HarleyTechTalk

I really don't know about air flow direction with the HFM, but believe that intake runner length has to be somewhat more than a stock H-D intake. Cant remember what their called but the air filters/runner that sit at a 90° to the intake forward may be the answere to the intake runner needed for the HFM, mounted just behind the AC , but I'm just speculating and can offer no concrete proof. I'll leave that to better minds . I'm just a guy in the trench's that's getting rusty.

Perhaps there are two elements in "series" (in the air flow) and the ECM can tell by which one changed first which direction the flow is?

While I am sure that something will come along sometime in the future that allows them to get it figured out I was just unaware of it at this time. That's why my interest in it. The long runner with a large plenum is what settles it down in automotive applications, as it acts basically like a shock absorber does on suspension. The plenum is typically 4 times the volume of the total of the intake runners to make it work out and there is no where near that amount of room on most motorcycle applications.

The HFM stands for Hot Film Measurement, but the original abbreviation is in German.
But it works in English as well.

The reason the HFM can detect air flow direction is; (grossly simplified)

There is a piece of silicon/chip area that is exposed to the air flow.
This chip is in the center heated by a resistor that draws a current = Heat produced.
There is a temp sensor on the upstream side of the silicon/chip, measuring the temp of the chip upstream the heating resistor.
The air flow is calculated by measuring how much current is needed to maintan temp X at the temp sensor.

There is also a temp sensor DOWNSTREAM of the normal airflow direction, measuring any flow in the opposite direction, in the same way as the upstream temp sensor.
(if the down stream temp sensor see a lower temp than the upstream sensor = air flow has changed direction.)


The built in electronics in the HFM does the math and sends out an air mass signal that corespond to the air mass that has passed through the HFM in the direction towards the engine.

How fast this is done I don't know (its OK for the pulsations in automotive applications but fast enough for a HD?), the technology has been around at least 10 years.

The Hot film has been around for a long while but I just have not seen one as you describe. It sounds like a good idea with the dual temp. sensors to try and determine airflow direction. Do you have any links to them being discussed, ie. who makes them? Would love to find out more about them and what year model automotive applications use them.

Steve , if it helps, the Mercedes Benz uses a NTC resistor along with the HFM.

Steve, try this link
http://rb-kwin.bosch.com/en/powerconsumptionemissions/electric_controls/sensors/enginemanagement/hfm.html

I think you can get the detailed answers I am sure you are looking for by contacting Bosch.

Quote from: Steve Cole on February 20, 2012, 04:38:32 PM
Now that you understand that MAP is just a reading of Absolute pressure and that once you reach an equal Absolute pressure in the intake to outside (Baro) NO more air is going to enter the engine. This is the way that it works and there is nothing you nor I can do to change than short of installing a forced induction unit.

With this in mind we can look back to another way HD has done the VE's and that is to base it from TPS and RPM. I think most everyone understands what these are and it is pretty easy to now compare the two methods of doing it.

Question #6

Why use TPS instead of MAP for VE's?

So what is the correct answer and whats the next question?

I would venture that MAP is a more accurate information input....

tps reacts to throttle changes quicker but then you better have a calibrated wrist.

Maybe we are still on 6th.

MAP is the best way to figure out the engine's condition in the term of load. 100% kpa is the most load that the engine is going to be under. 

Quote from: hrdtail78 on February 23, 2012, 01:45:18 PM

Quote from: Steve Cole on February 20, 2012, 04:38:32 PM
Now that you understand that MAP is just a reading of Absolute pressure and that once you reach an equal Absolute pressure in the intake to outside (Baro) NO more air is going to enter the engine. This is the way that it works and there is nothing you nor I can do to change than short of installing a forced induction unit.

With this in mind we can look back to another way HD has done the VE's and that is to base it from TPS and RPM. I think most everyone understands what these are and it is pretty easy to now compare the two methods of doing it.

Question #6

Why use TPS instead of MAP for VE's?

So what is the correct answer and whats the next question?

I'll take a stab at it. Maybe the MoCo used/uses TPSxRPM for reasons unrelated to total engine performance. Perhaps when the engine gets tested to meet EPA standards, it's done at known, and very repeatable parameters. i.e. 20%, 50%, 100% throttle etc at XYZ RPM, and at a standard altitude density. All they have to do is make it run dead nuts on 14.7 AFR at these settings, pass the test, and make it run as well as possible across the board???

Jeff

edited to correct the dyslexic "ARF" to what I meant to say...."AFR" !

