| Author | Message | ||
     Martin |
I have a non-stock T/B but used the stock TPS. If I reset it to stock specs with direct link it will tick-over but if I then goose the throttle it hangs-up at higher revs and comes back down slowly. I slotted the TPS mounts so I can 'swing' it and then fiddled about until I cured the 'hang-up' but then had nothing like full throttle being 'seen' by the ECU--max output from the TPS was about '70' when the butterfly was wide open. I tried setting the TPS to give '85' at actual WOT but then the situation off idle was worse than ever. I guess that my new T/B has a shorter range of movement (82 degrees)than the Buell one? and I need a different model but is there any other solution? Maybe I could buy some little electronic 'thing' from the wholesalers that alters the output signal? | ||
| Martin |
I just pulled out a standard T/B and it measures around 84 degrees of travel, so I am a little unsure what is really going-on. Would that 2 degrees make all the difference? | ||
| Blake |
The short answer is that it looks to me like you desperately need to perform a by the book proper TPS reset/rezero. The long answer follows:  I assume your "T/B" is short for "Throttle Body". Please forgive my lack of language comprehension; I'm also assuming that by "it will tick over" you mean that it will start okay. Let's back up a bit first to be sure we are all on the same page with the basics of TPS and Buell DDFI. Please forgive me if the following seems condescending. It's not meant to be. I am merely trying to think of others who might read this, others who may not be familiar with the Buell DDFI system or TPS subsystem specifically. The throttle position sensor (TPS) is an angle transducer that produces a voltage output that increases with increasing throttle opening. It is an electro-mechanical device similar to a volume knob on an old radio. As it rotates through its range, it produces an increasing voltage output. The TPS knob is simply driven by (connected to) the rotating axle of the throttle. So when the throttle rotates, the TPS rotates with it and its voltage output varies accordingly. The ECM (engine's computer brain) all the while continuously (thousands of times a second) reads the voltage to ascertain throttle position; it cross references that throttle position to to its fuel and ignition mapping tables, then provides the correct/optimum fuel charge and spark instructions to the engine (fuel injectors and ignition system). It is important to note that the particular magnitudes of the voltage output for the specific cases of throttle fully closed and fully open will vary from bike to bike. So, before the ECM can know the position of the throttle, it needs a reference. Kinda like how different folks start their day from different locations. Before we can provide driving instructions to someone to get them to a desired destination, we first need to know their starting location, right? And if we are mistaken as to their starting location, our driving directions will likely be horribly problematic. For instance, I could tell you that I have traveled westwards down the road one hundred miles, but if you don't know my starting point, you will have no idea where I am located after traveling those one hundred miles; I could be in Sheveport, LA; I could be in Dallas, TX; I could be in San Diego, CA; heck, I could be in London, England or Biejing, China. Without knowing where I started, my beginning reference location, you won't know where I am after traveling the 100 miles. Similarly, if the ECM doesn't know the exact reference voltage corresponding to when the throttle is 100% fully snapped closed with the idle screw fully disengaged, then the ECM won't know what position the throttle is in no matter what the increase in TPS voltage output might be. Any instructions the ECM provides to the engine will then be problematic. The ECM needs to know the starting/reference point for the throttle. In order to educate the ECM about the throttle's starting point and thus its corresponding TPS voltage output, we or a dealership service tech must perform what is called a "TPS reset" or "TPS rezero." In actuality this "TPS reset" has less to do with tuning or setting or resetting the actual TPS and everything to do with uploading to (computer geeks call it "flashing") the ECM digital memory chip the actual voltage for the throttle fully closed condition. The one exception occurs when for throttle fully closed case, the TPS voltage falls outside the specified range; then and only then will the TPS require its own special manipulation/resetting. Otherwise, it is very important that we don't ever touch the TPS or affect its operation/position/function in any way. So for the engine to run properly, it is vital that the ECM know the voltage for the throttle fully closed case. Why? Because the ECM bases its entire fuel and