Vertical Trajectory System (with some 3D printed parts)

Discussions on high/extreme altitude and mach busting rockets.

Moderator: Moderators

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1887
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: Vertical Trajectory System (with some 3D printed parts)

Postby SpaceManMat » Thu May 19, 2016 9:18 am

Looks good, although you seem to be doing a lot of work just use an existing controller. Will you reuses all the parts when you build your own control device?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Sat Oct 07, 2017 8:06 pm

SYSTEM INTEGRATION – ELECTRONICS
Been a while since I have done any significant work on this. Life gets in the way sometimes.
So with a bit more thinking, a bit more designing, and a bit more building, here is where I am at :)

Electronics have been integrated, but there are three loose-ends to go still. I need to find my final part (G-switch) which I have somewhere, but is currently doing a good job of eluding me :( .

Construction is basically discs of 1.6mm G10 F/G, separated by spacers. Before flight I will change the spacers to custom titanium bike spokes with plain cylindrical spacers.

During deployment this module will stay attached to the NC.

DSC07035resize.JPG
DSC07035resize.JPG (238.31 KiB) Viewed 333 times
DSC07034resize.JPG
DSC07034resize.JPG (230.03 KiB) Viewed 333 times


From Top to Bottom
The black cable connector plugs into the NC to interface to the TeleMega. It will provide logging of the servo signals for post-flight analysis.
Next is the flight controller and servo connectors.
Servo block (contains four servos and a few connectors)
Logging interface (converts the PWM signals to the servos into voltages the TeleMega can log)
Reference pulse generator and Control Interface PCA.
G-switch and external control/testing interface.
Two SBECs to provide power for the avionics and the servos on separate regulators.
Microswitches, with a pull-pin actuation, for turning everything on.
LiPo batteries (2S and 3S) for running everything.
Connectors for charging the batteries and connecting to the deployment altimeters that will control when the control system is operational.
Finally, an eyebolt to string it to the rest of the rocket.

You will notice that there is no PCB for interconnecting the wiring between the modules. I went with plain wires only for this breadboard.

I still have a 3D printed part to be made, and installed on the autopilot. More info on that later.

Next major step will be airframe integration, again, as I need a longer length of airframe than initially thought. I didn’t go particularly hard on miniaturising the assembly as it is really breadboard and will likely have lots of changes applied.

Following that will be some ground testing.

Now, if I can just find that G-switch...
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1887
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: Vertical Trajectory System (with some 3D printed parts)

Postby SpaceManMat » Sun Oct 08, 2017 8:57 am

Looks really good.
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Mon Oct 09, 2017 4:44 pm

Looks really good.

Thanks Mat.

Answering your earlier question I missed:
Looks good, although you seem to be doing a lot of work just use an existing controller. Will you reuses all the parts when you build your own control device?

This is my own controller :) Built using COTS parts. Assuming it works I will happily keep developing it. If more time becomes available I might venture into a new flight controller and custom firmware. I would just need to convince the firmware people at work to volunteer to help me out so I can keep doing the fun stuff. I can write software, firmware and VHDL, but it isn't something that I find amusing in my spare time. Would probably only go down that route if I needed a high-performance VTS and had to include things like gain scheduling. Otherwise KISS is good.
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Sun Oct 15, 2017 9:16 pm

SOME MORE 3D PARTS

We had a new 3D printer delivered at work recently. This has a smaller build volume than our other printer (145x145x175mm) but uses stereolithography as the printing system. Basically liquid resins are spatially cured by LASERs. Many different resins are available: Tough, flexible, clear, rubber etc. Print resolution can be set from 100um down to 25um. The other advantage of SLA over the more common FDM process is that there is no z-axis weakness in the part. It is an inverted SLA print process.
https://formlabs.com/blog/ultimate-guid ... -printing/

I needed a “cage” to limit travel of the autopilot unit during severe accelerations. The AP usually sits on four foam adhesive pads for vibration isolation. The usual outcome in a fixed-wing flight with a bad landing is that the AP tears the foam and travel is limited by the length of the attached wires. I designed an enclosure that would sit over the AP, with room to move around on its mount, but limit its excursions when jolted while still allowing for vibration isolation.
DSC07039resize.JPG
DSC07040resize.JPG

Left image shows the support structure added during the printing process.
Note that the small extension on one end is to hold the plugs captive on the AP, since they are not positively locked on.

Having the VTS on the bench proved a bit cumbersome, especially with the eyebolt stopping me from standing it up. I designed a stand that would take the VTS and support it while I worked on it. I also built in a voltage monitor so I can check battery voltage without getting the multimeter out, and added a switch to be able to have the VTS in active or fixed mode for testing purposes.
DSC07042resize.JPG

Note that the base has two diameters so it can stand the bare assembly or the airframe integrated unit (either way up).

I need to drill the airframe for the mounting points and fin hubs. Since the servos are fitted now I can’t use the same method I used before. I printed up a drill jig on the FDM printer for this one. It just slips directly over the airframe.
DSC07036resize.JPG


Continuing...
Work is progressing nicely. Electronics integration is complete. Testing individual components has commenced.

