HarleyTechTalk

I like the LM-2's from Innovate.  That doesn't mean I agree with everything they say.  I have talked to them many times, and it always comes down to.  I am working with motorcycle exhaust and they are working with cars exhaust.  I like a lot of products.  That doesn't mean that I agree with those manufacture 100% either.  Kury is a great example.  For me.  That catalog might as well be only about 4-5 pages long.

Now lets try and get some context with what you have posted.  How much does it matter that the sensor is sticking into the stream.  Does the exhaust have a continuous positive pressure?  Up stream of a turbo?  Up stream of a cat?  Or does the pressure bounce from positive to negative while running?  In a 4 inch long 2 in diameter drag pipe on a Harley.  Application matters.  Like I told you before.  Everything that I can find and read.  Tells me that what Glen said is correct.  I was asking you to show me doc of how and why he is wrong.  If he is wrong.  That means I am wrong, and If I am wrong.  I am more concerned with getting the right information than me being right.  If you prove it or somebody else does.  Doesn't matter to me.  It is a what is right, over who is.

They also sell this.  I can see it working in a turbo car but not in 95% of the application that I see.

https://www.innovatemotorsports.com/xcart/product.php?productid=16148&cat=271&page=1

BTW  I run LM-2's on my truck with Kooks long tunes and race cats.  The cat design is remarkably close to what the HD cat looks like.  The sampling issue that I have run into in the HD world hasn't been a concern for me.

The Innovate letter is 4+ years old now. I'll bet they know more now about how well shielding the broadband O2 sensors work is automotive exhaust systems.
The question is "how well do they work in a Harley open exhaust system".
Bob

Innovative Motorsport is about AFR gauges, would not be a choice for ECUs or serious data-loggers

Personally there are two many opinions on placement, and people using the Bosch LSU 4.2 data-sheet need to understand the bungs are not in stone...  Yes I generally still use a 1/2" straight bung, otherwise a 1" CUT down & angled.

If people pull out their compass, draw a 1-3/4" circle (size for most HD exhaust), then lay down the 4.2 where the threads just touch the outside of the circle it should be an eye opener.

I once asked a couple of old racers about it:
"Son you might as well call Eddy Murphy to come over and stick a banana up your tailpipe, cause that O2 is going to be one hell of a butt plug for performance" 

If people look at how the Bosch LSU 4.2 are used in other applications they are installed in the header pipe collectors (big diameters 3", 3.5" 4", 4.5" 5" ), and when installed in small diameter head pipe section they are angled pointing down stream.   

Interesting point on sensor protrusion hampering performance........any flow data coupled with dyno data to show that in detail?

Quote from: q1svt on October 19, 2015, 08:51:49 AM

Quote from: rigidthumper on October 16, 2015, 08:38:14 AM

IMO, some (if not all) of the accuracy drift that is documented by Bosch regarding internal pipe exhaust pressure is avoided by using a vacuum source to draw the sample through a tube across externally mounted 02 sensors. Lag becomes a factor, as well as tube placement.
If pipe mounted sensor blocks (Herko blocks or the ones from Bay Area Performance cycles) are used, they have a small amount of the exhaust volume coming out at a 90° angle, moving across the sensor. There has to be some pressure increasing with RPM, as I see some folks using vacuum at low RPM, but not at WOT.  Comparative transducers would be needed to see the delta between 02 sensor in pipe pressure versus 02 sensor in an aluminum block mounted on the pipe pressure.  Once we know what the pressure is, we can consult the Bosch specs to see if we need to adjust our corrections to the VE tables.
WB internally mounted sensors may get better "in the stream" readings, but the pressure bias should be considered when making corrections, if the pressure/temp is known. 

Or we could ignore the pressure induced accuracy drift, try to give the bike what it wants, and not get lost in the minutiae.

rigidthumper, could you expand on some of the points you made?  Many do not have access to the actual (non-public) Bosch documents, specifications and recommendations.  I understand that they are the company's interlectural property, but anything you can share generally will help move the conversation forward.

TIA

Here's the visual representation of the pressure induced accuracy drift:

I once asked some old racers what they thought about EFI, electronic ignition, and resistor spark plugs and wires.
After the laughter died down they told me "that high tech crap will never work."
My own testing done in 2007 showed absolutely no measurable power differences with O2 sensors in place or removed.

Thunderheader did testing back then and came to the same conclusion.
I'm sure the other exhaust mfgs have also came to the same conclusion but not sure.
I believe most all of the exhaust mfgs have shortened and/or relocated the O2 bungs so the closed loop system would work.
Bob

This one show the accuracy and which way (lean or rich ) the error is. Lambda >1 gets leaner as pressure increases.

note the % of error is applied to sensor current, not Lambda

There was a lot of good info in this thread on the same topic from a couple years ago http://harleytechtalk.com/htt/index.php/topic,59408.msg635674.html#msg635674

It would be cool if an inexpensive wideband could output the exhaust pressure, exhaust temperature, AFR and had full compensation for all of this stuff...

