Forum Replies Created
-
Posts
-
Yes, the computer will store fuel trims over a range of RPMs, I would imagine this is stored in a similar table to how the normal fuel calculations are done (watch some videos on how tuning is done, for example).
However, it is important to look at both STFTs and LTFTs at the same time no matter what.
I’m all for doing everything DIY, but shouldn’t a brand new Odyssey be under warranty? Why not take it to the dealer and have them fix it for free?
I don’t know if there would be any circuit diagrams available for it, but it is possible something is damaged on the backlight display. If you were going to attempt to repair this yourself I would suggest Louis Rossman’s videos on electronics repairs. He mostly does iphone and macbook repair, but has a really good free course on repairing electronics like that.
That being said without diagrams you would be fumbling around a bit unless there was an obvious burnt component or cracked solder joint on the circuit board. I suppose you could try to find one of those old school front light adapters that were popular on the Gameboy Advance as it was impossible to see anything on the screen at night? Unlikely it would fit as-is and would need to be modified, or if you wanted to get really creative and had access to a 3d printer you could DIY one.
Some of them were just like a book-light (which would probably decently as well), others had a built in magnifier too.
These codes are related to the electronic throttle body. You could review the 2010 Sebring video on the free channel or the “Inside the box” episode on the same car in the paid video section to get an idea of the direction you should go in terms of determining which part has failed or otherwise is causing an issue. It could be the throttle body, or one of the various position sensors related to the electronic throttle system, or an issue with the wiring or connector(s) as opposed to the sensors themselves, but no one can tell you where your issue lies with codes alone.
Not that it helps much, but there are “california approved” catalytic converters, still over a thousand dollars for the ones that come right off the block though.
The rear O2 sensor will bounce a little as far as I know, but should stay consistently rich and not follow the usual oscillation you would see on an upstream narrowband o2 sensor.
As I was discussing in another thread regarding a Mazda I do believe that at least some cars will use a rear o2 sensor will “calibrate” a front o2 sensor, but I haven’t found much information regarding what cars use this strategy. Are you seeing a big discrepancy between bank 1 and bank 2 rear o2 sensors?
Hi Tom,
New codes will not overwrite old ones, but if the conditions that set a CEL clear out they will likely clear themselves out if the conditions are no longer met. In this case we are no longer in a lean condition, we are in a rich condition, at least at idle.
So now we have a rich condition only at idle, the fuel trim may be just under the point where it will set a CEL. Is your STFT doing anything when you’re not at idle or are both at ~0%? I ask about the STFT because I believe that a reading of 6.45mA would be very lean, and -4.39 would be rich on deceleration, which I would think that on decel you would see a very lean condition (fuel cut).
What I was reading online is that the rear o2 sensor can be used to calibrate the front O2 sensor’s stoich reading, I have no idea if Mazda implements this in their ECM programming, but it is a possibility.
As for graphs, if you’re running forscan on a laptop you could always screenshot your data, or ideally just hit the save button after you stop live data, and you can save it either as a forscan graph or a csv.
I think we need to double check that total fuel trims as it seems like you’re running lean off idle and rich at idle which would make me think dirty MAF sensor… I’m a little suspicious that those MAF readings may be off. Did you ever check your calculated load %? I would like to see that data as well at idle with no accessories on (car warmed up). I assume when you cleaned the MAF you cleaned the 2 very small elements at the top of the sensor?
Hi Tom,
I’ve been thinking about it again recently, and I recently watched a video that mentioned that a rear o2 sensor can actually affect fuel trims on Mazdas. I do vaguely remember reading or hearing (not sure if it was from Matt) that rear o2 sensors are not purely for emissions/confirmation of catalytic efficiency. The Rear O2 sensor did seem to have activity on your data, so I don’t think that’s necessarily the problem here, but it is actually a potential variable (one that I believe we have already eliminated).
One area I forgot to ask about in terms of data is calculated load and MAF readings, this is one area that I struggle in as I don’t know the typical readings you would get from a normally running engine, but perhaps we can find something here. What is your calculated load at idle with no accessories on, as well as free revving, and while driving? What is your idle G/s or LBs/Min readings?
If you wanted to experiment yourself, they’re quite cheap. Supposedly there’s not too much difference between the cheap ones and the “name brand” ones, I know a lot of people suggest putting steel wool (if I remember right) into the catch can to help separate the air and oil. You just have to remember that it becomes a maintenance item if you do install one. Some of the more expensive versions come with dipsticks to check the level of fluid in them, but again you have to remember to check the damn thing.
I know that Subaru actually sells a specific “upper cylinder head” cleaner that is supposedly just like seafoam spray, since there is no cleaning of the intake valves they sell it with the description that it is for vehicles that have not kept up on oil changes.
ALSO, I do believe that new oils are formulated (again, this is primarily for turbo engines) to burn up in the combustion chamber without lowering octane levels. I believe Engineering Explained has a good video on this, the topic is “low speed pre-ignition”.
