Curious about TTS CLBs...

[sportygordy]

I especially like this bit of info:

"Notes on the Sensor:

You can typically tell if a sensor is starting to fail if it is very slow to respond to changes in A/F, like from blipping the throttle, or slow to show proper signals during warm-up (system power on but engine not running).  The response will continue to slow until finally there is no response from the sensor.  Running the sensor in "clean" exhaust for a while can partially restore a sensor.  See Note below about "cleaning".

The sensor contains a ceramic module and should not be subject to mechanical or thermal shock or it may be damaged.  Thermal shock would occur if liquid drops of water were to contact the sensor, like if a sampling trap for the DT2-AFR Remote sampler were to over fill.

The sensor is not designed for operation on leaded fuels, doing so will shorten sensor life.

Long term running in the rich region (Lambda < 0.95) will shorten sensor life, running very rich will dramatically reduce sensor life.

High exhaust temperatures (over 850 deg C, 1560 deg F) will shorten sensor life.

Engine oil consumption at a rate greater than 1 quart per1,000 miles will shorten sensor life.

Do not run the engine with the UEGO sensor installed without power applied to the controller and the sensor plugged in.

It may be possible to "clean" sensors if you run them in a modern, production car with good emissions, with the proper DT2-AF1, DT2-AF2 or DT2-AFG controller connected.  This must be done before the sensor has completely failed.

A good rule of thumb for sensor life is:  A sensor will last about as long as your spark plugs.  If your spark plugs are fouled by lead or very rich running, you sensor is likely to have failed also."

And I especially like this bit of info:

Typical production car O2 sensors rely on “nernst cell” technology, commonly called “Narrow Band”, “2 wire”, “4 wire” and sometimes erroneously described as “Wide Band”. This is a very cost effective method that outputs a voltage based on the oxygen content of the gas being sampled. It is accurate in the region surrounding stoichiometric operation and leaner. Unfortunately, in the rich region where high performance engines usually operate, their accuracy and repeatability is virtually non-existent.

The rich region output of a common O2 sensor is very temperature dependant, which renders it useless if an accuracy greater than 1.5:1 AFR is desired. This is immediately obvious given the fact that a single output voltage actually represents wildly different AFR’s depending on the unregulated and unmeasured sensor temperature. These sensors were designed for operating closed loop around the stoichiometric AFR (14.64 for gasoline), and for performance tuning they are useless.

The heart of the Performance Trends Wideband controller is the Bosch LSU4.2 Universal Exhaust Gas Oxygen (UEGO) sensor. This type of sensor is commonly referred to as “laboratory grade” and works on a different principle than the normal oxygen sensor you would find in your car. Its unique design makes precision AFR measurement possible over the entire operating range.

[Steve Cole]

Here is another good read of just how well these cheap sensors work. This was an independent test run on several units. Note that the Dyno Jet system was tested and this is just what most dyno operators are using for there results.

http://paceperformance.com/paceperformance/pdf/widebandshootout.pdf

The Results
We tested the top wideband air/fuel ratio meters available on the market. There was a surprising amount of variation between
the various units, in terms of both accuracy and response time.

Of particular note was the issue of calibration. Only two units appeared to be capable of calibration, and both used free air as a reference. One of the most accurate (Innovate) was self calibrating, while the other unit (NTK) required the user to turn a knob until the display read “CAL.” This raised the obvious question: If a unit is not capable of calibration, how does the user know when the sensor is going bad? We know from the
Bosch data that the sensors themselves change as they age.


As anyone can see that the units do not event agree with one another for the limited testing they got!

For those of you that would like to read what I've said all along here is the same discussion about these cheap wide band sensors.

http://cvogarage.com/index.php?topic=526.45

[ToBeFrank]

Here is a real world result of projecting (guessing) the VEs you can't tune with Vtune:

http://harleytechtalk.net/htt/index.php?topic=9170.0

[ToBeFrank]

Here is another good read of just how well these cheap sensors work. This was an independent test run on several units. Note that the Dyno Jet system was tested and this is just what most dyno operators are using for there results.

http://paceperformance.com/paceperformance/pdf/widebandshootout.pdf

This isn't testing the accuracy of the sensors. It's testing the accuracy of the controllers. Bosch says right in their documentation that the controller is the important part. As far as I'm concerned, this pdf proves Ed's statement that the controller is what dictates the accuracy of the Bosch sensor. It looks like Innovate does a good job. Dynojet, not so much. As bad as you say these sensors are, I'm wondering how any dyno tuner using the Bosch sensor gets a bike tuned?

