HarleyTechTalk

Quote from: rbabos on September 06, 2010, 07:23:18 AM

Quote from: FLTRI on September 05, 2010, 09:24:34 PM
Ron,
Do you not believe the intake air temp was actually 150?
Just trying to understand why you feel you need to lie to the system?
Bob

Of course it's 150. That's my point. It's preheated and not constant in temp. This causes the ecm to make incorrect adjustments for a given head temp.Sitting in stop and go traffic on a 90* day having the iat climb to 180+ and lean the afr. Where's the logic here? Wouldn't the tune be more stable in afr if the iat was reading the actual ambient air rather than what the sensor is getting cooked at in traffic. Are you telling me that tuning with the present iat location is not a lie in itself? Wouldn't a bike tuned to a relocated iat be a more accurate tune overall and adjust better to real ambient temps? You have the equipment, why not test it yourself? If I can hear and feel the differences I'm sure you could pull some really interesting data when monitoring the afrs.
Ron

I tried this quite a few years ago.
The ecu knows the temp from the IAT and adjusts accordingly,even at 150*.
Like they say, change the recipe and it's not a cake anymore.
A retune will fix it but it just my guess that you will be back to square 1.

Quote from: Rider57 on September 06, 2010, 08:55:09 AM

Quote from: rbabos on September 06, 2010, 07:23:18 AM

Quote from: FLTRI on September 05, 2010, 09:24:34 PM
Ron,
Do you not believe the intake air temp was actually 150?
Just trying to understand why you feel you need to lie to the system?
Bob

Of course it's 150. That's my point. It's preheated and not constant in temp. This causes the ecm to make incorrect adjustments for a given head temp.Sitting in stop and go traffic on a 90* day having the iat climb to 180+ and lean the afr. Where's the logic here? Wouldn't the tune be more stable in afr if the iat was reading the actual ambient air rather than what the sensor is getting cooked at in traffic. Are you telling me that tuning with the present iat location is not a lie in itself? Wouldn't a bike tuned to a relocated iat be a more accurate tune overall and adjust better to real ambient temps? You have the equipment, why not test it yourself? If I can hear and feel the differences I'm sure you could pull some really interesting data when monitoring the afrs.
Ron

I tried this quite a few years ago.
The ecu knows the temp from the IAT and adjusts accordingly,even at 150*.
Like they say, change the recipe and it's not a cake anymore.
A retune will fix it but it just my guess that you will be back to square 1.

Square 1.  Correctly calibrating VE tables.

Ron, I tried to explain all this, with what I'd thought was good mental imagery, a couple pages back.  Are you saying you didn't read it?

Right. Get the tables to get the effect desired in open loop. That's all Ron wants, is to have it ride right. So what if the IAT is this or that. Just get the garbage out to work by working with the garbage in. And if the IAT sens sig is proven the be susceptible to hot soak and you can't make a table mod for it, then move it.

That post contains a point I've touched on several times already in this thread.  There has been no "proof" that the IAT is ever indicating falsely, thus bringing about incorrect fueling calculations; merely supposition about it.

Ron, at the start you stated the IAT and head-temp sensors provided similar information at hot starts.  You said the new IAT showed a different resistance than the old.  How well does the new IAT correlate with the head-temp on a hot start now?  How did they both do in comparison with the head-temp on a cold start?

Quote from: glens on September 06, 2010, 09:32:10 AM
That post contains a point I've touched on several times already in this thread.  There has been no "proof" that the IAT is ever indicating falsely, thus bringing about incorrect fueling calculations; merely supposition about it.

Ron, at the start you stated the IAT and head-temp sensors provided similar information at hot starts.  You said the new IAT showed a different resistance than the old.  How well does the new IAT correlate with the head-temp on a hot start now?  How did they both do in comparison with the head-temp on a cold start?

