Quantcast

Level 1 help

The Rocketry Forum

Help Support The Rocketry Forum:

DANLK66

Well-Known Member
Joined
Jan 19, 2009
Messages
151
Reaction score
1
Hello,
I am working on finishing up my Mad Cow Sea Wolf for my level 1 launch. I am looking at motors and delays on roc sim. My question is what is the maximum velocity that is acceptable when the chute is deployed. Of course slower is better but is 20fps ok 30? 50? Were is the line. So far the H242 is the best deploying the chute at 17fps with a 6 sec delay. Can you adjust the delay on an Aerotech motor? Let me know what you think.



Thanks

Dan
 

hardinlw

Well-Known Member
Joined
Jan 19, 2009
Messages
372
Reaction score
1
15 to 20 fps is an OK landing speed. Deployment speed could be 50 fps or more without hurting anything. I usually figure that if I get a delay that is within a couple of seconds of the optimum one way or the other, I'm OK. Note that to get less than 30 fps you have to pop the chute within a second of apogee and the delays are just not that accurate. You cannot adjust an Aerotech delay.
 

bobkrech

Well-Known Member
Joined
Jan 20, 2009
Messages
8,354
Reaction score
31
Ideally you should deploy at apogee if you choose the proper delay or use an altimeter. At apogee, the deployment load is 1 G or less since the vertical velocity is 0 at apogee, and if the rocket is near vertical, the horizontal velocity is near 0 as well. However in the real world, the "nominal" motor ejection delay can be off by 1.5 to 3 seconds and still be within specification.

For each second you are off in the ejection delay, or altimeter error, your rocket is accelerating at 32 fps^2.

At +/-1 second from apogee your rocket is moving at 32 fps (22 mph) and is 16' below apogee..

At +/-2 seconds from apogee your rocket is moving at 64 fps (44 mph) and is 64' below apogee.

At +/-3 seconds from apogee, your rocket is traveling at 96 fps (66 mph) and is 144' below apogee.

At +/-4 seconds from apogee, your rocket is traveling at 128 fps (66 mph) and is 256' below apogee.

The shock load capability of your rocket and recovery system will determine just how far you can be off and safely recover your rocket. This is highly dependent on how fast your parachutes inflates and and slows down your rocket.

If your rocket has a descent rate of 16 fps under main parachute, and it takes 0.1 seconds to inflate your your main chute, the average shock load on your recovery system is (Vdeployment-Vchute)/(G x t deployment) where g is 32 fpsps (gravity).

For a 1 second error the shock load is ~(32-16)/(32x0.1) = 5 G

For a 2 second error the shock load is ~(64-16)/(32x0.1) = 15 G

For a 3 second error the shock load is ~(96-16)/(32x0.1) = 25 G

For a 4 second error the shock load is ~(128-16)/(32x0.1) = 35 G

If your chute opens twice as fast, in 1/20 of a second, the G loads are doubled.

Large HPRs typically employ a dual deployment system. At apogee the rocket either is broken into 2 aerodynamically similar pieces, or a drogue is deployed to stabalize the descent velocity to the 60-80 fps range. This significantly limits the main deployment shock to the 2 to 3 second shock levels.

The physical load the rocket and recovery system must withstand is simply the weight of the rocket x G. If you have a 5 pound rocket and you expect a 25 G deployment shock, your recovery system and all attachment points must be capable of withstanding a 5 x 25 = 125 pound load.

IMO the minimum design point for HPR recovery systems should be 50 G.

Bob
 

DAllen

Well-Known Member
Joined
Jan 18, 2009
Messages
4,820
Reaction score
756
Keep in mind one thing about RockSim is that it is rather optimistic with altitude predictions with improper variables entered into the system. Be sure to weigh your rocket fully loaded and determine the actual CG and use mass over-ride with those values entered once the rocket is complete. I will say that once that step is complete in RS, RockSim is quite good at determining optimal delays. I use RS all the time for that.

-Dave
 

DANLK66

Well-Known Member
Joined
Jan 19, 2009
Messages
151
Reaction score
1
Thanks for the help. I am finishing up the fin fillets and have to add the nose weight. Once this is done I can get an idea of the final weight so I can alter Roc Sim as you have suggested. I will keep you all posted once I finish.