[ToBeFrank]

I especially like this bit of info:

Which sensor do you use when you tune Bob?

[FLTRI]

Which sensor do you use when you tune Bob?

We use the LSU4. Since Dynojet's wideband O2 and air pump does not have to deal with positive exhaust pressure, vibration, or high heat, accuracy and reliability is far less of a concern.

IMO, preventative maintenance by checking with calibration gas and regular replacement is the key to successful use of these sensors.

We have found the problem comes when the EFI/O2 system on the vehicle uses these wideband sensors in the exhaust very close to the exhaust ports and there is no regular maintenance schedule for checking their accuracy or replacement. As time goes by the readings change. Exhaust vibration ("The sensor contains a ceramic module") over time shortens the life of these sensors. Also higher than normal EGT temps from traffic heat, lean burn, hard running, etc reduces the accuracy and life span of these sensors. While there may be readings still coming from a sensor in exposed to these conditions, the accuracy becomes compromised.

The above is based on my own personal experiences as used in the racing world and dyno tuning not on mfg papers or propaganda, or hearsay.
Bob

[Steve Cole]

Tobefrank
Guess you missed this part of the test report.............

We did do one interesting sub experiment: The dyno shop had a variety of old and damaged sensors around. With one of these sensors, the Innovate XD-16 would show an error code indicating that the sensor was bad. When we connected these same sensors to any of the analog
gauges (with the right connector), they read as much as 3 AFR off. Again, the obvious question is: If your gauge can’t tell you when a sensor is bad, how do you ever trust it?

Also did you look at the dynojet results as this is the systems used by Ed and most other motorcycle operators. The only difference is that the dynojet dyno system uses a remote mounted sensor in a pump which further delays the readings so the correlation between the AFR and RPM is further off. So one must know what lag time there is between the AFR reading and the RPM being shown on the graph.

[ToBeFrank]

Tobefrank
Guess you missed this part of the test report.............

We did do one interesting sub experiment: The dyno shop had a variety of old and damaged sensors around. With one of these sensors, the Innovate XD-16 would show an error code indicating that the sensor was bad. When we connected these same sensors to any of the analog
gauges (with the right connector), they read as much as 3 AFR off. Again, the obvious question is: If your gauge can’t tell you when a sensor is bad, how do you ever trust it?

No, I didn't. It is saying the controller that tested to be accurate to within .1 AFR (the Innovate) also correctly detected a bad sensor. Again, this shows the difference between a well done controller and a not so well done controller.

Also did you look at the dynojet results as this is the systems used by Ed and most other motorcycle operators. The only difference is that the dynojet dyno system uses a remote mounted sensor in a pump which further delays the readings so the correlation between the AFR and RPM is further off. So one must know what lag time there is between the AFR reading and the RPM being shown on the graph.

Sorry dyno tuners... it's time to find a new job.

[ToBeFrank]

Which sensor do you use when you tune Bob?

We use the LSU4. Since Dynojet's wideband O2 and air pump does not have to deal with positive exhaust pressure, vibration, or high heat, accuracy and reliability is far less of a concern.

IMO, preventative maintenance by checking with calibration gas and regular replacement is the key to successful use of these sensors.

We have found the problem comes when the EFI/O2 system on the vehicle uses these wideband sensors in the exhaust very close to the exhaust ports and there is no regular maintenance schedule for checking their accuracy or replacement. As time goes by the readings change. Exhaust vibration ("The sensor contains a ceramic module") over time shortens the life of these sensors. Also higher than normal EGT temps from traffic heat, lean burn, hard running, etc reduces the accuracy and life span of these sensors. While there may be readings still coming from a sensor in exposed to these conditions, the accuracy becomes compromised.

The above is based on my own personal experiences as used in the racing world and dyno tuning not on mfg papers or propaganda, or hearsay.
Bob

Thanks Bob. Good information.

[Steve Cole]

Well Frank I see you only want to read what you want to read into it.

