Weather effects on SAE Horsepower

[strokerjlk]

You tune a motor and it makes 100hp uncorrected. The customer leaves happy. He comes back from Sturgis where he ran it on the dyno to show everyone and it made 78.74hp uncorrected. The graph looks perfect just low. Was the Sturgis dyno wrong, was something wrong with your dyno, or is something wrong with the motor?

come on..... you know all you get at sturgis is STD

[Steve Cole]

The dyno doesn't measure the quality of the air. Get in some dyno shops and you can smell the exhaust fumes. Do you think it's going to make the same power when running on that air as it would good clean air at the same temperature? Do you think the correction factor is going show the bad air? As long as the temperature, humidity and pressure remain the same the dyno just doesn't know any better. There is just a group of things that can and do get used to make things look better or worse than they really are based on what one is looking for. Without having the supporting equipment to be able to see what and how things were really run, it's impossible to say for sure just by looking at a dyno sheet that says SAE corrected.

When you looking for changes that are down in the 5HP area you really need to know much more than is on the dyno sheet.

[1FSTRK]

Steve you are obviously a smart man but you often come here to tell us how complicated things are and many time with what seems to be the tone that we should not even try. The dynos are here to stay and some of us are willing to work to make it the best we can with what we have.  Are we going to have climate controlled Laboratory grade SAE certified graphs for every run posted here, NO. I understand that, but I do not buy into the BS theory that if it is not perfect there is no point in making the improvements that are possible. The posting in SAE rule here helped. If dyno operators would make the attempt to insure their dyno stack was recording the conditions of the air being ingested by the motor and not the temp and humidity of the computer stack itself it will be another good step. Another thing would be to final tune and post runs made within the SAE intended range of air temperatures of 59 to 95 degs F, and the intended range of dry air pressures is 26.58 to 31.01 inHg. Many times I see runs at 45 to 50 degs. and even though they correct on the graph at SAE 1.00 the temp is outside the accepted range of accuracy for the calculation. Will we all post sheets at the SAE J1349 standard of 77 degs, 29.234 inHg, 0% Humidity, no but if we try for those numbers and agree to stay within the accuracy parameters will things get better, I think so. It comes down to agendas, if the agenda is to brag, misrepresent, or sell poor products then standardization is bad. If the agenda is to share actual accomplishments, give and receive useful information then standardization is good.

[Merc63]

If I want to know how much HP my bike made on the dyno with no correction, at this very moment.  Just whatever the bike is putting down regardless of pressure, temp etc...  What would I use?

[1FSTRK]

If I want to know how much HP my bike made on the dyno with no correction, at this very moment.  Just whatever the bike is putting down regardless of pressure, temp etc...  What would I use?

Why would you want that number?

[Merc63]

I dont care about all that correction BS.  I just want to know how much power my bike is putting down at that exact moment, in the conditions it is in.

[build it]

Merc, uncorrected or actual. At the track, SAE and STD don't mean diddly. I've seen these cfs show higher and lower than actual in different situations. You still want all the sheets though.

[1FSTRK]

I dont care about all that correction BS.  I just want to know how much power my bike is putting down at that exact moment, in the conditions it is in.

That is the problem, it is not BS. To measure power you first have to define what power is and then you cannot stop with half the definition, so you are not getting anyone's definition of power using the uncorrected data graph.
Hp is calculated not measured. And we seem to forget that. When we look at the uncorrected graph on the Dynojet screen we are looking at the data that was collected and run through half the formula. You then go to the drop down box and pick which definition of HP you want the computer to use to finish the calculation, SAE, STD, DIN, EEC, JIS. You now have a graph of that shows whichever type of hp you specified. It would have been a lot less confusing if they had wrote the program so that when you chose uncorrected it labeled everything on the screen DATA instead of HP because that is all it is. You cannot even use the uncorrected data to see if your motor is still in good shape; look at my post #23 above. The bike was in perfect shape and yet it went from 100 uncorrected to 78.74 uncorrected and all you need to do to get the 100 uncorrected back is wait for the weather to change for the better. I know people that would be taking a motor apart if they lost over 22 points when they ran the second test and they would be wasting a lot of time and money.

[Steve Cole]

I am sorry if you took my post that way, that was not my intent. The idea is that people need to understand what is real and what is not IMHO. Just because you meet part of the SAE standard doesn't make it SAE. Yes, I have to work in a world that we follow all the specifications not just part of them. If you look at how things work and understand it then you have a chance to know what is real and what is not. Without the knowledge of right and wrong you have nothing to base your decisions on.

