New wideband controller ALM compared to Innovate LM-2

[Fred]

Edit: When I saw the "150ms faster than the LM2" I thought; holy sh*t someone has invented time travel, because the lm2 is ~100ms response time, guessing that based on my testing with the the LC1, so the ALM controller is reading AFRs from the future.

ROFL! :-)

[jharvey]

One thing I wonder is what the actual gas is when the sensor is reading. Basically I see several sensors that each claim to get to a reading, but I don't know what the actual gas is, so how do I know which sensor is more accurate, we could simply be looking a sensor variations.

In other devices I've see the use of acoustics and a small pinhole used to determine the difference between known gasses. It's commonly seen when purging air from a newly installed propane system in a house. You crack the valve at the feed slightly, and listen for the whistle, then when the pitch changes, you know the air has been purged and the propane is present.

I don't see that as handy for a road application, but it might be handy for bench test application. It sounds like the system is slightly pressurized. I wonder, if a small hole is added before/after the O2 sensor, could one notice by ear a change in pitch based on gas. If so, then the pitch could be logged with a microphone/scope/what ever to verify what gas is present.

[dr.mike]

I am quite convinced that regardless of controller design the bosch lambda sensor is not capable of resolving individual cylinder AFR at the collector, that is the holy grail. With that out of reach, resolving response time further is of no tangible benefit.

I have personally done this. Specifically, tuning a Ferrari 365GTB/4 tipo 251 engine. It is arranged as 2 banks of straight-6. It has 6 dual-throat Weber 40DCN carbs. I can close off the idle feed to one carb and see a full swing to ~free air for that cylinder at the collector up to about 2000rpm.

The limiting factor in sensor response speed is the diffusion rate into the measurement chamber. The more current the pump has to use to bring the chamber to 1.0 lambda the greater the diffusion rate. Assuming the pump is operating properly. The calibration resistors on the Bosch sensors are not actually to compensate for the pump. It is to normalize the manufacturing tolerance in the diffusion rate.

[toalan]

2k rpm is not enough IMO, in the best case scenario with a new sensor you can do 2k, but once the sensor starts to age the response time will suffer greatly. On my 4 banger I can resolve cylinder AFR at idle, I suppose I can use that data to figure out if I have problems with the injectors, going up to 2k is not going to give me anything more tangible. If I could go upto 8k rpm, then I could close loop control the thing and never have to touch a table again.

My approach is that if people want to measure individual cylinder AFR, then they can do so by running multiples of my lambda controller modules using I2C. I price the modules cheap enough to make that practical, but once you consider the cost of replacing multiple sensors, it is something that only diehard fanatics will do.

Yes, the calibration resistor is to compensate for variances in the size of the diffusion chamber, which directly effects the diffusion rate.

[dr.mike]

It's probably more of a "carb thing". But 2000-2500rpm will usually get you through idle and transition, and on to the main circuit. Which is about all you need (mostly). If you are looking to detect high speed misfiles, then, even a NB may no be fast enough.

High-speed distribution, we always did with EGT.

[antonmies]

LSU 4.9 is designed for gasoline, diesel and e85, LSU 4.9D is for diesels. It's the same sensor with different shell. Datasheets. I'd say that this legend that LSU 4.9's are for diesels only are FUD propaganda from big fat and lazy aftermarket WBO2 manufacturers not wanting to make further developments until totally necessary and now they start to get carved like a turkey. Their products are starting to be less advanced than automotive OEM and still selling it to the believers. I guess they still sell so many carbs in da states because of this "old tech is better" thing going on.

You can clearly see in the graphs that LC-1 starts to react at the same position than the ALM (therefore it's falseful to say it's due to pipe length) and always giving the right reading after the bigger transition phase is over. ALM measures the transition more realistically whereas LC-1 acts like a narrow band lambda.
If it would be constant lean-rich-lean.. I don't think LC-1 could even keep up with LM-2 already guided out of the playgrounds to eat pizza.

LSU 4.2 is a good sensor, but it's now being superseeded.

If you have any measured data similar to Ecotrons, please post it. No FUD, no assumptions, no speculations and no BS. Just business!

And if you see something wrong with Ecotrons measurements, make it better and document yours. They tried different ground spots, and even a 12VDC power supply to eliminate false readings and noise.

[Harry193]

For this reason , almost all wide-bands are really only useful for measuring steady-state AFR. i.e. constant speed / constant load.

Right there! after many long posts trying to tell people how fast the LC-1 is or how fast LM-2 is, now you are telling people, wideband controllers only work for steady-states.
Please tell this to professional engineers who work at GM and Ford. Because for them, the precise AFR readings are more critical in dynamic conditions than steady-states, because those are conditions under which the most emissions are generated. That's where the response speed and accuracy are really making difference. Keep in mind, the O2 sensors are created for emission controls at the first place.

If a wideband controller is only useful at steady-state, that's a garbage already!

ALM has been used by customers as a feedback device to control the AFR in real-time. check here:
http://www.ecotrons.com/ALM-helps-our-c ... tions.html

By the way, this is Ecotrons, I have to re-registered with a different user name to login. I can not login with my "ecotrons" name any more.

[toalan]

The only pro the 4.9 has over the 4.2 is that the nermest cell resistance is 300ohms @ 750C, the 4.2 is 80hms @ 750C. This means you can more precisely control the temperature of the sensor.

Both have the same lambda range and the same accuracy.

The 4.2 has larger holes for gas to enter, so it is a little bit faster responding.

I build units to work with the 4.2 and not the 4.9 primarily because in north America you can get the 4.2 for half the price of the 4.9. the other reason is that I have thousands of 4.2 connectors and pins already purchased. I have modified my units to work with the 4.9 in case 4.2 supply ever dries up or gets more expensive than the 4.9, all you need to do is change the ip vs 02 curve in firmware, change your target nermest resistance, and an overall reduction in the P I D constants controlling the heater power output based on the nermest resistance because the nermest resistance in the 4.9 is much higher.

My guess as to why Bosch is pimping the 4.9 more heavily than the 4.2 is from a production costs point of view, the 4.2 is heavier and takes more material to build, the cables are generally much shorter on the 4.9. They also save a few pennies shipping a smaller lighter 4.9 vs a 4.2.

[Harry193]

Since 2007 or earlier, all OEMs in the world start to use LSU 4.9 on their gasoline vehicles.
It's funny to see that there are still people trying to tell Bosch:"Hey, I know your sensor better than you! LSU4.2 is better than LSU4.9..."
In fact, LSU 4.2 is only sold in after-market now, no matter where, North America or Europe. No OEMs use it any more.
LSU 4.2 is cheaper, why? because it's obsolete.

[Fred]

Harry, econtrons, diyhansen, how many usernames do you need? You're clogging up the database man! If you register a fourth one I will delete all of your accounts, clear?

For the record, both of Harry193's posts were invisible until now, so Alan didn't have the privilege of reading the first one before he wrote his post, despite how it appears now that I've approved them.

As for credibility, I know who I trust, and it's not the guy who pretends to post about his own product as someone else, who then signs up to reply to himself, and then again as another name some time later. It's Alan, who has always been equally critical of his own products, just as much as those of someone else. Keep digging, Harrysen.

Fred.