Reading Dyno Graphs-Breaking the Dyno Code

[Dyno]

Reading Dyno Graphs-Breaking the Dyno Code

News Flash-We?ve purchased a new Dynojet 250i to replace our faithful but aging 250, providing the latest test equipment to our customers.

Dyno graphs have become part and parcel to the performance industry as a measurement of success and a means to compare that success with others. Yet, some still regard this data as a closely held and hard won secret. At Latus Motors Harley-Davidson we?ve decided to share 18+ years of data so that our customers (and anyone else on the internet) can view, analyze, compare and make informed decisions on performance choices using the same information we use to make our recommendations. We did not put our entire library on this site because of its size and repetitiveness, but have selected what we feel are interesting and informative files. At this time we believe this to be the largest available reference file on-line. Since it is intended to be a reference library, we have many categories not found on other sites (at least in one place) and may be somewhat overwhelming in its size. These include folders on a variety of tests/comparisons, exhaust brands/type, combinations by other makers/shops, things that did not work (ours included), our package combo?s and others. This library is alive and will be added to on a routine basis with information we believe will be interesting or valuable. We have built repetitiveness into some folders as this too can be useful to demonstrate variability. What you will not find (unless required for a test, etc.) are imported graphs from other dyno?s, specific tuning data (jet sizes, ignition settings, EFI maps) and information not relevant to a folder/graphs intended use. More detailed content descriptions will be found within each folder.

Having created this resource, interpreting the data may be something of a mystery.
What is shown on the graphs (when possible) is a horsepower trace, torque trace, AFR line, correction factors, smoothing setting, engine speed and run notes. Typically when making a run, we use STD corrections, 5th gear (including 6 speeds), forced scaling, smoothing setting 5 and the lowest RPM allowing a clean run. Most of the industry has gone to these settings but some are still using 4th gear and SAE corrections which will show lower numbers. We use forced scaling (HP and TQ scales on either side of the graph are the same graduations) to make the graphs easier to interpret. In this form, HP and TQ will always be the same (cross) at 5250 RPM. If this does not happen, check the scaling. Using 5th gear will give higher numbers, longer sample time and normally a smoother curve. Smoothing of 5 provides an easier graph to read but we may use a lower number to diagnose certain problems. The starting RPM is the lowest usable to provide the most data. Very important data exists from 1.5-3.0K RPM (torque dips) and graphs that do not display this information cannot tell the whole story (race motors, by design, may not allow this low of a pull). Conditions are the environment the graph was generated in and are used for correction factors. Correction factors are used to help cancel the affects of conditions (temperature and humidity) when comparing charts. The temperature shown on the graphs is the room temperature. The Engine speed (RPM) is required to calculate torque (HP X 5250 divided by RPM=TQ). The run notes are the most important to viewers since this describes the combination, differences between graphs and the effects of changes (when more than one is shown on the same page).

What do you want to see on a dyno graph? It is very important to take into account the intended purpose of the combination being viewed. Generally, touring bike owners will want a broad, flat torque curve that works best from 2-5K RPM and lighter bikes may be happier with that concentration of power from 4-7K. Drag bikes may not care at all about anything below 4K. What you do not want to see are spiky/jagged curves, huge torque dips (typically 2-3.5K), big differences from one graph to the next when no changes to the bike were made, large correction factors (like 1.2 or .90; normally .96-1.04) and really abnormal run conditions. Note that these graphs represent 100% throttle acceleration runs only and do not represent part throttle or steady state conditions. This is particularly important when looking at the AFR fuel line. A rich condition (below 13.1) at 3k rpm on acceleration may actually become lean (above 14.1) when held steady, with an applied load, at the same rpm.

The true value of this library will be the ability to compare a large number of combinations and to assess the affects of differences between them (particularly the exhaust folders). We invite you to view often and watch as this file grows. Use the Forum section of this web site to ask questions you may have when viewing the data.

Enjoy,

Mike Stegmann

[Mike]

Response to V-Twin Forums, Dyno/Drag racing, BigBoyz-tw37-95ci, dyno chart post by Adlerx, 01-06 and comments that followed.