Using your example, knowledge is what made the right decision but there is just way to much missing on a simple dynosheet to make big decisions from. DynoJet will even tell you that no two of there dyno's will give you the same results, close but not the same. I do believe in dyno's we have 5 of them but I also know the limitations they each have.

HP= torque*RPM/5252

The Dynojet dyno measures "rate of acceleration" as do most motorcycle dyno's and this in itself creates errors as they have no direct measurements of torque. That in itself puts it outside of the SAE specification. This is why you get various measurements for different gears when running on them. Finding the spot that best matches what your bike has to accelerate the mass is going to show the highest power. Take a dyno that has been unused over night in cool temperatures and I have seen power drops of as much as 10HP. While that good cool air should be making more power it shows less! To me that's a huge amount on a motorcycle measurement. All one needs to do is run the dyno with a bike in place to get the temperatures up in the bearings and that power shows right back up. So how do you tell people that have no idea of these this?

In a perfect world everyone would check all of this as part of the daily dyno preparation but it just doesn't happen. So when I see these dynosheets where one test is at 55 degrees and another at 95 degrees there is as much if not more going on that the correction has no part of to try and correct. Just as the other things I have tried to point out that you need to know about to make a decision.

The Dyno is a great tool when used to compare before and after in semi-controlled conditions. Now those conditions do not have to be a lab environment but there does need to be some control to get results that are believable and repeatable within there limits.

[redmtrckl]

So how about when an old dyno drum has the surface all worn down until it is almost slick? Or the bike has a worn out tire? Are there any slippages back there and wouldn't this affect the data?  Also tire pressure changes as things heat up. Seems there are lots and lots of variables that can happen even during a single tuning session that is not even connected to weather.

[Jamie Long]

A dyno is an awesome tool for testing and R&D however the biggest consideration to note is that while the actual dyno itself may be pretty consistent one dyno to the next or day to day, EVERYTHING else is relative. The term relative is an especially critical element when making comparisons, as things well.........change. Obviously there can be huge differences in air in terms of temp, pressure, and humidity, however air quality in the dyno cell is extremely important as well. One of the biggest variances we see one shop to the next is the actual technique they use when making dyno runs or evaluations, to be consistent you need to monitor everything! You need to make runs at the same engine temp and you need to monitor/graph ignition timing to make sure you have max timing during the runs. Do 5 runs at different engine temps and your max timing may be different each run, this can and will affect your numbers greatly. If possible in most cases you want to make the runs at 1:1 as this will affect the numbers. The better you control the conditions in the dyno cell as well as the actual technique the more consistent your data will be. Now based in this info you can easily see how comparing results from one dyno to the next are relative, and as far as comparing one bike to another thats a whole other story 

[Jamie Long]

So how about when an old dyno drum has the surface all worn down until it is almost slick? Or the bike has a worn out tire? Are there any slippages back there and wouldn't this affect the data?  Also tire pressure changes as things heat up. Seems there are lots and lots of variables that can happen even during a single tuning session that is not even connected to weather.

You can see tire or clutch slippage if you graph the run by gear ratio/engine speed on the X/Y axis. Tire pressure, belt/chain tension, and the temps of various components are all certainly factors that are relative.

[1FSTRK]

Both Jamie and Steve have given great examples of what would be taught in a dyno operator's class 101. I did not set out to hold such a class here. I would like to stay on the points of weather conditions and how to best use the dynojet dyno to achieve a reasonable amount of repeatability. We do not need to even address things the dyno does not do or what it does not measure, we need only to talk of what it does and how to best use it to achieve the most repeatable comparable results with regard to weather changes. Dyno jet used the title SAE and that is the format we picked for posting here. It does not matter if you call it SAE horsepower or Dynojet points as long as we determine the procedure and a range in which it will be repeatable.

[Jamie Long]

Both Jamie and Steve have given great examples of what would be taught in a dyno operator's class 101. I did not set out to hold such a class here. I would like to stay on the points of weather conditions and how to best use the dynojet dyno to achieve a reasonable amount of repeatability. We do not need to even address things the dyno does not do or what it does not measure, we need only to talk of what it does and how to best use it to achieve the most repeatable comparable results with regard to weather changes. Dyno jet used the title SAE and that is the format we picked for posting here. It does not matter if you call it SAE horsepower or Dynojet points as long as we determine the procedure and a range in which it will be repeatable.

