Jump to content

Link made CAN wideband with LSU 4.9 or NTK


hexdmy

Recommended Posts

I know that link recommends the Innovate wideband, however I've had very poor success with these in the past. I would like to see a reasonably priced Link wide band with CAN communication, and support for at least the LSU 4.9, even better to support the L2H2 (NTK # LZA-09-E1) and/or production NTK sensor (NTK #  LZA-08-H6). Any plans along these lines ?

Link to comment
Share on other sites

I don't think there are many people left in the tuning industry that would disagree with your experience with innovate lambda controllers.

You already have a few flavours to choose from that will do what you ask; For CAN based LSU4.9 controllers I believe the G4+ will already talk to KMS UEGO, Ecotron ALM or Motec LTC, For the NTK sensor your option would be Motec LTCN.  Even if link did build their own lambda to CAN controller I cant see it ever getting much cheaper than some of those listed above.  

Link to comment
Share on other sites

Adam, I am familiar with those three widebands you mentioned however I believe the price point is either too high or the quality is questionable. my thought is that if they could offer a CAN wideband including the sensor with a street price around 250 dollars that utilized the lsu 4.9 sensor, or for about 350 dollars that would support / utilize the NTK sensor (4 or 6 milliamp)they would have a real winner.

Besides eliminating the voltage offset, Link offering their own CAN wide band would allow them to implement warm up strategies that would protect the sensor and reduce sensor failures, in addition they could have more elaborate diagnostic routines that would help protect the motor in case of a lambda failure. With that in mind you could also comfortably run closed loop lambda all the time.

Link to comment
Share on other sites

Hi Chris 

A stand alone for that price point is not likely to happen as there just wouldn't be enough margin to make it viable. Given the limited volumes. If we were pumping out thousands of them then maybe.

What is more likely is building it in to the ECU so it takes the sensor direct.

Link to comment
Share on other sites

  • 4 months later...

Too pity! I think an quality external device which has CAN cababilities may cost one and a half  or twice the price from a LC-2 or something similary. That will be an great addition, because I'm a bit tired to always calibrate this analouge Widband. That means i have to weld in a second O2 Bung to calibrate the wideband in the LINK against my Motec PLM from the Dynapack. In my experience they differ alot, especially the AEM wideband. 

Another option that I would really like to see is an upgrade option for built in Wideband for all ECU, also Plug-In's. That could also be an good source of additional incom for Electronz. 

 

Link to comment
Share on other sites

Hi mapper,

just sell more FURY's ha ha :)

I run the LSU 4.9 directly to my FURY and also run another 2  different WIDEBANDS as extra inputs to compare data.

The other 2 wide band AFR's are Autronic A analyser and the other is NGK AFX

Regards

Dave. 

haha yes Sir, Furry's are great! I will also wire in one into my personal skyline as soon i find the time. 

Sounds interessting, I would love to hear some comparison results! 

Regards

Adrian 

Edited by mapper
Link to comment
Share on other sites

  • 1 year later...
  • 2 weeks later...

There has been discussion on this, but the LSU 4.9 sensor seems to be working well for most customers so far. Are you wanting the NTK because if an extended range or longer life or?

Scott

For what it's worth, I'm on my 3rd LSU 4.9 sensor connected to a CAN-Lambda & Monsoon ECU and still have not gotten even one to work.  I would try just about anything at this point to get my CAN-Lambda combo up and running.  If there were a compatible NTK sensor with even rumors of higher durability I'd be all over that personally.

Link to comment
Share on other sites

Similar to krohelm, I'm on my 4th sensor.

Sensor life is woeful at best, as a result I've resorted to using a different controller that accepts ntk sensors fed to the ecu via analog input.

If the controller could accomodate other sensors, i would revert back, as when it works it works well for the application.

 

Link to comment
Share on other sites

Similar to krohelm, I'm on my 4th sensor.

Sensor life is woeful at best, as a result I've resorted to using a different controller that accepts ntk sensors fed to the ecu via analog input.

If the controller could accomodate other sensors, i would revert back, as when it works it works well for the application.

 

We dont think Krohlem has a sensor problem - none of his have worked from new so it seems unlikely he got 3 faulty sensors and two faulty CAN Lambda devices.  The LSU4.9 is the most common OEM wideband sensor today and I dont see reports of massive emissions failures in the news so I think we can assume the sensors are well proven.  One thing I have noticed recently is there are a lot of counterfeit/clone 4.9's on the market so @wasted talent, depending on where yours have come from that is possibly something to consider.

Whilst the common NTK's are typically more robust to thermal shock they do have some drawbacks too; 1) about twice the cost, 2) much slower, 3) very sensitive to EMAP, so for those reasons I would generally recommend the 4.9 as more suitable for most users.  The NTK is better in situations like a dyno shop where it has to move it from vehicle to vehicle very often, it is also more resistant to lead poisoning so it is better for some of those odd fuels also.

