Let's put an end to the "Base Drag Hack"

[bad_idea]

Really? I haven’t seen anything really useful for an entire page. It’s mostly the same arguments on repeat.

Then you missed some very interesting posts. That's your loss, but not a reason to scuttle a useful thread.

 

[Daddyisabar]

I love the rocket science arrogance. It is very entertaining! Like Sheldon Cooper on the Big Bang Theory who is insulted when his sister calls him a " rocket scientist." Yes, BAN THE HACKS!" I am always up for a fun little which hunt.

Funny, I think all of these posts have been on topic. Why else would the mods behind the curtain let it go on?

So much intelligence, so little practical wisdom.

 

[NTP2]

Then you missed some very interesting posts. That's your loss, but not a reason to scuttle a useful thread.

i did read some interesting stuff, but it’s seems to have stopped.

 

[kjhambrick]

Not that I've added any useful ideas to the thread but I am still studying Munks Moment to see if there is a way to put it to practical use in a simulation program the way Barrowman 'just works' for "Slender Finned Vehicles"

So far no joy but I am not done looking and learning ...

-- kjh

 

[ThirstyBarbarian]

Some of it has been interesting, but I can’t think of anything actionable yet. At least not for most of us. Maybe the software developers have got some new ideas to think about, but if I just want to know how to figure out if my rocket is going to fly straight or if it’s going to skywrite, I haven’t got much out of this so far.

“You’re doing it wrong!”

“Oh, no! Tell me what to do instead!”

“I don’t know, but just not THAT!”

 

[Daddyisabar]

Bring back the Pendulum Falicy! It worked for 800 years and that ain't bad.

 

[tim cubbedge]

Let’s try to keep this thread on topic.

 

[SolarYellow]

Some of it has been interesting, but I can’t think of anything actionable yet. At least not for most of us.

Only use one of the tubes when building a Star Orbiter.

 

[Alan15578]

Not that I've added any useful ideas to the thread but I am still studying Munks Moment to see if there is a way to put it to practical use in a simulation program the way Barrowman 'just works' for "Slender Finned Vehicles"

So far no joy but I am not done looking and learning ...

-- kjh

Hey, it is great to read that somebody is doing research... I wanted to respond to the Munk Moment post, but I made the mistake of googling it. First, all I got was references to my favorite TV detective. Then I tried something like Munk Moment in fluid dynamics and i got references to pipe flows and conservation of momentum... Then I started paid work on an unrelated project and I have less time to educate the masses. (There might be an apparent mass pun there.) Party on.

 

[Jeff Lassahn]

Personally, I’m just killing time waiting for an answer to my questions. We’ve been advised to do away with the base drag hack and do away with the 1-caliber rule of thumb, which I agree are not perfect, but my question is, if we do away with those, what do we use instead? What information does our sim software give us that tells us if our stubby rockets are going to fly straight or if they need nose weight and how much? If we are tossing out our current tools, what are we using now?

I believe the notion of "replace the 1 caliber rule with a 10% of total length rule" has been suggested several times, and it seem to me it's a good one. So you could, you know, use that and stuff.

As for what to do about the Base Drag Hack, I don't have a recipe for you to follow, but I still believe there is value in understanding in more detail how well it works, why it works or doesn't, and so on. Be patient, if you insist that nothing is useful unless someone shows up with a complete working alternative solution then no progress on the better solution will ever be made.

 

[bad_idea]

I believe the notion of "replace the 1 caliber rule with a 10% of total length rule" has been suggested several times, and it seem to me it's a good one.

Indeed it seems that a proportion of length is a better basis for separating cg from our calculations of cp than calibers are.

I have generally used 8% as my lower bound, and have had no stability issues as yet. I have mind to probe even lower in the safety of an otherwise empty field if I ever get the time for park flying again.

 

[kjhambrick]

Hey, it is great to read that somebody is doing research... I wanted to respond to the Munk Moment post, but I made the mistake of googling it. First, all I got was references to my favorite TV detective. Then I tried something like Munk Moment in fluid dynamics and i got references to pipe flows and conservation of momentum... Then I started paid work on an unrelated project and I have less time to educate the masses. (There might be an apparent mass pun there.) Party on.

@Alan15578 --

Me too !

This is how I started searching for M. M. Munk

Max M Munk was one of the Pioneers:

Advanced Topics in Aerodynamics:  Max M. Munk (1890 – 1986)

NASA > Explore > Max M. Munk

Munk wrote "Aerodynamics of Airships" in Aerodynamic Theory, Vol 5, Division N, Dover Publications ... I am pretty sure I've got a copy of that book somewhere ( I love Dover books ) !