In the end the things we cannot control specifically give us the biggest challenges, in this example the biggest consideration is the air itself, this changes hour by hour, day by day. Regarding your statement specifically the best advice I can give to keep dyno results most repeatable is to have a well developed dyno cell or test area, this will in most cases allow you to maintain the temp within a reasonable variance, however you still simply have no control over pressure and humidity (short of a laboratory level test cell). So... keep your individual test standard's and evaluation procedures consistent, keep the dyno cell conditions the best you can with the resources available, and draw your conclusions off of the data. Remember motors can/will run different day to day based on temp, fuel, on and on as everything is relevant, testing on the dyno is no different. All you can do it try to keep everything as consistent as possible.

[Steve Cole]

Both Jamie and Steve have given great examples of what would be taught in a dyno operator's class 101. I did not set out to hold such a class here. I would like to stay on the points of weather conditions and how to best use the dynojet dyno to achieve a reasonable amount of repeatability. We do not need to even address things the dyno does not do or what it does not measure, we need only to talk of what it does and how to best use it to achieve the most repeatable comparable results with regard to weather changes. Dyno jet used the title SAE and that is the format we picked for posting here. It does not matter if you call it SAE horsepower or Dynojet points as long as we determine the procedure and a range in which it will be repeatable.

For what it's worth IMHO you would need to develop some minimum conditions that have to be checked and recorded. What those are exactly I do not know but to get to the point of reasonable repeatable results there has to be something that everyone can do. What is currently there with a DynoJet is not enough. SAE has set the standard to follow but if you want to cut back from there requirements, what should it be?

[strokerjlk]

The uncorrected info would show that you made the same power at beginning and end, but the corrected would put the 2.69 lost by weather changes back in and then would show a 2.69 gain that you made with the tuning.

Both Jamie and Steve have given great examples of what would be taught in a dyno operator's class 101. I did not set out to hold such a class here. I would like to stay on the points of weather conditions and how to best use the dynojet dyno to achieve a reasonable amount of repeatability. We do not need to even address things the dyno does not do or what it does not measure, we need only to talk of what it does and how to best use it to achieve the most repeatable comparable results with regard to weather changes. Dyno jet used the title SAE and that is the format we picked for posting here. It does not matter if you call it SAE horsepower or Dynojet points as long as we determine the procedure and a range in which it will be repeatable.

back to weather effects on sae HP.
here is a interesting comparison.
this is autoworkers bike . each time it was tuned by Bruce Bean and just rerun on my dyno.
so there is no reason to try and discredit me or Bean .  (usual MO of some)
bean could give a puck about numbers,and I dont give a chit ,what someone else's tune pulls up on my dyno.
I got two bottles of goooood bourbon from autoworker,and the three of use had a blast.
autoworker can give the details. but basically it was a piston swap and compression increase.
so we cant compare two diff sessions with the same exact combination.
but I thought you (1fstrk) would find the conditions interesting.
so how much did Autoworker gain?

[Jeffd]

can the 250i accurately predict what a bike will make that is in an area where say the max kpa that can be attained is say 93 vs an area that is 100?  seems like in an area where 100kpa is not possible it would be hard to duplicate what the bike can actaully do at 100kpa.

[Admiral Akbar]

You tune a motor and it makes 100hp uncorrected. The customer leaves happy. He comes back from Sturgis where he ran it on the dyno to show everyone and it made 78.74hp uncorrected. The graph looks perfect just low. Was the Sturgis dyno wrong, was something wrong with your dyno, or is something wrong with the motor?

Now the rest of the story, there was nothing wrong with either dyno or the bike. Both dyno runs were properly performed and gave accurate results, the bike made less power because of the conditions. This is where corrected data comes in.

The first run was in the spring at sea level with a high pressure system in the area with these conditions.
Temp 59.5, Pressure 31.0, Hum.5%      SAE correction = .915       
Uncorrected 100hp        SAE 91.5hp

The second run in August at Sturgis which is known for bad conditions resulted in this.
Temp 95.0, Pressure 27.5, Hum. 82%   SAE correction=1.162       
Uncorrected 78.74hp     SAE corrected 91.5hp

The same effect happened in the runs you made above which is why I pointed out that 2.69hp of the uncorrected gain was due to the weather change alone. I am not taking away from the tuning job or saying anything could have been done better, but what if the bike came in tuned and you made some fine adjustments that made a 2.69 hp improvement and during that time the condition got worse by the same amount? The uncorrected info would show that you made the same power at beginning and end, but the corrected would put the 2.69 lost by weather changes back in and then would show a 2.69 gain that you made with the tuning.

Your example above doesn't say jack..You can't assume because the measured numbers are 91.5 corrected that the motor is performing the same.. You need to run the bike back to where the original run was done to make sure the numbers didn't change.. You are trying to use a motor which can change performance  compare 2 dynos.. You'd be better off using an electric motor and comparing uncorrected readings..

