| Author | Message | ||
     Blake |
Bryan, You need to quit. More error than fact in your posts. You drinking?  The fact that temperature has an effect on density and thus pressure is not in dispute (at least by me). Temperature does not significantly affect the pressure. It affects the "volume" of air entering the combustion chamber. Really instead of "volume" we should say "quantity" or "mass." "Volume" strictly speaking is a measurement of capacity. The IAT sensor simply reports a voltage that varies based upon its temperature. The well understood IAT voltage output is simply read by the ECM and is used to factor up or down the mapping wrt how much fuel is supplied versus throttle position. There is never a pressure calculation. Nor a "volume" calculation. The ECM does not perform calculations of meaningful physical parameters; it doesn't need to. It simply reads inputs from the various sensors and reacts according to set mappings. Why do you believe that an O2 sensor output is any less immediate than a IMP sensor? Cruise down the road and yank the wire of your O2 sensor and see how quickly the bike reacts. Maybe it does operate off some kind of time delayed buffer. But it certainly doesn't need to due to any inability of the sensor itself to provide immediate readings. | ||
| Darthane |
I wish. I'm at work. I do blame some things on lack of coffee. Substitute mass for density. They are directly related, of course. How silly of me. Yes, a good idea to just say volume. Notice I did several times in the above post. The ECM may not be calculating air 'volume' on the fly but that parameter is in there somewhere. Without knowing the base volume of air in the cylinder it wouldn't be able to adjust the fuel/air ratio properly given the IAT signal. As we've both said now, the IAT is sending the temp signal to the ECM, which then adjusts the map. (Alright, here's proof that I've not had enough coffee and need to go home - I'm going to have the ECM 'talk') The ECM doesn't just go, "Hmm, the IAT says it's 12 degrees, I need to dump more/less fuel into the mixture." Somewhere in its little silicon brain it knows that X degrees = Y more air molecures per given volume Z. Now, the volume is most likely a fixed value, so you're right, it doesn't necessarily to be recalculated on the fly. And even if this isn't stored in the ECM, someone somewhere had to do those calculations and program the appropriate response into the ECM. You know, Ford's Hiroshima Powertrain division is right across the hallway. I should go grab one of those guys and sit down to have a F2F about ECMs, IATs, MAFs, and MAPs WRT FI's A/F mixtures. Holy acronyms, Batman, that's too many for the Army or even Ford! Yanking the wire from an O2 sensor is not an accurate representation of how quickly its readings affect fuel map. in that case you're causing an error code in the ECM (no signal from the O2 sensor, obviously). If the O2 sensor and ECM so quickly update the fuel map, then why do I have to ride my XB for 15-30 minutes after a ECM flash ini order for it to run properly again? Isn't that time to let the computer adjust the A/F ratio based on O2 feedback? What was the original topic for this thread anyways? LOL...I think we've wandered rather far afield. However, it's making an otherwise boring day interesting (Omoshiroiidesu!). Bryan edited by DArthane on March 08, 2004 | ||
| Unibear12r |
Ok. If a MAF uses a hot wire that varies the current through it because of the air flow by it,and the IAT varies the voltage across it because of the air flow by it,do they not send the same information to the ECM? I can't speak for engine applications but electronics wise it sounds like the ECM could use either device for both funtions-seems like its all in the design/programing of the ECM. I suppose I'm being too simplistic here. Yes I know that I stepped in where I should fear to tread but I'm really interested. | ||
| Opto |
A few ideas here: -This volume thing is missing the point, as far as I can nut it out the DDFI calculates the mass of the air (which includes the number of oxygen molecules) per stroke using the TPS, rpm (cam position sensor) and the IAT sensor only. It then looks up a table, depending on closed/open loop mode, engine rpm and cyl head temp, and works out exactly how much fuel needs to be injected to burn with the measured amount of oxygen. -My understanding of an O2 sensor is that it has a reaction time of approx 1 sec or more, that's why it's only good for steady-state closed loop conditions. -A MAP (manifold air pressure) sensor has an advantage of being able to measure air pressure before you start the engine, so the computer has an idea of ambient air pressure. "Not true. An intake leak anywhere in the system would affect the air pressure everywhere within the system. An absolute pressure sensor would be able to measure the change in intake pressure due to a leak further downstream." Yes exactly! The pressure sensor upstream of an intake air leak would see an increase in pressure and tell the ECM to reduce the amount of fuel being supplied. That is exactly opposite of what is required. The bike would run lean. If you cannot see that, imagine the extreme case... a completely dead intake, throttle closed with a HUGE leak aft of the throttle. With the intake pressure reading virtually equal to 1 atmosphere, the ECM would think that the bike was idling and supply and amount of fuel accordingly. But due to the leak the amount of air entering the combustion chamber exceeds greatly that small amount required for idling. The bike would run extremely lean. Sorry I just can't see that Blake, with the intake pressure reading virtually equal to 1 atmosphere this would tell the ecm that the motor is running at WOT, and it might inject fuel appropriately unless overridden by a TPS input. 