| Author | Message | ||
     Easyrider |
This topic is for Highscore, To keep his data in 1 topic and to stay out of mine.. | ||
| Drhodes1970 |
I kind of agree with you Easy. | ||
| Blackflash |
http://www.factorypro.com/dyno/true1.html | ||
| Highscore |
There has been a time, Marc Salvisberg, the owner of "Factorypro", and Mark Dobeck , the founder of Dynojet, were good friends. This time is long gone, but up today Salvisberg is crying out loud into the world he is making everything so much better than Dynojet. A little bit distressing in my opinion. But even this man has to agree ,that the Dynojet roller has set the worldwide industrial standard for performance testing, simply because 75% of the dynos around are made in this Dynojet-design. Being this standard does not mean instantly the Dynojet roller is perfect and the best dyno possible - not at all. But the fact, that this dyno is accepted so widely, shows on the other hand, that this dyno works and measures performance accurately enough to earn this reputation as a standard. So not only tuners like me, but a lot of independent institutes like motorcyle magazines, school and universities trust in the Dynojet. So which dynoe uses the German TÜV for instance, when you go there to make an EC-approval for - say - an exhaust? A Dynojet roller (I know this first hand, I have done such approvals). I cannot believe think these institutes are guided by ignorance. The utmost disadvantage of the Dynojet roller is at the same time its strongest advantage: it is a chassis dyno. On a chassis dyno you can test a bike as it is, without any dismantling, it is only necessary to put the driven wheel(s) on the roll and fix the vehicle. Especially for testing motorcycle this means a problem for the chassis dyno: there is just a single wheel with a very limited rubber surface on the tire to transfer the power on and into the dyno roller. Wheel spin and slip is not only a problem of realworl driving. But a dyno senses just the power applied to it. So huge spin must be avoided. The load, that means the inertia-moment, of the Dynojet´s roll is exactly calibrated for this purpose. As long as the tire is in good condition, the slip is negligble. So you make hundreds of test runs on the same bike without any substantial tire wear. This optimization of roller weight and diameter Dynojet has indeed performed mainly by try and error. With a Yamaha V-Max, the strongest bike of this time, as candidate. For a long time you could read this story on the Dynojet website. For me this development is a piece American pragmatism "not theorizing much around, just make a thing working". And the success of this design proofs, it works indeed. This optimization for slip-free power transfer from the tire to the roller on the other limits its weight and inertia moment. In fact the load the Dynojet makes to the engine, is not really high. Especially when testing really potent engines this is a problem, because the engine speed raises very fast, faster than in the real world. Here the Dynojet is prone to measure the put to optimistic indeed. Furthermore the Dynojet does not consider the rotating masses of the vehicle. The inertia moments of the crankshaft, gear shaft, chain/belt and rear wheel are part of the inertia moment as a whole, the dyno is sensing at its roller. This two imperfections make that the Dynojet displays a slightly higher engine output, but usually this failure is below 5%, which is the tolerance allowed for testing according SAE or ECE - standard. So when a bike cranks out -say - 140 HP on the Dynojet, it could be the case the power is in fact just 133 HP. But what does it matter for the job on the dyno? Except of the fact, that the numbers are probably to large - nothing at all! The Dynojet namely makes always the same failure, when it makes its failure. So when comparing different runs for any improvement by a modification, that little failure in accuracy has no effect. What is need for any comparison is repeatability of measurements. And exactly for this claim an intertia roller is pretty close to perfection -as long as the tire and the chain/belt of the bike is the same during the r&d process.But this job is manageable, I think. So when testing a motorcycle, as it is, fast and without dismantling, a chassis dyno like the Dynojet is absolutely sufficient. When someone goes really serious into r&d, there is no way beside of a true engine dyno, attached directly to the gear box. I own two dynos. Guess which type the second is? Highscore (Message edited by Highscore on December 29, 2009) (Message edited by Highscore on December 29, 2009) (Message edited by Highscore on December 29, 2009) (Message edited by Highscore on December 29, 2009) | ||