The aerodynamics of airships ... hmmm ... seems relevant for this topic ... I believe I will starting my digging in there ...

Interestingly, Munk worked for a while at Catholic University where James S Barrowman wrote his Master's Thesis, "THE PRACTICAL CALCULATION OF THE AERODYNAMIC CHARACTERISTICS OF SLENDER FINNED VEHICLES" ( Nasa.GOV )

Jim Barrowman mentioned the importance of Munk's work in his Introduction section but then JSB misspelled Munk as Murk in the bibliography !

20. Murk, M. M., The Aerodynamic Forces on Airship Hulls. NASA Report 184, 1924.

It's nice to know that JSB is human afterall

And this is JSB's and JAB's 1966 NARAM-8 Report: THE THEORITICAL PREDICTION OF THE CENTER OF PRESSURE by James s. Barrowman (#6883) and Judith A. Barrowman (#7489)

I am not yet into the aero stuff, just the background ... was that you I saw in my rabbit hole ? ( )

HTH

-- kjh

EDIT for spelling and grammar

 

[Rschub]

This thread has really run its course, let them have their fun.

There is no end to religious wars.

 

[bad_idea]

You'll run across Munk a lot if you research subsonic nosecone drag. A lot of the NACA work on nose cones seems to have skipped straight from Munk's airship equations to supersonic nose cones. I hadn't looked into his work beyond that specific application (which I admit I've not had the time to digest and put into practice), so @Jeff Lassahn 's post about Munk's moments and the applicability to saucers was very interesting.

 

[Buckeye]

Not to mention Cp technically can't be calculated for zero angle of attack. By definition, Cp location is pitching moment about the nose divided by normal force (moment divided by force gives you the distance at which that force is applied). For an axisymmetric rocket, there should be no pitching moment at zero angle of attack and no normal force, so you just get 0/0. For the math minded people, there is a point discontinuity at zero alpha when plotting Cp vs angle of attack. It can be approximated by sweeping angle of attack near zero degrees (like maybe 1, 2, 3, 4 degrees etc) and extrapolating back to 0 degrees. In a perfect world without wind or alignment errors, a statically unstable rocket would still keep going straight if there was nothing to get it to pitch.

I do agree that the base drag hack is certainly not entirely accurate and seems more like a case of artificially getting static margin over 1 caliber when you really just have to have some non-zero positive margin to be stable and enough margin to remain stable at non-zero angle of attack (Cp moves as you vary angle of attack). Something working in simulations does not mean it is what is actually happening.

Yes. This is the conundrum I ran into while attempting to calculate CP from CFD at zero angle of attack to compare with the Barrowman methods. For a symmetrical rocket, the net y and z forces are zero (or nearly so) and thus no moments. So, I used the classical definition.

When I pitch the rocket to an angle in attack in CFD, then there will be moments to calculate as you said. If we assume the CP lies on the central axis, then there can be 2 moment equations, and the solutions for x-location are not unique. CP will be defined as a line, not a point.

The concept of center of pressure on arbitrary bodies gets very foggy.

Edit: If I just pitch, and not yaw, my 3-fin rocket example, then there should be just one moment to worry about. I will start there.

 

[Rob Campbell]

How does choice of glue affect the CP of a rocket. This should really get things going.

 

[NTP2]

How does choice of glue affect the CP of a rocket. This should really get things going.

Well epoxy is smooth and slips easily through the air v super glue that pulls your down.

 

[bill_s]

Barrowman and the basic NASA equation I used in my CFD model will never predict CP behind the body. I think. So, for saucers and spools, Hack away.

Yes, because it's based on normal forces. But looks to me that simple symmetrical drag can't result in an effect like the CP behind the rocket either, merely right at the back surface, even if it does something. To get that effect requires asymmetrical drag. The problem is, merging that stabilizing effect into a normal-force based model doesn't actually place its force differential at the location of the component causing the effect. It has to be placed so that the stabilizing moment moves back the CP an equivalent amount, which means the effective CP of the effective area is always behind the total normal-force based CP, but only behind the rocket if it is extremely short, basically saucers and spools. Therefore the base drag hack could be overoptimistic for stubby rockets, the added massless object needs to be placed relative to the CP (and CG? it's complicated), not the end of the rocket.

As a basic rule of thumb, it would appear that diameter increases stability, not the demand for more of it.

 

[rharshberger]

Well epoxy is smooth and slips easily through the air v super glue that pulls your down.

You must not have RSO'd some college teams L1 cert rockets....the fillets are so rough they act as air brakes and probably contribute to highly turbulent air at the base of the rocket....

 

[boatgeek]

You must not have RSO'd some college teams L1 cert rockets....the fillets are so rough they act as air brakes and probably contribute to highly turbulent air at the base of the rocket....

