Rocksim and Rasaero were extremely wrong

[G2Rockets]

I have flown many rockets after doing simulations and have become fairly confident in the results and adjusting accordingly, BUT last weekend at SB3, I had something happen that is just weird.

I flew a mini darkstar on a I204 and a H255. I simmed both and was seeing around 8000 feet with both simulators(one just gets there faster than the other). The I204 was predicted to fly higher.

The actuals were...
I204 - 4700 ft.
H255 - 6,428ft.

It just goes to prove that no matter how well of a grip you think you have on this hobby, something comes along and proves that you don't know everything.

 

[troj]

Did you have your camera on both of these? If so, that created a tremendous amount of drag.

-Kevin

 

[G2Rockets]

Did you have your camera on both of these? If so, that created a tremendous amount of drag.

-Kevin

Good Point, Yeah I did, but that still leaves me baffled. I would have thought the slower burn of the I204 would have had a lower Cd than the faster burn of the H255. It seems to be the opposite, unless when the camera came off it was early enough in the flight that it had less time with that drag.

I wish I could have found the camera and seen when it came off.

John

 

[[POW]Eagle159]

I don't think a little camera can cause a 1/2 reduction of altitude though.... But that is interesting to see.

 

[cjl]

A camera attached to the side would make a huge difference, and if the camera fell off on the H255, that explains the extra altitude there too. All other things equal, an I204 will pretty much always outperform an H255, so there had to be another factor (such as the camera coming off).

 

[[POW]Eagle159]

I think a half mile IS a lot less altitude with a little camera... I don't think the LPR are that affected:y:

 

[Rocketman248]

I don't think a little camera can cause a 1/2 reduction of altitude though.... But that is interesting to see.

The Darkstar Mini is a very small rocket, so the camera is quite a large area relative to the size of the rocket. If you put one on a big 4" Darkstar Extreme, the difference would be much less drastic.

 

[Chuck Rogers]

The difference was the addition of the camera.

If the camera had a shroud covering it, take the average angle of the front face of the shroud, and the shroud frontal area, and turn it into a fin canister. Then rerun it on RASAero.

If the camera didn't have a shroud, and was just mounted on the side of the rocket, then measure the frontal area of the camera, and add a launch shoe with one-half of that frontal area (explanation below). Then rerun it on RASAero.

When you have a particularly large protuberance on a rocket, or a bunch of small protuberances, you've got a couple options in RASAero. You can add a fin canister, rail guide, or launch shoe (pick the one whose shape best matches the shape of the protuberances), and make sure the fin canister has the same frontal area, and the rail guide or launch shoe have one-half the frontal area of the protuberances.

In RASAero you enter the dimensions for one rail guide and the frontal area for one launch shoe, and then the software assumes you have two rail guides and two launch shoes. So, as an example, if you have a bunch of vertical forward-facing surfaces, add up the frontal area, then divide by two, and enter that frontal area for the launch shoe.

What seems like small protuberances on a rocket can have a big effect on the altitude, especially if they have vertical forward facing surfaces, and especially if you are going supersonic. Even a protuberance that you model with a fin canister can start to add up drag-wise.


Chuck Rogers
Rogers Aeroscience

 

[G2Rockets]

The difference was the addition of the camera.

If the camera had a shroud covering it, take the average angle of the front face of the shroud, and the shroud frontal area, and turn it into a fin canister. Then rerun it on RASAero.

If the camera didn't have a shroud, and was just mounted on the side of the rocket, then measure the frontal area of the camera, and add a launch shoe with one-half of that frontal area (explanation below). Then rerun it on RASAero.

When you have a particularly large protuberance on a rocket, or a bunch of small protuberances, you've got a couple options in RASAero. You can add a fin canister, rail guide, or launch shoe (pick the one whose shape best matches the shape of the protuberances), and make sure the fin canister has the same frontal area, and the rail guide or launch shoe have one-half the frontal area of the protuberances.

In RASAero you enter the dimensions for one rail guide and the frontal area for one launch shoe, and then the software assumes you have two rail guides and two launch shoes. So, as an example, if you have a bunch of vertical forward-facing surfaces, add up the frontal area, then divide by two, and enter that frontal area for the launch shoe.

What seems like small protuberances on a rocket can have a big effect on the altitude, especially if they have vertical forward facing surfaces, and especially if you are going supersonic. Even a protuberance that you model with a fin canister can start to add up drag-wise.