Thanks

Dan
 

Pantherjon

Well-Known Member
Joined
Jan 19, 2009
Messages
7,434
Reaction score
1
I don't manually input the 'loaded'(with motor casing and reload) into RS..I worry more about the 'empty' weight and CG..RS has the case and propellant weight correct in 95% of the time..I would make sure the weight and CG of the rocket ready to fly MINUS the motor is input into RS with the 'manual override'..Even then, RS is pretty optimistic about the altitudes..Case in point was my L2 rocket..I input everything in after weighing and measuring for the CG like 6 times and RS's altitude 'prediction' was almost 1500' more then the rocket actually flew..Probably because it was a little more windy at the launch then the atmoshperic conditions I put into the simulation..YMMV
 

MarkM

Well-Known Member
Joined
Jan 18, 2009
Messages
2,215
Reaction score
1
RS's altitude 'prediction' was almost 1500' more then the rocket actually flew..Probably because it was a little more windy at the launch then the atmoshperic conditions I put into the simulation..YMMV
You are correct in RS optimistic predictions on altitude. The main reason, especially when flying at higher altitudes, is that it is almost impossible to accurately input EVERY atmospheric variable RS allows/uses in it's prediction. How many of us know exactly the speed of the winds aloft at the various altutudes are, where thermoclines are located, etc. All these factor into how RS predicts altitude. I use RS altitude prediction merely as a guideline to give an idea what motors I could fly a particular rocket on during launch given the weather conditions.

That said, when the proper weight of the rocket and CG are inputed, RS is reasonably accurate in determining proper delay timing. However, in actual use motor delays can vary by as much as +/- 20% of their indicated delay so one has to be careful in choosing the delay.
 

DANLK66

Well-Known Member
Joined
Jan 19, 2009
Messages
151
Reaction score
1
So the answer is.......Longer shock cord:D
 

bobkrech

Well-Known Member
Joined
Jan 20, 2009
Messages
8,354
Reaction score
31
I don't manually input the 'loaded'(with motor casing and reload) into RS..I worry more about the 'empty' weight and CG..RS has the case and propellant weight correct in 95% of the time..I would make sure the weight and CG of the rocket ready to fly MINUS the motor is input into RS with the 'manual override'..Even then, RS is pretty optimistic about the altitudes..Case in point was my L2 rocket..I input everything in after weighing and measuring for the CG like 6 times and RS's altitude 'prediction' was almost 1500' more then the rocket actually flew..Probably because it was a little more windy at the launch then the atmoshperic conditions I put into the simulation..YMMV
It's simple to actually measure the weight and CG of the rocket so IMO why worry about the database. I'd manually override the calculated weight and CG values and used the measured values. The CP at zero angle of attack is easy for TS to calculate since it only depends on the shape of the rocket so leave that in the automatic mode.

The only CGs that counts for stability is the rocket with the empty motor casing (all parts less propellant) and the fully loaded motor casing because this is what happens in flight.

In most rocket using short APCP motors, the CG moves forward as the propellant burn and the tail gets lighter, however in shorter rockets with long motors, the CG moves backward as the propellant burns. To be on the safe side you would like to have a minimum 1 caliber stability for the rocket under both conditions. When you override RS I'd be conservative and use the least favorable CG value.

Bob
 

hardinlw

Well-Known Member
Joined
Jan 19, 2009
Messages
372
Reaction score
1
I agree with Bob about the weight. Usually, I add a mass object that gets the weight right and then move it around until the predicted CG matches the real CG. My impression is that the weight error is not so much in the accuracy of the database as in the things we leave out of the simulation, like the glue and paint. I've been helping a TARC team and their rocket came out several ounces heavy. I then considered the amount of 5-minute epoxy that had been consumed and there was a definite corellation with the under estimate of weight.

