I will say this: the math (not just in rocksim) certainly supports the notion that the F10 is about the perfect F altitude motor. The difficulty of course lies in getting a good straight boost through all 8 seconds of burn - any arcing will matter a lot more with an F10 than with an F37. I have no doubt though that both the F and G records could be taken with the F10 and G25, though, assuming a near perfect vertical boost. I'd love to see some actual attempts with the F10 though - I've never seen any real altitude attempts, though the Aspire claims a mile on that motor, and it is about as far from optimum as you can get. With a truly optimized rocket, I don't see any reason why at least 7500 feet couldn't be possible on an F10.
One way to think about it is terms of a work function. Work against gravity is constant depending only on the endpoints, while work on drag is not a state function, (in math speak, instead its a line integra) that very much depends on path, or in this case velocity. So if 4 seconds of a flight where say 5000 ft of altitude are gained is done at a Cd of .7 or so, the energy lost to drag will be almost twice that of the same rocket which stays below say 800fps ascending the same 5000 (and taking 7 seconds to get there). So the 45 percent increase in drag due to greater surface area is more than repaid. This is straight forward physics.
Actually, it's even a bigger factor than that. The Cd is just the coefficient in the drag equation F = 1/2 pV^2*Cd*A. So if the velocity doubles to put the rocket into the transonic region, you'll not only get almost a doubling of the Cd, you'll also get a factor of 4 in the velocity term, for a total of almost 8 times drag force for the same area. That's a lot of leverage to overcome the area increase. On the other hand, gravity losses are worse the longer it takes to get to apogee, but all this is accounted for in Rocksim, which shows a subsonic F rocket just crushing the TRA record.
Adrian,
You are right but we are talking about two different quantities, force and its time integral, impulse which removes momentum and work which is force integrated over distance and removes kinetic energy. The v squared term does cause a quadrupling of drag, but this is mitigated by the fact that the rocket is moving faster; if this analysis were complete the disparities in altitude would be huge. I do wish Rocksim would take these integtals as the data is all there....and it would be a very helpful data set.
I don't want this to come out wrong and I don't want this to come out smarmy (that was a disclaimer) but simulating something and flying it are two totally different animals. The TRA "F" record hasn't stood for as long as it has because it is a chump record. I have heard many folks say online and in person that they have an F rocket simmed to 8K or 9K. I also have stuff simmed that high. On game day, they just don't fly as high as they are simmed. Some sim files are just inherently bad.
The Ellis L330 is another weird one. My rocket "beLLwether" (built especially for that motor) sims to 28-29K. I've flown it three times on L330s, and each time I got about 23K. Ditto for the K250s. The single-use 29mm Aerotech G motor files seem to be notorious for bad simulations.
Bottom line is that every one of us (me included) has at one time or another gotten AGU (all geeked up) about some stupendous motor/rocket combo that goes a zillion feet high in Rocksim. I've found that my actual mileage DOES indeed vary -- and almost always to the downside.
BTW, I have found many of the AeroTech and CTI reloadable sims to be uncannily accurate. For some reason, the bugs seem to be in the aforementioned loads, with only a few exceptions. I'm not sure how Apogee loads perform? I've never flown one...
john,
Curious notion that some motors are inherently less simworthy. Forgetting that some of the RS engine files are in just plain in error, what do you see as the cause? I can see a few, but none easily verified. First the motors could depart from spec or have other QC issues--many of these were certified a long time ago. Some motors via asymmetries may be inclined to roll the rocket more (which RS can't account for),or eject more material sideways tho both seems unlikely, while others might depend to a greater extent on ignition and the initial propogation-the Ellis J228 comes to mind which blows up when lit from the top.
Personally dont trust RS above mach 1.6, and heaven knows some of the assumptions governing the computation of Cd are wrong/overly simplistic. This is my guess, some rocket/motor combos get into regimes where these assumptions are most prone to occur. Just a guess, and when all is said and done, simming is one thing, building/flying quite another.
I don't want this to come out wrong and I don't want this to come out smarmy (that was a disclaimer) but simulating something and flying it are two totally different animals. The TRA "F" record hasn't stood for as long as it has because it is a chump record. I have heard many folks say online and in person that they have an F rocket simmed to 8K or 9K. I also have stuff simmed that high. On game day, they just don't fly as high as they are simmed. Some sim files are just inherently bad.
I can't argue with your experience. And it seems like a rocket I can build in a day, on my first F flight, shouldn't be able to crush a world record. But the F10 design did exceed rocksim's predicts when flown on an E15, so I'm looking forward to seeing how it does.
Chris is right, though; if it's windy, the rocket won't be going anywhere near straight up for the whole 8 second burn, so that's a big factor.
Chris, thanks for the offer of the G25. I may take you up on that, if you're going to be too busy with school to build for it. Of course, I may be too busy with the Parrots and the A-F records, too. The motor is longer than the F by enough that I would build a separate rocket, and I think it would be a good idea to see if the F survives first. So maybe for Oktoberfest. Interestingly, when I just blindly put it in the same rocket as my F10 design, it barely goes any higher. But with a bunch more nose weight, I get it up around 9500 feet.
I will say this: the math (not just in rocksim) I have no doubt though that both the F and G records could be taken with the F10 and G25, though, assuming a near perfect vertical boost..
