+2, but if you understand your wideband well, it can be a very helpful tool.

 

So i looked over 4 LM-1 logs and four dynojet logs. I sampled every 300 rpm from 3000 rpm on. Then I averaged the results for both the dynojet and the LM-1.

The results are the same as my original post. After 5700 rpm the dynojet wideband continues to read richer than the LM-1. I also noticed that the areas where the LM-1 read lean (5700 and up) triggered the most KS on the logs.

Again, the data has limitations. The dynojet logs and LM-1 logs were not done @ the same time and place. The gas was the same and the car was the same.

Attached is the file.

 

Agree, we data log the car everytime we dyno or stree tune the car. also when you check the spark plugs you can see your timing mark
peter

 

Well guess what, I have an early LM1, my lm1 was bought back in 2002 or 2003 forgot exactly. There was never an option that I seen for a ntk sensor. Only thing I see is a jumper selector on the board inside that lets you determine what sensor you are using. There have been some people who have converted the VW connector to an ntk sensor and still wont work properly, probably has to do with the firmware. As far as the sensors being the same, this cannot be true, I have seen them both up close, Bosch is bigger, and also ntk is made by ngk and they look nothing alike. If they are the same, then I hope bosch does not make ngk spark plugs You would have to come with better facts than that to say they are the same lol make me a believer. I have a friend who has had his fjo wideband for 2 yrs and has replaced his ntk sensor how many times, 0 and all he uses is race gas. In those 2 yrs I have replaced my sensor 3 times and his is still as accurate as mine with the new sensor. I do not mind because the VW sensor is cheap so no complaint. One thing that I will say is that Innovative customer service is one of the best I have ever experienced with anything I have purchased. I had my wideband die out on me, missing screws, and also the 3.5mm analog jack was broken. All they said was ship it to us. I get the wideband back in 5 days and it was as good as new with a replaced analog jack. I was impressed, because this was after 3-4 yrs of owner ship, they didnt even ask for a dam reciept and I know my piece is beatup lol.

 

One step at a time. I'd have to look at your specific serial numbers and etc., but NTK support was distributed, to some extent, at one point. Also, NTK data and jigs were used for part of the patent work. There are only two engineers here, Klaus and me, and I can pull all the legacy source code up on my screen. A dual version is noted in the LM Programmer tool source, and NTK support code remnants are still in the current LC-1 firmware

As far as sensors, I'm not claiming they are the same - I'm saying a vendor who is heavily vested in NTK sensors sells a wideband with Bosch sensors, but disfigures the Bosch sensors to obscure this fact. They also sell the replacement sensors at NTK type prices.

I'm also saying that in our (well, Klaus' really) experience, users were not getting any percievable advantage with other sensors. Klaus has always been committed to the seemingly opposing goals of affordability and pro grade measurements. With NTK, we have to switch the connector and we lose the ability to monitor bung temp (a major cause of Bosch's short life reputation). And, although they are significantly more expensive, we found no improvement in accuracy, and almost no improvement in longevity with NTK. So Klaus said @#$% it, and focused on other things.

As noted above, we are heavily represented in pro racing and the related engine building. Were there a significant advantage in our providing NTK support in those markets, I am sure the support would jump up on Klaus' to-do list. We have been looking at another, highly lead resistant, sensor for a specialized application, but I doubt that support would ever go into our distributed commercial products. The sensor is very durable to lead, but very fragile in other ways.

As far as customer support, I can give one fact. As of today, we have yet to charge for a repair. And we are talking about many tens of thousands of controllers shipped. If you are doubtful of my facts, you are welcome to confirm them with Klaus himself. He is readily available on our support forum.

-jjf

 

Did you happen to notice the quallifier "I believe" I love the anger over this. An o2 sensor is a wear item. I don't care if it is a $7,000 Motec controller, if the sensor is old it is not going to be accurate. Load dyno's vs. acceleratometers. All factors that need to be taken into account. No I am not going to spend 5 hours researching it nor did I claim to be proclaiming absolute truth with blanket statements like it is a very very bad wideband. You win

 

The last thing I want to do is to join a squabble, but you are touching on one of the points I have been trying to make.

In a big picture sense, you are correct. The sensors age and can also get contaminated. Lead literally forms a silica type layer, water can damage a sensor, and so on. But unless you understand how your particular instrument responds and behaves to these sorts of changes, how do you plan a strategy to address it?