I'll drop another update shortly :)
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Bench Stand

Postby OverTheTop » Mon Oct 16, 2017 8:39 am

Bench Stand
Some very basic wiring was added to the base, for connection to the VTS.
A momentary pushbutton switch takes care of switching the system between fixed and active modes during bench testing. Having a 3D printed hole to mount it in makes things so easy!
DSC07043resize.JPG
DSC07043resize.JPG (226.58 KiB) Viewed 256 times
DSC07044resize.JPG

OnStand.JPG


A voltmeter module and switch was added to read the two battery voltages. The voltmeter module was glued in place using CA glue. These modules are great for having in place for voltage sanity checks when working on stuff. They can be purchased for $2-$5 on eBay, in whatever color you want. There are a couple of different types. Two-wire and three-wire. The three wire ones I have are good for indicating 0-100V on the input line, and require 5-30V to power them. The two-wire versions just measure their supply voltage and are typically only 5-30V or thereabouts.
Just search for “LED voltmeter module” on eBay
voltmeter.jpg
voltmeter.jpg (7.24 KiB) Viewed 255 times
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Fri Oct 20, 2017 7:50 am

TestPlanDone.png


Looking good so far. Now to dial in the required gains...
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1887
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: Vertical Trajectory System (with some 3D printed parts)

Postby SpaceManMat » Fri Oct 20, 2017 8:56 am

Coming along nicely. You see of TRF that Jim busted a servo?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Fri Oct 20, 2017 9:34 am

Coming along nicely. You see of TRF that Jim busted a servo?


I did see that. Interesting it gave up the ghost during the flight. Would be nice to know the failure mode.

Something like that is always a possibility. Hopefully it won't bite me!
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1887
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: Vertical Trajectory System (with some 3D printed parts)

Postby SpaceManMat » Fri Oct 20, 2017 5:24 pm

OverTheTop wrote:
Coming along nicely. You see of TRF that Jim busted a servo?


I did see that. Interesting it gave up the ghost during the flight. Would be nice to know the failure mode.

Something like that is always a possibility. Hopefully it won't bite me!


I imagine it broke a tooth. You should ask him what failed.
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Mon Oct 23, 2017 6:02 pm

Showing Restraint
As mentioned earlier, the AP needs to be vibration isolated from the airframe with foam pads, and it needs to be restrained during boost and impact to make sure the foam pads are not trashed, leaving the AP flopping about like a fish out of water.

The 3D printed cage was trimmed a little to allow for some wire access. The holes in it were drilled and tapped to M3. The top plate has holes for the cage added, as well as the existing holes for access to the LEDs, switches, and gain settings. The height of the cage is just a shade thicker than the spacing between the G10 to make sure it does not contribute to additional vibration during flight.

CageAndAP.JPG


APFitted.JPG


TopPlate.JPG



Note that the AP is on the top level of the sandwich so it is easy to access for adjustments.

Airframe integration up next...



Coming along nicely. You see of TRF that Jim busted a servo?


According to Jim his failure was due to the servo casing breaking. In fact, he thinks he may have possibly even damaged it during assembly and not noticed. Refer to link here for more information: http://www.rocketryforum.com/showthread ... nce/page13
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

strud
Engineer
Engineer
Posts: 914
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Re: Vertical Trajectory System (with some 3D printed parts)

Postby strud » Tue Oct 24, 2017 10:14 pm

Hi OTT

Looking good indeed

I would think it might be very good to add some tests in your plan to include sensor drift/cross coupling/off axis sensitivity to high accelerations since this sort of miss-leading signal could cause a significant drift in your integrated (eg velocity, displacement) and other metrics.

Have you got a rig that you can put this into to try to simulate this ? It would be quite difficult to generate linear accelerations (for example) over periods long enough to represent something like a flight.

CS

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Wed Oct 25, 2017 6:59 am

Hi Strud. Thanks for the comments.

I have been considering if there is any way of actually testing, with anything other than short-term rotations, the behaviour of the assembly. The short answer I think is "NO". Given this uses a commercial autopilot and so I have no way of tapping into the code, or real signals, I need to trust the code poets in their knowledge and skills, or at least hope they were working to a well thought-out specification. Bad products sold by the thousands does not do well for a company and that tends to encourage good efforts in design and product behaviour.

Two major unknowns for me:
    Low-pass filtering algorithm on the reference frame
    How the algorithm handles pegged accelerometers and gyros
Item one I am not so worried about. It needs to be sensibly handled or the fixed-wing aircraft using this would be not much better, or even potentially worse off than without assistance.
The second item, if allowed for in the code should ignore data that is likely to be nonsensical. It will be interesting to see how it manages in a sustained thrust situation, rather than the usual transient bumps. I think it will logically work.
Also, after burnout the expected "gravity vector" that the AP normally works with will be inverted due to drag. :?

What could possibly go wrong?! :wink:

The good news is that this will all be happening after boost and well up in the air. Any sub-optimal behaviour will end up with a chute deployed, although perhaps at less-than-ideal speeds. All actions will be logged so I should get good autopsy data regardless of what happens.

I still have quite a bit of bench-work to do with this yet so I am keeping an eye on any aberrant actions.
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1887
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: Vertical Trajectory System (with some 3D printed parts)

Postby SpaceManMat » Wed Oct 25, 2017 9:05 am

You planning to have cameras setup so you can monitor actual movements?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2457
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Vertical Trajectory System (with some 3D printed parts)

Postby OverTheTop » Wed Oct 25, 2017 9:26 am

You planning to have cameras setup so you can monitor actual movements?


Not at this stage, but it would not be too difficult to tape a couple of 808 keychain cameras to the airframe. I will see what happens between now and launch.
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)


Return to “Higher, Faster, Further”

Who is online

Users browsing this forum: No registered users and 2 guests