How It Works by Dr. Dyno, American Iron Magazine AIM, June 2010
Dyno Dos And Don’ts 2010
A new decade ushers in new dyno info
WELL, IT’S BEEN A FEW YEARS SINCE MY LAST Dyno Dos and Don’ts article, and many of you have been asking for more. All Harley models are now not only fuel-injected (infected, if you still love those carburetors), but also closed-loop emission controlled. Dynamometers and the way I use mine have both changed. A myriad of new tuning products promise to make your new Twin Cam run like old times, before electronic throttles. And, considering we’re experiencing times when it’s more important than ever to not blow money on half-baked products or waste it at the gas pump, it’s definitely time for a little more in-Doc-trination.
In this article, we’ll be looking at the changes in dynos and dyno tuning and, along the way, we’ll cover some dyno dos and don’ts to watch out for if you’re getting your bike dyno-tuned or even just tested. Mark Desorbo, the writer who started the Dyno Dos and Don’ts series in 2003, started his article with a quote from Ben Franklin. After meeting me, some of you may have been thinking, “I wonder if Fred knew ol’ Ben?” True, I’ve been doing this dyno gig for nearly 15 years now. And that’s during “retirement,” after I’d already graduated from Penn State, spent two years in the Navy (remember the draft?), and enjoyed a 20-year-plus career as an electronics engineer. But c’mon, people, I’m not that old. And neither is it true that I dyno’d the first Harley, or even the second or third as some have suggested.
TYPES OF DYNOS
Although commonly called dynos, the dual roller setups used for automobile emissions testing, as well as the one at the Harley factory in York, Pennsylvania, would be more accurately referred to as vehicle treadmills. Neither one is an actual dynamometer used for measuring maximum power. They both just offer a way to operate a car or bike under controlled conditions without going out on the road. Still, it’s worth checking out the one in York if you go on the factory tour. The rear rollers drive the front ones so the bike can be balanced just like when riding on the road.
Engine dynos are the real deal, and they measure power right at the crankshaft. In fact, they’re the most accurate type of dyno for directly measuring an engine’s power, but you’ll probably never see one. They’re tucked away in the soundproof rooms of engine manufacturers. By far, the motorcycle testing and tuning dyno you’ll see most often is the chassis dyno. The motorcycle is ridden onto this type of dyno, and, with the front wheel securely strapped down, the rear wheel powers a single large roller. The chassis dyno measures the actual power delivered to the pavement after transmission and tire losses (more on those losses later).
The most common chassis dynos are of the inertia type. They derive their accuracy and repeatability from simplicity. In the case of a Dynojet inertia dyno like mine, there’s only one moving part: an 875- pound, freewheeling drum riding on a set of industrial, low-friction bearings. One pulse per revolution from an electronic pickup feeds a computer the info it needs to continuously calculate horsepower at all speeds based on the velocity, acceleration, and weight of the drum. Engine rpm is read with an inductive pickup on the ignition system. Multiplying the horsepower by 5,252 and dividing by the rpm computes the torque curve. An inertia dyno can simulate the real-life situation in which the engine’s power accelerates the inertia of a bike and rider on the road.
Other chassis dyno manufacturers — Superflow comes to mind use a much lighter drum connected to an energyabsorbing unit that generates the load on the bike. The load is electrically adjusted to allow the motorcycle to rev through its rpm band. Either dyno gives you horsepower and torque curves. As is the case with most tools, the biggest difference isn’t the brand name of the dyno, but who uses them and how. I prefer the inertia dyno because it’s highly portable and no periodic adjustments or calibrations are needed to maintain accuracy.
THE DR . ’ S DYNO
I bought my Dynojet 150 in 1995 and I’m still using it today. I’m sure anyone with a newer one says it’s state of the art and, therefore, better than my 150. Well, all the newest Dynojets still measure horsepower and torque the exact same way as the 150. Seven years ago, I added Dynojet’s atmospheric and air/fuel ratio modules. That converted my 150 into a 200, as far as model numbers go. The current equivalent is a 200i. The i just means the air/fuel and atmospheric modules are integrated into the dyno housing. Then three years ago, I added a second air/fuel channel, making me the first Dynojet owner to do so. Guess that makes mine a 200-2 model. We’ll see what advantages that offers me and my customers later on.
