Rocket Theory, Disappointment with BackSlide experience, seeking input.

[BABAR]

@Rktman and @georgegassaway and @burkefj and @neil_w , are my go to guys for stability issues and glider stuf (Neil, you more for stability!)

Theory: Pure BackSlide recovery is easily achievable but cannot be made completely reliable (maybe 80-90%, but never 99+%)

note: credit to Alway brothers for development of concept,

I have done 4 “pure” BackSliders, defined as rockets which at apogee/ejection have NO change in physical configuration of the rocket (no chute, no streamer, no blades, no brakes, no elevons, no motor shift or motor ejection), but use a forward asymmetric side port to vent the ejection charge as a forceful “puff” to throw the rocket axis sideways. I use the term “stalling”, not in glider terms but in terms of throwing the rocket out of a near zero angle of attack trajectory where Barrrowman equations rule. Although most likely the rocket doesn’t truly come to a complete motionless stop, I think for practical purposes it is like hitting “reset” and effectively what happens next is similar to what would happen if you just dropped the rocket with zero velocity AND AT COMPLETELY RANDOM ORIENTATION from a hot air balloon.

The lateral “puff” definitely works to stop the forward trajectory, but the final rocket orientation AFTER the puff is completely random. It may be North, South, East, West, but more importantly may be straight up or straight down (very unlikely) or something in between straight up or straight down (highly likely.)

whatever orientation the rocket starts with, it begins to fall. Initially, because the velocity starts at zero, the fall is slow and the fins, body, and nose cone have little effect as initially the airflow is negligible. As it picks up speed, the airflow becomes sufficient to create a measurable and effective ANGLE OF ATTACK. MOST OF THE TIME (say 80%?) that angle of attack is greater than say 15 degrees (arbitrary numbers, but let’s just say Barrowman Equations don’t apply for high angles of attack and use 15 degrees as that threshold.). In these situations, Cardboard CutOut stability rules apply.

for MOST rockets, (medium length, decent size fins) the rocket is STABLE under both Barrowman AND Cardboard CutOut stability measurements, the rocket noses down and goes ballistic and lawn darts. For a dedicated BackSlider (CG forward of CP under BARROWMAN, but CG BEHIND CP under Cardboard CutOut), and the angle of attack is GREATER than 15 degrees, the rocket falls BACKWARDS, and the fins (now canards) kick in and the rocket goes into a horizontal BackSlide. Emphasize that BackSlider Rockets have long body tubes, those tubes having little impact on Barrowman Equations but much more effect on CardBoard Cutout calcs.

all well and good….but….

as I said, rocket orientation after “puff” is completely random (maybe not completely, but I have no reason to think otherwise.). So say 20% of the time the orientation is roughly nose DOWN (say 15 degrees or less off vertical nose down.). The rocket starts to fall, it still starts with very low velocity so NEITHER Barrowman NOR Cardboard Cutout calculations apply. As it falls, it soon develops airflow and an angle of attack. In THIS case however the developing angle of attack IS in the Barrowman range, and the rocket BECOMES stable, and unfortunately stable under Barrowman rules BECAUSE the rocket “just happened” to randomly begin pointed 15 degrees or less off vertical down angle when the “puff” rotation ceased.

my theory is that, while it may be possible to get better than 80% (maybe going with a more SuperRoc length to diameter of 60 or 70, but this is not EASILY done and brings up some challenging structural issues which may make it paradoxically HARDER to keep CG BEHIND CP under CardBoard Cutout rules, as you need to reinforce the forward tube), I am not sure it is POSSIBLE to get to the reliability of parachute and streamer models.

Of course, it could be that after a few flights the vent hole has gotten plugged up. We will see.

in any case, my recommendations for those pursuing this (if anyone other than @Dotini is interested) are as follows.

perform your testing in an isolated field, preferable with no spectators within rocket range. If spectators ARE present, have a strict RSO that calls a heads up flight and makes sure everyone has eyes on the rocket.

start with the smallest motors practical to achieve sufficient altitude for transition. This lesses the range of the rocket if it goes unstable, reduces the area of impact around the launch site for possible ballistic impact, and since peak altitude is lower, MAY reduce the impact velocity if the rocket goes ballistic (kind of depends on the time and distance it take for rocket to reach terminal velocity.)

use rounded or soft nose cones on these rockets (I find Nerf Darts work well for BT-5 rockets, they aren’t very aerodynamic, but some of my birds have gone nearly out of site on A3-4T motors, so aerodynamics not too much of an issue. Plus they are cheap, you or your neighbors likely have extra to give away. For the record @lakeroadster, I have tried the Whistling Nerf darts, unfortunately no audible whistle.) not that I have any intention of my rockets hitting any person or property, but if fecal turbine action occurs, I’d rather it be with a Nerf Dart nose than a hard pointy plastic or balsa nose.

