Servo Motor Deployment

[dragonsp]

OK, just to start this out, this may be a really crazy idea, but we wouldn't get anywhere in this hobby without them. I'm trying to do my part.
Hold on to your desk....
Dual deployment without any pyro or regulated items whatsoever.
Crazy, huh?

Well, my problem that started this whole thing was that us NAR Jr. L1 people can't use black powder and ematches because they are regulated, and we are obviously not 18 or older. This got me thinking to Air Command Water Rockets who I recalled had a non-pyro method of deployment.

They made a video (here) which I thought would be a good reference point. Everything seems straightforward but...( and I'm looking at you, robotics or R/C hobbyists)

a. I don't have a timer like the one in the video and...

b. I don't know much about servos. Well, a little, but I'm still fuzzy on some of the basics.

So, right now, I'm trying to find a way to get a similar assembly without spending too much money. My current (loose) plan is to purchase a servo, a servo controller, and an altimeter. I plan on connecting the output of the altimeter to the input in the controller, which will trigger the servo motor to turn.

So, a few questions to ANYONE whom knows much about servos in general.

a) How can I hook up a servo to a deployment charge? I know that they have a special cable, so should I separate the power and the ground cables from a connector, so I can still disconnect it and program it?

b) Once a motor is programmed to work at the right speed for the deployment charge, does it stay programmed in that way after you disconnect the power? Basically, would I need a special servo controller inside the rocket in order for the servo to work properly?

c) Lastly, would you allow a 14-17 year old to do this dual deployment with a high powered rocket?

Note: I know that there probably many other young'uns like myself who would be interested in this, and many other people as well, so I plan on documenting my tests and my new knowledge acquired with this form of deployment. One of my personal greatest disappointments with the NAR's system is that I can't do dual deployment. I'd like to help as many other people as possible.

Also: I plan on only using this for the main deployment, I'd only use motor ejection for a drogue. If this turns out to be reliable, I might consider using a full dual deploy system with this.

Thank you so much for your time and replies in advance.

dragonsp

[billspad]

Take a look at the article on page 12 of this CMASS newsletter.

[rcktnut]

From the NAR Jr. LVL 1 Rules: The qualification flight and all future flights must be single deployment only. This is due to regulatory requirements of ejection charges used in dual deployment systems. On board electronic devices are permitted as long as they are not used for deployment.


Even though you have a non pyro dual deploy method, you may want to see if this method is ok first. The rules may have to be re-worded first. Due to complexity, reliability issues NAR may not want Jr. LVL 1's using dual deploy period. You can't do anything anyway other than build the rocket. You need your sponsor to build the motor and load it in order for you to fly it.

Your sponsor would also have to go along with the idea, as ultimately he is responsible for the flight. I see nothing wrong with your idea, very good, just have to get it approved.

[sylvie369]

So, right now, I'm trying to find a way to get a similar assembly without spending too much money. My current (loose) plan is to purchase a servo motor, a servo programmer, and an altimeter. I'd program the motor with the programmer to a good speed and then attach the motor to the deployment charge of the parachute.

I think it'd be great if you worked on this. To "deconfuse" this a little, I think it'd help if you changed your terminology a bit.

- Most people refer to them as simply "servos", not "servo motors", and since your rocket has a rocket motor in it, it'd be less confusing if you stuck with just "servo" as well.
- It's not clear what you mean by "servo programmer". Standard servos themselves cannot be programmed. Digital servos are programmed, but not in a way, I think, that will be useful to you. Do you mean "servo controller"? (below)
- "deployment charge" is the phrase usually used to refer to the black powder that is ignited to blow out the parachute. You seem to be using it to refer to something else here, and I can't tell what.

I think that many of your questions can't be answered until these things are cleared up.

Normal servos take input in the form of electrical pulses usually from a radio control receiver. Those pulses tell the servo what position to move to, or to hold at (yes, when powered up, a servo requires pulses just to hold position). It's possible to send those pulses from other sources, including microprocessors and standalone servo controllers. When you send the pulses from a microprocessor, remember that you're tying up the processor in the sense that it has to continue to send pulses even when NOT trying to move the servo, just to keep the servo at one point. Since most processors can only do one thing that a time, that's a problem. Many servo controllers are designed to simply take over the load of sending those pulses, so the microprocessor only has to say when to move the motor, and the servo controller connected between the processor and the servo takes care of delivering the pulses between movements. Here's an example:

http://www.parallax.com/Store/Access...5/Default.aspx

Other servo controllers can themselves be programmed to move the servos either at a certain time or in response to certain inputs. Here's an example:

http://www.pololu.com/catalog/product/1352

[Andreas Müller]