I havn't logged it. I saw a 100 ohm difference between the two in the same cup of hot water. Stuck the damn thing in the hole and went riding. It is what it is and will still be relocated eventually. Wife just started riding so for now I've stopped experimenting and just ride it. First ride with the new iat was in cooler weather so I don't know if it was due to the new iat, or cooler air, but the whole riding day did not produce a single skipped beat in my light load areas.
Ron

Quote from: 1FSTRK on September 06, 2010, 08:17:27 AM

Quote from: rbabos on August 30, 2010, 03:50:52 PM
Anybody know of a source for twin cams for this item? Have them for vrods but no search results for tc's.
Ron

The zippers/ thundermax catalog has a kit to do what you want. The thundermax backing plate comes cnc'd with a spot to install the sensor. Seems like someone else thought this was a problem or they wouldn't make a kit to change it.. You can buy the extention harness  and sensor  separately

Fantastic. I will check this out. Beats hacking up the oem harness.
Ron

Quote from: glens on September 06, 2010, 09:32:10 AM
That post contains a point I've touched on several times already in this thread.  There has been no "proof" that the IAT is ever indicating falsely, thus bringing about incorrect fueling calculations; merely supposition about it.

glens
What test would one conduct to prove to you that this is a lean condition caused by the ecm correcting based on the IAT reporting a higher air temp than actually entered the cylinder?

In traffic at low throttle positions isn't it possible that the small amount of air flowing above and below the butterfly and into the motor does not pass directly through the IAT? It mounts directly in the center behind the throttle shaft. Maybe the port, cam combination that is responsible for the lower map pressures creates an air stream that does not pass through the sensor until the throttle is opened more. There is no way that all the air in a large throttle body is moving at very low TPs, and any slower moving air that is in contact with the TB will pick up heat from the TB.

Just a guess but if you thought about this maybe the engineers who could have mounted it wherever they wanted, figured where the put it would/did work?
Bob

Quote from: FLTRI on September 06, 2010, 02:46:17 PM
Just a guess but if you thought about this maybe the engineers who could have mounted it wherever they wanted, figured where the put it would/did work?
Bob

Or the bean counters felt by having it all in one neat package would speed up assembly time. Don't forget that exta wire made shorter will save money too.  . Remember the engineers screwed up the vrod also. An iat relocation kit was the fix for poor running in certain conditions. How dare those bastards try and fool the ecm. This can go on forever since obviously it works to a reasonable degree as is. I'm only stating it can be made better by eliminating the heat soaking. I feel we both have made valid points here for or against. Let's call it a draw and move on.
Ron

That's not fair.  I was working on this while you posted "let's move on"!

Quote from: 1FSTRK on September 06, 2010, 11:59:59 AM
What test would one conduct to prove to you that this is a lean condition caused by the ecm correcting based on the IAT reporting a higher air temp than actually entered the cylinder?

One thing that readily comes to mind would be to:

- Set the whole upper-left area to "14.6" AFR in the calibration and load it
- Attach harness to IAT in stock location
- Ride bike to operating temperature, or close-enough, and reset AFVs
- Datalog a specific route
- Attach harness to IAT in airbox
- Reset AFVs and do another datalog on that specific route
- Ideally, swap IATs and repeat the series

Look in the datalogs and compare O2 integrator activity, both at the start of the log and at the end.  It'd be nice to see what values the AFVs become, but perhaps not entirely necessary here.

If the system gets v-tuned for both locations, when running on whichever calibration, the "other" IAT location will undoubtedly provide much greater O2 integrator activity in a test like this.  I'm guessing that activity would be the least amount using the calibration for the stock-mounted IAT and with that location in use.

Not all engineers are idiots, and in most every other way the Dephi stuff shows good judgment calls in my opinion.  I tend to think they probably got this right, too, and likely not in an haphazard way.

Quote from: glens on September 06, 2010, 05:42:46 PM
That's not fair.  I was working on this while you posted "let's move on"!