We did do one interesting sub experiment: The dyno shop had a variety of old and damaged sensors around. With one of these sensors, the Innovate XD-16 would show an error code indicating that the sensor was bad. When we connected these same sensors to any of the analog
gauges (with the right connector), they read as much as 3 AFR off. Again, the obvious question is: If your gauge can’t tell you when a sensor is bad, how do you ever trust it?

The point that said "a variety of old and damaged sensors" sure sounds like it's more than one sensor to me!
So the Innovate and the others read the rest of the old sensors as much as 3 AFR off. That doesn't sound very good to me at all but maybe that's good enough for you. The point is the electronics do matter, but it's not the only issue with these cheap wide band sensors. Yes, the Innovate read the test gas at room temperature the best and it at least flagged one sensor as bad but it missed the other old sensors. This is just what I've been trying to let people know, there are problems with them.

Now you want to talk about a switching sensor and that it's only accurate in a narrow range, guess what, your right and I agree with you 100%. That is just why we do our calibration procedure in the range that they work so well in and this is what Vtune is for. Once the calibration is done in that range the AFR can be reset to what the customer wants/needs. Vtune only calibrates the VE tables so the ECM can calculate the fuel properly, it's still up to the customer to put the right AFR values in for his/her application and adjust the spark as well.

[wolf_59]

Quote from Steve Cole
I can backup each and every statement that I have said. If you read the instructions and do what is told you can tune 85% of the bikes running range. Yes, the VE table is based on TPS but let's look at that so that Frank and others can learn. If your at sea level the MAP kPa will run from about 25 - 100 kPa on a HD engine, based on the load on the motor. The TPS will be 0 - 100%. Once the MAP reads 100 kPa no matter how much you turn the throttle the airflow through the engine does not change at a given RPM. So since we measure up to 85 kPa and calibrate for that you have the information you need to dial the VE tables in by following the instructions. Let's use 1250 RPM for an example, at this RPM your going to hit 85 kPa about 10% TPS so the vtune data will get you to there and give you the proper VE for that. Before 15% TPS your at 100 kPa another words max airflow at this RPM. So what do you think will be the proper VE value to fill in for 15 - 100% TPS? Since its at max airflow it will be the same value over and over again! So you take the value you have at 10% TPS (85 kPa) and use it over again. You can look at the trend from the 5%, 7% and 10% TPS and follow that for the 15% TPS but then use that new value over and over again. This will get you within a few percent of the proper value which is just as accurate as you can get on the dyno with the cheap wide band sensors! Now let's move to 6000' instead of sea level. At this altitude you will see 25 - 83 kPa for the engine MAP values. Here you can tune near 100% with Vtune and have to fill no cells in. This process is covered in the instructions.

Excellent information, As heated or passionate as these discussion get there is always good info that gets passed around.

[philmcc]

dsanchez...this is Sailor285...I set my CLB's at 450....then did mmy tunes and ajusted the afr to what I wanted in each rpm and then adj the afr in the cells I wanted to change for instance..I run 13.8 at idle and 1250 then 14.2 at 1500 and 14.6 at 1750 and up to 4500... I have great power and good fuel milage... At 70 with no wind and 60 deg out I got 51 mpg ..At 52 deg and 15-20 mph heaqd winds running between 55 and 65 I'm gettin just over 45 mpg Hope this helps a little....Dan

[ToBeFrank]

Well Frank I see you only want to read what you want to read into it.

When we connected these same sensors to any of the analog
gauges (with the right connector), they read as much as 3 AFR off.

They specifically said the analog gauges were off. Innovate is digital. I read it as they stated it.

[Steve Cole]

Here is a real world result of projecting (guessing) the VEs you can't tune with Vtune:

http://harleytechtalk.net/htt/index.php?topic=9170.0

Yes Frank those are the results he has to this point, and most say that his combination should make about 120 Hp and he’s at 111 Hp That 92% of the way there with the best running this bike has had. He had a PC III dyno tuned on it before so he has something to compare too. Is it as good as we would like right now, NO, but what have I said time and time again……… 85% and he is well past that already with a good running bike that is very safe the way it is. Not bad for someone that has never tune before, that started with a calibration that was not for his engine combination and he did it himself on the street with no other help. Now we are looking at what he has done and making some corrections for him. These things will be explain better in future releases of our help files as we have not done it well enough it appears.

Well Frank I see you only want to read what you want to read into it.

When we connected these same sensors to any of the analog
gauges (with the right connector), they read as much as 3 AFR off.