I would like nothing better than for my theory to be proved wrong.

BTW, If you haven’t seen BackSlide recovery in action, it is REALLY cool.

I have solved one problem with BackSlide recovery, the problem with fins breaking off on landing due to the lateral velocity. Asymmetric fin placement very reliably causes the rocket in a successful BackSlide to come in with fins up. Since I have gone to this technique, I have had zero fin breaks.

@Rktman, Eric, have you had any failures with your BackSlider? If not, how many flights do you have? Did your fins break?

 

[BABAR]

Peter Alway's report said it always (pun unintended) worked except when motor ejected.

Report courtesy of @Mike Haberer

 

[Jeff Lassahn]

One thing that might help is to design your fins to make the stall angle as low as possible, so that chance that the rocket isn't in a stall condition after ejection is lower. The design airplane wings to be thick with a rounded leading edge to make them stall at a higher angle of attack, so maybe the ideal here is very thin fins with a sharp leading edge?

Also, anything you can do to make the orientation after ejection less random seems good. I would think that means using a delay that's either too short or too long so the rocket is definitely moving in a stable orientation either up or down when it goes off. Maybe also play with the location of the vent hole relative to the CG -- moving it closer to the CG means it causes less rotation and more sliding sideways motion, which might make the behavior more consistent.

 

[BABAR]

One thing that might help is to design your fins to make the stall angle as low as possible, so that chance that the rocket isn't in a stall condition after ejection is lower. The design airplane wings to be thick with a rounded leading edge to make them stall at a higher angle of attack, so maybe the ideal here is very thin fins with a sharp leading edge?

Also, anything you can do to make the orientation after ejection less random seems good. I would think that means using a delay that's either too short or too long so the rocket is definitely moving in a stable orientation either up or down when it goes off. Maybe also play with the location of the vent hole relative to the CG -- moving it closer to the CG means it causes less rotation and more sliding sideways motion, which might make the behavior more consistent.

Good thoughts! Thanks!
For durability purposes I think 1/16” is as thin as practical (goal is something just about anyone can build and fly, maybe a bit tougher than a Viking to build but certainly less than a typical heli or glider)

also the term “stall” may have been a poor choice on my part, I am open for a term that describes “abrupt departure from a straight trajectory parallel to the long axis of the rocket, I.e, the trajectory of a normal rocket on boost and an ill fated rocket coming in ballistic.” For now, maybe I will try “skew”.

I like the short delay idea. Initiating early “Skew” on ascent has a positive feedback effect, as it creates drag which in combination with gravity slows the rocket and makes it more likely I think to go horizontal. Late ejection charge with skew on descent may not be as effective.

for symmetrical fin rockets, there is no designated laterality, so hole position doesn’t matter as long as it is forward as possible. For asymmetric fin rockets, you can go side, top, or bottom. A side port is likely to give more “skew” as less resistance from the larger horizontal surfaces, but I am not sure if “more is better” and am not convinced the final orientation immediately after “skew” force stops (presumably due to rotational friction) is predictable.

certainly you are correct, distance from hole to CG affects magnitude of “skew”, again however optimal skew from vertical is 90 degrees. I CAN predict rocket orientation at “skew” initiation by using a short delay, the puff will occur with the rocket vertical. But I am uncertain whether the DEGREE of skew (ideally 90 degrees starting from vertical) is consistent or predictable.

I think I am going to build one according to the original published (and PATENTED!) plan, except I will place one fin perfectly dorsal, the others at the same ATTACHMENT points but with slight dihedral. I also may put a Nerf Dart in the nose cone, and I will balance the CG with weights as recommended, leaning a bit toward tail heavy and hoping that my dihedral ventral fin placement doesn’t degrade boost Barrowman Stability (corkscrew okay, skywriting not so much.)

the attached article mentions the Alway brothers played with asymmetric fin placement (I am curious, did they find or anticipate the same issue I have: you are highly likely to bust a fin on landing.). Anybody know what came of that

I am guessing Peter is @PeterAlway on this forum. So maybe this will reach him!