Spacetec Rocketry (www.spacetecrocketry.com) has some products that might be useful for you. Their Servo-Control is an add-on to an ordinary altimeter that simulates an ignitor to the altimeter (continuity, but no continuity after being "fired"), but controls a standard RC servo on the other end. It is best used with the Servo-Release-Mechanism SRM2 also from Spacetec (disclaimer: I'm not related to them, but I have used the products quite often with success, my L3 project used two SRM2 for deployment). The SRM2 holds the main chute in its bag until the altimeter "fires" the Servo-Control. My Atlas-Centaur 1:10 deployed the main chutes (3m and 4.5m) using SRM2s, and the Ariane 4 project of the swiss Team Ariane had a total of 6 SRM2 to deploy the main chutes of all 4 detachable boosters (each booster had 2-stage recovery) and the two main parts of the rocket.

With some altimeters, the Servo-Control isn't necessary: the Altimax line of altimeters (http://rockets.aquarix.de/node/4) has programmable servo channels. The timers of my own design (not for sale) also always have servo channels. I don't by any altimeters without servo channel, most of my rockets require them, so I expect them in my altimeters. Reduced component count also means reduced number of failure modes.

[troj]

Going from memory from an article I read several years ago...

Analog servos use varying width electrical pulses (pulse width modulation) to drive the servo. So, you'd need a controller to convert take the altimeter "output" and use it to trigger the servo. Several years back, there was an article in one of the rocketry magazines about this; as I recall, the electronics are not overly complicated.

Other options are to look at the hardware Andreas referenced, or the various shields available for the Arduino. A quick search reveals motor shields designed for driving various motors, including servos.

If such a system were developed, documented, and shown to be reliable, my guess is NAR would consider rewording their jr Level 1 policies to say that mechanical dual deploy methods are permissible.

Even ignoring the junior Level 1 aspect of it, it's a pretty cool project, and something I think a lot of folks would find interesting. Might even be worth submitting a writeup to try to win NAR's annual R&D award.

-Kevin

[rockets4kids]

Although you can use a servo, there is another option that may be easier, cheaper, simpler, and lighter.

Simply rig two #4 brass screws on a plate about 1/4" apart and span the distance with a piece of 36 gauge nichrome wire. Use this as a catch for whatever spring mechanism you desire. You can melt the nichrome directly with a high-current altimeter output, or you can use a relay/mosfet and any output. To keep things light, you will need a high-discharge LiPo battery to melt the wire. I use a 25-40 C 160 mAH LiPo that is about the size of a small stick of gum and weighs about 5 grams and can deliver a full 7 amps no-load current.

I have not done any flights with this rig yet, but it has performed flawlessly in ground tests.

[KevinDunn]

Simply rig two #4 brass screws on a plate about 1/4" apart and span the distance with a piece of 36 gauge nichrome wire. Use this as a catch for whatever spring mechanism you desire. You can melt the nichrome directly with a high-current altimeter output, or you can use a relay/mosfet and any output. To keep things light, you will need a high-discharge LiPo battery to melt the wire.

I devised a similar scheme using a rubber band and nichrome wire. The idea was that the rubber band would wrap around the chute to hold it closed. Motor deployment breaks the rocket and the parachute is rubber banded in the closed position until the altimeter releases it. The altimeter heats the nichrome, which breaks the rubber band, releasing the chute. I used the AIM USB altimeter because you can program the time that the channel fires (e.g. 0.5 sec, 1 sec, etc.) The nichrome has to be pretty thin to get hot enough. I think I used 39 gauge. It worked in ground tests, but then I started using conventional pyro DD and the project got put on the back burner (pun intended).

[DAllen]

I've always thought that mechanical methods for deployment would be a bit heavy. I like rockets4kids idea however maybe instead of nichrome what about fishing line? Would that melt at a lower temperature than the nichrome?

Now you got my wheels a turnin'.

[rcktnut]

Fishing line and nichrome rang a bell after thinking about it. I thought there was a product available a few years back that was used for non pyro DD. Here it is:http://www.chutetamer.com/introduction.shtml

[rockets4kids]

I've always thought that mechanical methods for deployment would be a bit heavy. I like rockets4kids idea however maybe instead of nichrome what about fishing line? Would that melt at a lower temperature than the nichrome?