Quote from: whittlebeast on October 20, 2015, 06:20:56 AM
It would be cool if an inexpensive wideband could output the exhaust pressure, exhaust temperature, AFR and had full compensation for all of this stuff...

Hmmm,  make it in 12 millimeter that would be amazing.

Quote from: Jamie Long on October 19, 2015, 06:04:50 PM
note the % of error is applied to sensor current, not Lambda

There was a lot of good info in this thread on the same topic from a couple years ago http://harleytechtalk.com/htt/index.php/topic,59408.msg635674.html#msg635674

So, the output voltage of the sensor is affected by this error but not lambda?  How does the ECM know how to scale the error to get the correct lambda?

Thank you rigidthumper & Jamie for the posts...

Quote from: Jamie Long on October 19, 2015, 06:04:50 PM
There was a lot of good info in this thread on the same topic from a couple years ago http://harleytechtalk.com/htt/index.php/topic,59408.msg635674.html#msg635674

great information and insight how biases haven't waned


So, for people that don't want to wade thought the technical jargen for their DIY tuning it seems that ToBeFrank pretty well sums it up:

Quote
Note that it is unlikely the error in real world use will be 0.2 AFR. Keep in mind that the pressure is pulsing up and down in the exhaust. Combine that with the fact that the controller will be averaging several samples from the sensor to report one AFR sample, ... It's also possible the wideband controllers are doing some algorithmic sampling work to further increase the accuracy.

If I am tageting an AFR goal in the upper RPM's of say 12.75, I might want to input 12.65... well until that combined wideband, exhaust pressure, and exhaust temperature sensor come to market 

Quote from: q1svt on October 20, 2015, 08:04:12 AM
If I am tageting an AFR goal in the upper RPM's of say 12.75, I might want to input 12.65... well until that combined wideband, exhaust pressure, and exhaust temperature sensor come to market 

....and what output difference have you seen by the difference in target of 12.75 compared to 12.65?

Like I asked before and I will ask the same question to Frank.  How does the ECM or even the controller know how to compensate for the error?  Either of them have any input of pressure.  I understand that controller can take the input 4 times a second and turn that into an output that is 10 times a second.  I believe that is what Frank is pointing to, but that doesn't say anything about the error that is brought up.

Like we have agreed on.  The controller does nothing but put out a voltage.  How that voltage is dealt with is the job of the ECM.  The voltage scaled over to a lambda value is a set value that comes from the controller manufacture.  Changing the value in the tuning software is what is needed to use one controller over another.  In the tuning device that allow you to do this.  I have never seen any input for the error value.  It is as simple as:  If the voltage is this, this is your lambda value.

http://www.innovatemotorsports.com/support/manual/LC-2_Manual.pdf 

section 5.3.1 para .2 has the config and with this unit you can actually change it if you can't change it in the tuner software you can match up the output.

I need to correct my previous post where I am stating LM-2.  I actually use the LC-2's.

Quote from: hrdtail78 on October 20, 2015, 08:47:08 AM
Like I asked before and I will ask the same question to Frank.  How does the ECM or even the controller know how to compensate for the error?  Either of them have any input of pressure.  I understand that controller can take the input 4 times a second and turn that into an output that is 10 times a second.  I believe that is what Frank is pointing to, but that doesn't say anything about the error that is brought up.

Like we have agreed on.  The controller does nothing but put out a voltage.  How that voltage is dealt with is the job of the ECM.  The voltage scaled over to a lambda value is a set value that comes from the controller manufacture.  Changing the value in the tuning software is what is needed to use one controller over another.  In the tuning device that allow you to do this.  I have never seen any input for the error value.  It is as simple as:  If the voltage is this, this is your lambda value.

http://www.innovatemotorsports.com/support/manual/LC-2_Manual.pdf 

section 5.3.1 para .2 has the config and with this unit you can actually change it if you can't change it in the tuner software you can match up the output.

I need to correct my previous post where I am stating LM-2.  I actually use the LC-2's.

The proof is within your very own grasp...

Setup a test using one of your trusted Innovative Wide Bands AFR controllers, and your DynoJet 250 and optional AFR unit.  Please insure that all components are on current software, firmware, and don't forget the computer too. 

Publish your findings.  That's it you can independent prove your postings for all to clearly see. 