I don't have my logs with me (I'm traveling) but I tried several G25s in "Gambit", which was the first rocket I built to go specifically after the G record. I got about 5K on a laser-straight boost. It was minimum diameter, close to optimal mass, etc. I flew several G25s in that rocket and only one lit perfectly and boosted perfectly. It was *very* straight. I had apogee deploy, and watched it from the bluffs above the north site. I was sure I had nailed it. When I recovered, I found it was not even on the same page as the sim...
When I get back next week, I can look up the details including rocket length and weight, date flown, conditons, etc. I know that it didn't come close to what was then the G record. More to the point, it bore no resemblance to the simulation.
J
When I recovered, I found it was not even on the same page as the sim...
Yes, the date would be particularly interesting. Maybe you were using an older version of Rocksim at the time that didn't model transonic drag as well as it does now?
I will say this: the math (not just in rocksim) I have no doubt though that both the F and G records could be taken with the F10 and G25, though, assuming a near perfect vertical boost..
I don't have my logs with me (I'm traveling) but I tried several G25s in "Gambit", which was the first rocket I built to go specifically after the G record. I got about 5K on a laser-straight boost. It was minimum diameter, close to optimal mass, etc. I flew several G25s in that rocket and only one lit perfectly and boosted perfectly. It was *very* straight. I had apogee deploy, and watched it from the bluffs above the north site. I was sure I had nailed it. When I recovered, I found it was not even on the same page as the sim...
When I get back next week, I can look up the details including rocket length and weight, date flown, conditons, etc. I know that it didn't come close to what was then the G record. More to the point, it bore no resemblance to the simulation.
J
While I don't doubt your data, I do doubt that 5000 feet is anywhere remotely near the potential optimum for a G25, as with G80's last year (with less impulse and a far shorter burn), people were getting in excess of 6000 feet pretty easily. Seeing as the G25 has an extra 20Ns, as well as a 5.3 second burn, it should be capable of far more than 6000 feet.
Just wish I had flown my f10 last w/e instead of fussing with the 2 stage flight--weather was great/ One question I have for John W is whether Gambit sported a conical nose? I really don't want to get into a NC debate, but there have been any number of science fair, NAR R&D projects. etc that make one point abundantly slear--elliptical cones for subsonic flights have signif less drag. This is one of the areas where RS unequivocally mucks it up--using Datcom methods, and the assumption that foredrag is proportional to wetted area, pointy headed rockets still sim higher, and one can go so far as to make the cone as blunt as one cares and still het better altitude predictions. The latter flies smack in the face of our collective eperience. The former has been verified ad nauseum.
This may be one of those "regimes" that when using a particular motor, usual design practices may lead to a situation where the RS results diverge from observed, owing to its inherent misbehavior, and less about the motors simworthiness. I hope someone understands what I'm trying to say 😳
One question I have for John W is whether Gambit sported a conical nose? :
John, Gambit had an Estes body tube. I can't recall the size, it is whatever is just a sniff bigger than PML 29mm. I remember because I used PML 29mm aiframe for the coupler. Anyway, IIRC the cone was indeed conical that day, because it was all I could find.
FWIW, I used a Sears-Haack (spelling?) on my F and G, both subsonic. I was way (way, WAY) supersonic on H, J, K, and L flights, and all used conicals. My I flight was subsonic, and had an ogive. I guess I'm saying I buy into the supersonic-for-a-significant-part-of-the-flight means conical, and subsonic means something else.
JW
JW,
Not sure, and not really wanting to rehash a debate here. Warren and I have been there, done that, and like most debates on pointy headed subjects, no real conclusaions reached IMO.
I bought a lathe in part to do custom power series cones. It would be interesting to swap back and forth on a High perf supersonic bird to see what gives. I know in the high Mach numbers, say above three the cones pull away unequivocally. Theres some interesting posts on Nakka's sugar shot to space project that look at NC's in some detail, using what I think is still a Newtonian plate based sim engine. Really for that kind of work iirc, one need's a full blown Euler or equiv. based FEA model. Right, that will be my next purchase--small supercomputer and some good software.... 🙄
Back to the real world....
It would be interesting to swap back and forth on a High perf supersonic bird to see what gives.
Yeah, Chris got some good data on his Cd as a function of velocity. It would be really interesting to see the same plot with two different nosecones. Along those lines, Estes makes 24mm ogive nosecones in a short version and a long version. I should fly both of those with a Parrot to see how the drag compares for an E record bird. The nice thing about the accel data is that it takes any variation in motor impulse, launch rod stickiness, etc. out of the equation.
Adrian,
I've got ellipticals in 25 and 30 mm as well you are welcome to try. Ogives are a fairly sucky Cd. Excellent and simple R/D here:
Thanks for the link. The wind velocity was so low, though, that I'm not sure how relevant it is for F or G records. Maybe if they had used much larger models to get a better match with the Reynold's numbers. The most surprising thing to me was the inability to measure any difference in fin shape (though the actual numbers seemed to be missing)
Well this was hardly groundbreaking study, and as you note, done at low velocities.
But you'll find I think if you do a thorough search, that the data and general conclusions hold all the way up to transsonic velocities. The absolute Cd's will drop with velocity just as RS and others simware predict. A lot of this was known pre WWII from research done at Langley and the like.
JS
BTW, the egglofter design was pioneered by yours truly; to my knowledge,noone had used a transition section the entire length of the rocket and the prevailing practice was to stick an egg capsule on yop of a BT20. At the time to think they would someday be using ut for Bowling Balls was well unthinkable... 😆
Latest Post: Hello World Our newest member: amiecannan4271 Recent Posts Unread Posts