Link to comment
Share on other sites

Having burned through a big stack of wideband sensors I've found the way to make them last is to have a delayed start after engine is running for the controller / heater.

Having the heater turn on while the engine is off is absolutely the worst thing for them, first crank and cold air past sensor = cracked. 

Since I have added a relay to turn on my controller, powered by ECU with two conditions of either 15 second delay after first start, or 50 deg minimum engine temperature.

I have had no further issues. 

With the Fury etc is there an option for delayed start or any extra heater controls? 

I think the difference in lifespan between OEM 4.9s and aftermarket ones, is that generally the aftermarket controllers have no visibility of what the engine is doing, it can only REACT to exhaust temp changes etc. 

It would be cool if the wideband controller integrated into the ECU could (via TPS or rpm or whatever) anticipate increase in EGT and proactively back off the heating of the sensor or something like that, to mellow out the "spikes" in sensor temperature.

At the moment am I correct to assume that it's just constantly aiming for a goal sensor temperature with no input from TPS etc?

Even just as simple as having wideband heater control linked to Accell enrichment and decel fuel cut would probably help with sensor life I'd imagine. 

Edited by Davidv
Link to comment
Share on other sites

 

With the Fury etc is there an option for delayed start or any extra heater controls? 

Yes with internal Lambda or our lambda CAN device there is a built in start up delay that times out after RPM is detected.  I couldnt tell you what the time period is off the top of my head however.

It would be cool if the wideband controller integrated into the ECU could (via TPS or rpm or whatever) anticipate increase in EGT and proactively back off the heating of the sensor or something like that. 

At the moment am I correct to assume that it's just constantly aiming for a goal sensor temperature with no input from TPS etc. 

Yes your assumption about "goal temperature" is correct but in my experience I have never seen the need to have any further controls or influences on the heater setpoint.  I have ETAS controllers on my dyno that display sensor temp so I do regularly get to observe the EGT influence.  

For the LSU4.9 the normal control temp is 780°C, but the sensor is rated for continuous use with exhaust gas temp up to 930°C with up to 250hrs accumulated at 1030°C, so its going to need some fairly serious external heat influence to achieve that.  I would probably be more worried about my molten pistons than my O2 sensor by that point....

Edited by Adamw
typo
Link to comment
Share on other sites

Ahh okay thats interesting about the temperature rating.

I've been thinking about getting a Fury for a 6 cyl turbo project in the works, onboard lambda is definitely awesome.

Do any of your spec sheets about the sensor list anything about best practice for sensor warmup procedure or whatever? 

I dont think I've ever seen any documentation directly from Bosch about the 4.2 or 4.9.

Link to comment
Share on other sites

Bosch has some pretty detailed information here, including graph over time of heater voltage:  www.bosch-motorsport.com/media/catalog_resources/Lambda_Sensor_LSU_49_Datasheet_51_en_2779147659pdf.pdf

Observationally, CAN-Lambda waits 5-7 seconds after engine rpm exceeds 400 before engaging the sensor.  

My air is still pretty cold after 5-7 seconds, I'd love to have a virtual output control when rpm is 0 and when it's "as measured" for CAN output to id 958 (iirc that's where can-lambda 0 gets rpm).  That way I could delay heating further.

Link to comment
Share on other sites

I blew four 4.9 sensors last season (its winter in my world right now).

However im currently using a different brand ECU but thats beside the point. Lets call it X. It got Bosch controllers built in it though.

I narrowed it down to beeing a heater problem and blowing sensors like this got me a little puzzeled really. Cause after all the ECU are using Bosch controllers to control Bosch sensors, so you would think there would be no problems at all. I meen you would think Bosch knew how to handle the heating of their own sensors.

That was until Andy Wyatt (the founder of Adaptronics) made me aware that infact it is NOT the Bosch controller that decides how heating is controlled but the ECU manufacturer. In other words it isnt hardcoded into the controller but a function of settings in the ECU. Then it all made a little more sense to me.

OEM`s are mainly using 4.9`s and in those applications they lasts for years and years for what its worth...

X have now fabbed up a new firmware with different heating algoritm for me to test when the next season starts. Well see how it goes.

Edited by Steve
Link to comment
Share on other sites

Do any of your spec sheets about the sensor list anything about best practice for sensor warmup procedure or whatever? 

I dont think I've ever seen any documentation directly from Bosch about the 4.2 or 4.9.

LSU4.9 Tech info:  https://1drv.ms/b/s!AiYbYlZQuRHPixMg31pUx3HxmQI2

 

However im currently using a different brand ECU but thats beside the point. Lets call it X. It got Bosch controllers built in it though.