Quote from: 1FSTRK on September 06, 2010, 11:59:59 AM
What test would one conduct to prove to you that this is a lean condition caused by the ecm correcting based on the IAT reporting a higher air temp than actually entered the cylinder?

One thing that readily comes to mind would be to:

- Set the whole upper-left area to "14.6" AFR in the calibration and load it
- Attach harness to IAT in stock location
- Ride bike to operating temperature, or close-enough, and reset AFVs
- Datalog a specific route
- Attach harness to IAT in airbox
- Reset AFVs and do another datalog on that specific route
- Ideally, swap IATs and repeat the series

Look in the datalogs and compare O2 integrator activity, both at the start of the log and at the end.  It'd be nice to see what values the AFVs become, but perhaps not entirely necessary here.

If the system gets v-tuned for both locations, when running on whichever calibration, the "other" IAT location will undoubtedly provide much greater O2 integrator activity in a test like this.  I'm guessing that activity would be the least amount using the calibration for the stock-mounted IAT and with that location in use.

Not all engineers are idiots, and in most every other way the Dephi stuff shows good judgment calls in my opinion.  I tend to think they probably got this right, too, and likely not in an haphazard way.

glens:  You guys have beat me down. I don't want to play any more.
Ron

You had me going for a minute, then I got to the part where engineers never make a mistake. I flashedback to a company that had a timken bearing  setup that worked for about 40 years and changed to a roller bearing setup, that they upgraded to a lefty bearing setup, and then for the ultimate performance upgrade offered a timken bearing setup.
I didn't say that the original design didn't work on the stock bike. But like everthing else that needs a scaled adjustment when we make performance changes to these bikes, I say maybe the IAT falls it to this category.

What thread were you reading where it was said engineers never make a mistake?

The first person who does the test I outlined above and posts the results in this thread will certainly have the gratitude of many.  I'd probably do it just for kicks, but my IAT is part of the MAP sensor, so I can't.

Ron:  http://delphi.com/shared/pdf/ppd/sensors/et_airtemp.pdf

Get an "intake air temp" sensor for mounting in your "airbox" instead of the "manifold air temp" sensor which you have.  The "intake" sensor has 4x quicker response time.  Other than that (as a result of the packaging to insulate from mount-point heat and pressure) there's no difference in the specifications between them.

Good info glens, now Ron just has to get Steve to modify the software to use airbox temps rather than manifold temps. 
Bob

Good info glens
  You may want to check your stock harley sensor. It appears that Harley is misusing a Intake Air Temperature Sensor as a Manifold Air Temperature Sensor already.

Quote from: FLTRI on September 07, 2010, 09:15:00 PM
Good info glens, now Ron just has to get Steve to modify the software to use airbox temps rather than manifold temps. 
Bob

I think an iat correction table would do the trick. Then I could set the upper limit of heat soak and not be influenced by it. In other words cap the hot resistance limit to a realistic temp . Like that's going to happen? Let's get one thing straight here. The reading the iat is sending to the ecm to pull fuel when (heat soaked) is not the actual temp the cyls are getting filled with or it would not exibit a lean condition.  You've got the equipment, prove me wrong on that. This has been proven over and over in the auto industry to cause the exact same issues I've seen with my tune. I was bored today so I came back to play again.
Ron

Ron, sounds like you have hit the limits inherent in the Delphi system, as designed for Harley. This is really good input from the field that should be taken up with them, IMO.  Remember, the ECM is an electronic component that has theoretical limits based on the physical construction.  Much like outgrowing the old 8086/286/386...Pentium processors for PC's, perhaps Dephi and MoCo need to look at the next generation of hardware design that could incorporate the additional tables required to tune for additional parameters, much like autos do today.

In this aspect, I believe MoCo still lags the electronic controls best practices.  I am sure that you and others would come up with other data points needing to be monitored and managed by the ECM.  I would agree.  But, todays ECM is limited, and there isnt any way a SESPT or TTS or any other tuner pack can make it any better.  The ECM is at its theoretical max, and needs to be upgraded to a next generation system in order to do what you propose to do with it.

