Using Innovate LC-1/LC-2 to Simulate NBo2
09-28-2016, 07:02 PM
#1
Senior Member
Thread Starter
Using Innovate LC-1/LC-2 to Simulate NBo2
Just got off the phone with Innovate MotorSports support. For both the LC-1 and LC-2, they do not recommend and in fact firmly discourage using either of the 2 analog outputs to simulate and replace the NBo2 sensor signal. Direct quote: "Their controllers were never intended to simulate an OEM NBo2 sensor". Their reason is that despite being able to change the LC controller's "Response Speed" in Log Works, a timing mismatch may exist between the LC controller and the ECM and result in damage the ECM. Went at it several different ways and got the same answer. Not sure about the ECM damage, but the timing mismatch makes sense (see below).
So would like to start a discussion on this subject with some questions for anyone using a LC-1 or LC-2 in place of the NBo2:
1. Are you successfully using an analog simulated NBo2 signal?
2. Which analog output is being used (0-5v reprogrammed, or the 0.1v to 1.1v reprogrammed or untouched)?
3. What Response Speed setting is being used in Log Works (Instant(LC-1 default?), 1/12(LC-2 default?), 1/6 or 1/3)?
"Instant" does not appear to be a Response Speed choice on the LC-2 (??? is this correct). So at the 1/12 setting, the LC-2 controller would be providing one analog o2 sensor sample every 83ms. The $8d code running in the '7730/'7727 ECM reads the NBo2 every 12.5ms. That indicates there is indeed a timing mismatch between the controllers and the ECM, with the LC-1 controller being 3 times slower and the LC-2 being 6.6 times slower. Said another way with the 1/12 setting and the LC-2, only one of approximately 7 sensor reads will provide a new value. Either would seem to have the potential of greatly affecting the frequency of INT & BLM updates and BPW correction frequencies and amounts.
If looking at this correctly, it doesn't seem things would work as intended due to the o2 data lag.
Thoughts/comments?
PS ---
Found the attachment on Speed Talk listing WB controller/sensor combination response times. Very interesting in that the LC-1 controller response time was (Avg 37.1ms [likely Instant???]) which is significantly faster than the LC-2 (85.1ms = approx. the 1/12 setting of 83ms above).
So would like to start a discussion on this subject with some questions for anyone using a LC-1 or LC-2 in place of the NBo2:
1. Are you successfully using an analog simulated NBo2 signal?
2. Which analog output is being used (0-5v reprogrammed, or the 0.1v to 1.1v reprogrammed or untouched)?
3. What Response Speed setting is being used in Log Works (Instant(LC-1 default?), 1/12(LC-2 default?), 1/6 or 1/3)?
"Instant" does not appear to be a Response Speed choice on the LC-2 (??? is this correct). So at the 1/12 setting, the LC-2 controller would be providing one analog o2 sensor sample every 83ms. The $8d code running in the '7730/'7727 ECM reads the NBo2 every 12.5ms. That indicates there is indeed a timing mismatch between the controllers and the ECM, with the LC-1 controller being 3 times slower and the LC-2 being 6.6 times slower. Said another way with the 1/12 setting and the LC-2, only one of approximately 7 sensor reads will provide a new value. Either would seem to have the potential of greatly affecting the frequency of INT & BLM updates and BPW correction frequencies and amounts.
If looking at this correctly, it doesn't seem things would work as intended due to the o2 data lag.
Thoughts/comments?
PS ---
Found the attachment on Speed Talk listing WB controller/sensor combination response times. Very interesting in that the LC-1 controller response time was (Avg 37.1ms [likely Instant???]) which is significantly faster than the LC-2 (85.1ms = approx. the 1/12 setting of 83ms above).
09-29-2016, 11:40 PM
#2
Member
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
I used an LC1 to simulate the narrowband on a 7730 running SAUJP4 for years and it worked great, I fed analog output 1 into the narrowband pin on the ECU the default for that output is 0.1v to 1.1v and i that's what I ran. I never messed with the response speed, just left it at the default settings.
