Air fuel meter for jetting

[Jdedad]

Anyone tried using air fuel ratio meter for jetting ?

http://www.autospeed.com/A_0217/cms/article.html

http://www.scirocco.org/tech/misc/afgauge/af.html

[GPracer2500]

Great topic!! O2 sensors are the way to go for getting the best a/f data. Forget reading plugs. Forget jetting by "feel". Measured a/f ratios are where it's at. And I love the idea of a low-priced DIY system.

BUT.....a wideband O2 sensor is needed for the kind of tuning we want to do. This is the type of O2 sensor used by dynos equipped with a/f meters.


Both of those guides are essentially for making visual volt meters for use with a narrow band O2 sensor. The vast majority of cars use narrow band O2 sensors. All they are really good for is telling you if you're rich, stoich, or lean. They aren't suited to discerning levels of "rich". And that is what we need. When tuning for best power a/f ratios we need to accurately measure the rich side of stoich because best power comes at mixtures slightly rich of stoich.

The second guide suggests that a simple home built meter for reading a narrow band O2 sensor can achieve the resolution range useful to us. He compares his unit against a commercial unit to develop his a/f ratio vs. voltage table. That is misleading! Both his homebuilt unit and his commercial unit are sensing from a narrow band O2 sensor. Narrow band O2 sensors are not capable of resolving the a/f ratio very far from stoich.

Narrow band sensors are more like a switch that "flip" between lean and rich. They might output small differences in voltage once we're in the rich range, but not with nearly the accuracy we require.

Cars with closed loop EFI systems make do with narrow band sensors because they only need to know: "am I rich" or "am I stoich" or "am I lean". They constantly oscillate back and forth between rich and lean. They don't care how rich they are--once they detect any level of richness they lean the mixture until a lean signal is produced. Then they richen until a rich signal is produced and the cycle starts all over again. This can happen many times per second and it's through this scheme a car's EFI system controls the a/f mixture when in closed loop mode.

But that doesn't help us much. A narrow band O2 sensor could tell us if we are above or below stoich--which is something, I guess. But that won't help you get your jetting precisely correct. It might help you keep it from being way off on the lean side though. The resolution of a narrow band sensor isn't good on the lean side either, but being lean of stoich by any amount is good to know.

I suppose one could develop a technique for jetting with a narrow band sensor. You could jet down until the meter showed lean of stoich and then richen slightly. But you'd never know for sure how much richer you'd need to go. Once a narrow band sensor is in the "rich range" it looses ability to accurately distinguish between a little rich and way too rich.

What we really need is a wideband O2 sensor that can accurately measure levels of richness. Those sensors need a controller to maintain part of the O2 sensor at a specific temperature. You could build one of those yourself but it's not going to be the simple device found in those two guides.

You're going to wind up with something that looks more like this (these are both the guts of wideband controllers)....
http://www.plxdevices.com/M-Series-Controllers/M-300_PCB1.jpg

http://wbo2.com/2d1/im/2d1buyoptstny.jpg


...instead of this....
http://us1.webpublications.com.au/static/images/articles/i2/0217_17lo.jpg


I've kept my eyes open for a low cost home-use a/f meter that would actually be worth having (i.e. wideband) for some time now. There are a few "popular" ones (none are really all that popular) such as the Innovative LM-1 and Dynojet's WideBand Commander. I'm sort of keen on the PLX Devices M-300.
http://www.plxdevices.com/products/m300/M300.jpg

I'm not aware of any wideband a/f meter option that doesn't cost a couple hundred bucks. :mad:


Again, great topic. I wish home-use wideband a/f meters were more commonplace.



[note: Throughout this post I referred to rich, stoich, and lean. I'm using rich and lean to describe above or below a stoichiometrically correct mixture. But remember that rich and lean are relative terms referenced to some "changeover point". Depending on the context of the discussion, that changeover point could be the stoich a/f mixture or the best power a/f mixture, or even some other a/f mixture. When talking about O2 sensors, stoich is the appropriate changeover point between rich and lean.]