Adding a load control module to a Dynojet 200i yields a 250i model, supposedly the best of the company’s line. The claim is that the load control is needed to hold the engine at a constant rpm for tuning the map cells of fuelinjected bikes. My issue with the module is that it can too easily be used to create loads that not only don’t exist anywhere on the road, but also, in my opinion, abuse the motorcycle. Maybe it’s a feel I have for machines, but I can tell when an engine isn’t happy. Last summer, for the first time, I heard a Harley on a nearby 250i dyno being loaded to a constant rpm at full throttle. That engine sounded horrible, like the dyno had the bike in a submission hold. I couldn’t make myself do that to any motorcycle. In fact, when I give a dyno training course on a 250, I never even turn on the load control. All I use it for is to slow the bike between inertia runs. Argue all you want, but for me, full load at full throttle is a full dyno “don’t.” There are no hills steep enough in the real world to hold a Harley at a constant rpm while making 50-plus horsepower at or near full throttle. So, why do other instructors teach adjusting the fuel maps to conditions that don’t exist? I’m not sure. I think we should leave full-load tuning to big diesel engines designed to pull semis up mountains.
An inertia dyno is an inherently stable measuring device. But the operator still has control over several factors that allow “adjusted” results, either through ignorance or by design. Let’s review a few of these factors, starting with the correction factor.
Many common measurements have to be corrected to be accurate. What does this mean? Well, let’s say I want to lose weight. I’d only be fooling myself if I stepped on the bathroom scales in full riding gear one week, and then in my birthday suit the next. I didn’t really lose the pounds that no longer register on the scale. Is the scale wrong? No, I introduced an error, the weight of the clothes I wore the week before. However, I could weigh the riding gear, correct for it, and get my true weight loss (or otherwise). The same goes with measurements on a dyno. It always measures actual power correctly, but it’s affected by the “weight of the clothes,” which in this case is the weight of the air, specifically oxygen density. By adjusting the actual measured, or uncorrected, power by a correction factor that takes into account temperature, barometric pressure, and humidity, dyno test results done under very different atmospheric conditions can be directly compared. All modern dynos automatically measure the conditions, compute the correction factor, and apply it to the measured power. But the operator gets to choose which correction factor to use. Dynojet recommends SAE, the one I use exclusively. Some guys set the Dynojet to uncorrected if it’s higher, which is pretty dumb. The biggest fake-out one is STD. It sounds kind of official and usually reads higher than SAE. But I’ve noticed that on a hot day in Sturgis, SAE and STD read the same, meaning every bike measured using STD in Daytona will “lose” power in Sturgis. The dyno operator will blame it on the altitude, but, really, he’s the problem. STD is a dyno don’t. SAE is the do.
The next factor is smoothing, which is used by many everyday systems. Take a gas gauge on a car, truck, or motorcycle, for instance. All of these vehicles have a float in the tank that rides on the surface of the gasoline. When the vehicle is moving, the constant sloshing of the fuel probably gives the float quite a ride. Suppose the tank’s half-full. If the gauge read the actual float action, the needle would likely swing between one and three-quarters full. We’d have to do a visual interpolation to figure out that we have half a tank. To make the needle steadily point to half full, quite a bit of electrical dampening, or smoothing, is applied to the signal from the float before it gets to the gauge. Similarly, because of a lot of factors, a dyno graph would be a pretty ragged line if smoothing wasn’t applied. Dynojet offers smoothing factors from a minimum of one to a maximum of five. In some cases, a low factor may help when diagnosing a problem, but maximum smoothing should be used most of the time. Anything that creates raggedness in the curves artificially gives higher horsepower and torque numbers because the computer reads the highest peaks of the curves. A dyno operator may want you to believe the highest spikes are the real numbers, but just like believing we had three quarters of a tank left with an undamped gas gauge, we don’t have what we might think. Setting smoothing at 5 is a dyno “do.” So, smooth (a nice name for deceptive) dyno operators may fake the results by using the correction factor and smoothing to inflate the numbers. For some examples involving actual bikes, check out Dyno Dos and Don’ts: Part III (July 2006). Did you know that some smooth operators might also deflate the numbers? But why, you ask? Simple: it’s to show a bigger-than-actual increase from tuning or installing parts. In the most deceptive case, as I witnessed on another dyno in Sturgis, the guy did it to show an increase where there actually was none. It all has to do with how the baseline testing is done, and, in particular, how much the operator warms up the bike.