09-30-2016, 11:22 PM
#3
Junior Member
Join Date: Apr 2015
Location: Minnesota
Posts: 30
Likes: 0
Received 3 Likes
on
2 Posts
Car: 1989 GTA
Engine: 5.7 TPI
Transmission: 700R4
Axle/Gears: 3.27 (Stock)
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Although it's a bit different of a setup, I am using the Analog 2 out on my Innovate MTX L. The 0.1v to 1.1v default, at the default settings. 7165 seems like like it so far, but I only have maybe a week of driving and a bit of data logging so far. Still need to go in and fine tune after installing the FIRST TPI.
10-01-2016, 01:50 AM
#4
Senior Member
Join Date: Aug 2002
Location: ARIZONA
Posts: 948
Likes: 0
Received 4 Likes
on
4 Posts
Car: 92 Trans Am Conv
Engine: LB9
Transmission: T5
Axle/Gears: 3.08
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
That difference and the fact that it looks for a swing probably won't have too much effect or noticeable effect.
10-16-2016, 09:37 PM
#5
Senior Member
Thread Starter
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Thanks to those who have replied.
There are some who say their NBo2 simulation is working OK, but there are many who have not posted here who apparently have not had that success as indicated in other posts and forum sites. That's why I wanted to start a conversation. Perhaps I should have asked for unsuccessful uses and symptoms.
Regarding the effect: BPW calculation and correction occurs 80 times a second (every 12.5ms), with the corrections based on a current o2 sensor voltage .vs. calibration boundaries and swing points. Getting new o2 voltage 12 times a second (every 83ms) with which correction decisions are made, versus 80x per sec. or nearly 7 times slower, would seem to have a significant effect.
Sorry to get down in the weeds, but things that would appear to be affected just to name a few would be:
- BPW will still be calculated and corrected 80x per sec., but the same unchanged BPW correction values will be applied to a changing BPW approximately 7 times before being changed and won't reflect current o2 sensor voltage during that time.
- The frequency of INT +/- and BLM update will be similarly delayed because they are dependent upon o2 voltage errors that will occur 7 times less frequently.
Maybe I'm over-thinking this, but GM surely used the 80x per sec frequency to have everything occur concurrently for a reason.
Regarding the effect: BPW calculation and correction occurs 80 times a second (every 12.5ms), with the corrections based on a current o2 sensor voltage .vs. calibration boundaries and swing points. Getting new o2 voltage 12 times a second (every 83ms) with which correction decisions are made, versus 80x per sec. or nearly 7 times slower, would seem to have a significant effect.
Sorry to get down in the weeds, but things that would appear to be affected just to name a few would be:
- BPW will still be calculated and corrected 80x per sec., but the same unchanged BPW correction values will be applied to a changing BPW approximately 7 times before being changed and won't reflect current o2 sensor voltage during that time.
- The frequency of INT +/- and BLM update will be similarly delayed because they are dependent upon o2 voltage errors that will occur 7 times less frequently.
Maybe I'm over-thinking this, but GM surely used the 80x per sec frequency to have everything occur concurrently for a reason.
10-17-2016, 06:45 AM
#6
Member
iTrader: (1)
Join Date: Jan 2009
Location: Alabama
Posts: 237
Likes: 0
Received 1 Like
on
1 Post
Car: 87 Firebird, 89 Trans Am
Engine: 406, LB9
Transmission: T56, 700R4
Axle/Gears: 3.23 Torsen, 2.73 Posi
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
I'm using an AEM UEGO and not an Innovate gauge but I will say that I have not noticed any issues as I tune my 89. AEM doesn't give any details that I'm aware of on how often they update their analog output.
11-05-2016, 05:53 PM
#7
Senior Member
Join Date: Sep 2001
Location: Cathlamet, Washington
Posts: 564
Likes: 0
Received 2 Likes
on
1 Post
Car: 87 Formula
Engine: 327
Transmission: 700r4
Axle/Gears: 3.23
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
I had been using the narrow band simulation to feed the ecm on my jeep. (4.0 supercharged setup) After reading this post I put the stock sensor back in and just use the Innovate to monitor afr. Big difference in the way the jeep runs. Has smoothed out a bunch and mileage has gone up a bit.