[zrpilot]

I have been wondering the feasibility of using the Innovative LM1 as well. Here are my questions:
1) IS it possible to get accurate A/F readings using a tail pipe inserted tube (like 3 feet) attached to some kind of remote mounted O2 sensor?
2) I'm assuming once the sensor issue is resolved that one can simply ride the quad and study the A/F readings off line?
3) Are there other measurements that need to be taken as well, such as RPM?
4) How do you know that the engine is being ridden properly to make the reading meaningful? Do you have to "load" the motor or is light Acc/decc OK?
5) Basically, if I bought the Innovative, would it work?

Thanks

[speedjunkie13]

I agree that a wideband O2 sensor would definitely help with tuning issues. However, my question is would you be using this similar to an exhaust gas analyzer and just sticking it down the silencer, or would you be installing it permanently by drilling and tapping into the header pipe?

I'm sure you could take out the sensor and plug the hole once your have it dialed in, but I'm just curious what your thoughts are.

[zrpilot]

I know that sticking it down the exhaust system would require that the pick-up tube be quite long because in a single cylinder engine, there is alot of reversion and that would introduce atmospheric air into the system and influence the accuracy.

[GPracer2500]

I think welding a bung into the head pipe is the best way to do it. But Innovate makes a clamp-on bracket thingy (below) that goes in the end of the silencer. I guess that should work fine--but I don't really know. Going in through the silencer you'd have to make sure there were zero air leaks at the silencer/header joint.

http://www.rocketindustries.com.au/images/brands/innovate/IM3728.jpg

You might even be able to add-on a tube to the end of that adapter so it extends further towards the engine. [shrug]

I imagine calibrating the sensor is trickier with any kind of tailpipe insert. [another shrug]

[zrpilot]

Originally posted by GPracer2500
I think welding a bung into the head pipe is the best way to do it. But Innovate makes a clamp-on bracket thingy (below) that goes in the end of the silencer. I guess that should work fine--but I don't really know. Going in through the silencer you'd have to make sure there were zero air leaks at the silencer/header joint.

http://www.rocketindustries.com.au/images/brands/innovate/IM3728.jpg

You might even be able to add-on a tube to the end of that adapter so it extends further towards the engine. [shrug]

I imagine calibrating the sensor is trickier with any kind of tailpipe insert. [another shrug]

This is from the Inovative Web site Forum board there:
"Logging Harleys(which have a deceent overlap and large reversion effects on the first 6"-10" of their pipes) prompted me to tap the end of my exhaust clamp to a 1/8" NPT thread, I then made different lengths of copper tubing extensions to install in it(depending on the exhaust and distance I need to reach). There's some other aparatus(vacuum system and free standing mount) I made for static testing(be it bench idling or on the dyno)... but just the tubing(copper self anneals and will go up any pipe shape) getting up in the exhaust cleared everything up beautifully for road tests(idling at lights, etc.).

[zrpilot]

Originally posted by GPracer2500


I imagine calibrating the sensor is trickier with any kind of tailpipe insert. [another shrug]

From the manual for the LM-1

"The meter will now calibrate itself by using air as a reference gas with known oxygen content.
After the calibration period is over (2-3 seconds), the instrument is ready to operate."

[Jdedad]

OK I'm back again, I did not expect such a detailed response so quickly you got me off guard so basically after looking at this closer it seams the narrow band sensor is only able to measure +/- .5 around ideal "14.7" with 0-1 volt range ware the wide band can Accurately measure 9~20 to 1 with 0-5 volt range with fast response time, however a light in the dark is better then no light at all.. that said our good buddy's at K&N are pushing a narrow band one wire version claming to read 12~17 to 1 the same range as the DIY kits I have seen, I am a Process Development Engineer for the Hard Drive industry and I can tell you from first hand experience things are never what they are rated to be, that can be good or bad, meaning a sensor with a spec only intended to work narrow band around ideal may also work a bit more then that only the vendor will not grantee it, most components have a over engineering factor of safety "FOS"built in that ensures the devise will work safely in the range intended this number is a ratio used by the government and others to put a spec on the point of failure, I have no idea if this is the case here since this is used as a smog control devise I assume it will have something like this, I have looked a many analog wave forms in the past and I have never seen one turn off and on like a switch, I suspect its some ware in-between the two, one other observation is all of the graphs I have seen are hand made depicting the theoretical operation, not the true waveform screens shots.


http://www.knfilters.com/airfuelmonitors.htm