I resemble that remark!

 

[NTP2]

You must not have RSO'd some college teams L1 cert rockets....the fillets are so rough they act as air brakes and probably contribute to highly turbulent air at the base of the rocket....

Well I haven’t RSOd anything but I’ve seen enough college kids work that I’m not surprised…

 

[RobertH3]

Attempting to re-usefulize this thread : )

Mine is a question, not an argument. I'm looking for knowing and don't have a stick in the game.
I agree there will be no perfect answer given that a calculated stability is dependent on a lot of things.
Mainly whether you use the cardboard cutout (Weathervane), Barrowman, or Rocksim / RASAero / OpenRocket calculations.
If you use software and adjust everything to a real world built and painted model, it works. No brainer. Every model I have designed this way has been utterly stable. The only fudge is drag coefficient (Cd) and you have to measure altitude for that to get a correct Cd for a correct calculated altitude. Although base drag is drag.

Here's a rephrase: "Rocket design software does not model Fat Boys/Warlocks/Mini-Maggs/etc. satisfactorily.
What needs to be modeled in order to take these instances into account?"
The length-width ratio sounds good but this is definitely a fluid dynamics-type question.

Seems like a really good place for the next iterations of software to go. Or maybe an old-school Estes Wind Tunnel

Cheers / Robert

 

[lakeroadster]

Seems like a really good place for the next iterations of software to go. Or maybe an old-school Estes Wind Tunnel

Well Robert, except:​

The nature of the base drag will be dependent on the Reynolds’s number and will not scale from low-speed wind tunnel tests.

So for folks hobby building on a budget, OpenRocket or Rocksim utilizing the Base Drag Hack, yields data they can use to build and fly short fat rockets, odd rocs, spools and pyramids.​

 

[Tractionengines]

The nature of the base drag will be dependent on the Reynolds’s number and will not scale from low-speed wind tunnel tests.

How many "big box store" 20inch box fans does it take to make a .8mach wind tunnel test area 1ft sq x 2ft long?

All my college books and notes are packed away...

 

[NTP2]

How many "big box store" 20inch box fans does it take to make a .8mach wind tunnel test area 1ft sq x 2ft long?

All my college books and notes are packed away...

I’m going to go out on a limb and say, more than your willing to pay for.

 

[kjhambrick]

I just want to watch this thread.
This weekend I am launching a fury of fire 4" for the first time. I added nose weight to make it 1 caliber with a I180. It is a short squat rocket, it would of barely qualified for the base drag hack. I added bb's and epoxy then foam filled the nose cone. It weighs almost 5 pounds, and sims out great in open rocket. I will let you know.

Did you get to fly your "Fury of Fire", @Wrightme43 ?

-- kjh

 

[Jeff Lassahn]

I highly recommend playing around with a low speed wind tunnel. I built one out of some cardboard and a fan that was way crappier than the design from Apogee (https://www.apogeerockets.com/education/downloads/Newsletter252.pdf) and it was fun and interesting.

The concerns above about operating at low Reynolds numbers are real, but not necessarily a deal killer. For stability analysis, mostly what matters is where flow separation occurs, and for a lot of rocket designs that's pretty consistent across speed ranges (the rocket's smooth until there's a sharp corner at the back, so flow separation happens at the back).

If you've got a rocket "flying" stable in wind tunnel airflow, you can gently tap it with your finger and feel it actively correct to straight flight, which is kinesthetically satisfying. It also gives you a good sense of when something is actually stable vs when it's just kind of pointing in some direction.

I played around with finless rockets in a wind tunnel a while ago (basically I stuck a nose cone on a body tube and added weight to various places to move the CG around). So here's some things I found out:

1. the CP of a finless rocket is about 1/4 of the length back from the nose at low angles of attack.
2. the CP moves backwards at higher angles of attack.
3. at 90 degrees (flying sideways) the CP is at about the halfway point.
4. It's not "unstable" at 90 degrees, if you balance it at the halfway point it actively seeks an orientation of 90 degrees and tries to stay there. (this is the Munk moment from above I believe)
5. If you put the balance point between the 1/4 point and the 1/2 point, it will attempt to stabilize itself at some angle to the airflow that gets larger as the balance point gets farther back. Again, it actively seeks a specific angle. If you tap it it will move in a cone shape holding a constant angle of attack.

I rechecked the rule that the low angle CP of the body is at the 1/4 length point by building and flying a couple of little finless rockets on 1/2A motors. They need A LOT of nose weight to get the CG up far enough, so they're pretty under-powered, but it can sort of happen...

