Triumph Rat Motorcycle Forums banner

O2 Sensor location problem

6K views 32 replies 8 participants last post by  bigbofo 
#1 ·
So, I bought this nice Innovate A/F meter kit with gauge and all.

G2 Gauge / LC-1 Kit - P/N: 3801 - #3801

So far so good. Now comes the installation:

It requires a O2 sensor to be welded into the exhaust system. The manual and customer support recommend to put it into the header pipe, about 12" away from the head.

Now, as we all know, we got THREE header pipes, all could potentially give a different a/f reading due to O2 content of exhaust.

For tuning purposes, this would mean you need to install THREE bungs (one in each header) to allow an accurate exhaust a/f examination.

This would leave me with the dilemma of either installing said three bungs and swap the sensor over to monitor each cylinder exhaust individually and ruin the look of the header pipes (and bike as a whole) - or settle for a sensor installation in the rear after the collector of the 3-into-1 system and get (potentially) an inaccurate reading (which could be pretty meaningless overall).

Question to the community - has anyone installed this LC-1 thing successfully? If so, where did you place the sensor (and in addition to that - how and where did you mount the gauge?)???

Any assistance would be greatly appreciated.
 
See less See more
#3 ·
No experience or special knowledge to offer but I'd have thought after the collector is getting too near the tail pipe, esp if the silencer is just a glass pack type?

But surely fitting to one cylinder should be fine? I'm thinking carburettors are much less precise & much harder to fine tune than fuel injection in any case, so not sure what could be achieved with each cylinder?

I guess you want to ensure everything is tip top in the carbs/coils/plugs/valve clearances etc. to be sure the gauge info is consistent. If you haven't already, I suggest fitting some Iridium plugs - they seem to stay very clean & reliable imo.

Be very interested to hear what results you get if you fit it :)
 
#4 ·
ok i might be way off here but am thinking u want to mount the sensor in the header pipe as far as my t/bird we have bolts installed in headers for this purpose but they are arnd 6 inches from the head port not sure if this is what u neeed or if the headers can adapt to others in the tripple range
 
#5 · (Edited)
Unfortunately, the bolts, intended for CO sniffers, are much smaller than the wide band O2 sensor (18mm, IIRC).

I've been wrestling with the same question and am leaning toward the #1 pipe, a few inches before the balance pipe. This should give a reading primarily for the #1 cylinder and should make it easier to isolate any effects. (The main reason I'd do it is to aid tuning.)
 
#6 · (Edited)
Interesting!

It sounds like you have a good idea. Keep in mind what you want to accomplish. If you are getting proper jet sizes and needle positions, would you install different jets in different carbs due to a different reading between the headers? If not, then one is probably good enough.

It would be interesting to get readings from the bung you install, then take the sensor and put it in the muffler exit and see what difference you get, if any.

One idea that MIGHT be worth checking is an adapter for the CO sniffer bung. If the sensor does not have to be right in the stream, that might be good enough. The ratio of the different gases is probably going to be the same just outside the stream, within an adapter, as it would be within the stream.

Another thing would be to simply replace the CO sniffer bungs with the larger O2 sensor bungs. If you ever needed to use CO you could presumably get an adapter to go the other way. This is assuming your CO bungs are not being used...

It might be an idea for someone to buy one of these and rent it out to others on the forum so they can get their jets right, assuming a muffler reading is good enough. 3 or 4 rentals might pay for the thing.
 
#7 ·
It might be an idea for someone to buy one of these and rent it out to others on the forum so they can get their jets right, assuming a muffler reading is good enough. 3 or 4 rentals might pay for the thing.
I was thinking the same thing. This could be a very valuable tuning tool if it works; and you really only need it once.

I would try an adapter in the existing CO bung on cylinder 1 first. That would be the cleanest way assuming it can pull a reliable reading. It would also make it easy for renters to use it.

I wonder if Bosch sells a comparable 02 sensor with threads that match the CO bung?
 
#8 ·
I decided to get one myself, a different model that is not as expensive:
http://www.innovatemotorsports.com/products/MTXL.php

Amazon had it for around $170.

It requires a O2 sensor to be welded into the exhaust system. The manual and customer support recommend to put it into the header pipe, about 12" away from the head.
The manual for the one I bought said it was acceptable to shove it up the tailpipe as long as it is far enough in to not be "contaminated" by clear air being drawn into the pipe between exhaust pulses. That is how I intend to use it, just to get my jetting right. I am not using it for a permanent installation. I might be talked into lending it out if I can find a way to protect myself from it being broken when in someone else's hands. Probably charge a bit to pay for the bother, along with a deposit. Or I might just sell it at a reduced cost to the next guy in line. I will have to see how I like fooling with it and what vehicles I can use it on.
 