No idea how gm computers will take the simulated narrowband but there's definitely a difference.
No idea how gm computers will take the simulated narrowband but there's definitely a difference.
Trending Topics
11-06-2016, 05:54 PM
#8
Senior Member
Thread Starter
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
I had been using the narrow band simulation to feed the ecm on my jeep. (4.0 supercharged setup) After reading this post I put the stock sensor back in and just use the Innovate to monitor afr. Big difference in the way the jeep runs. Has smoothed out a bunch and mileage has gone up a bit.
No idea how gm computers will take the simulated narrowband but there's definitely a difference.
No idea how gm computers will take the simulated narrowband but there's definitely a difference.
Thanks for responding.
05-21-2018, 11:24 PM
#9
Supreme Member
Join Date: Jul 1999
Location: Alberta, Canada
Posts: 1,453
Likes: 0
Received 57 Likes
on
31 Posts
Car: 1989 Camaro-1LE
Engine: TPI(s)
Transmission: 5 speed (MM5, MK6)
Axle/Gears: 3.45, 3.73
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Interesting read.
I've just added an LC-1 onto my setup for tuning. Swapped out the Narrow band for the Wideband, and setup Analog 1 to be narrow band.
Now the results in TunerproRT data logging. I setup channel 1 in Autoprom to record the data. After a flub-up on calculations, I have what I think is the right formula for Wideband (X * 0.0146875) + 7.350000.
Here is what I see.
The Analog 1 output is very fast. The O2 swing is very short. Changes in O2 go from 222mv to 359mv, and then back down. The stock O2 would swing from 220 up to 700mv sort of swing.
Here is the issue. The analog out sits in the 200-350 range bouncing back and forth. The computer "thinks" it's ok due to the O2 swing. But the Wide bad says 15.6 AFR. And the BLM will settle in around 128.
I've adjust the MAF tables ($6E) to try and bring the Wideband value down but no luck.
I was going to try and slow the output down to see if I could get a larger swing in the O2.
In short ... I'm stuck at 15.2-16.2 AFR at idle. If I try to richen it up, then the BLM moves.
I may have to mount the Wideband in a separate location and keep the stock O2 in place.
Mark.
I've just added an LC-1 onto my setup for tuning. Swapped out the Narrow band for the Wideband, and setup Analog 1 to be narrow band.
Now the results in TunerproRT data logging. I setup channel 1 in Autoprom to record the data. After a flub-up on calculations, I have what I think is the right formula for Wideband (X * 0.0146875) + 7.350000.
Here is what I see.
The Analog 1 output is very fast. The O2 swing is very short. Changes in O2 go from 222mv to 359mv, and then back down. The stock O2 would swing from 220 up to 700mv sort of swing.
Here is the issue. The analog out sits in the 200-350 range bouncing back and forth. The computer "thinks" it's ok due to the O2 swing. But the Wide bad says 15.6 AFR. And the BLM will settle in around 128.
I've adjust the MAF tables ($6E) to try and bring the Wideband value down but no luck.
I was going to try and slow the output down to see if I could get a larger swing in the O2.
In short ... I'm stuck at 15.2-16.2 AFR at idle. If I try to richen it up, then the BLM moves.
I may have to mount the Wideband in a separate location and keep the stock O2 in place.
Mark.
05-22-2018, 06:16 AM
#10
Moderator
iTrader: (1)
Join Date: Mar 2002
Location: Chasing Electrons
Posts: 18,400
Likes: 0
Received 215 Likes
on
201 Posts
Car: check
Engine: check
Transmission: check
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Use the O2 sensor window tables in the tune to enrich the AFR. This is what the ECM targets for O2 swing.
Can also change the LC1 analog NB output swing.
RBob.
Can also change the LC1 analog NB output swing.
RBob.