 

[Daddyisabar]

Blinding me with Science!

 

[Ronz Rocketz]

I suspect your biggest variable is wind and other atmospheric conditions which is why it works 80% to 90%. I'm not sure how you would overcome the wind variability without resorting to a streamer.

Some of my best streamer recoveries occurred in windy conditions.

 

[BABAR]

I suspect your biggest variable is wind and other atmospheric conditions which is why it works 80% to 90%. I'm not sure how you would overcome the wind variability without resorting to a streamer.

Some of my best streamer recoveries occurred in windy conditions.

Respectfully disagree, or at least we are not on the same page.

I think the "skew" is the equivalent of a 3D roulette wheel. If IF we reliably KNEW the orientation at initiation of skew (e.g., @Jeff Lassahn 's idea of either short delayd early rejection [up vertical] or long delay late ejection [down vertical]) I STILL am not certain that the magnitude of the skew is consistently calculable . 10 degrees? 45? 90? 180? 360? I think the mantra, "Round and round and round she goes, and where she stops, nobody knows," apparently from this
https://en.m.wikipedia.org/wiki/The_Original_Amateur_Hour
Applies. Final position/orientation is random

Winds would I believe just make it random squared, which I suspect although perhaps mathematically couterintiutive is that random multiplied by an other number except possible 0 is still equally random.

Agreed a small streamer WILL alter the equation, and if it balances the nose will result in a horizontal recovery. Not sure if rocket acquires much LATERAL velocity, therefore uncertain if the qualifies as "augmented or assisted BackSlide" or simply "streamer with a relatively large tail surface" recovery.

I usually am NOT a purist, but in this case like Horizontal Spin part of the attraction is a safe recovery with no moving parts or motor eject.

 

[Dotini]

I think I am going to build one according to the original published (and PATENTED!) plan, except I will place one fin perfectly dorsal, the others at the same ATTACHMENT points but with slight dihedral.

I think this is a really good idea, that is, building exactly according to their plan and establishing a baseline for further experimentation. So I have a question or two. Their plan calls for a 3" coupler between the fin unit and their one piece 34" tube. Will you immediately compromise on this? Is this coupler one of those very heavy wall Totally Tubular couplers and essential for CG location? Will you order a 34" tube or use segments with a coupler moving the CG forward?

 

[BABAR]

I think this is a really good idea, that is, building exactly according to their plan and establishing a baseline for further experimentation. So I have a question or two. Their plan calls for a 3" coupler between the fin unit and their one piece 34" tube. Will you immediately compromise on this? Is this coupler one of those very heavy wall Totally Tubular couplers and essential for CG location? Will you order a 34" tube or use segments with a coupler moving the CG forward?

I haven't ever used a store bought coupler on a scratch build, I make my own out of same size BT, cut out a long segment to fit, go with two layers. In this case may double as engine block. I have the 34" tubing (I do a lot of scratch work so got a big box as most economical. ) need to find a rounded BT-20 in my spare box, or figure a way to put a Nerf Dart in one (not sure how that will effect CG, but when these fail it is sadly epic and aside from rear eject designs i don't like things coming in pointy end first!)

After three or four great Turbinator flights, the last one didn't "skew", and came in ballistic. Trashed one BT-80 Tube. If i rebuild in this size, i think i will use heavy wall tubing for the BT-20 curved fins and make them longer.

I can also paint them. The thin walled tubes were too flexible, I think the paint would just have wrinkled off.

 

[John Kemker]

I had a LOC Bruiser (not EXP) that tended to fall sideways and "paddlewheel" until ejection. Very consistent behavior.

 

[Ronz Rocketz]

I had a LOC Bruiser (not EXP) that tended to fall sideways and "paddlewheel" until ejection. Very consistent behavior.

I’m sure if you actually tried to show someone this behavior, it wouldn’t do it.

 

[John Kemker]

I’m sure if you actually tried to show someone this behavior, it wouldn’t do it.

Of course not! It'd go nose-down at a high rate of acceleration!