timing mapping (electronic engine control instructions) on that one critical reference parameter. If that parameter is inaccurate, even by a few tenths of a degree, the bike will not run well, especially at idle and for other small throttle opening scenarios. If it is off a lot, if the ECM thinks the throttle is 1/2 open, but it is really just at idle position, the bike will run horribly everywhere and lose a LOT of power for full throttle operation. So, it sounds to me like you need to perform a TPS reset on your ECM. More accurately, I like to describe this as an "ECM recalibration for TPS zero." To do that, you need the VDSTS, a friend with one, or a dealership willing to work with your non-stock throttle body. To further clarify, the voltage output of the TPS for the throttle fully closed (idle screw backed off, throttle snapped closed) must also be within the specified range - I cannot recall what that particular voltage range is, somewhere around 0.5 volts +/- I think? Get a proper TPS reset performed on your ECM, that may solve your problem. If not, with a relatively stock machine, you might consider reinstalling a stock TB. With a relatively stock engine, why use a non-stock TB? | ||
| Court |
>>>I could be in Dallas, TX; I could be in San Diego, CA; heck, I could be in London, England or Biejing, China. You could be in New York too; you're always welcome here. That's a wonderful recitation of how the TPS works. Make sure it gets archived in the appropriate section of the KV or I'll fine you. Court | ||
| Martin |
Blake and Court in one thread: thats like an invitation for Tea at the Palace! For tick-over, read 'idle'. I have a twin Throttle-body set-up with richer than standard injectors so I think that I'll have to set the mapping up for idle air/fuel first. What I succeeded in doing with Direct Link (VDSTS on Viagra) was to perform a re-zero but that leads to the problems I described. I have a feeling that the TPS mounts to the new T/B in a slightly different orientation to the standard one and that may be the basic flaw. I have found a company that makes electronic rather than electro-mechanical TPSensors for automotive use that sound interesting but that might be tackling the wrong symtom and not the cause. Thanks again, Martin | ||
| Blake |
Why do you have dual aftermarket throttle bodies on a relatively stock motorcycle? The system certainly wasn't designed for a radical departure from OEM configuration like dual throttle bodies, so I would agree that an extensive remapping via direct link is probably in order. What is the TPS voltage when the throttle is fully closed including having the idle screw fully backed-off/disengaged? I have trouble imagining a throttle position sensor that is only electronic with no mechanical aspect to it. | ||
| Martin |
The dual T/B was in the mistaken search for higher gas speed at low revs. Alex from M-Tek put me right on that one but I'm just persisting with the folly because I hate to be beaten. The 10% bigger injectors were fitted because I couldn't add enough fuel with a PC111 to straighten-out the A/F curve and still saw 80rwhp (once and never again!) I agree that the solution to the 'hanging-on to revs' lies somewhere in the volt reading at idle, compared to zero and also the reading at WOT. When I get home tonight I'm going to do a bit of further investigation. The 'electronic' TPS was a 'hall-effect' rather than a 'wiper' mechanism, which I presume means the use of non-contact triggering rather than a traditional rheostat, but it is many years since that particular physics lesson for me. www.willow.co.uk/hall.pdf once again, thank you for your interest. Martin | ||
| Blake |
I can relate, on account of I'm stubborn too sometimes. It seems that even a hall-effect sensor would need some kind of rotational mechanism. Not sure how a hall-effect sensor would provide a proportionally variable signal. I'm no electronics expert though, that's for sure. Let us know what you discover. | ||
| Al_lighton |
Two degrees can make a huge difference, especially on the lower end of the scale. That is ONE reason why the throttle angle positions are so close together in the fuel tables down low. (The other reason is the need for controlling it more precisely down low, where the emissions testing is prescribed) Until you've run out of adjustment bandwidth using directlink, I think it is folly to go further down the path you're on. I'd be looking at just larger injectors and/or changing the fuel pressure regulator before I'd go messing with the entire throttle body and TPS sensor. I realize you weren't getting the right results with the Power commander that way. But DirectLink is much more than a power commander. But if I WERE messing with the throttle body and the TPS, I'd be finding a way to make it output the same voltage delta over the 85 degrees of rotational travel (which is essentially just what you asked for). THAT is what the ECM is calibrated to see, and we can't affect that voltage-to-degrees conversion with DirectLink. We can change the lookup table values, but we can't change the algorithms. If you want to use Directlink and the stock ECMs, you need that conversion to match exactly. Al | ||