| Easyrider |
Lets Talk about tuning now... any questions... | ||
| Justa4banger |
OK my question... is there an actual amount of fuel added in % when compared to the AFV values? Meaning if i show 100F and 110R... does that mean it is adding 10% more fuel to the rear only? and 0% to the front... Honestly its nice to see AFV numbers, but unless they relate to something specific i can't see them actually doing anything usefull except show you a confusing number. Also can the ECM adjust the open loop tables in relation to the closed loop tables? If the closed loop tables are adjusted using the O2 sensors to a ratio of adding 5% more fuel, does that change also adjust in the WOT maps? | ||
| Cram_grebsivlas |
Sorry - Mark Dobeck and Marc Salvisberg were never really "good friends". That is so far from the truth that if Mark D and Marc S were reading this, they would be chuckling uncontrollably. Neither one was a good loser, though. I always say that the best deal for the mc world would have been combining the two. Both have their strong points. You seem to "know" how different True hp is from djhp numbers - are you looking at a large sample of True vs. djhp numbers? As far as DJHP numbers, I thought that everybody had already read the Sports Car International http://www.truehp.com/magazine/dynojet_dyno_inflat ed_hp_reason.html interview where M Dobeck confessed that he made up the dj HP scale to make a stock VMax read closer to crank hp on a magazine test VMax..... but he didn't know that that mag bike that he was comparing the stock VMax to was an overbored 1400cc "magazine sleeper"' It's much more than +/-5% error. Ask TUV Measurement and Standards department if they know what calculations dj used to create hp number? Probably not. I just emailed a few TUV certification facilities as to what the chassis dynamometer requirements are, so, maybe we can sort that out, too. As far as Factory Pro's EC997 dyno doing things differently than the commoner dynojet dynos? How can one explain how much better and quicker fuel and ignition tuning can be when one uses a 4 or 5 gas EGA vs. an o2 sensor that just measures residual o2 in the exhaust that can get there many ways? It's tough if you haven't - or can't do it. The US EPA has disregarded o2 sensor readings for "AFR" and uses 5 gas EGA to determine AFR. Hmmmm.... just like us. How can one simply accept that the potentially significant tuning error involved in tuning something to Best Power ob a dj dyno with the mentioned insufficient dj loading and how can one compare that to being able to do steady state, inertia sweep and controlled acc rate sweep, if one hasn't done it on a different dyno? I guess that one would have to have a dyno that could do all those things and be trained to tune (relatively easy) or learn on one's own (way harder). I've see a couple of the brightest, most stubborn people in the tuning industry insist that they needed no training and both of them, for the want of 2-3 hours of EGA training, failed to grasp the full meaning of CO% readings, much less the additional info the other 3-4 gases would give when tuning fuel and ignition to optimal real world settings. There was a US exhaust system manufacturer who complained about the EC997 - "When I do testing on your EC997, it tells me that I have to do tuning. When I do testing on a dj, it shows me I have horsepower increase and I'm done." Bummer for his customers. Anyway, that dyno went to Canada and served for many years tuning championship riders' bikes. After that tuner "semi-retired", it went to Computrak. The Insufficient dj Loading that you mentioned. What about combustion material temps during a dj run? 300f to 400f less than the real world (gsxr750)? How can one account for that in the real world? One can only guess. EC997 Loaded testing virtually eliminates loading errors. If one knows that there's error in testing, like high hp bikes and inertia sweeps, then FIX it - I'm sure TUV has identified and corrected their errors, right? No, I thought not. As far as tire slip - a tire under CONTROL, has up to a 3.75% creep rate at high power dyno loading. On a dyno with good traction, it's there and in the real world on the road surface, it's there. There is no real wheel dyno in the world with a 100% power transfer rate as the loss is mainly in tire flex. Factory Pro has the patent on the slotted drive roller - It holds about a third more hp than a used knurled drive roller - and the creep rate is still around 3% to 4%. The EC997 easily holds Steady State, 175 True (True x 1.15 to 1.20 is about 205 to 210 djhp) and has held up to 225 True (260-270 djhp). Steady State is where you need to be when tuning for proper