Comparison with range calculation by Bearman and Harvey

This is the place for any questions and discussions about TrajectoWare.

Moderators: fschmidberger, dtutelman

Post Reply
Rogers
Trusted user
Posts: 4
Joined: Sun Dec 19, 2010 9:09 pm

Comparison with range calculation by Bearman and Harvey

Post by Rogers » Sun Dec 19, 2010 10:00 pm

Hi,

Bearman and Harvey (Aeronautical Quarterly, 27, pp112-122,1976) mention their golf ball trajectory calculation based on their windtunnel from which they determined the drag coefficient measurements on a conventional golfball and with initial conditions : initial speed =57.9 m/s , elevation=10 degrees, and for a spin =3000 rpm they found a range of 150 m.

I inserted these initial conditions in the program and found a range of 186 m. That's quite a difference!

In this program it is not quite clear which type of golfball is used. Can this difference in range be due to their choice of a conventional golfball with dimples.

Regards,

Roger

dtutelman
Site Admin
Posts: 107
Joined: Mon Apr 09, 2007 8:55 pm
Location: New Jersey, USA
Contact:

Re: Comparison with range calculation by Bearman and Harvey

Post by dtutelman » Sun Dec 19, 2010 10:33 pm

Rogers wrote:Hi,

Bearman and Harvey (Aeronautical Quarterly, 27, pp112-122,1976) mention their golf ball trajectory calculation based on their windtunnel from which they determined the drag coefficient measurements on a conventional golfball and with initial conditions : initial speed =57.9 m/s , elevation=10 degrees, and for a spin =3000 rpm they found a range of 150 m.

I inserted these initial conditions in the program and found a range of 186 m. That's quite a difference!

In this program it is not quite clear which type of golfball is used. Can this difference in range be due to their choice of a conventional golfball with dimples
Roger,
Interesting discrepancy. Thanks for bringing it to our attention.

I tried the numbers and got 176m, not 186. So there's also a discrepancy between your use of TrajectoWare Drive and mine. Don't know where the problem is.

But that's still a big difference with Bearman & Harvey. So who's right?

Their input numbers correspond very closely to a modern driver with 12* loft and a clubhead speed of 89mph. I used to have about a 90mph clubhead speed in my mid-60s. And I find it hard to reconcile my experience with even 176m, much less 150m; I cleared 200yd (183m) with any clean contact.

Then I tried the same inputs with Max Dupilka's Traj program and got 184m. Closer to your answer, but was not obtained with TrajectoWare Drive.

Certainly TrajectoWare Drive assumes a normal, modern golf ball with dimples.

I would ask what Bearman and Harvey were smoking that day (it clearly wasn't balata). But, to be honest, their paper is very widely respected in the golf industry, so I guess I should be worried. And I would worry if their results were not so much at odds with my experience. As well as all the data we used to validate TrajectoWare Drive. As well as another tried and reasonably true trajectory program. Guess I'm not that worried after all. :)

Cheers!
DaveT

Rogers
Trusted user
Posts: 4
Joined: Sun Dec 19, 2010 9:09 pm

comparison with results from Bearman and Harvey

Post by Rogers » Mon Dec 20, 2010 9:09 am

Dave

I checked my calculations and found also 176 cm. Maybe I was smoking yesterday evening, although I don't remember. Your TrajectoWare in my computer gives the same result, don't worry.
Anyway I have the same feeling. An initial speed of 57.9m/s or 208 mph is a fine speed and shows an experienced golfer. Then a range of 176m is reasonable.
I raised the question just because I was worried about their result, but as you mentioned there paper is well respected so I thought I should be careful.

Thank you very much for your answer.
Regards,

Roger

Rogers
Trusted user
Posts: 4
Joined: Sun Dec 19, 2010 9:09 pm

comparison between range Trajectoware and Bearman (ctnd)

Post by Rogers » Mon Dec 20, 2010 10:43 am

Hi Dave

I was thinking that the only reason for such a discrepency in range values between the TrajectoryWare software and the data by Bearman must be the difference in drag and lift coefficients used in the calculations.

Can you inform me about the coefficients used in Trajectoware. With reference to the often mentioned Negative Magnus effect I am also particularly interested in your drag data around the critical speed.

Regards,

Roger

dtutelman
Site Admin
Posts: 107
Joined: Mon Apr 09, 2007 8:55 pm
Location: New Jersey, USA
Contact:

Re: comparison between range Trajectoware and Bearman (ctnd

Post by dtutelman » Mon Dec 20, 2010 1:40 pm

Rogers wrote:I was thinking that the only reason for such a discrepency in range values between the TrajectoryWare software and the data by Bearman must be the difference in drag and lift coefficients used in the calculations.

Can you inform me about the coefficients used in Trajectoware. With reference to the often mentioned Negative Magnus effect I am also particularly interested in your drag data around the critical speed.
That's not my specialty; perhaps Frank can say something about it.

But we did not do much inventing with the ball's drag and lift coefficients. We started with the algorithm -- complete with coefficients -- from John C. Adams spreadsheet. (See http://sections.asme.org/canaveral/Spec ... megolf.pdf for Adams' presentation on his work. It does spend some time on the golf ball coefficients for lift and drag.) Then we tweaked it just a little to give the best fit to the data we were working from. Not much tweaking was needed.

Note that Bearman & Harvey published in 1976. Golf balls have come a long way (pun definitely intended) since then.

DaveT

fschmidberger
Site Admin
Posts: 58
Joined: Tue Apr 10, 2007 4:13 am
Location: Ravensburg, Germany

Post by fschmidberger » Wed Dec 22, 2010 6:13 pm

Hi Roger,

as Dave says, our calculations are based on the work of John C. Adams.

It is possible to tweak this calculation by yourself. However to do this you must be able to compile a C++ source code to a DLL. I have done this for Visual Studio 6, but it should be possible to do it with Visual Studio Express too.

If you are interested in the source code for the DLL, let me know. But be warned, the support on this will be very limited.

Cheers
Frank

Rogers
Trusted user
Posts: 4
Joined: Sun Dec 19, 2010 9:09 pm

Post by Rogers » Sun Jan 09, 2011 12:23 pm

Hi Frank,

Unfortunately I am not that kind of programmer, I used Matlab a time ago but that is as far I can do. Moreover I think the TrajectoWare program works fine, my basic concern is about the numbers which were put in. How were they obtained, where were they published, for what type of balls were they obtained? Is it possible to extract them from TrajectoWare?

I saw in the reference to John Adams work a figure on drag coefficients versus Reynoldsnumber. Considering the shape of the latter curve it seems to me that these drag coefficients are taken from the work of Bearman and Harvey. Is that correct?

There are two other changes to the software I would appreciate.
1. I would like to compare the calculations with the case without gravity (this will be very simple to imply).
3. I would like to be able to blow up the ball trajectory for smaller speeds so that I could study e.g. the particular case of Negative Magnus effect as explained by Bearman and Harvey.

With best regards,

Roger

Post Reply