| Firemanjim |
What injectors did you use for the better flow? | ||
| Martin |
Blake-I think you are right about the sensor, but it is 'electronic' in so far as it has no mechanical wiping 'contact' I'll contact the company. Al-thanks for that pointer, I hate to keep firing e-mails in your direction when you are so busy. Fireman Jim- I found some 330cc ones aftermarket ones meant for cars that came in a Ducati parts bag! (Standard XB9 was 295cc on my '03 bike)that were shorter than the standard but that didn't matter as I could lower my fuel rail to suit on the aftermarket T/B. I understand that there is a Bosch BMW part that fits with minor mods but I never found a number for it. | ||
| Martin |
Al-Please hide all sharp objects before reading this! So I just spent an interesting hour trying different approaches to what is an insuperable situation! No amount of playing around did anything other than what Al suggested! I need to either give up on the whole thing or find a TPS that gives 85 units of output over 82 degrees of travel. I have a feeling that a simple resistor that knocked off 3/85 of the reading won't work as it would have a disproportionate effect on the all-important bottom-end but I could be wrong. Just out of interest, does anyone know what the output range actually is? | ||
| Blake |
I really doubt that the last few degrees matters much. Take a look at the fuel mapping at 83o versus 85o; I bet it is nearly identical. I think to make a system that was laboriously configured for a single TB with all its strange split/dual tract dynamics also work for a dual TB configuration, you will have a LOT of work to do, namely mapping each cylinder from scratch. I suspect that there is a lot of pulsing pressure wave communication between the two intake tracts as they split off after the TB. Just guessing, but it seems like that could be a significant challenge to your effort. Whatever it is, good luck and please keep us posted. | ||
| Martin |
Maybe I should drill a big hole between the runners beneath the throttle body! Willow, the sensor company, won't sell single units, which is a shame--that's another blind alley I could have talked myself down... | ||
| Martin |
You know how it is, you think about a problem for days until you've almost decided to quit and then a part of your brain you don't normally use slips you a shot of clarity and you're off and running. I spent a happy couple of hours stripping off the throttle body and finding that it wasn't getting full opening anyway and that I could file the stop to get a full 85 whatevers of movement. never one to pass up the opportunity to change too many things at once I stuck the standard injectors back in and a spare TPS on I had lying around. One reset later and 85 whatevers showing at WOT and I have to wait until a more civilised time to start the thing and find itss made b*gg*r all difference! | ||
| Blake |
You need a film crew there to record all this.  I'm enjoying living through it with you just reading about it. | ||
| Martin |
So my girlfriend's shouting 'get your pants off' but its not my birthday, she's seen the garage keys in my hand as soon as I get back from work and knows that 'I'm just going to start it up, I won't be long', is highly likely to turn into something dirty (unlike, it has to be said, my last 8 birthdays...we met 9 years ago) Five minutes later and I'm in the garage, jumper on against the cold and a forecast of snow but I don't care, because it fires first go and settles down into that long, lazy idle and I know, I just know, that I'm going to make Al proud. Someday. But not today, because I goose the gas and the revs rise and fall-back, but not straight away enough. I let the motor tick over, doing its best 'fish-boat diesel two-step tango' as it sqeezes some life back into a battery and a cleverer brain than mine tries to juggle temperature and speed and fuel and air ....and then it hits me...not just that ECU's are essentially feminine, because they can multi-task, but that they are always right! If a new, zeroed to Pearl TPS and standard injectors misbehaves in exactly the same way as the old set-up, it has to be something else! I recall seeing an idle trim function on the Direct link! there is a tunnel and a white light! I can't get on the dyno for a couple of weeks as I have to earn my 'free' dyno time so I switch off, close the door and go back in to watch 'Overhaulin'. Another poor American guy who just wanted straight paint and his American racing mags blasting is crying into Chip Foose's shoulder over the bill he's going to have to find to insure his too new '56 Chevy and I know how he feels. Now I've been too quick in the garage and there's still time to get changed again and go out to her friend's birthday party. And they say every cloud has a silver lining. I'm martin and I've been underhauled. | ||