 

[PeterAlway]

Tom / Babar linked me to this and a related thread. I'm delighted to see people are playing with the backslider. First, I kind of have a pet theory that might be wishful thinking--I feel like we had a higher success rate than we deserved from random orientations. Since the pitch over at ejection has to go through a 90 degree angle of attack at least twice before it reaches the 360 degree point where the rocket lawn darts, I suspect that the rocket probably settles into its backwards glide before it can go 360 degrees. Again, this is a pet theory, but if someone could take decent video of the ejection repeatedly, you could confirm or reject it.

I don't recall what was up with Bob's asymmetrical fin experiments, though I think they might have just been replacing one fin with a different shape but the same total area as the other two, in hopes of stopping any spinning in roll. I don't think that idea had any noticeable effect, so the work Tom/Babar has done is original as far as I know. It's been 20 years, though, and we sort of called it quits on the project after our R&D report. The patent application was 100% derived from the R&D report. We pretty much declared victory with the backslider and moved along

We were both well aware that backsliding can work if you eject the nose. Way back around 1970 or so, there was a report in Model Rocketry magazine that if the Infinite Loop (the original tube fin design) had a separation, the bottom section would glide back. AVI, the 70's company that took over the MPC rocketry inventory, sold a kit made from MPC parts called the Lineaus Gigantus that popped the nose, and with no other recovery system did a backwards gliding recovery. We called that style of recovery a nose-blower, and the no-moving-parts version a backslider. We decided to investigate the no-moving-parts backslider because it was simpler to understand (how would you even describe the center of pressure of a model with a nosecone attached by a shock cord) and because when I'd seen rockets recover by backsliding when there was no recovery, it seemed so miraculous. John Kemker comments above that he had a model that behaved like a backslider before ejection, and I had seen that before.

Someone asked if they could fly one with an ejecting nosecone. It seems highly likely that one could, since nose-blowers do work.

I like that Babar has concocted an anti-damage fin configuration. We went from square fins to tapered of elliptical fins in hopes that landings would be easier on those shapes.

We used those tube couplers in our design mostly so that we could swap out fins for the R&D project. There's no magic to building that way, and it makes sense to simplify the design as long as the CG/CP/CLA relationship is right. We chose the specific coupler type out of convenience. Essentiallly, Bob could order large quantities of 34" BT=20, large quantities of heavy 2.75" couplers, and large quantities of 2.75" BT-20 from Totally Tubular at the time. It made it easy to churn out a lot of models, and to test out different fin areas.

By the way, it seems that backsliders are a thing amongst people who make 2-liter bottle water rockets. Without delays or jejection charges, it seems that they can get backsliders to work (you just need a sufficiently vertical boost so that the rocket starts falling backwards instead of arcing over). It's always a delight to see people still referring to Bob's and my work. I didn't know this thread was going on, because I pretty much stick to the scale board, so thanks to Babar for letting me know this was here. I'll keep an eye out in case anyone wants to know more.

 

[boomtube-mk2]

Long ago I launched an Estes "Long Tom" where the ejection charge didn't do its job with the upper-stage remaining as launched.
It arched over until it was parallel to the ground wherein it began to spin end for end all the while staying flat to the ground.

Came down as gently as a feather and I always imagined; What if you could make that happen reliably every time?
A new and different recovery system could be developed for light-weight rockets above a certain length.
Trouble is, that was without a doubt a one in a million happenstance never to be repeated again.

 

[BABAR]

Eric

Good to know you are 6 for 6


Ya know how you always find what you're looking for in the last place you look? Because you stop looking for it after you find it!

One thing all my failures have in common, they were all on the LAST flight. Because wasn't flyable after that.
Made it to 5 on two, 3 on one. Very small sample size.

OTOH I didn't do any measurements or balancing, so some of that is on me.

I want something practical, easy to transport.

 

[Dotini]

I want something practical, easy to transport.

I've begun construction of a new backslider which follows the Alway design very, very closely. Would you like to see pix here, or in a new thread? I have gone to the extent of ordering special heavy wall coupler, so it'll be a few days before I can fly it.

Note:
I waterproofed and compartmented this 46" cardboard box and made a handle for it. It fits in the trunk of my subcompact together with the range box.