combustion chamber and exhaust system material temps. I can only try to elevate the overall level of tuning and hope that the people who understand tuning more than settling for inertia sweeps and "tuning to an AFR" use or buy more EC997 dynes and high speed EGA's or similar systems. [wink!] The gist of this is that one doesn't have to settle for cold "too short" dj dyno tests, unnoticed, so-so traction at high hp levels and it's not necessary to settle for some "AFR" number tuning when you don't have to - and just because 75% of the people have dynojets, doesn't even come close to saying it's the best tuning dyno - it says it's an "ok" dyno and they have better sales people. If you could use a dyno that didn't have the disadvantages of a dynojet, wouldn't that be the best choice? Now, let us get back to tuning. I get to do an 09 Buell when we get back to work - Woohoo!! :-) Best regards and Happy New Year - Sorry I hijacked the thread for a moment - but I really do have a Buell to tune.... :-) Great informational posts, other than what concerned me! <wink!>. I agree with you on most of what I've read and appreciate the large amounts of time and care that you've taken to provide education for the rest of the forum members. Marc Salvisberg Factory Pro Tuning (Message edited by cram_grebsivlas on December 30, 2009) (Message edited by cram_grebsivlas on December 30, 2009) (Message edited by cram_grebsivlas on December 30, 2009) (Message edited by cram_grebsivlas on December 30, 2009) (Message edited by Cram_grebsivlas on December 31, 2009) (Message edited by Cram_grebsivlas on December 31, 2009) | ||
| Cram_grebsivlas |
Thanks for the pms, guys - AF1 in Texas is a pretty decent tuning facility with a lot of Rotax experience - Marc Salvisberg | ||
| Chameleon |
I agree with Marc. He has the scientific data to back-up his claims posted in plain view on his website (even if the site looks like something out of 1997). Marc, please do post up a before and after performance measurement of that Buell you'll be tuning... hopefully it's an 1125R. My 1125R is nearing the end of its factory warranty and I'm considering getting it tuned, mostly to smooth it out, particularly down low... But I certainly wouldn't mind more power. I see that there's an EC997-equipped tuner not too far away (53 miles) from my home. Oh, and just because something is the entrenched and de-facto option, doesn't mean it's the best... Look at Microsoft Windows for a prime example. | ||
| Highscore |
I am not here to educate anyone, my only intention is to give information. An engine is a mechanism, so its operation is not governed by "hidden secrets" or "voodoo", but solely by the common physicals law around, valid for that region of the universe we live in. Indeed my intention is to explain these fundamentals here in a language, anybody -hopefully, I am no native speaker - understands, without formulas and weird maths, as the way, engineers talk about this subject usually. My goal is to show up the interaction and dependence of underlying real world phenomenons which are correlated within this engineering formulas, in a more likely intuitive manner, And what is this good for? Isn´t this only but just plain theory? Not at all: Such an "scientific approach" makes it possible to convince this "expert field" named "tuning" pretty similar to cooking: "Use the right stuff - combine it in the proper manner - you get the results you desire". My concern here is "knowing how" instead of plain "knowing that". On the back of such a scientific approach starting with fundamentals, the description of a tuning item will not end with phrases like "it is a xxxx-pipe", designed by .......". It is not the brand mark which makes good stuff working, it is the design and technical function. In this sense I try to give here neutral and objective access to the matter. And we shall see, there will be more than a single solution for design. So there is in fact no absolute "best" - for example - exhaust, an exhaust is "good" in respect of the application it is designed for. "Application" means here the operation area of a bike. So a pipe for the drag strip looks different than on for the street. And it may be,it cranks out even some HP more than the street pipe. But this quality alone does not make the drag pipe for a better exhaust. Of course we shall talk here a lot about dyno graphs, simply because such a graph is the real world proof of a design and thereby also of the theory, the physics behind. I use for this purpose Dynojet roller graphs, simply because there is no need to argue about these graphs, this a standard anybody knows (Marc - I come later back on dynos). Furthermore I want to lay the focus of dyno-discussions