| Bomber |
Martin -- you gotta post more, bud -- this is crackin me up to beat the band -- go back inside and take your pants off (we don't need a write up on that partm though) at the risk of asking astupid question (I know, there ARE no stupid questions, only stupid askers, but I digress), there IS a throttle-return spring or similar device in this bitsa, isn't there? | ||
| Martin |
Now, I am on a detox diet for April, as the nurse tells me if I don't lose weight I will have to go on medication for life to reduce my blood pressure. That may seem irrelevant but I have spent a couple of hours with a bunch of increasingly drunk people who make perfect, but increasingly loud sense to each other and whom are totally incomprehensible to a stone-cold sober me. I now have a much better grasp of how my motor feels as the increasingly confused injection is watering it's beer and dodging his round. Yes, there is a return-spring of some sort but I am shutting it off manually, something I found increasingly difficult to achieve with a female pub-buddy who was trying to make a point about how tough women are. 'That's why' she said ' since childbirth is so painful, God give the job to women''Oh, no',came the reply from 'Five-second Phil'(who isn't a drag-racer...)'it's because he's a bloke' I can't fight that kind of logic so I've left them to it. Goodnight America | ||
| Henrik |
Martin; great writing. Thanks. Keep us posted on the motor. Henrik | ||
| Jlnance |
You could be in New York too; you're always welcome here. Actually he couldn't be. 3 guesses why.  | ||
| Martin |
It was one of those days, when you visit the doctor to reassure yourself about the third testicle you seem to be growing and end up in the hands of Nurse Diesel having it removed. That's just one reason why I haven't yet swung my swollen saddlebags over the bike and taken it to the Dyno, yet. I did notice that the Direct Link screen where the TPS and AFV reset live also contained menu items for idle trim whilst running (not that I'm doing much of that at present) and with ignition on. Since I also noticed a readout on another screen that can be configured to record O2 sensor output and I thought that maybe I could adjust the mixture at idle without a dyno, since idle is a low, if not no-load condition. It was fun to watch the cylinder-head temp climbing, whilst I waited for the engine to warm-up, safe in the knowledge that I would have the absolutely correct conditions for my adjustment. I adjusted myself several times during this process and may have a new career if rap-star crotch-grabbing is ever an Olympic sport. As the temp rose, the 'hanging throttle' symptom became less and less pronounced, so what I think is happening is that the ECU is manipulating the fuel and/or accel tables to cater for the cold start and that is causing the symptoms. Since there is nothing I can do about that( a situation for which Al is probably sacrificing a virgin in grateful thanks for right now) I shall wait until I am able to get back in the saddle and see what effect the re-mapping on the dyno has. I did notice that there is no 'cell' for 1050 rpm, which set me thinking.... | ||
| Steveshakeshaft |
Martin, you should've invested in a MegaSquirt system!  Blake, I can imagine a circular ramp rotating and the ramp height being sensed by a proximeter probe. Sort of like those used in vibration monitoring systems. They can have excellent resolution down to .000's of an inch. I guess that would be the basis of an "electronic" TPS device. To see the type of device I mean, look up Pepperl Fuchs. They make 00's of different types of proximeters. HTH. | ||
| Martin |
Now those of you who have been following this saga will know that I am currently healing after the local practice Nurse (why they can't give me someone who doesn't need to practice I don't know)removed a large section of scrotum (mine, not hers)to send away for testing. I don't mind the swelling, its the pain...On Wednesday I got my results, there is no sign of malignancy but I have a severe misfire above 3500 revs. That'll teach me to use a dyno-cell next-door to the doctors. Anyway, I hate to say it but Alex at M-tek has probably got a lot to smile about as here is the first dyno-trace from my twin t/b system compared to an old one with the single. If we ignore the obvious horrendous misfire and concentrate on the shape of the curves, we can see that the motor is choking above 5000 revs, probably due to the smaller individual throttle-bodies. There is certainly a lump in the motor's trousers in the mid-range which, apart from the misfire, makes a very pleasant