So it will not come as a supsrise, we will be applying Segal Law principles to them 
remember ... it is 'like a man that wears two watches' how does he know what time it is?

https://en.wikipedia.org/wiki/Segal%27s_law

Is the error enough to make a huge difference either way?  .2 afr is not going to make a break any of these motors.  It becomes a game of averages.  If one firing cycle is 13.5 and the next is 13.7 and the wideband records 13.6 do you think there is a true issue there?  As this all becomes a mute point if the data either one of the sensors is recording is not good.  The narrowband will just be recording bad data faster.  It doesn't matter either way because if what you are recording and applying isn't making the engine happy even though it's what the ECM or everyone else tells you it needs to be then you have to use your brain and think maybe what your doing isn't right.

Quote from: hrdtail78 on October 20, 2015, 08:47:08 AM

....and what output difference have you seen by the difference in target of 12.75 compared to 12.65?

I think you missed this question as you were asking me to perform work with my equipment.  It isn't a one way street. 

I have done the very test you asked for.  I have tested LC-2's, LM-1, 250 sampling, wego 3.........  I have tested AT-100 in an open loop verifying with other sensors.  I have tuned a vision with with their WB's as I was reading with different devices.  I have hours and hours into these types of testing.  I am more than happy to compare my test with yours.   But it isn't going to be that I take more time and get this all into a way that I can present it.  To be the only guy posting data so others can learn from my hard work.  With out others also having something to bring to the table.  I also have data of reading pressure in the exhaust.  I always find it funny that people want to talk about the pressure and what it does and doesn't do.  What pressure are seen in the exhaust of these system from people that have never even read exhaust pressure. 

What would this prove and not prove?  I understand what it has proved to me, but I don't understand what it has to do with the ECM or controller can and can't do or what input it is getting or what input it isn't getting.  What they are not getting is any pressure feed back from the exhaust.  There for it isn't correcting for the pressure in the exhaust.  Seems like more of a misdirection or changing topics more than anything else.

No need to just share the data and prove your points... it's not a hard concept to comprehend.

edited: sorry you asked... you keep raising the point how important the controllers are and how they are different (actuacy) and how the effect the end tune.  There are two controllers Innovative and DynoJet.

They are different.  Just as the NB's and WB's have different holes in the shield.  DJ 250I has no way to free air calibrate, but I do send in my stack once a year.  AT-100 has no way to free air calibrate.  Wego III has pots on the outside that is used for free air calibration but does cut the resistor out of the connector.  Innovate also has a procedure for free air calibration.

So, one big difference I see by just a quick glance is some controllers have a way with dealing with an aging sensor and some don't.

You make the statement that you wouldn't use Innovate for data logging or to control the ECM.  Why, and what do you think is better?

A few observations:
Jason has no horse in the race.
Jason has purchased proper equipment to educate himself directly rather than assuming forum opinions are correct.
Jason has done a ton of testing, data logging, and analysis. I've learned from his gratious sharing of his findings.
Jason has offered to the members here, without bias, his actual results from independent testing with most all available tuning systems.

q1svt, on the other hand has questioned and tried to discredit virtually whatever Jason and others have shared from their experiences.
Not sure why he is so aggressive toward dyno tuning folks willing to share their own personal and unbiased opinions based on actual experience.
Bob

Jason you are arguing against yourself...

If for the moment everything you say about the Bosch LSU 4.2 is correct:
too slow
inaccurate
uncontrollable
impacted by exhaust gas pressure
etc
etc
etc


THEN; me as a customer asking about a dyno tune, you are telling me that getting a dyno tune is a big waste of money...

Not because of your tuning skills, but simply the Dyno equipment sucks because it uses Bosch LSU 4.2
too slow
inaccurate
uncontrollable
impacted by exhaust gas pressure
etc
etc
etc


I cannot make it any simpler than that.

Quote from: hrdtail78 on October 20, 2015, 09:41:54 AM
You make the statement that you wouldn't use Innovate for data logging or to control the ECM.  Why, and what do you think is better?

Here is my response posted many posts ago...  believe Joe-Lyons said it earlier today too.

Quote from: q1svt on October 18, 2015, 02:18:51 PM
I think this is the best reply comment posted to the issues with a dyno & bike running Bosch LSU 4.2's

Quote from: rigidthumper on October 16, 2015, 08:38:14 AM

Quote from: q1svt on October 16, 2015, 06:07:22 AM
There are a number of Wide Band posts regarding Bad Sampling Results...

Without valid data, how can the following be eliminated as the cause?

"So what would happen if a dyno operator was using a broadband [system] along with a device to vacuum-pull a sample from the pipe to and across the sensor tip, but he did not include the sensor itself in the same amount of clean-air vacuum, leaving it instead exposed to "outside" air pressure?  Would his readings have a "built in" skew one way or the other?"