more on the shape of a graph, its gradient and progression instead on pure peak HP-#. And I shall talk most time about the torque-graph. The torque-curve is just another expression of the power-curve. I prefer the torque-line because its "up" and "down" is a direct picture of the cylinder charge: More or less thrust or moment at the engine drive line means instantly and proportionally more or less pressure applied to the piston crowns. Looking for power means nothing else as searching for more charge, a better cylinder filling. The torque graph is the finger print of that. So this project is intended as a technique-neutral - if you want - open source approach to engine tuning. So this not my thread here, anybody is invited to upgrade and sharpen this story by comments and contributes. Highscore | ||
| Highscore |
So let us talk about dyno: Marc, you are right when you described the Dynojet roller as an "as-it-is"-dyno, far away of being perfect. But are therefore its readings just fake? Fore sure not. We do not need to argue about, that the Dynojet simple applies to less load to a healthy-strong engine. Also your comment regarding the HP-inflation by the Windows-PEP is correct. For this reason my own Dynojet still works under DOS. I use the Dynojet for a quick snapshot of engine performance and for side-by-side comparations. When you know the limits of the Dynojet, for this purpose this tool works absolutely sufficient well. BTW: The German TUV uses Dynojet in this way, when comparing the effects of aftermarket good to engine performance. Of interest is here just the proportion and percentage of change by an aftermarket but, not the "real" and "true" engine output. For this the TUV has different dyno, the real sophisticated stuff. "Sophisticated stuff - this is the key word. I know your dyno, Marc, first hand because I have visited Kainzinger in Germany to set up the fuel injection of a Yamaha R1 using my "Kastl":  The R1 was supplied by Kainzinger with a Powercommander, My Kastl´s analysis technique enabled my then to detect and compensate two flat spots of the Dynojet fueling. Your EC-dyno was really a sweet tool during this job, it was fun to work on. Nevertheless I like here to stop the discussion about "perfect dynos" You should know, Austria is the home of the real sophistic" dyno stuff, because here is the headquarter of the AVL, a company with 8000 employees worldwide, building dynos and developing test equipment for the automotive industry since 1945. So it is quite easy to dimple an amazing amount of money inside a dyno chamber. There is always a dyno somewhere, which is better. Sorry, Marc, if I talk now against your business, I am really convinced that the Dynojet roller, when used within its limits with care, shows up to 80-85% of the information, a perfect(er) dyno delivers.So for the most part of dyno work, even for development, this dyno is absolutely sufficient. When I feel a need for more load to be applied to the engine, I go to a dyno compartment with the "real stuff": eddy current, water cooled generator, dissipating its current to the public electricity network, 800 cubic-meter test cell volume air conditioned to 20° C (Sorry,no pic by hand, wait until next week). More important as the dyno itself is my opinion the guy operating the dyno and making the job there. A dyno is just a source of information about the condition of an engine. But information without interpretation is worthless. This is the job of the operator, to draw the right conclusion from the picture the dyno shows. For me exhaust gas reading is an essential element of this picture. Without this analysis a dyno run is more a carousel ride without any deeper value. So more important for the quality of workman ship as the brand and kind of the installed dyno are the people, operating the dyno. And if the gas analyzers are either badly maintained or not in use at all, I would say, such an dyno center should be avoided no matter how much money the management has spent for the equipment. Let us come back to "load": When we talk about engine load and its measuring by a dyno, we have to separate two kinds of this load: steady and unsteady. At steady load the engine load and speed remains constant over a certain time. This is condition of cruising and holding a constant vehicle speed on the highway. Un-steady load means now the opposite: Engine speed and load changes continuously. This is the condition of accelerating a vehicle. About which kind of vehicles are we talking here? Mediocre powered cars and scooters? No, we are talking about performance motorcycles. At the moment the engine power, installed to a vehicle, is sufficient strong, steady load occurs only when