I narrowed it down to beeing a heater problem and blowing sensors like this got me a little puzzeled really. Cause after all the ECU are using Bosch controllers to control Bosch sensors, so you would think there would be no problems at all. I meen you would think Bosch knew how to handle the heating of their own sensors.

That was until Andy Wyatt (the founder of Adaptronics) made me aware that infact it is NOT the Bosch controller that decides how heating is controlled but the ECU manufacturer. In other words it isnt hardcoded into the controller but a function of settings in the ECU. Then it all made a little more sense to me.

OEM`s are mainly using 4.9`s and in those applications they lasts for years and years for what its worth...

You are correct in that the CJ125/135 chipsets dont control the heating but Bosch does give guidelines for how that should be done to achieve the quoted sensor performance and life.  Unfortunately many aftermarket manufacturers seem to ignore those guidelines.  As Steve has pointed out here also many of the standalone controllers dont have any way to know when the engine is running so they cant do the recommended "reduced heating during water condensation phase".

In my experience though the CJ125/135 based controllers that followed the recommended implementation usually give good reliable performance - some even without the reduced heating phase.  

Then there are some manufacturers that have thrown the Bosch book out the window and do some really unique sensor control/interface in an effort to get better response time but they often consequently get very poor life too.  Inno* is one such example that seems to fall into this category.   

Link to comment
Share on other sites

Hi- If its any help with regard to 4.9 sensor life - i have an ecotronz ALM with can set up on a link G4+ . I have run this on my car for a year now ( weekends mainly ) and it has worked faultlessly . One feature worth noting is the unit has a 30 second delay built into the sensor heating during which time the display on a separate gauge shows battery voltage . Cant say for sure if this has helped sensor life but it goes towards what others have posted .

 

Link to comment
Share on other sites

Hi all,

a very big hurdle as far as the use of the very much preferred NTK sensor is simply the availability of the information as far as the software and the code to write the software to the NTK control chip.

This information is simply not available to who ever asks for it making it virtually impossible to write to the hardware chip that the NTK sensor is controlled with.

I have used lots of different brands of wideband meters for many many years and have had my share of sensor failures, mine are more so just plain worn out, use of leaded fuels such as avgas, 

Excessively rich mixtures on a hot sensor that destroys them as they get cold shock when very hot.

Dropping them, melted wires, and the biggest one the unknown, loan it out , used to work, get it back doesn't work anymore.

Letting the battery run flat with the sensor hooked up.

Lots of things will destroy these sensors and or shorten there life span, incorrect angle in exhaust allowing H2O to come into contact with the heated element causing failure.

As we all know H2O is a by product of the combustion process and can not be eliminated.

So unless you have millions of dollars to get the chips coded and a full reconfiguration of the main ECU or another CAN MODULE i doubt we will see much happen here.

The best advice i can offer is to avoid the above mentioned sensor killers.

Calibrate your meter regulary, avoid over rich mixtures.

Poor tuning of engines that cruise at 11:1 instead of stoich or a little leaner if possible will save the sensor life also.

Also in this day and age, is that really a genuine, LSU 4.2 or LSU 4.9 sensor.

The other issue is if the sensor is connected to a controller then an analogue output voltage from the module to the ECU is not of high standard, the sensor may be fine and the meter is the culprit.

And this is where we come full circle back to the Innovate LM-1 with no analogue volt output earth/sensor ground.

Earth looping.

This issue was big enough for Innovate many many years ago to design and build a GIA, GROUND ISOLATION AMPLIFIER to stop ground loop feed back effecting readings.

There are wideband meters on the market running NTK sensors, with the intellect (i definatly don't have it) 

and the ability to buy the chips,maybe the code could be reverse engineered to get the base code to program the chip, who knows.

Anyway just my long 2 cents worth.

Regards

Dave.

P.S BY THE WAY DAMIEN KING THIS IS YOURS HA HA HA HA, SHOULD BE THERE SOON.

INNOVATE GIA.jpg

Edited by Dave Kriedeman
Added photo of GIA
Link to comment
Share on other sites

  • 11 months later...

Dave, you are right on the money when it comes to sensor failures. 

Just to add my 2c

I have a Ecotrons CAN Lambda with their supplied Bosch 4.9 Sensor installed in my GTR. It's about 2-3 years old and it is still running on the original sensor. It's mounted at about 2:30. Not ideal, but the best I could do. No issues though, so hopefully it lasts even longer.

The Link CAN-LAmbda seem to be okay but I have two customers that are having sensors fail prematurely. One that I will be requesting warranty on shortly actually.

Innovate..... I will never purchase another one of their products again.. LM2, MTX, etc, all rubbish. Controllers fail, Sensors die etc. I've got two dead LM2 controllers. Cost me $50 to freight them to the US, and they sent them back and said there was nothing wrong with them. Plugged them in and they failed. 

I agree 100% with you also regarding battery voltage dropping while they are plugged in - They die! 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...