| Author | Message | ||
     Sharpmetals |
does anyone have an experience with Dobeck techlusion ? have heard good things, and they do have buell applications, plug and play, no laptop needed. just wondering if anyone out there is running one with any posotive results ? thanks... | ||
| Brewtus |
I used to run one but with mixed results. The problem is that it only adds fuel. Most applications require you to remove fuel in some RPM and TPS ranges. It is very hard to adjust even with a wide band O2 data logger. I would avoid it. | ||
| Nell |
i'm riding a '04 xb12 and i'm wondering if your app. was similar? also have you heard of the direct link from TECHNORESEARCH.COM? | ||
| Al_lighton |
We sell them, because some folks want them. There are cases where they can help minor fueling problems with gentle changes on them. But I generally don't recommend them, because I think there are better alternatives, for the following reasons: 1) As already indicated, it can only add fuel, and that is very rarely all that is needed. This is especially true on Buells that have interaction between closed and open loop. 2) It paints with a pretty broad brush. They liken it to a carburetor with the pilot, needle, main, and accel pump circuits. But Carburetors and fuel injection aren't the same, and I think the approach is over simplistic. Spot changes over narrower RPM and throttle position changes are often needed, and it isn't capable of that. 3) It lulls folks into thinking that you can tune the bike without a dyno. That may have been partially true with carburetors, but good tuners will argue that point anyway. Fact is, to get good results, a dyno tune with a wide band O2 sniffer at minimum, or a 4 gas analyzer, is needed. 4) It adds another failure point to the bike. With wire splices, additional electronics, etc, come additional failure modes. All of the above is why I prefer Directlink. DL isn't perfect, there are a bunch of things I'd change if I could. But approached properly, it is capable of doing an excellent job of making a bike run right, and there isn't anything added to the bike to make it happen. Al | ||
| Mwbob |
Al, I've heard that it is best to tune the bike with the closed loop "disabled". Is this true and how do you achieve this? Also, when you use the wideband 02 sensor, does the signal from the 02 sensor to the ECM pass through the laptop, bypass the laptop or not go to the ECM at all? If the 02 sensor input to the ECM is cut, how does the ECM react? Thanks, Bob | ||
| Mwbob |
My '06 XB12Ss with the Techlusion runs much better than it did stock. The problem that I am experiencing is that it doesn't run consistently well. I'm convinced that this is due to the closed loop operation of the ECM. I've read that the FI Harleys don't have the closed loop operation at all. I'd really like to see how the bike runs with the closed loop out of the picture. I use my bike to commute into Dallas and while most of the time I don't have too many traffic jams, the other day I spent about 40 minutes in stop and go traffic. The bike ran just awful. There simply is no excuse for this with a $10k+ motorcycle with less than 2k miles on it. It is really frustrating having to spend hundreds of dollars making a new bike run properly. The Techlusion helps, but the ECM has a mind of its own and is determined to starve the bike to death to satisfy the EPA. I think that Al has the best solution and I may end up going with that, but I am a tinkerer at heart and I'm not giving up yet. I'll try to post my results. -Bob | ||
| Al_lighton |
You absolutely want to disable closed loop when doing tuning, for two reasons. One, when tuning, you're trying to tune the map. If the bike moves into closed loop, it isn't running on the map. But you also don't want the ECM to change it's AFV when you are tuning it, otherwise you'll chase the tune. If you're tuning at sea level, you want to do an AFV reset to 100. If you're tuning at something other than sea level, there is argument for getting the AFV to the nominal value that it would have at the tuning altitude, and tuning the closed loop domain to an AFR that would yield that AFV. That takes some experimentation. It is a lot simpler at sea level to be sure. You can disable closed loop by disconnecting the O2 sensor. But it will inconsistently log a fault and light the red light on the dash. The bike will default to running on the open loop map with it disconnected. I don't think the ECM does anything different than just run open loop at that point. I'm not aware of any other way to disable closed loop operation. I think you are correct if you are having consistency problems. The AFV is likely learning to a different value depending upon the cruise RPM/throttle position, which then scales the map richer or leaner. I've never been able to figure out how a Dobeck would circumvent this. Tuning with a wideband sensor is done independently of the narrow band sensor and ECM. You use a completely external system. The dynos have wideband sniffer and 4-gas options built into them, with the results displayed on the same computer screen as the dyno plots and controls. We do our map development with special modified headers with bungs welded onto them so that the sensor can be screwed directly into each header pipe. But when a customer bike is tuned on a dyno, it is typically done with copper tubes snaked up through the muffler and up into each individual header pipe so that each cylinders gas output can be sampled independently. A vacuum pump pulls the gas samples to the wideband O2 sensor, located off the bike. Directlink has a companion wideband O2 system, here: http://www.americansportbike.com/shoponline/ccp0-p rodshow/9310.html. It spits out an RS232 signal that goes into the same host computer running Directlink. DL gets RPM and throttle position data directly from the ECM, the wideband O2 system feeds the computer the AFR data. With a system like this, you get real time correlated RPM, throttle position, and AFR data on the same computer screen. There are other systems that do this, but you have to splice into the sensors for TP and RPM, DL is the only one I'm aware of that gets the RPM and TP data directly from the ECM. AL (Message edited by al_lighton on May 29, 2007) | ||