you are in the "public road mode" to keep you driver´s license. In each other situation, when you twist the throttle, a performance motorcycle does just one thing: ACCELERATION, its only real limitation is the air drag at high speed. So something like "steady load" occurs only in this situation. Unfortunately the engine works different when in steady load or on acceleration. When accelerating and operated in a "transic" mode, wherein the speed changes fast, and the engine output differs also accordingly to these changes. For this reason when the engine "snaps" thru its range, the engine prefers a different "care", especially regarding its fueling, compared to its needs at steady speed without any huge changes of load points. So for setting up the real strong engines, we need a dyno which is capable to measure the engine output even under acceleration. This invention is not very old. The first dyno system which made this possible was the "inertia dyno", consequently also called "acceleration dyno". The inertia dyno is no invention of Dynojet, this system was created in middle of the 80th by Syntec, a company from Switzerland. This company makes workshop tooling especially for trucks. Truck engines have a huge and bulky flywheel. So this company born the idea: Let us measure the power of a truck engine by calculating the increase of its flywheel´s rotational speed". I guess, Dynojet has never paid license fees to this company. The principle of measuring engine output by measuring the rotational speed of a flywheel and calculating its acceleration by time , has been one of the first examples of a true computer-based measuring system. In our times it is hard to believe but there has been a time without computers. The first computers came out last century in middle of the 80th. Before this time and without the computer the data acquisition and ongoing calculation of acceleration was simply impossible. The only way to measure engine out before the development was to "weight" its thrust or momentum by attaching a brake to its crankshaft, against which the engine "worked". The brake was usually a water turbine throttled down, until the engine speed dropped slightly, the torque applied by the engine to its shaft was measured by a weighing machine. This procedure worked sufficiently over decades, until in the 80th the fabulous turbo-charged F1-engines appeared, cranking out up to 1000 HP from just 1500 ccm displacement in the 80ths. Suddenly the engine department developing these engines was faced with the principal shortcoming of "weighing" the engine power on a brake: this method offered just a limited collection of separate load points for measuring, so it does not show the full picture of the engine´s operation over its whole range. So there were engines with brilliant dyno+#, which nevertheless simply did not "come out the corner". When accelerating out of a corner, the engine is in "transic mode" and there is no steady load. And here things are, especially in the case of turbo-charged engines, completely different compared to its "intransic" mode, when engine load and speed are constant. Consequently Ferrari was one of the first companies which purchased this new Syntec inertia dyno system. At this time this was the only dyno able to measure the engine out put constantly over the whole range without any interpolation of "brake-points". This dyno "draw" power graphs of a new appearance and shape, offering details which just are been hidden in that old simple "brake-days". Sounds likely as I am a fan of Dynojet too. But this is definetely not the case. In fact Dynojet excluded me from the Dynojet-family, because of "being too intelligent". What was my fault? The Dynojet-family is nothing else but a MacDonald´s-style franchise network. Being outstanding within such a network causes instant trouble because you disturb balance and symmetry within the system. My first dyno now was such a Syntec. I purchased it some month before Dynojet "invited" me to buy one its dynos as obligatory to become a Dynojet distributor. In this early days I ran both dynos side by side (sorry - again no pics at the moment). The Syntec used a true rail road wheel as inertia flywheel, driven via an intermediate gearing directly by the gear box shaft. Because an inertia moment depends more on flywheel diameter than on weight the Syntec puts a considerable load to an engine. So a run with an FZR1000 Exup, on of the screamers of this time, lasted with the Syntec aprox.three times longer than with the Dynojet. With this "true" (all roting masses calibrated) Syntec on my side I had a rather steep learning curve in these early days. So I started to create my own jet kits which performed