the MTX-L.
Other output goes to EEHack via pin D27.

Watching the gauge while I drive, jives with what I see in the data logs. So I am confident it is working correctly.
I used it in the past, in the same way to tune my Jeeps running 16197427 pcm's.
Loaned it out to a buddy, and he also had good results using it on his LS swap in a Jeep.
For the $ the MTX-L is a quality piece IMHO.

I would love to get one of the new Innovate dual WB's, maybe some day when I get some disposable hobby/play $?

[91ss]

The MTX-L seems to work very well. I have both outputs in use, stock o2 sensor was removed, and pcm is being fed an analog signal from the MTX-L. Other output goes to EEHack via pin D27. $? Don't have schematics here. What was D27 used for before? It's not recommended in the ford camps to replace the NB with a pseudo NB signal from a WB. In general, it adds a conversion delay and the ford eecs sample with an expected delay as it switches it's LAMBSE's back and forth. If it doesn't detect the mixture yet switch cause of the delay, it continues to swing the LAMBSE further than needed in order to detect a switch happening. Guys are always told to weld in a dedicated bung especially when they seem to be chasing tuning ghosts. I would think the GM PCM's work similar.

[steveo]

Cool. Curious, how does EEHack accept the analog input? In mustang world, we tend to use the formally used egr input into the EEC for analog or the separate digital output straight into the laptop.

the rarely used egr position on pin d27, or the ac pressure sensor input, both are in the stock datastream

unfortunately they aren't great for linearity but they're good nuff with a bit of tweaking

[babywag]

Can't speak on any technical aspects of simulated NB signals. I've read the arguments from guys saying it doesn't work/not recommended.
I've done it before, and compared tunes after the fact with a NB installed and the WB in a dedicated bung.

This was just the easiest method/solution on the car for me.
It isn't a track car, and just a stock engine with 140k on it, that I wanted to run better. Fuel isn't the same as when they were built/tune developed by GM.
It's pretty obvious with the WB installed that they're running on the lean side.
My initial data & calculations showed it was off by ~4% running the stock tune. With a little tuning it runs much better, and I'm happy with the results.
Might not be the "right" way, but it works for me.

[yoheer]

old thread to the top
I need some explanations
Ive got an innovate wb sensor. Just the sensor with no controller. Is there a way to make it work directly? And if yes then is there any manuals how to do it?

[WASyL]

Nope, you need to have WB controller.