| Alessio66xb12r |
this is interesting...so if i run my bike with 02sensor disconnected the ecm can ALWAYS follow the map ( there is no more closed loop operation ) right ?? and now with the 02 connected there is no sense for "rapide bike" or "power commander" owners making maps values under 3500 rpm with light throttle charge ?? Al what you think about ? bye Alessio | ||
| Mwbob |
Al- As many other bikes run open loop exclusively, do you think there would be any benefit to abandoning the closed loop function altogether? I have been thinking about throwing together a regulated power supply capable of outputing 0V to 1V DC(in .1V increments) across, say 11V to 14V DC range and putting that output into the ECM instead of the output of the 02 sensor. Maintaining a constant voltage into the ECM would (I'm thinking) "lock" the closed loop into an unchanging state. Varying that voltage would enrichen or lean the closed loop fuel output. I have to mention here that I'm in a "blue sky" thinking mode as we've had about a month of steady rain every day that has kept me off the bike, LOL. Any thoughts? | ||
| Al_lighton |
Other bikes that don't run in closed loop still have other methods of correcting for temperature/density/altitude. Since a Buell doesn't have a manifold pressure sensor, or any other mechanism for compensating for air density, disabling closed loop should only be done for those knowing that they are operating at a consistent density/altitude. The Buell RACING ECM (not the race kit ECM) that is only available to licensed racers from Buell is an open loop only system. There is a factor called "Learned Fuel" that is input to scale the map. Nice name, eh? Sounds a lot like Adaptive Fueling Value by a different name. The difference between them is that one learns from the engine operation during closed loop, the other is a manually entered scalar, one that is figured out from lookup tables based on altitude/barometric pressure and temperature. One might say that running an open loop system like that would be not a lot different than running a Mikuni Flat slide carb (Non-CV type), but that would be incorrect. The amount of fuel drawn in by the venturi effect is still air density dependent. FI systems are not like that . They will squirt the same amount of fuel regardless of density, and that can result in significant error in some cases. So you should only disable the closed loop and learn functions when you are in a known, constant environment. Besides, you'll get a fault that will light the light and mask any other faults. Al | ||
| Al_lighton |
"and now with the 02 connected there is no sense for "rapide bike" or "power commander" owners making maps values under 3500 rpm with light throttle charge ?? " Ya know, I've never really understood that part of the Power Commander (PC) tuning instructions. They say not to change anything "in the box", and that box is clearly in the closed loop learn area. But if you swap a free breathing exhaust and intake on to a bike, you're very likely going to affect the injector duration needed to hold 14.7:1 across that domain when running in closed loop. If the injector duration commanded by the ECM from the internal map doesn't match the injector duration the ECM needs to hold 14.7:1 in closed loop, that is definitely going alter the AFV. I don't understand how making NO changes there on a PC is going to prevent AFV skew. Since the PC doesn't change the internal ECM map for the AFV calculation, then the only way to prevent AFV skew is to change the injector duration the ECM commands during closed loop. Since the PC intercepts the injector signal and alters it, and the ECM reacts to the O2 sensor based on that altered signal, it seems that you could get the AFV to stabilize by altering the closed loop commanded injector duration. But there is one major problem. The PC is oblivious to temperature, and the ECM algorithms use temperature, both intake and rear cylinder head, to alter the injection durations. So any map changes you'd make in the PC to prevent AFV skew would be good for one temperature only. That is the biggest complaint I've heard about running on PCs, if you tune them for hot, they're off when cold. If you tune them for cold, they're off when hot. The beauty of the Directlink system is that all the changes are happening inside the control loops so temperature is still compensated for. I can't speak to Rapid Bike, I don't have any experience with it. But I just don't see how any "in between" box can properly coexist with the adaptive closed loop/open loop system across all temperatures. And I certainly don't understand why leaving "the box" alone is a good thing. Al | ||
| Buell_bert |
It is to bad they cannot make a cheap CO sensor (reliable of course) to work with the O2 sensor and compt. This would help trim the fuel at different air densities, altitude, fuel quality, air temp, throttle position and so many other variables. But I am new to this and really think it is cool to think about it. Maybe they all ready have it. | ||
| Mwbob |
Al wrote: "One might say that running an open loop system like that would be not a lot different than running a Mikuni Flat slide carb (Non-CV type), but that would be incorrect. The amount of fuel drawn in by the venturi effect is still air density dependent. FI systems are not like that . They will squirt the same amount of fuel regardless of density, and that can result in significant error in some cases. So you should only disable the closed loop and learn functions when you are in a known, constant environment." Very good point and even though you shot a hole right through the middle of my carefully crafted theory, my hat's off to you for catching that. -Bob |