Quote from: akjeff on February 23, 2012, 05:55:21 PM

Quote from: hrdtail78 on February 23, 2012, 01:45:18 PM

Quote from: Steve Cole on February 20, 2012, 04:38:32 PM
Now that you understand that MAP is just a reading of Absolute pressure and that once you reach an equal Absolute pressure in the intake to outside (Baro) NO more air is going to enter the engine. This is the way that it works and there is nothing you nor I can do to change than short of installing a forced induction unit.

With this in mind we can look back to another way HD has done the VE's and that is to base it from TPS and RPM. I think most everyone understands what these are and it is pretty easy to now compare the two methods of doing it.

Question #6

Why use TPS instead of MAP for VE's?

So what is the correct answer and whats the next question?

I'll take a stab at it. Maybe the MoCo used/uses TPSxRPM for reasons unrelated to total engine performance. Perhaps when the engine gets tested to meet EPA standards, it's done at known, and very repeatable parameters. i.e. 20%, 50%, 100% throttle etc at XYZ RPM, and at a standard altitude density. All they have to do is make it run dead nuts on 14.7 ARF at these settings, pass there test, and make it run as well as possible across the board???

Jeff

On top of that.  I would have to think.  TP/RPM is a lot less complicated and easier to deal with.

It's easier to v-tune by blind feel...

To me the answer to #6 is what's good enough? We know the problems it causes and we know it's not accurate but is it accurate enough for what's currently being done? I know without a doubt that MAP is more accurate but as I said before, you need to look at what's going on in the intake manifold while the engine is running. If TPS is believed to be good enough and in someone's mind easier to do then why change it. Since there are two different groups at HD in charge of it, that could be the answer but I just do not know.

What I do know is that if you run a bike on the dyno with a steady load on it you will find that the TPS reading in each gear is different, for the same load and RPM. So that means when tuning what gear are you going to use, as it matters. So the mixture is going to be different based upon the gear your in and yes it can be measured! By using MAP, 40 kPa is 40 kPa regardless of the gear! Just as 100 kPa is 100 kPa. Now TPS gives a measurable error based on the gearing of the bike and the heaviest bikes all get MAP referenced tables starting in 2010. So it could be that due to the weight of the bike and gearing they needed to switch and the lighter bikes can get away without it. Lot's of guessing as to why but no solid answers as to why.

The EPA test are very close to being a tight set of conditions but there is a rider on the bike and there is some room for him to influence the test results. The rider has to control the bike by following a speed trace on the dyno which has been setup for the test. In this test there is a high and low trace on the screen and the rider must keep it between the lines. Now the trick comes from how smooth the rider is on the throttle, if he varies a lot between the lines with little throttle movements (long slow changes)or does he vary a lot with big throttle movements (quick large changes). It all plays in to a pass or fail. Rider57 can probably give a better description of the EPA test schedule than I've done here.

What you all need to learn here is the basics of how it really works and then you can makeup you own minds as to why, knowing what's really going on. The whole reason for me to start this was for you to learn and while I'm not the best at doing it as a teacher I keep seeing the same problems over and over because people just do not understand how things work. It's not only the DIY guy either.

This is just the easy stuff so far and it gets more complicated when we start going into all the interactions that occur in a running engine, that have to be handled, but I'm not sure if you guys want to get into it that far? If you do I can try and explain more of it but it's up to you all.

Bring it Steve. I like learning new stuff.

Quote from: Tsani on February 23, 2012, 07:25:28 PM
Bring it Steve. I like learning new stuff.

+1

This thread has convinced me to quit mooching, and make a donation to the site. Was beginning to feel the need to wear a ski mask when I logged on for free! Thanks again for the education Steve!(and everyone else, for that matter)

Jeff

Please do keep it going Steve. As a truly uneducated person on these subjects, I am not qualified to offer any opinions or contributions to the discussion. I have, however, been following along with keen interest and am learning thanks to your lead.
Extremely interesting subject and discussions!
Roger

  Please do keep it comming, getting back to the basics sometimes gets lost along the way of overcomplicating the simple.

Thank you, Steve, for taking the time.

I have NOT posted until now, because I do NOT want anything I say to derail threads like this.  Thank you Steve, and all the other members for posting up.  I too like learning.

I didn't read the "whole" thing so forgive me if these answers have been adressed.
First many Speed density CAR systems do not use MAP as a VE table axis. ( Map or more correctly Vacuum is always used on fueling.) Vacuum is the difference between current MAP and Baro.
( Baro can be derived even if the system ONLY has a MAP sensor. Many vehicles have both, MAP and Baro, but some make due with only a MAP Sensor and derive Baro at key up where MAP=Baro when the engine is not running and then use a correction scheme to adjust Baro based on engine load and speed during engine run.)