In other words, do you just perpetually guess, or do you approach it systematically? It seems to me that if you just guess, confidence in readings should always be pretty low. On the flip side, if you employ a strategy based on a particular instrument, I think confidence can remain pretty high.

I've had some experience with this myself. Awhile back I did quite a bit of work with lambda measurements on small airplanes. The engines run 100LL, which, despite the "LL" designator, is quite a highly leaded fuel. They also generally run pig rich for massive chunks of time. Bosch sensor life is rated in this fuel for about 200 hours, but the spec does not consider how rich these engines run. Actual life is more like 70-100 hours, and a small number of sensors fail in a much shorter time period (I think possibly because of shock cooling, but I've never really experimented to determine precisely why). For my research, I needed good readings. Consistancy was more important than absolute accuracy, but either way, I needed a predictable amount of variation in my measurements.

In automotive use, sensor management with our widebands is pretty simple. But I did not want to assume that what I knew works in cars would work on an airplane. So, I elected to confirm my sensors with calibrated gas before each flight. Only after going through quite a few sets of sensors did I decide to trust the error mechanism in the wideband to monitor sensors for my experimentation. After that, fine, use the sensor until an error is thrown. When an error is thrown, recalibrate. Once the sensor can no longer be calibrated, throw it in the box (and try not to think about just how many sensors my little research project drowned in lead!) But the key thing is that I developed and confirmed a strategy that consistantly yielded me usable results for my purposes.

I'm not saying that everyone should run out and buy calibrated gas, but I am saying you'll potentially get a lot more out of your wideband if you have a strategy to deal with the limitations of your particular instrument. The one strategy I would not use with sensor life is to look for 14.7 at idle. Stoich is the one thing even the mungiest of sensors seem to always find well (this actually makes sense when you look at how the sensors work). But there are other strategies, for example, at the same location, under similiar ambient conditions, with the same engine, you could cross check to several specific EGT readings (assuming you use the same timing each time as well). But again, a good strategy probably starts with understanding your particular wideband.

-jjf

 

Very good points and well said jfitzpat . It seems like methanol may even shorten the lifespan of these sensors further then leaded fuel. At least this is wht I have heard from dyno operators that have reported going through sensors more since they have been tuning cars with meth the last couple of years. Once again so that no one gets their panties in a wad. This is what I have been told by a few people who regularly run a dyno. Not definitive evidence, just secind hand info. I know my own VW sensor only last 10 months or so before dying with my UTEC tuner on my car.

 

I have actually heard this annecdotally as well. We even kicked around some possible theories. However, in some limited testing I haven't seen any data to really confirm it. But again, the data I have is limited. I may not have a big enough sample to see the effect or, it may be a simple thermal issue.

People doing water and meth injection are typically looking to effectively raise octane and run more boost, etc. So there is often more heat. I doubt that people are typically exceeding the rated max gas temperature for the sensors, but it is pretty easy to exceed the rated max bung temp for the sensors. An often overlooked part of the Bosch spec. It is typically about 300 degC cooler.

Exceeding this temp scews your readings a bit and also really fatiques the sensor. I think the fact that we monitor both temps and throw an error if _either_ rated max is exceeded (and it is typically the lower max bung temp which is a problem) is why some folks claim we are 'overly temperature sensitive'. They are right, you can typically run while exceeding the Bosch spec in this regard, but at the cost of a little accuracy and a lot of sensor life.

However, while exceeding max bung temp does happen, and it does wear down sensors, any connection to meth injection is pure speculation on my part. Again, I don't have enough data to really say.

-jjf

 

Hi,

Joe does not know but I have some data on methanol engines. Most tuners of those engines run them pig rich. Rich in the sense of Lambda, but also of course in the AFR sense as methanol has a stoich AFR of only 6.4. Methanol also has, compared to gasoline, a very high latent heat value. It takes a lot of heat energy to vaporize it. As a consequence, on a cold engine, liquid methanol literally runs out of the exh. ports on those engines. The liquid methanol hitting the hot sensors causes thermal shock and the sensor ceramics shatters. This is the same failure as with water shock. That's why the sensors die quickly in methanol engines when no special measures are taken to protect the sensor from liquid methanol.

- Klaus

 

Thanks Klaus! Thermal shock was my theory for the occassional short live sensor on the Lycoming TIO-540.

-jjf

 

Finally a logical answer. I never really understood why they were wearing so quick. This certainly explains it. Thanks

 

I am happy to see Klaus on the EVO forums. Welcome to my thread