3. How is it better to surmise the state of mixture/combustion via intake pressure versus actually sensing the state of combustion via O2 sensor? No answer. The answer is that it is not better. The O2 sensor is the single absolute best sensor/indicator of the state of combustion and air/fuel mixture in the engine. All automobiles use O2 sensors. All FI systems use O2 sensors as far as I know. I wish Buell would release some more indepth details on their DDFI so us curious Buell owners wouldn't have to go fishing in the dark simply trying to understand our FI system. I think a lot of us have an appreciation for innovation and how our bikes work, otherwise we might have bought a more conventional motorcycle. | ||
| Darthane |
Unibear12r, I'm not precisely sure how a MAF works, so Wycked could be dead on or off the edge of the map for all I know, but I can tell you with absolute certainty that a MAF and IAT do NOT send the same signal to the PCM (for one thing, a MAF has two inputs to the PCM, at least in automotive applications. The IAT is usually a simple variable resistance sensor). How, you may ask, given that I just admitted to not knowing precisely how a MAF funtions? Simple - I've yet to see a modern FI car that doesn't have both (often combined into one unit) and as cost-conscious as automakers are you can be DAMNED sure they aren't putting unnecessary sensors in there! LOL | ||
| Wyckedflesh |
Darthane on my Mitsubishi it was a single input and honestly that was what I based the anology on. Since reading the manual on the Mitsu gave the same impression as the manual for the DDFI. | ||
| Darthane |
Really? Fords, Mazdas, GMs, and Chryslers (at least those I've seen) all have two, plus a power and ground, making the MAF a four circuit application. Well, again, I'm not positive how the thing works, and it wouldn't surprise me in the least if there are multiple types of sensors. I know a lot of things about a lot of stuff, but even less about far more. You know, that sentence made perfect sense in my head. | ||
| Benm2 |
Nope. Alot of motorcycle systems run with no O2 sensor. That means the systems are running "open loop", with no means to post-adjust. That in reality makes them really fancy carburetors, offering digital adjustment of fuel curves based on inputs. | ||
| Unibear12r |
I'm thinking that part of the MAF uses a "hot" wire. The hot wire varies in heat content due to the air flow. The changing heat of the wire changes the resistance which changes current flow which is the data used along with other inputs to determine mass.( Yes a lot of semi- educated guessing here). The IAT also varies resistance (guessing?) due to heat and the air flow. They might not send the same signal but it seems you could use either data curve to gather the same info. as I think both deal with heat & air flow. Darth- I think I know just enough to sound lost. | ||
| Darthane |
Somewhere around here I've got an internal schematic for the MAF used by Ford...ahh... Well, I guess now I'm even more confused. The MAF signal to the PCM is measuring instantaneous voltage whereas the IAT signal is measuring resistance, both based on air temperature (it even uses the same note to show the temperature range per signal return for both). And the damned thing's got a power and ground that I have absolutely no clue as to its function...maybe that's supplying power to heat the wire? I would think that would have to be a regulated line, though, and it's not. LOL...I'm still sticking by my statement, though - there IS a reason to have both, otherwise they would NOT be there unless both were necessary and thus one could not reproduce the function of the other. My guess is that the MAF portion of it requires a baseline from which to operate, whereas the IAT is really more of an ambient air temperature reading. In fact, that's gotta be it, because you can still have sufficient airflow mass regardless of the temperature of the air, therefore the engine really needs to know both the temperature of the incoming air for density's sake, and the amount of incoming air (based on temperature drop of a baseline heated filament due to airspeed). -=quite possibly getting blissfully farther and farther from the mark=- | ||
| Opto |