better as the Dynojet originals. Of course, naive as I am, I informed Dynojet about this progress because I thought they are there happy about this technological input. But this was obviously not the case. My "final failure" happened as I refused to sell a Dynojet-kit "as it is" for a bike, because it performed worse than stock with it. The answer was: "If you do not like to sell the stuff, we find someone other". And so it happened. I told this story in detail to show you, Marc, that I know pretty well what it means to compete as a single person with this "industry leader". Since my exclusion I designed and manufactured my own stuff. And of course this is better than the Dynojet-stuff. It isn´t very hard to reach this goal, because Dynojet is much better in commerce than in techniques. Regarding the "buddy-story": As one of the first persons here in Europe, who bought a Dynojet dyno, the company invited me in 1991 to the AMA spring opening at Daytona. Of course Mark Dobeck was also there. I just remembered his words as I kept them in my memory. But let us forget the old stories and look into the future. I would appreciate if we could continue our discussion here. Kind regards Christian Skorianz (Message edited by Highscore on January 02, 2010) | ||
| Easyrider |
How do YOU determ the amount of load, and how much load the dynoroll (rear wheel) needs when measuring the exhaust gas. | ||
| Cram_grebsivlas |
Hi Highscore - You seem very earnest and I am sure that you are. As I am and so is Easyrider.  You can be sure that there are many companies that are capable of doing similar dyne systems as AVL - in several different countries. I designed and evolved our dyne systems to be as close to 100% flexible in tuning routines as possible. The EC997 does Steady State, low inertia Sweeps and Programmed Acceleration Sweeps, addressing your explanation of why it's ok to do all testing at very quick (too quick in my testing). I've compared incremental acc rates from Steady State to quicker than a dj. You can be sure that there is a difference in both tuning AND total hp AND shape of power curve as one increases the acc rate. A common result is that the quicker the acc rate, the flatter the power past the power peak is. That explains why I can tune something in steady state that revs with good power to the redline and then test it on a dj and the power doesn't appear - though, on the track, it more works like a Steady State test indicates. That doesn't happen all the time, but the Steady State always seems to get the high rpm right. The way you explain that everything is always accelerating so that's how you tune sounds kind of like what dj told people when telling them that their inertia dyno was better than a brake dyno. They were too self absorbed to notice that decent brake dynos can do sweeps, and average dyno customers hardly knew the difference between Steady State and a Sweep, anyhow. Instead of accepting an estimation of the dj being 80% to 85% as "ok" and "kind of close", I just evolved our dyno so that you can set your base fuel and ignition maps just like the factories do (steady state) and then, adjust the acc map correction tables in any one of the controlled sweep rate test that you desire. Best of both worlds - I don't know why Kainzinger didn't show you an EC997 sweep test, but he is one of the smartest.... I just thought of something that steady state picks up easily: Reversion. It's easy to see and diagnose in Steady State - but it's hard to see in a quick sweep. For example, the 99-02 R6 had a unique exhaust system problem that caused a problem at full throttle / 8k-9k and I only saw one exh manufacturer identify and correct it. In this case, equal length headers were not what was required. Funny thing was that I told 4 north american manufacturers and only one could verify it and address it. I'm not trying to sell you a dyno or tell you how large our dyno room is or more many hp's worth of fans we have. Stop telling me how OK your dj is and I'll stop addressing it. What I can tell you is that both of you have tons more specific experience designing exhaust systems and "stuff" for the Buell. I don't specialize in any one particular bike - but I do specialize in dynos and test routines. And I've gotten just about every type of tuning thing and software that you can think of to tune - even stuff I'm not supposed to have.(oops! :-) When it's goofy or "impossible to tune", I get it.... So - want to talk about Buell stuff? I'm all ears. If there's a specific question about modifying Buells, I'm going to be asking you guys. Best regards, Marc | ||
| Easyrider |
Marc, You are more MY kinda man.. (-: Let me know how I can help, or send me a ECM?.... |