The last Speed density I calibrated used IACTPS, as a VE breakpoint. That is a percentage of the IAC position vs. TPS. ( At small throttle openings and IDLE IAC has more "effect" than TPS. As throttle increases IAC becomes less of a factor.) And at any given MAP with different Throttle positions, VE is effected again as turbulence about the throttle blade changes flow. You can have the same MAP, ERPM with varying throttle posistion during an engine cycle due to changes in ROAD Load. i.e. hills and valleys. So MAP varies alot at the SAME TPS AND RPM while running down the road.
The problem with using either "all TPS" or MAP is precisely becasue YOU can get different reading for MAP at the SAME TPS ERPM load point. If you accelerating/steady state cruise/UP HILL/or DOWN
you will not get a consistant reading vs. MAP alone! And conversely, using only TPS and ERPM as table points means the MAP is not consistant at each speed and load! Nether (MAP or TPS)  Based VE are perfect, but I have used BOTH on different applications.( and additional "corrections" factors need to be used depending on which is choosen as base! )



Using only TPS also presents the same problem. But by using MAP as an input to the base fueling and using an "adjusted" TPS( in this case by also using IAC counts for expected air rate) you can get "closer." But!! BIG BUT, using speed density you will NEVER be correct 100% of the time! There are simply too many variables that effect the VE other thna MAP! If you use a given VE at even IDLE, as the INTAKE heats up ( with engine heat) the air lag time in the manifold will walk the fuel control up as well! There are tables to "try" and negate some of those effects but it is never perfect!

All that said, there was a question regarding AIRFLOW through a MASS airflow sensor.
Reversion in the intake track ( s valves open and close and overlap effects are present) does cause an INCORRECT reading in the MASS sensor output. Some neww sensor are working on "directional flow" but the older sensors need to be corrected for this reversion!!
Guess how "they/we" did THAT!The engine application is run with the correct induction parts and connected to a STANDARD AIR MACHINE ( measures all air in and out! Of the intake)
( I might add as an aside, fewer cylinders produce MORE reversion in a common plenum. There is more time for the airflow to be effected between cylinder event as th enumber of cyclinders goes down!)
So back to the problem of reversion with MASS air sensors.! The area of reversion is related to speed load and throttle, so you can map ( as in MODEL and KNOW where reversion is taking place)
and choose ignore the MAF SENSOR. That is why MOST MAF sensor application also HAVE SPEED DENSITY! Tables. When the SENSOR reading in an area of KNOW reversion ( as well as some engine acceleration events)  is outside of a calibrated "window" the Sensor output is ignored and Speed density calculations are used instead of the sensor input! So any application without a bidirectional MAF sensor still uses speed density "sometimes."!!

And to answer Steve's first question "what is kPa" it merely the UNIT of measure for MAP. Just as a speedometer can be used to measure MPH or KPH or ft/sec, kPa is a unit of measure, and nothing more! In this instance it is the unit of measure for the "converted" output of the MAP sensor. The OUTPUT of a MAP sensor is voltage! That voltage is converted inside the ECM software to a UNIT of measure.
And any unit measure of PRESSURE could have been used, it just happens to be kPa in this software.

Very interesting thread  THANK YOU Steve  and FBRR and ALL that have posted in keeping this a civil and educational thread ...

FBRR

I think it would be nice if you could join in here if you have the time and help to teach those that want to learn the basics. At some point we can get into all the reversion that goes on in the intake tract as well as how fuel sharing and correction terms go on in a HD application but I'm not so sure how fast people can understand it all. Both you and I have been doing this for a very long time and I know that I assume people understand what I'm trying to say here. I've been told by a few people that they do not understand, so I'm trying to start with the very basic things of a Speed Density system and then work forward. I think you would be a great help in explaining things and it would give people possible another way to have it explained.

Just so everyone understands no system is perfect but the better you get it the better it will run in the long run. Doing this at a factory level allows plenty of ways to test and create correction terms that get things closer but remember that those were all done based on how the equipment was when it got built. Most all of you have changed something on your bike and that is what you are trying to correct for, so you cannot expect all the factory presets to be calibrated correctly and some never will be. So it all boils down to what you as the bike owner feels is good enough in the long run.

FBRR...thanks!
'Effective' communication.