#10 ·
The manual for the one I bought said it was acceptable to shove it up the tailpipe as long as it is far enough in to not be "contaminated" by clear air being drawn into the pipe between exhaust pulses.
Something struck me as wrong about this. Today, in a flash of insight, I thought I had it. After doing a little research. I'm convinced it will not work. As it turns out, the O2 sensor works by comparing the O2 content of the exhaust to that of the atmosphere. To that end, the external part of the sensor has to be exposed to atmospheric air. How it would work shoved up the tail pile escapes me.
 
#11 ·
Perhaps only some kind of sensors work that way. I did not dream this up myself; I got it out of the manufacturer's manual. Presumably they would know. I have to admit some doubts myself, i.e.
1) Will the pipes melt the sensor wire insulation?
2) Will the diameter of the sensor even fit, or if it does will it create unwanted back pressure?
... and so forth. So I will just have to see how this goes. I will let you know.
 
#12 · (Edited)
I did not dream this up myself;
I didn't mean to imply that. I knew you were quoting the literature. Still, I have my doubts... Wouldn't be the first time marketing and engineering weren't on the same page...

(Checked their on-line support info and didn't find a statement to that effect, although they do mention an adapter that allows you to install the sensor into the end of the tail pipe. Optional at extra cost, of course...)
 
#18 ·
I am pretty sure that is not how it works. The voltage generated by the sensor is proportional to the difference in O2 content between one side of the sensor material and the other (an oversimplification...). The external part of the sensor is vented to the atmosphere.

The calibration procedure is just to calibrate the CPU to the value generated when both sides of the sensor see the same amount of O2.

Here is more info than anyone I know could ever want on wide band sensors.
 
#19 · (Edited)
Checked their on-line support info and didn't find a statement to that effect,
The manual says, "Alternatively, you can also use the optional exhaust clamp (part number 3728) to sample exhaust gases at the end of the tail pipe." See page 5:
http://www.innovatemotorsports.com/support/manual/MTX-L_Manual_1.1.pdf

Since the thing really doesn't fit in the end of the muffler anyway, I may try to make a small adapter, a short piece of pipe, that fits into the hole and have the sensor mounted to it at right angles which would keep the outside of the sensor in the air stream rather than inside the pipe. That would take care of any problems you mention (assuming there are any) and allow me to easily switch between the two pipes. Only other things I'm wondering about are 1) air leaks at the muffler/pipe junction, and 2) is the sensor designed to run hot, and would it have trouble heating up enough (since they mention end of pipe readings this seems unlikely to be an issue).

<later>
Ah, on reading the link you supplied it says the sensor has a heater to make the temp correct, so I don't have to worry about that. I suppose I could use a little high-temp silicone to fill any holes at the pipe/muffler junction.

One other thing I wonder about is the reading procedure. Those charts you see, are they done with the engine in a no-load condition? Obviously, one has to open the throttle more with a load, than one needs to without a load, to maintain a given rpm. And the amount of load matters too. Different throttle openings translate into different carb circuits in use. I suppose a load is needed anyway to check the main jets as I don't think it is good for an engine to have the throttle pinned while sitting in your garage! I just need to find a place where I can get away with going that fast...
 
#22 ·
I am not an expert by any means, but dont you want to achieve a 14:1 air/fuel ratio (or close to) under all conditions for ideal performance? At least that is how I understand it...

So you got main/pilot jets, shims, fuel screw and more to achieve that and play with.

The whole purpose of this thread was to ascertain the best procedure on WHERE and HOW to install the O2 sensor for BEST results.

I believe consensus is to get the best reading from the header pipe about 6-12 inches from the head. Which means you would need a bung in each header pipe for most accurate tuning. If you just put a bung in the collector or the muffler you will achieve a composite reading from all three cylinders - which could be close enough if you are happy with an approximate reading.

Could be completely wrong here, but this is the outcome of my research. Happy to be challenged any given time. :D
 
#23 ·
After having a read thru' the Innovate site, it looks like the sensor could be much further back. The accessory item on the far left of this picture:



is a device they offer to fit/clamp the sensor in the end of the tail pipe. That suggests to me that any point, say, on the header side of the muffler should give reliable readings? Innovate also seem to say that the sensor is best in a cooler area of the the exhaust flow so that it's own heater control works better to keep it at the correct temperature.
 