 

[PeterAlway]

That yellow one with the curly fins (I presume for spin during glide) looks to have a lot of fin area. Also I'm a little worried about what the paint is doing to the CG. I do like the idea of the curly fins for spinning glide, though. Appealing idea!

 

[BABAR]

I've begun construction of a new backslider which follows the Alway design very, very closely. Would you like to see pix here, or in a new thread? I have gone to the extent of ordering special heavy wall coupler, so it'll be a few days before I can fly it.

Note:
I waterproofed and compartmented this 46" cardboard box and made a handle for it. It fits in the trunk of my subcompact together with the range box.

View attachment 468983View attachment 468982

Nicely done! While most of the damage to my rockets has occurred from non-nominal flights, the amount encountered from just tossing them in the trunk after flights is not inconsequential.

these long rockets with long sculpted fins also likely need the extra TLC.

 

[BABAR]

That yellow one with the curly fins (I presume for spin during glide) looks to have a lot of fin area. Also I'm a little worried about what the paint is doing to the CG. I do like the idea of the curly fins for spinning glide, though. Appealing idea!

Peter, did the spinning rockets in your study REALLY backslide? My experience and I think @Dotini ’s reports seem to fit flights where the rocket is definitely horizontal, but is not developing much lateral velocity, as opposed to the “pure” non spinning backsliders, which are definitely flying in a tail-first fashion at moderate velocity, very similar to a true glider.

my spinners are all coming horizontal orientation but trajectory either straight down or a tight spiral. For both Dotini and me, the rockets spin ALL the way down, they are designed to do so. His definitely show Magnus force, which is LATERAL to BOTH the rocket and the downward trajectory, they literally go sideways.

so I am wondering if Horizontal Spin and BackSlide are significantly different. Both definitely take advantage of the fact that earthward drag is greatly increased when the rocket is horizontal to earthward vector.

 

[PeterAlway]

Peter, did the spinning rockets in your study REALLY backslide? My experience and I think @Dotini ’s reports seem to fit flights where the rocket is definitely horizontal, but is not developing much lateral velocity, as opposed to the “pure” non spinning backsliders, which are definitely flying in a tail-first fashion at moderate velocity, very similar to a true glider.

my spinners are all coming horizontal orientation but trajectory either straight down or a tight spiral. For both Dotini and me, the rockets spin ALL the way down, they are designed to do so. His definitely show Magnus force, which is LATERAL to BOTH the rocket and the downward trajectory, they literally go sideways.

so I am wondering if Horizontal Spin and BackSlide are significantly different. Both definitely take advantage of the fact that earthward drag is greatly increased when the rocket is horizontal to earthward vector.

It's been literally 21 years since we did those experiments, so my memory is rusty, but I recall that it was my sense that they did indeed backslide. I find your idea here interesting. I wonder if there is a spectrum of behavior from straight backsliding to whatever accidental spin our models had, to what you are getting with those large curly fins.

Peter Alway

 

[BABAR]

It's been literally 21 years since we did those experiments, so my memory is rusty, but I recall that it was my sense that they did indeed backslide. I find your idea here interesting. I wonder if there is a spectrum of behavior from straight backsliding to whatever accidental spin our models had, to what you are getting with those large curly fins.

Peter Alway

Horizontal spin horizontal recovery is much easier to achieve than pure backslide, in my experience. It takes surprisingly little fin adjustment to get the rocket to go horizontal. Getting effective RPM to generate demonstrable Magnus effect OTOH seems more difficult (translated: I ain’t done it yet, but @Dotini has a solid handle on it!)

specifically I think the length to diameter ratio required is much greater for Backslide recovery.

there is an unfortunate overlap in the use of the term “Backslide.” It seems to be most often referred to with rockets, usually but no always high power, that achieve near perfect vertical boosts, terminating in a nose up Apogee and a slightly delayed deployment. The rocket therefore literally stops, nose up, for a fraction of a second, then falls straight down (tail first) for a (hopefully) short distance before deploying the laundry. I think I heard @burkefj verbally describe this in his recent excellent 1/5 scale Pershing launch. To keep things clear, I guess I should use the full description “backslide RECOVERY” instead of just “backslide” to reduce confusion.

so in general rocketry, transient “backsliding” is not uncommon. True “backslide recovery” seems pretty rare (but maybe we can fix that!)