How to estimate a constant drag coefficient of a boat?

[Panos_leof]

TL;DR Summary

Ways to calculate drag coefficient for resistance analysis without knowing drag force and velocity.

Hi there,

I have modeled with MAXSURF a few variations of a prehistoric logboat in order to test a few theories. Now, I am trying to look at resistance-performance over a range of speed. To do so, I need to include a constant drag coefficient without knowing the drag force and velocity (both will be estimated after the resistance analysis).

So, CD = (2*FD)/(ρ*v2*A) is not an option since I am missing FD and V. Also, I cannot solve Cd in terms of the Reynolds number since Re and Cd can relate through velocity.

Is there any other workaround? For now, I just need a constant drag, then I will just use speed^2 and projected area.

Thanks in advance

 

[PhanthomJay]

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4174551/#Sec2title
Maybe this site helps.
There are a lot of factors at play.

 

[anorlunda]

The drag critically depends on the shape and the surface roughness. Those things are very hard to express mathematically, so the usual approach is empirical.

The timber industry has a long history of pulling logs in the water. The tug boat industry has a long history of pulling barges.

I'll try some searches to see if they have any standard data. Busy today, so give me 72 hours.

Edit: It didn't take as long as I thought. I found this:

If the vessel is a barge, sometimes a simplification is adopted...

Calculation of transverse underwater hull area is pretty simple in case of barges, which generally have a rectangular section shape. If the width of the Barge is B, and its draft is T, then the underwater transverse section area is simply B x T. If there are cuts around the bilge of the barge, these can be deducted from the area. The current force is finally calculated using the standard formula

Current force = ½ x water density x (current speed)2 x underwater transverse section area

 

[Dr. Courtney]

Drag coefficients are best determined experimentally.

Most theoretical predictions of drag coefficients have significant errors, often 10% or more.

It's not usually hard to get experimental measurements much better than that.

Of course, you need to have the actual object of interest.

In the case of a real boat, my approach would be to get it to a constant velocity (with propulsion) at the top of the velocity range of interest, remove the propulsion, and get a good video as it slows. Then I would use Tracker to analyze the video and infer the drag from the slowing.

 

[Panos_leof]

The drag critically depends on the shape and the surface roughness. Those things are very hard to express mathematically, so the usual approach is empirical.

The timber industry has a long history of pulling logs in the water. The tug boat industry has a long history of pulling barges.

I'll try some searches to see if they have any standard data. Busy today, so give me 72 hours.

Edit: It didn't take as long as I thought. I found this:

That's spot on. Thanks

 

[Panos_leof]

Drag coefficients are best determined experimentally.

Most theoretical predictions of drag coefficients have significant errors, often 10% or more.

It's not usually hard to get experimental measurements much better than that.

Of course, you need to have the actual object of interest.

In the case of a real boat, my approach would be to get it to a constant velocity (with propulsion) at the top of the velocity range of interest, remove the propulsion, and get a good video as it slows. Then I would use Tracker to analyze the video and infer the drag from the slowing.

That's an interesting approach thanks! Well, I actually have the surface model in a digital form though. So, do you believe that the use of an approximate Cd would not allow me to later infer the actual one?

 

[Panos_leof]

Thanks for your help guys. I'll try to test a few of the things proposed and integrate them into MAXSURF resistance software. I'll keep you posted for the results.

 

[Dr.D]

Effective hydrodynamic drag is strongly dependent on wave making which in turn depends upon hull length. Wave drag is a separate issue from viscous drag.

 

[Dr. Courtney]

That's an interesting approach thanks! Well, I actually have the surface model in a digital form though. So, do you believe that the use of an approximate Cd would not allow me to later infer the actual one?

Infer? How?

Use of an approximate Cd from theoretical considerations does not usually prevent an accurate experimental determination - except perhaps by convincing the relevant parties not to bother with the actual experimental measurements.

My experience is that experimental measurements of drag tend to be larger than theoretical predictions. Those hoping to avoid the "bad news" tend to argue against the usefulness of actual measurements.