#24 ·
Yeah, the adapter pictured above snags a bit of the exhaust from several inches up the pipe and feeds it to the dirty side of the sensor, allowing the clean side to stay in the atmosphere. Should work fine unless it's a 500CC single.

And, yes, you have to load the engine. CV carbs move the slides in response to air flow, so without a load you won't get much past the pilot transition circuits. A dyno is ideal, but you can do it on the road, too, especially if you can capture the data stream and somehow correlate it to RPM and throttle position.

I'm pretty sure all wideband sensors have integral heaters.
 
#25 ·
Yeah, that makes sense. Probably get on a long straight road (without cops), sit up straight in top gear for max wind resistance and set the tach at each thousand rpm and take a reading, maybe speaking into a voice recorder. Not sure how to figure out what the throttle position is, though. I happen to have a throttle lock so maybe I could make some marks on it or something.

I was thinking of putting together some kind of logger with one of those Arduino boards, since the guage does outputs for a logger, but this project just keeps getting more and more complex so I may just try some simple-minded manual method as I just described.
 
#27 · (Edited)
I was looking at MikeIrl's thread on adding a throttle position sensor. That would be ideal for logging, since it is just a voltage from a resistor or similar. I like that idea too although hooking into the throttle shaft is a pain and I'm not yet buying a new ignitor so it would be kinda pointless. I don't know about manifold vacuum, maybe I can find some pressure sensor somewhere. Of course hooking that in would be trivial, just use the sync ports. Thanks for the suggestion.

BTW what lifts the slide higher, high manifold vacuum or low vacuum? I need to re-read about CVK carbs too...

Just got my new first-attempt jets. Put the needles up two washers and put in 115 mains. That should do for a start. I'll be able to play a little with idle mixture but with winter I probably won't be doing much full throttle testing for months. Also it's about 2 hours drive to the other side of the Cascades to find a nice safe place for testing. Maybe I'll spend the winter messing with my logger.

I think I figured out a simple/cheap device to get the wideband sensor sampling exhaust. I'll give that a try too and post here when I try it out to see if it works.

<later>
Ouch! MAP sensors are expensive! Maybe I can find a junkyard sensor. Or maybe I should just go to get a dyno test done!
 
#29 ·
BTW what lifts the slide higher, high manifold vacuum or low vacuum?
CV stands for constant velocity. The slides rise in response to the drop in pressure caused by an increase in the speed of the air passing through the throat of the carb. The rising slide causes the air to slow, until a new equilibrium is reached between the low pressure (above the diaphragm) and the spring pressure. High manifold vacuum (low pressure) correlates to a low volume of flow and lower velocity, so less slide lift.

Junk yard MAP sensors should be easy to find. They are also used to monitor vacuum control on emission controls, like EGR valves.
 
#28 · (Edited)
Paul, if you go for a MAP sensor, be sure to get a an analog one (giving a 0 to 5V dc output). Some of them give a frequency output. A 5V regulated power supply would be needed to power it too. One of those auto USB power supply (5V) converters would do I think. Here's a handy info link:

http://www.crxsi.com/info/driveabil...ns/Manifold-Absolute-Pressure-MAP-Sensors.htm

And a data sheet on some common Bosch analog sensors:

http://www.bosch.com.au/content/language1/downloads/Map_Sensor_Technical_Specification.pdf

I think some sort of damping arrangement in the connecting vacuum pipe might be needed also to damp out the intake oscillations. I suppose something like the constrictor tube that carbtune fit in their vacuum gauges might be the thing?

Oh, and you obviously want one suitable for a naturally aspirated engine. Turbos etc. use much wider range MAP sensors as they need to cover a few bar positive pressure as well. It would still give a signal but smaller with lower resolution.

I must say I'm a bit tempted by this Innovate kit, esp already having the TPS signal that would make logging easy at various rpms & throttle positions. But the signal interface & logger boxes double the cost to over $430 + shipping. It's a bit hard to justify the cost for me, as my bike seems to run well & get very good economy too.

If a temporary tail pipe O2 sensor works reliably on our bikes that might make the difference for me as it would then make it easier to use on other bikes too.

Be interesting to hear how you get on with your exhaust sampling set up Paul?
 
#30 ·
#31 ·
looks like a moderately priced toy for Christmas.
 
#32 ·
Just got an email back from Ecotron saying the later ALM-S does still have the same rpm & aux (MAP or TPS) data logging functions as the earlier unit.

I think a buddy will go halves so I'll probably get the ALM-S with the gauge option.

Need somethingto play with in the off season :D
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top