Chuck Rogers
Rogers Aeroscience

Chuck,
Thanks for the information. It is so easy to neglect the small camera on the rocket. It doesn't seem that it should make that much difference, but it does. I still think the second launch lost the camera early. This would explain why the H255 went higher than the I204. This actually helps alot as I have a 6x upscale of the estes super neon being built for LDRS. Troj keep reminding me that I will have an enormous amount of drag on that rocket. This actually helps visualize just how much.

John

 

[Chuck Rogers]

For the upscale Estes Super Neon, take one of the cylindrical tube fins and get the circumference.

circumference = 2 * 3.14159 * (diameter / 2)

Multiply this by the number of tubes (6). Note that this isn't the 6x upscale, it is because there are 6 tube fins.

Take that big circumference (circumference of one tube times 6), and find the diameter for that circumference.

Enter that diameter into RASAero as a GIANT launch lug. You'll get a pretty accurate drag coefficient prediction for the rocket.

The launch lug model used in RASAero assumes that there are two launch lugs in line with each other, but the first launch lug makes most of the drag. From previous flight data I've seen little difference in the drag between single launch lugs and two launch lugs when mounted on the same diameter rockets. So this launch lug based model should be pretty accurate for the upscale Neon rocket.


Chuck Rogers
Rogers Aeroscience

 

[[POW]Eagle159]

Chuck,
Thanks for the information. It is so easy to neglect the small camera on the rocket. It doesn't seem that it should make that much difference, but it does. I still think the second launch lost the camera early. This would explain why the H255 went higher than the I204. This actually helps alot as I have a 6x upscale of the estes super neon being built for LDRS. Troj keep reminding me that I will have an enormous amount of drag on that rocket. This actually helps visualize just how much.

John

6x UPSCALE !!!!!!! OMG that is big.... It would be cool to see a build thread or something. :wink:

 

[gpoehlein]

Another way to figure it is the way we used to do it in Rocsim before they added tube fins - Just put 18 fins on the rocket that are the same shape as the side view of each tube fin. So, if the tube fins are square cut, 3" long and .976" diameter (for a BT-50 as an example), each fin would have a 3" root, 3" tip, no sweep and .976" span. Thickness of the fin is the same as the thickness of the tube. So if you spread 18 of those fins around the circumference of the rocket, you will approximate the stabilizing factor and drag of the tube fins.

 

[G2Rockets]

For the upscale Estes Super Neon, take one of the cylindrical tube fins and get the circumference.

circumference = 2 * 3.14159 * (diameter / 2)

Multiply this by the number of tubes (6). Note that this isn't the 6x upscale, it is because there are 6 tube fins.

Take that big circumference (circumference of one tube times 6), and find the diameter for that circumference.

Enter that diameter into RASAero as a GIANT launch lug. You'll get a pretty accurate drag coefficient prediction for the rocket.

The launch lug model used in RASAero assumes that there are two launch lugs in line with each other, but the first launch lug makes most of the drag. From previous flight data I've seen little difference in the drag between single launch lugs and two launch lugs when mounted on the same diameter rockets. So this launch lug based model should be pretty accurate for the upscale Neon rocket.


Chuck Rogers
Rogers Aeroscience

Chuck,

How confident are you in those figures?

I am coming up with a cd of 52.
Scratch that I went back and refigured and it was a cd of 204.

Did i do it correctly?

John

 

[Chuck Rogers]

Chuck,

How confident are you in those figures?

I am coming up with a cd of 52.
Scratch that I went back and refigured and it was a cd of 204.

Did i do it correctly?

John

You correctly followed my incorrect instructions.

Yes, that CD value is incorrect. The mistake on my part was that the launch lug drag has a very large interference drag factor; the small launch lug on a large rocket affects the flow over the rocket over a wide area. The actual increase in drag is much larger than just the drag of the launch lug itself. Of course, when the launch lug (the tube fins) is the same diameter as the rocket body, and the tube fins are mounted on the bottom of the rocket, this effect is just not present. While there will be interference drag between the tube fins and the rocket body, it will be nothing like the interference drag between a small launch lug and a large rocket body, relative to the small frontal area of the launch lug.

Looking up some wind tunnel data for a flow-through near-cylinder configuration, with 20% mass flow blockage (20% of the mass flow can't make it through the cylinder and spills over the front lip, probably very conservative for a tube fin), the drag coefficient based on the frontal area of the tube fin would be 0.10. Since this number is probably very conservative, I'd use an interference drag factor of 1.0. Since there are 6 tube fins, and they have the same frontal area as the body (the reference area for the CD), I'd take the CD for the body alone and the regular fins, and then add 0.60 to the CD for the 6 tube fins.


Chuck Rogers
Rogers Aeroscience