Nope. Alot of motorcycle systems run with no O2 sensor. Thanks for that Ben, and it makes me wonder why a manufacturer would go to a lot of trouble to fit FI and then not quite finish the job. I wonder what they actually gained over a carburettor, just a bit more top-end power? Unibear, if you put an IAT sensor on your back verandah it would measure air temp, but it would not know how fast the wind was blowing. | ||
| Blake |
Opto, Thanks for trying to clarify the situation. NONE of the repliracers use O2 sensors. Bryan, You have a great imagination. edited by blake on March 09, 2004 | ||
| Darthane |
LOL...you'll have to pardon me, but which portion of my exposition are you referring to? We've kind of moved beyond the intake leak thing now. I'm gonna see if I can pin down one of the PT guys today and talk to him about MAFs...that one's bothering me now. Bryan | ||
| Smitty |
One thing the MAF can do to help to determine air quanity that the IAT cannot alone is determine how humid the air is by comparing the actual air temp from IAT and the voltage reading from the hot wire in the MAF sensor. (The ability to absorb heat varies by the humidity)allowing a more precise measurement of air charge. Another thing is that some mfgr are using the sparkplug as in the cylender sensor checking on each explosion and adjusting the next fuel charge. My guess: The XBs don’t need a lot of extra sensors and an idle air control because the load at idle is constant. Unlike a car with a cooling fan air conditioner power steering pump and other accessories varying the load on a closed throttle plate. | ||
| Blake |
Your conception of how an ECM uses the various sensor readings seem very imaginative, all the "calculations" and such you describe going on. You know how a look-up table in a spreadsheet works? There is no "calculation" involved,. | ||
| Smitty |
Blake I’m assuming you mean me Yes exactly. How far does it need to go. Ducati, Bmw need a fast idle lever to warm up. The XB does not. The old phrase K.I.S.S. comes to mind. Tim | ||
| Darthane |
Blake, you're right, and notice that I said: "And even if this isn't stored in the ECM, someone somewhere had to do those calculations and program the appropriate response into the ECM." I was covering my ass. LOL | ||
| Bigbird |
Ah, fuel injection. Now there's a subject I REALLY know a thing or two about, including typical sensor construction and operation. I've had some great conversations about DDFI with the guys in tech services too. But I'm not telling...too freakin hot in here for me  | ||
| Unibear12r |
I thought it was obvious that I was trying to catch up to you guys here- not prove anything, other than that outside the box thinking might be in order here. We are talking Buell, so maybe they are doing something unexpected. Like monitoring the cylinder with the spark plug? Just trying to keep an open mind and learn something here. Thank You | ||
| Chainsaw |
JQ said
The XB's can show the FI trouble code by jumping a wire across 2 plugs on the data link, it will blink the code on the engine light. If your O2 sensor or rear injector is out, or both, you can find out without the dealer. Maybe your Suzuki is equipped similarly? Do you have the SV service manual, or were you referring to your Buell (I thought you had sold it)? | ||
| Unibear12r |
Come on Bigbird whats a little heat to be worried about?  You would have to do better than I am! I would be very interested in what you had to say on the subject. | ||
| Darthane |
Aww, come on. We're (mostly) speculating. Enlighten us - if you actually work with FI then you're doing better than us! I just know that the damned MAF/IAT Ford uses is ridiculously sensitive...had six in a row that were bad last year because someone dropped the box 2 feet. | ||
| Dynarider |
Personally it sounds like you all are trying to baffle the rest of us with Bullshit.  | ||
| Benm2 |
A carburetor has one or two "tuning" circuits available, even an open-loop system has many more. Wierd air/fuel feedback from a dynojet dyno (for example) can be used to produce a nice, even line with enough tinkering. Interesting thing, though, is that NON-Buell Open-Loop FI uses secondary butterflies to replicate the "progressivity" (made that up  ) of CV carbs. | ||
| Unibear12r |
I've been to baffled TO Bullshit!  Ok I did a little bit of extra study on the DDFI. The IAT sits in the air box inside the filter but way away from the intake so its unlikely that it could do anything but sense air temp. only. I thought it was in the intake. Ok I've only had the air box open once and I didn't realize that strange "thingy" was the IAT! Before reading the service manual and the last few posts I thought you HAD to have something like a MAF or MAP full time and an O2 sensor was optional. Now I can see where the MAF/MAP sensors might only be needed under certain conditions and that the O2 sensor is more important than I thought. I am less baffled now | ||
| Cyclonecharlie |
Well I have read this whole post and Im impressed,and I also have a hellova headache.Now I remember why I like carbs. so much. Fantastic reading though,makes you think..........Later Charlie | ||
| Bigbird |
