Roll & Attitude control

Discussions on high/extreme altitude and mach busting rockets.

Moderator: Moderators

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2595
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Roll & Attitude control

Postby OverTheTop » Fri Aug 21, 2015 2:41 pm

FY31-AP from HobbyKing is my choice of poison at the moment. It makes for an interesting project as it is like the much used "black box" as far as functionality. I have been trying to get into the headspace of the programmers who wrote the firmware for fixed wing aircraft to check that they haven't programmed in any gotchas that would happen due to my use in a different craft. Nothing ringing alarm bells so far, except for the beginners mistake of coding without allowing for input range checking. The rocket dynamics could cause it to have a few sets of kittens along the way if they have not done sensibility checks on the sensor data :D
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

ogivemeahome
Rocket Crew
Rocket Crew
Posts: 443
Joined: Tue Jul 20, 2010 8:06 pm
Location: Brisbane

Re: Roll & Attitude control

Postby ogivemeahome » Fri Aug 21, 2015 9:46 pm

cryoscum wrote: ...providing the required torque at high velocities.
33.jpg

G'day Nic. Possibly you've considered this - move the pivot point aft relative to the fins for a more balanced system.
i.e. The airflow provides more force forward of the pivot point to reduce the load on the servos.
Awesome work!
A smooth sea never made a skilled mariner. (English Proverb)

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Sat Aug 22, 2015 5:50 am

Hi Ian

That is the obvious thing to do, but the problem is that the Cp of the canard shifts forward with increased speed. Accordingly that pivot is actually on the balance point at approx mach 2.5, accordingly giving the servos the best chance of working right when the forces in play are high. At the lower speeds the canard pivot is indeed off-centre, but the forces in play are also much lower, so the servos can cope. It's all about finding the best compromise as there is no 'right' answer...

Cheers
N
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2595
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Roll & Attitude control

Postby OverTheTop » Sat Aug 22, 2015 10:28 am

That is the obvious thing to do, but the problem is that the Cp of the canard shifts forward with increased speed.

I think you might have it the wrong way around. At slow speeds the Cp of a wing (what the canard is) is approximately 25% chord. As you go up in speed, especially supersonic, the Cp shifts rearward. How much of a shift depends on the wing design. So the Cp shift is towards the rear as speed increases, but the Cp is always forward of the midpoint on the canard.

As you said, you aim to pivot the wing around the Cp to minimise the torque requirement on the servos, which increases with speed.
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Sun Aug 23, 2015 4:54 pm

OverTheTop wrote:I think you might have it the wrong way around. At slow speeds the Cp of a wing (what the canard is) is approximately 25% chord. As you go up in speed, especially supersonic, the Cp shifts rearward.


Yeah, you're of course right when speaking about a wing shape in clean air. Mach tuck is the same for a canard as for the rocket as a whole... Until the effects of the NC start screwing with it, i.e. the very reason why I believe so many N5800 flights failed. See below:
1.jpg


As you can see the overall rocket Cp moves both backward and forward during the same boost, with the reversal happening above 1000MPh. My understanding is that the NC's shock wave increases in severity, to the point where the airflow over any control surfaces in the shadow is all messed up. The canards are not acting in a hypothetically perfect environment and, in this example, traditional mach tuck only occurs for the first 1.5 sec of the flight and then it reverses to a point far beyond it's starting Cp, i.e. the nett effect is a forward shift when the rocket is going at it's fastest. The graph of course refers to the whole rocket and I don't know how well the principle transfers to just the canard, but looking at the canard in pure isolation is likely a mistake too.

Admittedly, and crucially, the effectivity of the canards become highly suspect once you go above this speed and my head is too flat to find a good way of predicting the effects.
Guess we'll just have to try it, but I've been thinking about jettisoning the canards later in the boost when they can do no more good and just cause drag... (My sims seem to indicate that the amount of drag they add deletes all the gains you get from going straight up, and that's just with them there in static form. Once you actually start steering the rocket it'll get much worse.)
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2595
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Roll & Attitude control

Postby OverTheTop » Sun Aug 23, 2015 9:38 pm

Thanks for clearing that up Nic. I always question my sanity when I think you have said something incorrect. It's not often you are wrong but you are right this time :)

I have deliberately chosen the double-wedge airfoil to make the CP more predictable, especially at supersonic speeds.

I am still considering the planform of the fins, so I you have any thoughts please let me know.

I think if we keep the excursions relatively tame the majority of our assumptions will hold. You are pushing a much higher speed than I anticipate trying at this stage, so yes, you do have some other factors like shock wave/body/fin interactions to deal with.

I am still progressing with mine. I am doing a 4" version to fly in my soon to be rebuilt Velociraptor and this will be my main test bed for the VTS. Conveniently the Velociraptor will take the VTS and also the nosecone I have printed for the Apache. It is nice to have things interchangeable.