05-22-2018, 10:21 PM
#11
Supreme Member
Join Date: Jul 1999
Location: Alberta, Canada
Posts: 1,453
Likes: 0
Received 57 Likes
on
31 Posts
Car: 1989 Camaro-1LE
Engine: TPI(s)
Transmission: 5 speed (MM5, MK6)
Axle/Gears: 3.45, 3.73
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
I worked with the LC1 and tried to simulate the Narrow band O2. If you adjust the sampling rate down to 1/6th of a second it is better. It still runs too lean at idle ... yet the BLM goes to 128. For a trick, I fixed the BLM at 128 and then adjusted the MAF tables until the wide band came to 14.5 ish. idle was much better. I can't seem to do this when the BLM is active. I need to find where to set the "desired" Air fuel ratio.
Mark.
PS. I'm working on a pair of Superams.
05-23-2018, 09:10 AM
#12
Moderator
iTrader: (1)
Join Date: Mar 2002
Location: Chasing Electrons
Posts: 18,400
Likes: 0
Received 215 Likes
on
201 Posts
Car: check
Engine: check
Transmission: check
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
These three tables. Note that some masks have parameters for when in idle. But I didn't see any for $6E:
RBob.
Code:
;----------------------------------------------
; UPPER ZERO ERROR REF FOR SLOW o2 R/L
; ARAP
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C459 ; mvdc Air Flow g/sec
;----------------------------------
LC459 FCB 140 ; 619 0
LC45A FCB 148 ; 655 8
LC45B FCB 152 ; 673 16
LC45C FCB 152 ; 673 24
LC45D FCB 148 ; 655 32
LC45E FCB 144 ; 637 40
LC45F FCB 132 ; 584 48
LC460 FCB 128 ; 566 56
LC461 FCB 125 ; 553 64
;----------------------------------------------
; LOWER ZERO ERROR REF FOR SLOW o2 R/L
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C462 ; mvdc Air Flow
;----------------------------------
LC462 FCB 120 ; 531 0
LC463 FCB 128 ; 566 8
LC464 FCB 132 ; 584 16
LC465 FCB 132 ; 584 24
LC466 FCB 128 ; 566 32
LC467 FCB 124 ; 549 40
LC468 FCB 112 ; 496 48
LC469 FCB 109 ; 482 56
LC46A FCB 106 ; 469 64
;----------------------------------------------
; Fast o2 Rich/lean Treshold vs Air Flow
;
; FAST o2 RICH IF LC4AD + LC453
; FAST o2 LEAN IF LC4AD - LC453
;
; (SUB OFF LC454 IF AIR DIVERT)
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C46B ; mvdc Air Flow g/sec
;----------------------------------
LC46B FCB 130 ; 575 0
LC46C FCB 138 ; 611 8
LC46D FCB 142 ; 628 16
LC46E FCB 142 ; 628 24
LC46F FCB 138 ; 611 32
LC470 FCB 134 ; 593 40
LC471 FCB 122 ; 540 48
LC472 FCB 120 ; 531 56
LC473 FCB 118 ; 522 64
05-23-2018, 07:02 PM
#13
Supreme Member
Join Date: Jul 1999
Location: Alberta, Canada
Posts: 1,453
Likes: 0
Received 57 Likes
on
31 Posts
Car: 1989 Camaro-1LE
Engine: TPI(s)
Transmission: 5 speed (MM5, MK6)
Axle/Gears: 3.45, 3.73
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Ok so loosely speaking, the mv reference is equated to air fuel ratio on the narrow band then? So I want to have a richer mixture, I need a higher O2 value which then calculates out to a lower AFR number?
These three tables. Note that some masks have parameters for when in idle. But I didn't see any for $6E:
RBob.