Personally it sounds like you all are trying to baffle the rest of us with Bullshit. Not everyone Dyna, just you! But then again I can't baffle you with something you already know so well Ok, so a couple of you want me to get involved. I'll take the bait...what's the question? So far one of the points I have picked up on is the question of whether or not a MAP and/or MAF based system can be more precise than DDFI. Is that the crown jewel of information that this thread has been condensed to? If so there really isn't an easy or exact answer. In theory a MAF/MAP system can be more precise than DDFI. In reality there are too many variables to only consider that factor. Basically it works like this: The greater the number of sensors providing data to the ecm, the more precisely combustion events can be controlled. HOWEVER...The sum of all of that data is only as good as the mapping program contained in the ecm. In other words you could have enough sensors to tell the computer anything and everything about combustion events, but if the mapping resolution (number of cells or blocks of mapping data) is relatively sparse the injection system will not be able to correct in increments that are fine enough to yield proper performance and driveability under all conditions that are possible to encounter. Conversely a system like DDFI that is relatively sparse in terms of sensors and data input can still be made to yield excellent driveability and performance under almost all circumstances through the careful construction of "high resolution" mapping. The best of both worlds would be carefully designed mapping combined with more data available to the ecm, but as DDFI has demonstrated it can be largely unnecessary. If the XB's had undesirable performance characteristics I would agree something more should be done. But it's hard to argue with it when it works right. IMO probably the only place DDFI lacks is in the fact that running it at sea level, then putting it in a trailer and driving it to the grand canyon (at nearly 8,000 feet) will make it run poorly until it relearns. The addition of a baro sensor would have eliminated this problem. I won't go into more specific details than that right now because I'm sure my explanation will generate questions, accusations, and hostility. So fire away and I'll do my best to answer, within the boundaries of my feeble knowledge base One other issue I would like to address is the comparison of a Power Commander to the Screamin Eagle race tuners kit. That's like comparing a Volkswagen to a Porsche. Yes, they both come from the same basic idea, but believe me the Power Commander is the Volkswagen in this case. If you think the race tuner kit isn't worth it you should compare specs and abilities. There really isn't any comparison. The race tuner kit provides individual cylinder timing adjustment, cold idle speed control, and a whole host of other functions not available on a Power Commander. Not to mention the fact that the Power Commander is a "piggyback" computer that interrupts, alters, and then re-sends the signals that originally came from your bike's ecm and/or sensors. It can't possibly be as good or respond as quickly as having the re-worked code installed in the ecm itself. I really wish they had a race tuners kit out for the XB, complete with the ability to attenuate oxygen sensor voltage (hint hint hint) | ||
| Unibear12r |
Thanks Bigbird you just confirmed what I have learned and suspected over the last couple of days. Just before reading your post I was trying to figure out how DDFI worked with altitude without a MAF/baro. etc. I knew it did but couldn't see how until I realized I had a misconception about the O2 sensor. I thought it only sampled between 3k & 5k rpm. Don't know how I got that. Actually the system goes closed loop to the O2 after warm up and not above 5k rpm or WOT. Ok I figured the mapping had to be more complex but how much so? It seems like you could extensively reduce restriction in the intake and the ECM would be fine with it but not so with the exhaust unless you go to the race module. | ||
| Unibear12r |
I was just trying to think of all the posts that helped me come to understand this when I realized it was all of them. Thank you BadWeb. | ||
| Opto |
Thanks for having a go Bigbird, can I start with a few questions? 1) What is the quickest way to let the DDFI relearn, for example after changing from a stock airfilter to a K&N? 2) Does the engine "stumble" briefly as the new "relearned" settings are locked in? 3) If not, when do the new "relearned" settings start to be used, gradually as you ride, or after an engine stop/restart sequence? 4) What are the significant differences between a stock ecm and a race ecm? 5) After running an XB12 with Drummer, K&N and race ecm, the tech tells me the ecm is trying to lean out the mixture but it can't go any further, AFV was 88%. I started running a stock ecm and stock filter with Drummer, take it in and tech says "AFV 100%, now put the K&N in and bring the bike in to check the AFV". I have not been charged for these AFV checks. Does all this sound right to you, like we're on the right track? Any answers appreciated very much. |