This one is stable enough to fly straight:


This one is slightly unstable, bad enough that it shimmies with increasing amplitude but doesn't go into full-on tumbling mode:


So that suggests the CP really is just a little forward of the 1/4 length mark.


So, what does this mean about the base drag hack?

It means that for normal shaped rocket bodies it's too strong. It puts the CP of the body a bit more than 1/2 of the length back, like this (CG on these models doesn't mean anything, they're just thrown together to show the CP change):


If you use a smaller base drag cone with a very short length (I'm using 0.1cm here) and a diameter half of the body diameter, you get a much better result:

 

[Buckeye]

I highly recommend playing around with a low speed wind tunnel. I built one out of some cardboard and a fan that was way crappier than the design from Apogee (https://www.apogeerockets.com/education/downloads/Newsletter252.pdf) and it was fun and interesting.

Nice! Finally, some observed evidence that is on point.

So, what does this mean about the base drag hack?

It means that for normal shaped rocket bodies it's too strong. It puts the CP of the body a bit more than 1/2 of the length back, like this (CG on these models doesn't mean anything, they're just thrown together to show the CP change):

If you use a smaller base drag cone with a very short length (I'm using 0.1cm here) and a diameter half of the body diameter, you get a much better result:

Yes, another good piece of evidence that the Levison cone is too powerful. That's the risk of the method.

It is crazy that a tiny, 1 mm long cone on a 300mm rocket will move the CP that much. Barrowman to the extreme.

Though, it does add to the argument that the Hack can be useful to objects with no fins. Once fins are added, not so much.

 

[kjhambrick]

Glad to hear you're taking a stab at the papers.

But they are not off-topic. Once you start understanding what is happening at the
bottom end of the rocket, you'll need to consider if the Hack's supposition squares
with known base drag conditions from the second the motor lights, to burn out,
and the coast.

You'll also need to consider how significant is the the change in CG as the motor
burns down. Remember, the stability margin is based on the distance between
the CP and CG - and that initial margin is "static". The CG quickly moves forward
on most motors and so your margin improves quickly.

Think about those rockets that for just a split second act unstable when coming
off the rail, and then a split second later are suddenly flying straight. Is that a
rocket that had an iffy margin until the CG shifted forward from loss of motor
weight at the bottom? I'm not saying that's the case for all those unstable
launches, but it is for some.

Sorry ...

It's taken a while to get back to your post.

Blindly following Rules of Thumb can be dangerous.

Here is an example.

I plan to build a 'reusable' 4-inch fin can and a nose cone with an HED Av-Bay.

They will be resusable components in a modular RMS 38/x rocket I've been working on in OR.

In addition to the fin can and nose cone, I plan to make a set of Modular Recovery Bays in variable lengths for RMS 38/240 ( 9-inch ), 38/360 ( 11-inch ), 38/480 ( 12-inch ), 38/600 ( 14-inch ), 38/720 ( 16-inch ) and 38/1080 ( 21 inch ).

The motor serves as a coupler between the fin can and the recovery bay and there will be an internal min-diameter motor retainer located appropriately for the length of the particular motor casing and a Threaded, Plugged Forward Closure 'somewhere' inside each recovery bay.

This is my RMS 38/1080 Rocket along with sims for RMS 38/1080 J825R, J520W, J575FJ and J570W in the 21-inch Modular Recovery Bay:


Looks like some fun flights -- no records here -- but what is going on with the J825R ???

It has about the same total impulse as the J570 ( 975 Ns -vs- 973 Ns ) but Apogee is less than 2400 ft ???

The on-screen CP -to- CG relationship looks good according to the 1 Cal / 10 % RoT at 2.32 calibers / 11.3 % stability margin.

But what is up with the Altitude and the Max Acceleration ?

OR gave me a hint in the red exclamation point on the sim-line: "Large Angle of Attack Encountered (179.7 degrees)

Oops !

The rocket is flying so fast on the J825R that the CP moves ahead of the CG before burnout and then the rocket starts sky-writing just before it reaches Mach 2.5:


I 'fixed' it by adding 33 grams of nose weight:


It looks better but that low Stability Margin at burnout ( about 0.35 cal ) sure makes me nervous !

So what is the rule of thumb when you're dealing with a CP that has moved due to Supersonic effects ???

I dunno, I imagine I could calculate moments ...

But I think in this case I really do need to add a little more nose weight

-- kjh

 

[Wrightme43]

Did you get to fly your "Fury of Fire", @Wrightme43 ?

-- kjh

Yes sir, flew perfect. Recovery was perfect. It was just beautiful.

My Journey 75 hung up about 55 foot in a tree. Luckily I work for a equipment rental company, so I ran back to work and towed a boom lift back to get it down.

Let that rocket eating tree starve!