Code:
;----------------------------------------------
; UPPER ZERO ERROR REF FOR SLOW o2 R/L
; ARAP
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C459 ; mvdc Air Flow g/sec
;----------------------------------
LC459 FCB 140 ; 619 0
LC45A FCB 148 ; 655 8
LC45B FCB 152 ; 673 16
LC45C FCB 152 ; 673 24
LC45D FCB 148 ; 655 32
LC45E FCB 144 ; 637 40
LC45F FCB 132 ; 584 48
LC460 FCB 128 ; 566 56
LC461 FCB 125 ; 553 64
;----------------------------------------------
; LOWER ZERO ERROR REF FOR SLOW o2 R/L
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C462 ; mvdc Air Flow
;----------------------------------
LC462 FCB 120 ; 531 0
LC463 FCB 128 ; 566 8
LC464 FCB 132 ; 584 16
LC465 FCB 132 ; 584 24
LC466 FCB 128 ; 566 32
LC467 FCB 124 ; 549 40
LC468 FCB 112 ; 496 48
LC469 FCB 109 ; 482 56
LC46A FCB 106 ; 469 64
;----------------------------------------------
; Fast o2 Rich/lean Treshold vs Air Flow
;
; FAST o2 RICH IF LC4AD + LC453
; FAST o2 LEAN IF LC4AD - LC453
;
; (SUB OFF LC454 IF AIR DIVERT)
;
; 02-20-1997 Dissassemby of ARAP Lines= 9
;
; TBL = .226 * mvdc
;----------------------------------------------
ORG $C46B ; mvdc Air Flow g/sec
;----------------------------------
LC46B FCB 130 ; 575 0
LC46C FCB 138 ; 611 8
LC46D FCB 142 ; 628 16
LC46E FCB 142 ; 628 24
LC46F FCB 138 ; 611 32
LC470 FCB 134 ; 593 40
LC471 FCB 122 ; 540 48
LC472 FCB 120 ; 531 56
LC473 FCB 118 ; 522 64
05-24-2018, 03:18 PM
#14
Moderator
iTrader: (1)
Join Date: Mar 2002
Location: Chasing Electrons
Posts: 18,400
Likes: 0
Received 215 Likes
on
201 Posts
Car: check
Engine: check
Transmission: check
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
RBob.
05-24-2018, 09:16 PM
#15
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Those tables work in $8D. Though it seemed that I had to move them quite a bit more than I would have thought to get the resultant AFR that I wanted.
The old Grumpy adage definitely applied... give the engine what it wants.
for example, I have the high MAP values I to the 900 mV range with the upper tolerance literally over 1V just to get 13.5:1 AFR for moderate (non-WOT) acceleration.
The old Grumpy adage definitely applied... give the engine what it wants.
for example, I have the high MAP values I to the 900 mV range with the upper tolerance literally over 1V just to get 13.5:1 AFR for moderate (non-WOT) acceleration.
05-26-2018, 11:52 PM
#16
Supreme Member
Join Date: Jul 1999
Location: Alberta, Canada
Posts: 1,453
Likes: 0
Received 57 Likes
on
31 Posts
Car: 1989 Camaro-1LE
Engine: TPI(s)
Transmission: 5 speed (MM5, MK6)
Axle/Gears: 3.45, 3.73
Re: Using Innovate LC-1/LC-2 to Simulate NBo2
Well I had a good run on Friday. Pulled hard up to 5300 (which is where I had it set to shift with the old setup). I had the MAF values a bit too high and after 3500 rpm it dropped to 9:1 AFR. So need to adjust the tables to what the engine wants. Plus I think I screwed up a few scalars AND tunerpro definition might have had some upper limits set. So all that is fixed and the tables look linear once again.
It rained ... so my testing was cut short. Part throttle drivability is improved. Feels crisp. Have some mid range tuning to look at.
Spark is good ... no knock at all. My son runs the same setup with iron heads, and he always gets some knock in wide open throttle tests.
The LC-1 wide is nifty. I had to upgrade the firmware to the latest I could find. It kept dropping connection which looked like O2 swing to rich. The 1.20 firmware is much improved for the connection.
Mark.
It rained ... so my testing was cut short. Part throttle drivability is improved. Feels crisp. Have some mid range tuning to look at.
Spark is good ... no knock at all. My son runs the same setup with iron heads, and he always gets some knock in wide open throttle tests.
The LC-1 wide is nifty. I had to upgrade the firmware to the latest I could find. It kept dropping connection which looked like O2 swing to rich. The 1.20 firmware is much improved for the connection.
Mark.