Latus Motors Harley-Davidson policy normally is to not respond to internet forums or discussions with respect to comments about us (or others), positive or negative. Having seen the sometimes emotional and surprising replies regarding Adlerx?s dyno chart posting we felt it necessary to clarify our dyno methodology and reasoning. A copy of this information will be posted on our web site with our dyno library. This is not intended to be a critique of other methods and will be our only response.
For the record, Adlerx?s chart (run 014) represents a fifth gear/STD corrected run, 102.29hp/114.35tq. The SAE correction yields 99.29hp/110.9tq. Note that the uncorrected (actual) were 104.08hp/116.36tq. These hp/tq numbers were repeated over several runs. A fourth gear run was not made and so that data is not available. Partial runs were made to try to level the fuel curve but doing so resulted in torque losses and increased hesitation. Since the Power Commander 3R (Adlerx?s is not the USB version) tunes both cylinders as one, it is sometimes necessary to skew the fuel curve in some areas to achieve the best balance between cylinders. Fr/Rr cylinder offset had no real affect. Achieving the best torque curve may mean a technically imperfect fuel curve.
The hesitation is common for MM bikes with cams and free flowing exhaust. It is thought to be a function of injector targeting, injector timing vs. intake valve opening, valve overlap/duration, loss of manifold pressure and other influences. Calibrations for stock cams are timed to fire the injectors at, or about, the time the intake valve opens (2 deg btdc-crank angle). Most aftermarket cams open the intake from 10-25 deg btdc. The intake valve may now open a substantial amount of time before fuel delivery by the injector. MM throttle bodies target the injector spray onto the radiused portion of the intake port floor. The original intent was that the fuel would splatter and be picked up by the high velocity air (high manifold pressure) in the intake runner and port. Another affect of increased valve timing/duration/overlap is reduced manifold pressure at low rpm?s. MAP losses (at low rpm?s) are also increased by the loss of cylinder scavenging as a function of a free flowing exhaust. The net affect is late fuel delivery, pudling in the port, low port velocity to carry it into the chamber and hesitation. Most tuning devices have no control over injector timing (not to be confused with ignition timing) and targeting is fixed when using the stock/modified stock throttle body (the Screaming Eagle MM t-body corrects the targeting issue and does improve the hesitation). Injector targeting seems to have the greater impact vs. timing. Changing the TP sensor voltage changes many of these timing events (including ignition timing) and can be used as a tuning aid. Cam timing and pressure drop are a function of the components selected. That selection is often a compromise between the above affects and the desired hp/tq expectations. More tuning may improve Adlerx?s overall tune but from a practical time/cost/number of runs vs. amount of perceived gain, the tune was considered acceptable. Adlerx?s numbers should be considered exceptional and not the norm. Reviewing the large number of charts in our library would reflect this. The library is by design large to allow just such review. Adlerx is to be commended for his achievement.
Gearing does have an affect on dyno numbers. The 3.37 vs. 3.15 primary ratio change imparts a leverage advantage and is seen by the dyno. The dyno gear ratio is listed on the Latus library graphs so that gearing correlations may be made. Mechanical efficiency and parasitic changes also affect hp/tq numbers. Drive train mass (heavy rear wheel assemblies) has a large affect. Horsepower values will decrease as drive train mass increases since the motor has to spin up both this mass and the drum. Also, the affect of oil drag is measurable both as a function of volume and viscosity and has the same affect as mass on hp values. This is why heat soak and consistent test temperatures are important. It is the inertial dyno?s capability to measure these that make it most useful in translating how those same real affects will be found on the road or race track.
Regarding our dyno methodology, it is not our policy to ?inflate? the numbers. We publish STD corrected numbers even though uncorrected (actual) numbers are often higher. Published graphs are of those that are repeatable. One time only, non-repeatable graphs are not published unless for specific test reasons (such graphs are not unusual due to heat soak, active cold start circuits, oil cling or other influences that tend to decay or stabilize as runs are made). There is no advantage to inflate numbers and Adlerx?s was not a Latus build, and so, no incentive to do so. Reviewing our Dyno library would reveal many builds that made less than expected numbers. Bikes run on other dyno?s (Cohen Motors, Mikuni Calendar show, shootout contests and trackside dyno?s) have routinely given hp/tq numbers within 2% of those shown on the Latus dyno, when similar conditions could be achieved. 2% is considered to be within the normal resolution of these machines (ie; +-2hp/100).