How do you know the dynos gave accurate results? With experienced operators they dynos gave consistent numbers which has nothing to do with accuracy..

Max

[1FSTRK]

You tune a motor and it makes 100hp uncorrected. The customer leaves happy. He comes back from Sturgis where he ran it on the dyno to show everyone and it made 78.74hp uncorrected. The graph looks perfect just low. Was the Sturgis dyno wrong, was something wrong with your dyno, or is something wrong with the motor?

Now the rest of the story, there was nothing wrong with either dyno or the bike. Both dyno runs were properly performed and gave accurate results, the bike made less power because of the conditions. This is where corrected data comes in.

The first run was in the spring at sea level with a high pressure system in the area with these conditions.
Temp 59.5, Pressure 31.0, Hum.5%      SAE correction = .915       
Uncorrected 100hp        SAE 91.5hp

The second run in August at Sturgis which is known for bad conditions resulted in this.
Temp 95.0, Pressure 27.5, Hum. 82%   SAE correction=1.162       
Uncorrected 78.74hp     SAE corrected 91.5hp

The same effect happened in the runs you made above which is why I pointed out that 2.69hp of the uncorrected gain was due to the weather change alone. I am not taking away from the tuning job or saying anything could have been done better, but what if the bike came in tuned and you made some fine adjustments that made a 2.69 hp improvement and during that time the condition got worse by the same amount? The uncorrected info would show that you made the same power at beginning and end, but the corrected would put the 2.69 lost by weather changes back in and then would show a 2.69 gain that you made with the tuning.

Your example above doesn't say jack..You can't assume because the measured numbers are 91.5 corrected that the motor is performing the same..

But you can assume that the motor is not performing the same and the two different dynos running the same type of sensors and software just guessed the number 91.5 by coincidence. The math behind this software is taught in grade school and making copies of a program is taught there as well. You make these dynos out to be rocket science, and as far as the sensors I am sure they are not laboratory grade but I can go to Walmart and pick up ten Chinese digital weather stations for $12.00 a piece and all ten will read the exact same readings. I know the Dynojet sensors are at least that consistent. That is pretty good Max, the dynos must be wrong because the bike power can vary and the bike power must have varied because the dynos could be wrong.

You need to run the bike back to where the original run was done to make sure the numbers didn't change.. You are trying to use a motor which can change performance  compare 2 dynos.. You'd be better off using an electric motor and comparing uncorrected readings..

Not if you read this thread. The post was intended to show the value of corrected readings over the uncorrected which include the weather variables.

How do you know the dynos gave accurate results? With experienced operators they dynos gave consistent numbers which has nothing to do with accuracy..

Max

And you are the one talking accuracy,I am talking repeatability and comparability. 95% of the dyno info is collected and posted in Dynojet format, I do not care if dynojet reads and reports in peanuts instead of HP as long as they can record consistently they can produce comparable data. I have seen it done and all the while I keep reading on the internet how they cannot, it is like going to the shooting range and every guy that cannot hit Sh*t  is talking about how his gun must be off or it is bad ammo.

[BVHOG]

It's amazing with the inaccuracies of these dynos that I see new cvo bikes on GMR's dyno at high elevation making right around 80 horse and a tad over 100 torque, within a couple of what I see here at 1000 ft elevation on a completely different dyno. 

[1FSTRK]

back to weather effects on sae HP.
here is a interesting comparison.
this is autoworkers bike . each time it was tuned by Bruce Bean and just rerun on my dyno.
so there is no reason to try and discredit me or Bean .  (usual MO of some)
bean could give a puck about numbers,and I dont give a chit ,what someone else's tune pulls up on my dyno.
I got two bottles of goooood bourbon from autoworker,and the three of use had a blast.
autoworker can give the details. but basically it was a piston swap and compression increase.
so we cant compare two diff sessions with the same exact combination.
but I thought you (1fstrk) would find the conditions interesting.
so how much did Autoworker gain?

It is clear from the uncorrected data sheet that the dyno recorded an increase of 6.49. We can clearly see the weather conditions for the blue run were warmer with less pressure so we know the blue run would have been higher if the bike had been run under the red run conditions. By going to dynojets version of SAE correction the software corrects the uncorrected data from both runs up or down to a single set of conditions thereby making them comparable.  The corrected graph shows the changes made to the bike actually produced an increase of 2.93. As Jim pointed out this does not guarantee the state of tune for either run. The dyno cannot predict how far out of tune the weather change caused the bike to be, it is the dyno operator's responsibility to optimize the tune for each set of conditions so the software can correct for the uncontrollable effect that weather has on a tuned motor. This is one place operators that are new to the game are at a disadvantage, the carbureted motors this made very apparent because nothing was automatic. When the weather changed you lost because of the weather and because the tune was wrong, you then rejetted, and checked timing, for the new conditions and made a run and you were right back to the same corrected numbers.