As it’s warming up, a Harley will typically pick up 3 to 5 hp between the first and third or fourth runs on a Dynojet. So, if the operator can get away with just one run as a baseline, he can show an increase of several horsepower over the baseline by changing spark plugs or installing a “clean” air filter element and then just making more runs to fully warm up the bike. That was the Sturgis bad apple’s moneymaking approach. He could just as well have said “hocus pocus,” made more runs, and gotten the same results, but I guess he’d have a harder time collecting an extra hundred bucks for that. I haven’t seen him since I wrote those earlier dyno articles, but there may be others like him out there. Do watch out.
When a useful increase really does occur, the 6-8 hp gain from, for example, replacing the stock air cleaner with a high-flow unit can be made to look like 10 hp, 12 hp, or more. I’ve even found that some owners actually want their final run compared to the first one so they feel like they gained more horsepower. Maybe they figure they want more for their money, but I just won’t do it. The third run is your true baseline, not the first. Lately I avoid the whole situation by not even doing a run before the bike is fully warmed. I cruise the bike at about 100 mph, warming up the engine, transmission, and rear tire. Then the first two runs are usually within 1 hp, giving me a true baseline. So a dyno “do” is to look for two virtually identical runs as a baseline. As we’ll see next, also make sure both before and after runs are done in the same gear.
TRANNY & TIRE LOSSESI discussed this in detail in the July 2006 article and the only change since then is the introduction of Harley’s six-speed transmission, so I’ll try to be brief. Whether five or six speeds, OEM Harley transmissions are direct drive in high gear. All the lower gears introduce the loss of two gear meshes. So, high gear should show the highest power on a chassis dyno, right? Yes, if we weren’t also measuring via a tire that adds rolling resistance loss. An H-speed rated (130 mph) tire, as typically used on Twin Cams, shows very little loss below about 110 mph. If peak torque occurs at those speeds, it will read highest by up to 4 ft-lbs. in high gear.
The peak horsepower reading is another matter since we don’t hit it until higher tire speeds, where tire loss may equal or exceed the lack of loss in direct drive. Result: peak horsepower may read the same or lower in high gear. On sixspeed Harleys, the tire gets going so fast in sixth (over 150 mph), that it isn’t even safe to do a full rpm test, making peak horsepower testing in sixth a dyno don’t. And all this assumes a properly inflated tire. Low tire pressure may knock off a dozen horsepower, but that’s okay, I love selling that special Dr. Dyno high-horsepower Harley air. So, is checking your tire pressure before you come to my dyno a dyno do or don’t?
DUAL AIR/FUEL MODULE ADVANTAGES
These next two items are what’s new with my dyno and testing procedures. I’d call them definite dyno dos and recommend them for all dyno testing and tuning. When I upgraded my dyno with Dynojet’s air/fuel module, it added a whole new level of precision to my tuning, and I wondered how I ever lived without it. I soon realized I could cut my tuning time on fuel-injected bikes almost in half if I had a second channel, one channel for each cylinder, so I contacted Dynojet about it and ended up buying the first second air/fuel channel they sold. All serious Harley tuners should have two. If not, they should tune one cylinder and then move the sniffer to the other cylinder and tune it. The graph of Jimmy’s bike shows what happens when the second cylinder doesn’t get tuned. He had his bike supposedly tuned elsewhere, but wasn’t satisfied because it was only half tuned. All I did was finish the job.