They specifically said the analog gauges were off. Innovate is digital. I read it as they stated it.

Really, then what happened to the “variety of old and damaged sensors” since it only found one bad? Also how many motorcycle dyno guys are working with the Dynojet product and how many are using the Innovate or other brands? The problems are real and that's all I want people to know.

[Tre_11 FLHX]

I have great power and good fuel milage... At 70 with no wind and 60 deg out I got 51 mpg ..At 52 deg and 15-20 mph heaqd winds running between 55 and 65 I'm gettin just over 45 mpg Hope this helps a little....Dan

Philmcc-Just curious, what Bike/build are you obtaining the above on?  My 09 UC has never seen better than 35, even when new.  Last tank was 31 MPG,  I am currently re-tuning (with Help from Mr. Herko) I have discovered that a lot of my issues with TTS have been self inflicted (attention to detail).  As I begin to understand how the system operates, the better my scoot runs.  MPG have yet to be tested as I am still in the tune phase.


All,
I am interested in purchasing what ever I need to set the WOT AFR.  Any recommendation would be appreciated.

As always TIA,
Mike

[sportygordy]

Here is another good read of just how well these cheap sensors work. This was an independent test run on several units. Note that the Dyno Jet system was tested and this is just what most dyno operators are using for there results.

http://paceperformance.com/paceperformance/pdf/widebandshootout.pdf

This isn't testing the accuracy of the sensors. It's testing the accuracy of the controllers. Bosch says right in their documentation that the controller is the important part. As far as I'm concerned, this pdf proves Ed's statement that the controller is what dictates the accuracy of the Bosch sensor. It looks like Innovate does a good job. Dynojet, not so much. As bad as you say these sensors are, I'm wondering how any dyno tuner using the Bosch sensor gets a bike tuned?

I caught that too. Steve has a very bad habit of jumping to things without comprehending them. This is evident in several other forums. When he says the Bosch Wideband sensors are junk, i think most of us really understand the real deal here. His PDF attachment clearly proves the Bosch Wide Band Senor's are not junk and actally very good performance tools and the Narrow bands are the 'real' junk. Plain and simple, anyone after performance will use the Bosch sensor and a tuning system capable of wide band sensing, just as you see on every Dyno Tuners garage. Unfortunately the TTS is not a performance package. I guess it's it's o.k. for the EPA guidelines and getting good gas milage, but its autotune flunks in the performance arena.   

[FLTRI]

Maybe I'm a bit thick in understanding what the point is here, but let me toss my $.02 in here.
AFAIK the narrow band (switching) sensors the automotive industry relies on for true and consistant output is highly accurate in the 14.0-15.0 AFR range.

TTS instructs the tuner to set the target AFR table to 14.6 and V-tune, over 2-3 run sessions, will calculate VE changes to get to that 14.6 AFR everywhere except basically WOT.
Then, after V-tune has recalibrated the VEs to 14.6, and the subsequent calibrations have been installed, the AFR target table can be physically changed from 14.6 to whatever the tuner desired to run the engine at. Then the software recalculates the VE tables to end up with the actual AFR listed in the AFR table.

While this is certainly different from what tuners normally do (manually tune VE tables to desired AFR) if the V-tune software calculations are correct, the end result s/b the same.

My take on the inexpensive Bosch wide-band sensors is as long as these sensors are routinely checked for accuracy with test gas and replaced as necessary (change sensors when changing sparkplugs was mentioned in an article), they work just fine. It is when they are permanently mounted to high vibration exhaust pipes for extended periods, without routinely checking their accuracy, that expected AFR can change without the rider being aware.

I have always maintained auto-tune systems using these sensors should be taken out of auto-tune and the sensors removed once the system has accomplished the tuning process. That way there is no concern of sensor voltage output change or failure causing running issues.

[Steve Cole]

Not only have we done the V-tune calibrations and back checked them with Lab equipment we have hundreds of customers that have tuned there own bike after a so called dyno tune to have a better running bike but one that also gets better fuel economy. How do you think we learned about the problems with the cheap Bosch sensors? The hard way, back in automobiles of the late 80's and early 90's. They just do not do what you’re trying to say they do and it’s plain to see if you do nothing but read the Bosch documents as to why. Dave Darge and I have been personal friends for over 15 years now so I know him well; as a matter of fact I have done plenty of calibration work for him over the years. The stock switching sensors are very good at what they do and the Delphi ECM code is very good at what it does. You cannot with any accuracy tune with a cheap wide band sensor and the current aftermarket units any closer than one can do with a switching sensor when done properly. The reason I call it "Cheap" is to set it apart from the "Expensive" sensors. The cheap sensor works as advertised by Bosch perfectly but it does not work like some here are trying to say.