IMO, some (if not all) of the accuracy drift that is documented by Bosch regarding internal pipe exhaust pressure is avoided by using a vacuum source to draw the sample through a tube across externally mounted 02 sensors. Lag becomes a factor, as well as tube placement.
If pipe mounted sensor blocks (Herko blocks or the ones from Bay Area Performance cycles) are used, they have a small amount of the exhaust volume coming out at a 90° angle, moving across the sensor. There has to be some pressure increasing with RPM, as I see some folks using vacuum at low RPM, but not at WOT.  Comparative transducers would be needed to see the delta between 02 sensor in pipe pressure versus 02 sensor in an aluminum block mounted on the pipe pressure.  Once we know what the pressure is, we can consult the Bosch specs to see if we need to adjust our corrections to the VE tables.
WB internally mounted sensors may get better "in the stream" readings, but the pressure bias should be considered when making corrections, if the pressure/temp is known. 

Or we could ignore the pressure induced accuracy drift, try to give the bike what it wants, and not get lost in the minutiae.

There are issues with a sampling tube process, there are issues with drilling exhaust pipe & using Blocks, and the bikes on-booard ECM...

Quote from: FLTRI on October 20, 2015, 09:52:08 AM
Not sure why he is so aggressive toward dyno tuning folks willing to share their own personal and unbiased opinions based on actual experience.
Bob

I have a great fondness for engine tuners and have been using them for about 40 years.

I have only found one that on an open forum repetitively talks about:
how DIY tunes are *&%&^$#&*^*(
that anyone that uses TMax is one dumb )(&(*&*^^%&^%%^&^&
that the only way to get the a good tune is with a dyno other wise they are a *(&(*&**&^*&*&

I am a DIY tuner
I use TMax
I believed you are no more skilled than many of us.

So, how does it feel when the table is turned around on you 

Quote from: q1svt on October 20, 2015, 11:57:11 AM

Quote from: FLTRI on October 20, 2015, 09:52:08 AM
Not sure why he is so aggressive toward dyno tuning folks willing to share their own personal and unbiased opinions based on actual experience.
Bob

I have a great fondness for engine tuners and have been using them for about 40 years.

I have only found one that on an open forum repetitively talks about:
how DIY tunes are *&%&^$#&*^*(
that anyone that uses TMax is one dumb )(&(*&*^^%&^%%^&^&
that the only way to get the a good tune is with a dyno other wise they are a *(&(*&**&^*&*&

I am a DIY tuner
I use TMax
I believed you are no more skilled than many of us.

So, how does it feel when the table is turned around on you 

I haven't clue who you are referring to:
I never call names.
I have always maintained Tmax has its place...at the race track where the vehicle is at high loads/TP/rpms most of the time.
I have never indicated anyone using Tmax is dumb.
I have always maintained a proper dyno tune will produce and quantify results from modifications where as the DIYer has no practical way to do so on the street.

As far as skill? I have experience...many years of tuning HDs of all states of modification with virtually all tuning systems. I regularly see many build variations and have witnessed the good, the bad, and the ugly. Does this relate to skill level? Not necessarily.
Now I will agree I am not the sharpest pencil in the drawer but I have learned from rote so I have learned over time what to do and how to do it.

Now that you know some of my experience...what's yours...as related to tuning HDs?
Thanks,
Bob

Quote from: joe_lyons on October 20, 2015, 09:06:47 AM
Is the error enough to make a huge difference either way?  .2 afr is not going to make a break any of these motors.  It becomes a game of averages.  If one firing cycle is 13.5 and the next is 13.7 and the wideband records 13.6 do you think there is a true issue there?  As this all becomes a mute point if the data either one of the sensors is recording is not good.  The narrowband will just be recording bad data faster.  It doesn't matter either way because if what you are recording and applying isn't making the engine happy even though it's what the ECM or everyone else tells you it needs to be then you have to use your brain and think maybe what your doing isn't right.

That's what I think as well Joe.

Quote from: Hilly13 on October 20, 2015, 12:45:23 PM

Quote from: joe_lyons on October 20, 2015, 09:06:47 AM
Is the error enough to make a huge difference either way?  .2 afr is not going to make a break any of these motors.  It becomes a game of averages.  If one firing cycle is 13.5 and the next is 13.7 and the wideband records 13.6 do you think there is a true issue there?  As this all becomes a mute point if the data either one of the sensors is recording is not good.  The narrowband will just be recording bad data faster.  It doesn't matter either way because if what you are recording and applying isn't making the engine happy even though it's what the ECM or everyone else tells you it needs to be then you have to use your brain and think maybe what your doing isn't right.

That's what I think as well Joe.