The use of STD/5th gear is in response to an industry trend towards this. Dyno libraries from other web sites reveal that both SAE and STD are used by contributors (ref; Wood Auto, Horsepower Inc., Headquarters, T-man perf., Hemi-designs, Revolution Perf., and others). Since we are often comparing (or being compared to) results with others on a national level, we must use similar values. The STD correction was originally introduced to account for high under-hood temperatures found in cars. Some racing organizations realize that race cars have a higher mechanical efficiency than production cars. Motorcycles tend to be more like race cars from this standpoint. As such, the logic is that the STD correction better reflects this difference. The real value in a dyno is the use of it as a comparator (the ability to compare affects of changes) and in this sense the type of correction used is irrelevant so long as the same correction and dyno are used. In the interest of allowing others to compare, future graphs on our web site will include the hp/tq numbers for an SAE/4th gear final run (after tune, no changes). These will appear in the run notes but not in graph form.
The use of 5th gear is to obtain longer sample times. Dynojet recommends the use of 4th gear to keep drum speed low and run times short to help speed up the tuning process and reduce wear and tear on the dyno. As a predominantly Harley tuning center, the 5th gear drum speeds are not nearly as high as those routinely seen by shops tuning import bikes. The shorter spin-up/spin-down (run) would save time but we feel that a longer sample time gives more accurate tuning information and resolution. Pulls are made from the lowest RPM the bike will allow without bog, buck or stall. This information is important and results in a more complete tune. There are graphs showing the affects of 4th vs. 5th gear runs in our library under ?Tests and Compare-misc?.
The dyno itself is a Dynojet upgraded to a model 250. Each upgrade included software
updates from Dynojet to account for drum mass changes, accessories and Winpep software revisions. The latest Winpep software version available is currently used. A 250i version has been purchased and is scheduled for installation early ?06.
The dyno room is by Industrial Noise Control, specifically designed for dyno applications. It is located inside the service department with the intake and exhaust from outside. It is a ?pull through? system using two INC fans mounted overhead which tends to be self balancing. Four additional fans are inside the room for engine and exhaust cooling. After installation, an outside environmental company was hired to measure air flow throughout the room. This was to insure exhaust gases would be properly purged, check for air pressure rise or drop, dead air pockets, adequate incoming air and balanced air flow. This testing showed a minor pressure drop inside the room as compared to the ambient air in the service department, when both overhead fans were running. This was corrected by changing the drive pulley ratio on one of the overhead fans to reduce its speed. As is reasonable, the inside fans had no affect on pressures since they recirculate the air inside the room. These tests were important for health reasons and to insure dyno readings were not skewed by either a positive or negative pressure. A minor depression is desired to insure contaminants are expelled. Conditioning of the air is not practical due to the volumes involved but heat from the bike does change room temperature to some degree. The dyno sits in a pit to ease loading/unloading of bikes. The pit is open around the dyno to allow ?on lift? type work on the bike.
Last, Portland is blessed with a near perfect Dyno environment. We typically have low to moderate humidity, fresh marine air, moderate temperature and low altitude (our dyno is set to 150 feet altitude). This translates into the best dyno numbers possible (i.e., we routinely have really ?good air?). Even with corrected numbers, environments in high altitudes or ?poor air? (high air density altitude, humidity or temperature) may struggle to realize similar values. As a tuning tool, the dyno is the best tool we currently have to fully optimize motor/bike combinations. As a measuring tool, it can tell us the affects of component changes as well. What the dyno cannot do is change or correct the way a motor processes air, this is a function of the component selection.
This methodology is the result of 16+ year?s dyno operation time, constant training, working with industry leaders and the daily learning opportunities that come from every unique bike finding its way onto our dyno. There are many very good tuners and operators across the country that may have techniques or solutions that we may have yet to discover. We make no pretensions of believing we are the best or most knowledgeable. Conversely, Latus Motors Harley-Davidson?s history, experience, ability to pull intellectual resources from some of the industries greatest talents, pride in our work and overwhelming dedication to give our customers the best possible product we believe puts us among the industry leaders.
Latus Motors Harley-Davidson