In my rig I have a metal hook that catches on the nichrome and I melt the nichrome. You could indeed use a nylon catch around the nichrome, in which case you would only need to heat the nichrome hot enough to melt the nylon. In this case the nichrome would be re-usable many times. My rig using a wooden dowel and rubber band to drive a foam piston is perfectly suited for larger MPR flights.

[blackbrandt]

Would this also work? What my idea is is to use a cluster, and do a shotgun recovery. One motor fires the apogee charge, and the other motor fires the main charge. Is this legal for a junior HPR?

[jadebox]

Would this also work? What my idea is is to use a cluster, and do a shotgun recovery. One motor fires the apogee charge, and the other motor fires the main charge. Is this legal for a junior HPR?

I think the rule specifically forbids dual-deployment recovery, but, obviously, what you're planning doesn't involve using a separate ejection charge. So ... I dunno if it would be allowed. I'd ask someone at NAR for a clarification.

BTW, I tried that technique on a model rocket once. It sort of worked. But, the difference in delays between the two model rocket motors wasn't enough to really benefit from dual-deployment.

-- Roger

[blackbrandt]

I am talking about a 5 second delay and a 15 second delay.

[rockets4kids]

Would this also work? What my idea is is to use a cluster, and do a shotgun recovery. One motor fires the apogee charge, and the other motor fires the main charge. Is this legal for a junior HPR?

I cannot comment on the legality of such a set-up, but you would need an awfully long delay on one of those motors for any sort of dual-deploy to be actually useful.

One option you may wish to experiment with -- and one which may be legal -- is to rig a full-size parachute with motor deployment. You could then rig secondary deployment to occur during descent. The secondary deployment could be nothing more than separating a motor tube. If you can achieve a number of successful launches using this method, you may be able to convince the RSO to allow you to fly with a minimally sized motor-driven chute followed by an equally sized secondary chute.

Finally... do any of the Junior HPR rules apply when you are launching MPR? Rather, are there any age-dependent rules when launching MPR? You should really be testing all of this stuff at the MPR scale anyways before moving up to HPR.

[dragonsp]

I think it'd be great if you worked on this. To "deconfuse" this a little, I think it'd help if you changed your terminology a bit.

- Most people refer to them as simply "servos", not "servo motors", and since your rocket has a rocket motor in it, it'd be less confusing if you stuck with just "servo" as well.
- It's not clear what you mean by "servo programmer". Standard servos themselves cannot be programmed. Digital servos are programmed, but not in a way, I think, that will be useful to you. Do you mean "servo controller"? (below)
- "deployment charge" is the phrase usually used to refer to the black powder that is ignited to blow out the parachute. You seem to be using it to refer to something else here, and I can't tell what.

I think that many of your questions can't be answered until these things are cleared up.

Normal servos take input in the form of electrical pulses usually from a radio control receiver. Those pulses tell the servo what position to move to, or to hold at (yes, when powered up, a servo requires pulses just to hold position). It's possible to send those pulses from other sources, including microprocessors and standalone servo controllers. When you send the pulses from a microprocessor, remember that you're tying up the processor in the sense that it has to continue to send pulses even when NOT trying to move the servo, just to keep the servo at one point. Since most processors can only do one thing that a time, that's a problem. Many servo controllers are designed to simply take over the load of sending those pulses, so the microprocessor only has to say when to move the motor, and the servo controller connected between the processor and the servo takes care of delivering the pulses between movements. Here's an example:

http://www.parallax.com/Store/Access...5/Default.aspx

Other servo controllers can themselves be programmed to move the servos either at a certain time or in response to certain inputs. Here's an example:

http://www.pololu.com/catalog/product/1352

Thanks so much for the links and suggestions! They were really helpful and your information was exactly what I was looking for. I'll fix the original post to make it more clear.

I cannot comment on the legality of such a set-up, but you would need an awfully long delay on one of those motors for any sort of dual-deploy to be actually useful.

One option you may wish to experiment with -- and one which may be legal -- is to rig a full-size parachute with motor deployment. You could then rig secondary deployment to occur during descent. The secondary deployment could be nothing more than separating a motor tube. If you can achieve a number of successful launches using this method, you may be able to convince the RSO to allow you to fly with a minimally sized motor-driven chute followed by an equally sized secondary chute.

Finally... do any of the Junior HPR rules apply when you are launching MPR? Rather, are there any age-dependent rules when launching MPR? You should really be testing all of this stuff at the MPR scale anyways before moving up to HPR.