I've said it time and time again that the sensors work as Bosch says they will when use as Bosch says. The issue is they are not being used like Bosch calls outs by the aftermarket. If used with good quality electronics with the proper pressure and temperature measurements they work as advertised by Bosch. So if the test equipment does not measure pressure and temperature along with the AFR you are not going to get accurate results.

I guess a tuning package that is tuning the motors, by the bike owner, with no other equipment involved and by someone who has never tuned to within a few percent of max power and getting good gas mileage  is now call tuning for EPA guidelines. 

[Blackbaggr]

Steve...

Another question...if I'm reading correctly it appears that the PZ176 file has also had changes to the AFR table in addition to the timing. Assuming that is correct is it OK to cut & paste the AFR as well as timing changes to the calibration that I created from the older PZ176 calibration. I'd think that both of these changes may assist me in my goal of better mileage. I'm thinking some of my problem is actually my problem and I'm running harder than I should in the areas uneffected by the CLB table. 


OR...

Would it be better to take the new PZ176 file and start over ?

[Steve Cole]

We always recommend using the latest files to start with. I would take your VE tables and CLB tables and copy them into the newer file and go from there.

[sportygordy]

My take on the inexpensive Bosch wide-band sensors is as long as these sensors are routinely checked for accuracy with test gas and replaced as necessary (change sensors when changing sparkplugs was mentioned in an article), they work just fine. It is when they are permanently mounted to high vibration exhaust pipes for extended periods, without routinely checking their accuracy, that expected AFR can change without the rider being aware.

I have always maintained auto-tune systems using these sensors should be taken out of auto-tune and the sensors removed once the system has accomplished the tuning process. That way there is no concern of sensor voltage output change or failure causing running issues.

Bob,

Just curious, how often have you or do you change the Bosch sensor on your dyno table? And how often do you preform free air calibration?

[Herko]

Not Bob, but Bosch sensors can check good and go through their free air cal just fine, then be off by a significant amount in the 11 to low 14 AFR range. I've found they need to be changed fairly regularly.

[sportygordy]

Not Bob, but Bosch sensors can check good and go through their free air cal just fine, then be off by a significant amount in the 11 to low 14 AFR range. I've found they need to be changed fairly regularly.

Herko

I thought FLTRI was Bob.. I'm just trying to find out what he (Bob) does on his dyno table. I have input from a few other dyno owners and would like to compare their sensor maintenance schedules against his. Just working on another future post in regards to the 'junk' Bosch sensors. BTY, do you own a dyno table? and is your reply in reference to your hands on experience?

thanks,

sportygordy

[eddfive]

The single most important event a dyno tuner has is the handling of O2 sensors.  In my case they get free air calibrated before every bike and are allowed to warm up for 20-30 minutes before install.  After the 20-30 minute warm up I check free air again.  Free air needs to be just that no Hydrocarbons from exhaust fumes in the air where testing.  If during the free air calibration the position of the calibration knob has changed too much since the last calibration then I toss them and get a new one.  I go through sensors about 1 or 2 every 60-90 days.  The other important part is how they are handled and installed into the bike.  I have designed a special collector/chamber with the copper tube so that it does not "kill" the sensor with the vacuum pump.  I also have a very high dollar vacuum pump as I do not have a Dyno-Jet I have a Dynostar.  Vacuum pump maintenance is also vital if using the copper tubes.  My vacuum pump pulls a continuous 28 in hg all the time.  On the newer bikes I pull the 2-wire O2 sensors and install my 5-wire sensors.  At the end of the tuning session I again pull the 5-wire sensors and check the free air calibration, if it is the same or very close when I started I call it good.  It really boils down to the details.  I have very good success with this method and have "killed" a bunch of sensors to get this all figured out, I have a box full of rejects.  Is this method right or wrong I do not really care as it works for me and my customers are happy.

[Herko]

"BTY, do you own a dyno table? and is your reply in reference to your hands on experience?"

Yes and yes.

John