Speaking of the Nike Apache, I was going to go with trailing edge control surfaces on the fins but mechanically they are a nightmare (ok, maybe slightly too dramatic). I'll probably put the VTS with the canards in the payload bay section of the Apache when the time comes.

I have also backed away from using sliding pins for mechanical advantage, in the interest of keeping the system as simple mechanically as possible. My canards do not have the force or the speed that yours need to deal with, so I can simplify. I might get the servo mounting core printed this week. I will use PLA for that, and probably get the fins done in ABS.

Here is where I'm at on the VTS currently:
VTSpic1.png

This is only the servo housing. There will be a couple more slices of PLA that get bolted together to hold the electronics and batteries.
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

Lister
Engineer
Engineer
Posts: 967
Joined: Fri Jun 05, 2015 5:36 pm
Location: Bendigo, Victoria

Re: Roll & Attitude control

Postby Lister » Sun Aug 23, 2015 9:43 pm

Cant wait to see one of these in flight 8)

kir
Parachute Packer
Parachute Packer
Posts: 13
Joined: Wed Aug 19, 2015 6:57 pm
Location: Warner QLD

Re: Roll & Attitude control

Postby kir » Tue Aug 25, 2015 4:29 pm

Nic, what's the range of the torque values needed to control your rocket as applied to the canard axis ?

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Tue Aug 25, 2015 9:16 pm

Kir, where are you from? Please introduce yourself. We often have spammers attempting a way in by asking this sort of question as a first and only post.
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns

kir
Parachute Packer
Parachute Packer
Posts: 13
Joined: Wed Aug 19, 2015 6:57 pm
Location: Warner QLD

Re: Roll & Attitude control

Postby kir » Tue Aug 25, 2015 9:28 pm

Sorry, if I'm looking like a spammer, I'm not :).

I have actually introduced myself in the "Self Introductions". I'm a new guy here from Brisbane. I'm currently working at UQ, doing some machine learning research. This thread looks interesting to me, because I have some experience making controllers for UAVs, but I never was working with speeds >1M.

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Tue Aug 25, 2015 9:40 pm

Much better :)

I have a spreadsheet with the calculated torque requirements, but it is based on ideal scenarios where the rest of the vehicle and its effect on the canards are ignored. The reality is likely to be stranger than the spreadsheet indicates. I recently switched desktop PC's to a new one, so I'll have to go dig out the data. The red/green images earlier in this threat comes from this, but Strud (the creator of said spreadsheet) would be the right guy to involve in these discussions.

Cheers
Nic
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns

kir
Parachute Packer
Parachute Packer
Posts: 13
Joined: Wed Aug 19, 2015 6:57 pm
Location: Warner QLD

Re: Roll & Attitude control

Postby kir » Tue Aug 25, 2015 9:48 pm

Basically I'm asking, because it is not obvious to me if a standard servo (or, "expensive" one) could be used to directly drive the canard.
Now I got it, red/green works :)

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Tue Aug 25, 2015 10:00 pm

Screen grab from the earlier part of the thread:
When using a linkage ratio of 1:13-ish the numbers are all with the torque capabilities of the servo, noting that this is based on the canard sitting in clean air, not 10" behind a 7.5" Von Karmen nose cone.
The servo spec is in the box to the left and the torque numbers are in the green area (incorporating the linkage)
1.jpg

2.jpg

(click on the image to see an enlarged version)

I think this answers your question.
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns

kir
Parachute Packer
Parachute Packer
Posts: 13
Joined: Wed Aug 19, 2015 6:57 pm
Location: Warner QLD

Re: Roll & Attitude control

Postby kir » Tue Aug 25, 2015 10:14 pm

yes, thanks for the charts. But does it really necessary to have 15° ?

User avatar
cryoscum
STOP... Hammer Time The Prodigy
STOP... Hammer Time    The Prodigy
Posts: 1802
Joined: Wed Oct 06, 2010 2:54 pm
Location: QLD, NSW & VIC

Re: Roll & Attitude control

Postby cryoscum » Tue Aug 25, 2015 10:30 pm

At 200 mph yes, at 2000 mph definitely no. This is the big difference between the RC world and the rocketry world, i.e. the insane extremity of the velocity variations in 10 seconds.
AMRS L3 | NAR L3 | QRS 089 | MDRA 224
AMRS Technical Advisory Group

Total impulse for 2016: 32,458 Ns (thus far)
Total impulse for 2015: 84,231 Ns
Total impulse for 2014: 40,757 Ns
Total impulse for 2013: 62,927 Ns


Return to “Higher, Faster, Further”

Who is online

Users browsing this forum: No registered users and 3 guests