back-roads ride. Between the runners is a flattish wall that I am thinking about drilling through so that each cylinder can 'see' more throttle area. My instinct is to make this rather large, but I wonder what the general consensus is? Such a 'hole' will probably cause horrendous flow turbulence so maybe a smallish one will work better. so far as the dyno went, I just opted for 100% runs to play with Direct Link and get a feel for the process. Despite the comprehensive instructions I still feel that I am only scratching the surface, a concept that leaves me feeling very uncomfortable as I try every form of underwear known to man to try and get comfortable. I would not recommend a thong. Anyway, I'll wait until I have a chance to get back in the garage and track-down the misfire, before making any decisions about putting a hole between the intake legs. If only my doctor had been so cautious...  | ||
| Alex |
Martin, Iīm in no way smiling about Your results. Instead I like to encourage You to keep up the work (I know You will do it anyway ). Many improvements in my head lay out resulted from mistakes I accidentally made (thatīs why I neglect to use CNC equipment for head porting. Machines make no mistakes to learn from). Is it correct that the Direct Link has not been used with the old single runner design? Do You think You could lift the mid range power with the Direct Link to the same numbers as You get with the dual runner design? Best regards Alex | ||
| Martin |
Well it is a dip that is just begging to be filled! What do you think about the idea of making a hole between the two runners so that they can 'Communicate'? | ||
| Alex |
Regarding the torque dip: it would be interesting to see if the dual runner setup produces more mid range torque according to smaller square area than the stock single runner setup(like what we know from using smaller carbs; still carbs are different animals from injections). I would not expect it with an injection system but who knows. So if You use the Direct Link with the stock setup and end with at least the same torque numbers the answer would be: no, smaller diameter injection systems (read throttle bodies) on a XB donīt improve mid range torque. Could You test that for me if You find the time? Regarding the "communication hole" between the two runners: to be honest, I donīt know what is going to happen. I know that filling a Harley or Buell engine through one common runner normally results in different cylinder fillings. This often results in different AF readings between the cylinders especially with carburetted engines (we all know that). So the major advantage of Your single runner system is an independent cylinder filling. By breaking the wall between the two runners this advantage is probably gone. I wouldnīt do it. To be honest Iīm basically a fan of the single runner design (and obviously the Buell engineers are too. Take a look at the RR setup). But it should not restrict air flow. If You find a single runner setup with enough flow capacity Youīre on the right track. The major problem is the space between the cylinder heads. Keep us informed. Best regards Alex | ||
| Martin |
OK Alex, I'll put my drill away! Although it is in no way a direct comparison, I would have thought that there would be some useful data from Al Lighton's work with a direct-link tuned std XB9 with maybe just the open air-box, to see how much of the dip he manages to fill. So far as space between the cylinders goes, the steeper RR ports will certainly help with packaging as well as flow. Maybe my cylinder heads are where I should be drilling? (Only joking) I notice that the Dutch drag-bike has a big Weber feeding into the standard manifold with no division and can't help but agree with you that individual runners are better. Ah, well, back to the garage. Martin | ||
| Alex |
Martin, let me correct myself. Iīd like to say: I would not expect big advantages from a "communication" hole. But if I wanted to know for sure I would certainly drill it. Only I donīt want to advice You to do so. So if it doesnīt work You can not accuse me for giving bad advice Regarding the Dutch drag bike: last season it was about 1 second slower than our drag bike. We use a single side feed carb Well, our bike has a Sportster engine still it uses modified XB heads. Did I mention that it uses a single runner manifold and a single runner carb But what do I know. For a drag bike You donīt need too much low rpm throttle response (unless You want to impress Your neighbors by using the bike for shopping) as it will normally see full throttle most of the time. For Your street bike demands are different. So go on with the single runner R&D work (and drill that f...... "communication" hole, we all want to know what happens  Best regards Alex (Message edited by Alex on April 29, 2006) | ||
| Martin |
There is no 'F' in hole (well not yet, anyway!) |