Again, it’s all anecdotal evidence that I have seen, and primarily on turbo engines. I would like to see a video series on this as it is a common upgrade when turbocharging an engine that is not factory turbocharged, I believe that understanding everything about an engine and how to diagnose one is fundamental to modifying an engine/car as well. I don’t know how much of the audience of SB/SBQM is of that same mindset though as it is not the typical focus of the channel other than Matt’s own modified Trans Am that makes an occasional appearance as a study.
The theory is that an oil catch can will collect any oil vapors or moisture that might be flowing through your PCV system, acting as a filter for the crankcase pressures being recirculated into your intake. direct injection engines do not have fuel injectors washing “gunk” off of your intake valves because the fuel is sprayed directly into the intake.
This is especially bad on turbo direct injection engines which tend to have more crankcase pressures that can carry oil vapors that get recirculated right back into the intake. You can see this on videos of engines getting their valves cleaned by being walnut blasted or similar cleaning methods.
You can definitely see where gunk builds up in older engines in the upper intake manifolds where fuel isn’t sprayed. A lot of this is theory, and evidence of benefits that I have seen would be anecdotal at best. As I mentioned though, this is only for DI engines that have no form of port fuel injection, some manufacturers like Toyota and Subaru use both DI and port fuel injection on some of their engines.
Oh yeah, I forgot Forscan was good for some Mazdas as well. I wish I got to use it more, but unfortunately my laptop took a crap. Most of what I’ve worked on happen to be Fords.
Remember that a vacuum leak would show up as a lean condition (increased fuel trims) at low rpm, but improve at high rpm. Sometimes forum posts for a specific make and model can give you direction as an idea of where to start or what to look for with common issues, but they’re not the end all be all.
Also remember that sensor 2 should always show rich except if I remember right at WOT it will tend to follow or trail sensor 1. It is interesting the computer is throwing a code for a “too rich” condition on sensor 2.
You might wait for confirming the data with Forscan, but you could try running live data with only LTFT and STFT and seeing if you can get a reaction out of the propane into the intake.Somehow we went from a very lean condition on and off idle to a rich condition on idle but not off of idle
Hi Tom,
Trying to write this up to hit some of your questions before you get into replacing anything, unfortunately I don’t have much time at this very moment.
This is interesting because according to your new fuel trims, a -20% LTFT with a 0% STFT would indicate a rich condition, not a lean condition, yet you still have codes for lean condition?
I would not replace that O2 sensor without some confirmation it is bad. Remember that wideband sensors do not function like regular o2 sensors. See Matt’s video on wideband sensors on youtube. It may be bad since you’re not getting any change from the sensor even with propane, you MIGHT be onto something here, but I would make sure that it is bad since those sensors are quite pricey.
You will go into open loop on deceleration when you let off the throttle because the car is cutting fuel as the wheels are driving the engine (as opposed to the other way around).
Also, a CSV file is basically an table with data on it, there should be a way to convert this to a graph. I did a quick google and it looks like there are some online ways to do it.
Hi Tom,
I replied to your intro post, but I posted a link to another forum with a suggestion on your MAF housing issue so I believe it has to wait for admin approval first.
It sounds like you are not dealing with a vacuum leak as typically in the case of a leak you will see a fuel trim improvement at higher RPMs. As for why the LTFT isn’t going anywhere, every manufacturer does their programming for when LTFT adjusts differently, so keep an eye on it, but in one case I believe Matt got burned on a reading where he only looked at the LTFT and the STFT was stuck lean or something to that effect, so it’s always important to look at both for a total fuel trim.
I had 3 suggestions for repair on my other post which I’ll repeat here for good measure if you wanted to fix the stripped screws, but the propane test was a good idea, assuming it would have fed enough to reduce the trim, did you try feeding a small amount of propane directly into the intake to see if it would show up on your fuel trims just as a sanity test? The first suggestion was what I linked to on a Mazda RX8 forum where they drilled out the plastic housing and used threaded inserts instead, you can google “DIY: Fix your MAF screw holes for $2.26” and the exact thread I was looking up will come up. I am assuming this will work on the Mazda 6 housing, but you will need to compare to what the 6’s housing looks like and see if this will work for you. My other suggestions were to either melt new plastic in there or use some sort of soft epoxy to fill in the hole, then carefully drill a new, straight hole and use your screws to cut into it.
You could have a lack of enough fuel or a bad reading from your MAF among some other possibilities, but a vacuum leak is actually unlikely in this case (or at least it is not the only contributor). Are there any drivability problems or just the lean condition? What do your MAF sensor readings look like on and off idle? Also, I would graph the O2 sensor signal to see if it is oscillating, it would be odd for it to be stuck at 0.9v when in a lean condition as that would indicate a rich condition or bad o2 sensor, unless you are looking at the rear o2 sensor anyway (B1S1 = Upstream, B1S2 = Downstream).
I guess you can only edit for 5 minutes, so I’ll just double post.
As Matt mentioned the rear O2 sensors should never oscillate, if there’s any factor besides a bad cat that is causing the rear sensor(s) to go lean you could run some live data viewing the rear o2 sensor voltages and try to reproduce it.
I would suspect the cat is right on the threshold of how well it’s working for setting the code. As Matt has mentioned in at least one video that P0420 and P0430 codes are usually pretty accurate.