Rotating the air cleaner so it points up to vertical pulls the front and rear mixtures together so the carburetor can be jetted correctly.
We can again see the benefits of simultaneously measuring air/fuel on both cylinders by taking a look at Kevin’s chopper. For the last few years, I’ve noticed that some right-angled air cleaners popular on choppers create a major disparity in air/fuel ratios between the front and rear cylinders at full throttle. Kevin brought me his gorgeous homebuilt red chopper to dial in his Mikuni carb with the right-angled air cleaner pointing forward. Since he was going to have to hang out for a couple hours before I could get to his bike, I told him about the air/fuel issues I usually see with this type of air cleaner and that I would most likely need to point it up to make it work. If that wasn’t an option for him, there was no point in him waiting. He said that would be okay, so here are his bike’s air/fuel graphs. The first graph (above left) is with the air cleaner pointed forward as many bikes run it. Note that the black line, the front cylinder, is very rich. Notice that at 3000 rpm it goes to 10:1, as rich as the AFR module can read. If, as with most dynos, we only saw the front cylinder, we’d conclude much leaner jetting is needed. But with the dual AFR capability, we see the rear cylinder (green line) is already a little too lean. It proves jetting can’t fix the problem. The graph above (right) shows what happens when you point the air cleaner up. It pulled the air/fuel lines together making them both a little on the rich (safe) side. I jetted the Mikuni down some and could have gone more, but by that time we were having trouble getting the clutch to hold.
With all the discussion about peak horsepower and torque we skipped over the most important consideration. We almost never ride at peak horsepower or even peak torque. There’s no place we can. That throttle may get twisted full for a few seconds a few times a day, and maybe hit half way occasionally, but it spends most of its time cruising closer to a quarter open. While I do tune all throttle openings when I tune a fuel-injected bike, I’ve now added a cruise test to my baseline runs. In my opinion, if the AFR isn’t close to 14:1 on both cylinders, the bike needs tuning. Much leaner and it will run hot and sluggish; much richer and the mileage will be below the mid- 40s that I like to see. I’ve also added idle mixture and deceleration pop to my battery of baseline tests.
CONCLUSIONNext month, we’ll delve into what I call the O2 Harleys, not the 2002 models, but the 2006 and later models, starting with the Dyna, that use O2 sensors in the exhaust. Bet you’re thinking “Please, not more of this open/closed loop stuff. I’ve already read a dozen articles on it.” Yes, and some of the best are probably the ones right here in AIM that are written by Donny Petersen. Great articles for sure, but then I saw this statement by Chris Maida in the March 2010 Ness Big Sucker install story: “Since [these] bikes have O2 sensors in their exhaust header pipes, the stock ECM adapts to this air cleaner change. That means we and you will not have to do any EFI map changes after this install.” Chris knows a smoother, more responsive hog is just under the pigskin of his test 2008 Street Glide, so I asked him what he meant by his “no tuning needed” comment. His point is that the H-D download or calibration some might say you need with an air cleaner or exhaust change doesn’t do much more than increase the rev limiter limit. He also knows that the O2 sensors will readjust the air/fuel mixture to what it was before the air cleaner or exhaust system upgrade. I agreed but added, “Yeah, and it’ll run just as lean as it did before.” Chris has no argument here. Adding fuel to the lean mix set by H-D will definitely make the bike run a bit better, as well as a little cooler. So, yes, you won’t need tuning for the upgrade. However, that doesn’t mean a bike that would like more fuel to begin with wouldn’t still like more fuel at the end.
Okay, are Chris and I talking open or closed loop, or maybe both? It’ll probably take a couple of months, but we’ll straighten out all the loops and find out that whereas maybe half of the earlier fuel-injected Harleys didn’t need tuning, all of these new ones can run as is, but would like a bit more fuel. AIM