But, what do I know, I am only a computer architect guy, making a living with these things for decades.  lol

Dennis

Dennis, with a proper networking stack and hardware, our ECM could probably saturate a halfway-decent connection serving static web pages.  The task it has has got to be well within its capabilities.

I'm sure H-D worked with Delphi a little on the system, but would be surprised to find out it was much more than some optional stuff on an off-the-shelf product.

Delphi would probably be glad to work with anyone so long as there was a sizable order forthcoming.  But getting them or H-D to just help make the ECU more tunable?  Not likely, because then they'd start showing up modified like that on the street, and that's quite against the law.

Maybe Ron just needs to get a cam more suitable for his usage habits?

Quote from: rbabos on September 08, 2010, 04:05:12 PM

Quote from: FLTRI on September 07, 2010, 09:15:00 PM
Good info glens, now Ron just has to get Steve to modify the software to use airbox temps rather than manifold temps. 
Bob

I think an iat correction table would do the trick. Then I could set the upper limit of heat soak and not be influenced by it. In other words cap the hot resistance limit to a realistic temp . Like that's going to happen? Let's get one thing straight here. The reading the iat is sending to the ecm to pull fuel when (heat soaked) is not the actual temp the cyls are getting filled with or it would not exibit a lean condition.  You've got the equipment, prove me wrong on that. This has been proven over and over in the auto industry to cause the exact same issues I've seen with my tune. I was bored today so I came back to play again.
Ron

Lets keep in mind how Ron has calibrated his VE tables.  Seems most of the problems he has posted on has been conditions he says he can't vtune. 

Quote from: hrdtail78 on September 08, 2010, 09:05:05 PM

Quote from: rbabos on September 08, 2010, 04:05:12 PM

Quote from: FLTRI on September 07, 2010, 09:15:00 PM
Good info glens, now Ron just has to get Steve to modify the software to use airbox temps rather than manifold temps. 
Bob

I think an iat correction table would do the trick. Then I could set the upper limit of heat soak and not be influenced by it. In other words cap the hot resistance limit to a realistic temp . Like that's going to happen? Let's get one thing straight here. The reading the iat is sending to the ecm to pull fuel when (heat soaked) is not the actual temp the cyls are getting filled with or it would not exibit a lean condition.  You've got the equipment, prove me wrong on that. This has been proven over and over in the auto industry to cause the exact same issues I've seen with my tune. I was bored today so I came back to play again.
Ron

Lets keep in mind how Ron has calibrated his VE tables.  Seems most of the problems he has posted on has been conditions he says he can't vtune.

Fair enough, however if I wasn't tuned on the ragged edge this iat behaviour would have been harder to detect and may never have gone looking for the cause.  For me, heat soaked is like an on/off switch between good running and not so good. It will still have an effect on a perfectly (at a given temp) tuned engine as well. Might not be as dramatic, however the symptoms will be there to some extent. They all will lean with heat soak beyond the tuning temps.
Ron

But then again, if you were to (could?) leave the area closed-loop this "heat-soaked IAC" might not be any sort of problem at all for you.

There is one other possibility which I don't remember being discussed, and that is that when the mixture gets so rich the O2 sensors can report it all the same as being lean.  I don't guess you've had that problem though.

Do you have a large overlap on the cams and maybe fresh charge is getting through at these operational areas for you?  That would cause closed-loop "tuning" problems, I'm sure.

Is this merely a pipes/cams combo that doesn't work well there and if so, are you sure it's worth it on a street-going bike?

The harley engineers have corrected the problem on the v-rod by moving the IAT to the air box and retuning the ecm. They must have had a reason that they didn't change the cam and pipes.
Why does it seem that rbabos could not be helped by the very same solution?

I think his whole problem is using the TTS... he should switch to a SEST and Smart Tune it..and enjoy the ride...







( Now that should start something.. )