[Steve Cole]

It is clear from the uncorrected data sheet that the dyno recorded an increase of 6.49. We can clearly see the weather conditions for the blue run were warmer with less pressure so we know the blue run would have been higher if the bike had been run under the red run conditions. By going to dynojets version of SAE correction the software corrects the uncorrected data from both runs up or down to a single set of conditions thereby making them comparable.  The corrected graph shows the changes made to the bike actually produced an increase of 2.93. As Jim pointed out this does not guarantee the state of tune for either run. The dyno cannot predict how far out of tune the weather change caused the bike to be, it is the dyno operator's responsibility to optimize the tune for each set of conditions so the software can correct for the uncontrollable effect that weather has on a tuned motor. This is one place operators that are new to the game are at a disadvantage, the carbureted motors this made very apparent because nothing was automatic. When the weather changed you lost because of the weather and because the tune was wrong, you then rejetted, and checked timing, for the new conditions and made a run and you were right back to the same corrected numbers.

Let's for the moment say it just a DJ measurement. The highlighted part of what you said is one of the reasons why conditions between runs need to be taken into account and needs to remain as close to the same as possible. The engine is really changing performance and since the engine is running open loop at WOT (where the measurement is being taken) there are NO  corrections taking place with HD fuel injection, much the same as a carburetor. Sure you can change the tune to make it look better in those conditions but when the conditions change yet again your right back out in left field for the new conditions, so to speak.  This is why some rules like SAE has done need to be applied if you want to make it so you can compare and understand what is on the dyno sheet.

[Admiral Akbar]

It's amazing with the inaccuracies of these dynos that I see new cvo bikes on GMR's dyno at high elevation making right around 80 horse and a tad over 100 torque, within a couple of what I see here at 1000 ft elevation on a completely different dyno.

First off you are using the terms wrong..

http://en.wikipedia.org/wiki/Accuracy_and_precision

Comparing a populations gets you an idea of what the precision is of the 2 dynos. You could infer that since the populations cam from the same source that measurement accuracy or inaccuracy between the to dynos is the same..

You can't even come close to this by saying since 2 numbers on 2 to different dyno have matched that both exhibit the same accuracy..  You could if you moved back and forth between them a couple times.. How else would you guarantee the standard hasn't changed..

I worked in test for the last 18 years.. If I had reported 2 testers had the same accuracy based on a test of one unit because the numbers read the same for one test on each tester, I would have been laughed at cuz they figured I was joking..

Now the fact that the numbers did read the same shows promise but it still need validation..

Max

[BVHOG]

It's amazing with the inaccuracies of these dynos that I see new cvo bikes on GMR's dyno at high elevation making right around 80 horse and a tad over 100 torque, within a couple of what I see here at 1000 ft elevation on a completely different dyno.

First off you are using the terms wrong..

http://en.wikipedia.org/wiki/Accuracy_and_precision

Comparing a populations gets you an idea of what the precision is of the 2 dynos. You could infer that since the populations cam from the same source that measurement accuracy or inaccuracy between the to dynos is the same..

You can't even come close to this by saying since 2 numbers on 2 to different dyno have matched that both exhibit the same accuracy..  You could if you moved back and forth between them a couple times.. How else would you guarantee the standard hasn't changed..

I worked in test for the last 18 years.. If I had reported 2 testers had the same accuracy based on a test of one unit because the numbers read the same for one test on each tester, I would have been laughed at cuz they figured I was joking..

Now the fact that the numbers did read the same shows promise but it still need validation..

Max

Holy "Potty mouth" are you good at complicating something simple, just an observation was all I was making, not saying any of these machines are test grade by any stretch, just two very different dynos in very different environment coming up with similar results.

[Admiral Akbar]

Holy "Potty mouth" are you good at complicating something simple, just an observation was all I was making, not saying any of these machines are test grade by any stretch, just two very different dynos in very different environment coming up with similar results.

Sorry Bob.. I shouldn't have a used a quote from you(Bob) to start the discussion.. I was also meaning both 1FSTRK, you and the audience as "you".. It was a bit of a lead in to the discussion about "populations".  Maybe I was making it too technical..

Max