Based on that experience, I'll throw out two other factors that will impact the RS accuracy, temperature and CD. After having accounted for the actual weight, RS said they needed an E28 to reach 750'. Not even close. The biggest engine I had in my box was an F39 reloadable and that only got to 634'. The default temperature for RS is 59 deg F. We were flying in temperatures in the upper 30's. They put in actual temperature and barometric pressure and still predicted a higher altitude that was being obtained, but not by much. Adjusting CD upwards a bit the actual altitudes from several flights were matched reasonably well. The rocket was not painted but had several coats of sanding sealer and had been sanded smooth, so we felt that calling the finish "smooth" in RS was reasonable.

Back to the drawing board. They cut out 4" of overall length from the first design and built a second prototype, being careful of how much glue was being used. The resulting rocket was about 3 oz lighter. This one also got painted before flight. In similar temperatures, it is going much higher. In fact, with enough ballast to fly at what was the empty weight of the original rocket, it flew to 796 feet.

What was learned from this? With lower air temperatures, the air is denser. RS accounts for this and computes higher drag, hence lower altitude, but you have to put in the right temperature. This suggests that the delay that works in warm conditions may be long for colder conditions. Also, in the denser air, the drag on the parachute is also higher, so it will hang longer. It appears that they need to ballast to one weight and use one size chute to qualify here in Connecticut and will probably need more ballast and a larger chute if they make the flyoffs in Virginia.
 

bobkrech

Well-Known Member
Joined
Jan 20, 2009
Messages
8,354
Reaction score
31
ROCKSIM calculates the Cd automatically and this is a source of the over optimism.

Most regular rocket have an average Cd value between 0.6 and 0.75. I would suggest that you run either wrasp or webalt with the actual prepped rocket weight, airframe diameter, thrust curve and both Cd values to bound a preflight altitude estimate. I would also use the best temperature estimate for the day of the flight to get the estimate a bit closer. (The difference in air density from a 77F (25C) summer day to a 32 F (0C) winter day is 9.1%!)

Then for each Cd value change the rocket weight by +/-1% in the sim to determine the change in altitude for a 1% weight change to determine the approximate -dA(feet)/dW(grams) constant for the rocket.

To really dial in a rocket, you want to make a minimum of 3 flights with the same engine on a calm day with an altimeter in the rocket to determine the apogee. (The choice of engine should be based on the previous sim to get you a10% above your goal.) You must record the weight to a gram before each flight, and record the air temperature and the altitude attained to the nearest foot. Then you can simple adjusted the Cd in the simulator to match the altitude, and lock in the value. Once you have the Cd locked, if the altitude is close to your objective, rerun the snensivty sim and adjust the weight based on the new sim sensitivity value and relaunch. You'll be amazed how close you'll get to your estimate.

Bob
 

hardinlw

Well-Known Member
Joined
Jan 19, 2009
Messages
372
Reaction score
1
The method you describe is basically what the team is doing. Rocksim also allows you to override the Cd and that is what the kids did to match their earlier (unpainted) data. We got three datapoints a couple of weeks ago on the painted rocket (would have been 4 but they forgot to turn the altimeter on for the first flight), but the wind was gusty and I don't think they can separate the effect of wind (weather cocking) from the effect of changing the weight. If mother nature cooperates, they will get a few more datapoints tomorrow and dial in the Cd and weight sensitivity. The current estimate based on Rocksim is 31' per ounce at a 15.4 ounce baseline.

I used the flight they made with the altimeter off to emphasize the importance of a preflight checklist, which one of the members had already made, but the ones on the field didn't think they needed it. The previous outing, they learned about dressing for the weather instead of fashion. Who would have thought that flying rockets could be so educational? :D
 

bobkrech

Well-Known Member
Joined
Jan 20, 2009
Messages
8,354
Reaction score
31
Last year a team I mentored flew ~30 practice flights and really analyzed their flight data. Their motor CATOed at the finals destroying the lower end of the rocket. Unfortunately they did not bring their spare rocket, but since it happened early in the day, they went to the local hardware store for stuff they didn't have, rebuilt the lower half of the rocket, leaving it 6" shorter, weighted it, reballasted it, and on their relaunch hit 759'. Unfortunately they did not check out their shock tube airframe mounting and has an airframe separation on ejection. The egg capsule was successfully recovered but stayed up longer than expected due to the reduced weight. They still finished 34 overall, and breaking their stretch of top 25 finishes.

Bob
 
Top