Yep. That's exactly my plan once I get all of the parts and assemble them. If it doesn't work, then no damage will be done to the rocket. If it does work, then I might move on to a system of two equal-sized parachutes, one deployed at motor ejection, and another deployed at a lower altitude.
To answer your second question, the strict rules are not used on us when flying mid-power. I'm thinking of getting either a new 2.6" kit to modify, or using a few LOC 54mm tubes I have lying around.

I've done more research, and have got a basic plan. I'm going to buy an altimeter, a Mini Maestro servo controller, a generic servo probably from the same store, and a resistor at the local RadioShack. I'll need the resistor to reduce the voltage that the altimeter sends to the microcontroller when the rocket reaches the programmed altitude. I certainly don't want to sizzle the microcontroller during flight! The microcontroller says that it detects voltages from one of the inputs from 0 to 5 volts, so even though it might not make a difference, I would rather be safe than sorry with a 9V battery.

My current question is whether the altimeter will still fire the lower altitude ejection without continuity. My current plan is to use the Adept 22 altimeter, so does anyone know if that will be a problem?

Once again, thanks so much for the replies. All of this interest is giving me more incentive to continue on with this project!

[dragonsp]

So, I've got most things sorted out, but I still have one problem. On the wiring schematic for the input (here) I need a way to connect the switch in the circuit with the deployment charge sent from the altimeter. Right now I'm drawing a blank, so any ideas?

Also: I'd prefer not to use nichrome wire in the system, just in case it might melt some of the electronics, payload bay, etc.

Thanks in advance!

dragonsp

[clreynolds]

I need a way to connect the switch in the circuit with the deployment charge sent from the altimeter.

Sounds like a relay (remotely powered switch) to me.

A quick google search found theses.
http://www.futurlec.com/RelDip.shtml

[troj]

Something to consider in your plan is that your controller (or switch) may require constant power, rather than just momentary power, in order to keep the servo moving the full way.

Going from memory with Christmas tree bulbs, some altimeter provide power on their outputs for a set amount of time, some just leave power applied the entire time.

Whatever altimeter you're using, you can test its behavior with bulbs.

-Kevin

[Sailorbill]

Something to consider in your plan is that your controller (or switch) may require constant power, rather than just momentary power, in order to keep the servo moving the full way.

Going from memory with Christmas tree bulbs, some altimeter provide power on their outputs for a set amount of time, some just leave power applied the entire time.

Whatever altimeter you're using, you can test its behavior with bulbs.

-Kevin

If you use a relay such as clreynolds suggests get a dpdt and use one set of contacts to latch the relay closed when it is energized. You could also put a timer in the latching circuit that would open after a set time so the Battery would not be totally drained.

[dragonsp]

All right, I think I have everything down, and a picture is attached with the basic wiring. So, when the altimeter fires the "ejection charge", it will use a relay to close a circuit, which the micro controller will sense and will then turn the servo, releasing the rubber band, and open the hatch which will then deploy the 'chute. I plan on using a similar setup as in the video in the first post.

Well, at least it's not rocket science, right?

Over the summer, I'm going to build a Madcow Discovery to be the test rocket, and all of the materials. I plan on testing the system on the ground by wiring everything and then putting the altimeter in a shoebox and using a shop-vac to suck the air out to simulate flight. Since my current choice for altimeter is the Adept 22, I won't have to worry about an accelerometer during the testing.

[dragonsp]

A quick update:

All of the parts have arrived, except the servos that I have yet to order. I did a couple tests with the altimeter, which works fine, but I'm stuck on a few things (again).

First: The alleged "7-9 volt" relay from RadioShack doesn't work with a 8 volt battery. Yes, I ran out of fresh nine volt batteries. But, I have hopes that since the altimeter has a capacitor to provide more power to the ejection charges, the additional current might close the circuit. I might do some testing tonight or tomorrow, so I'll post those results. If it doesn't work, I might have to resort to a lower voltage relay.

Second: After cutting up a few soda bottles, the plastic seems to be much stiffer than in the video in the first post (http://www.aircommandrockets.com/construction_7.htm). This is a big problem, so I am not sure what to do. I thought I could either stop by the grocery store and see if they have some better bottles, or I could find a flexible plastic to use as the ejection plate. Thoughts?

Thanks in advance!

dragonsp

[John Beans]

In my humble opinion, this sort of thing is a great idea. Especially if it's small enough for lots of rockets, and easy/reliable enough to use quickly.
I've sketched a mechanical "chute delay" device many times, either with servos or muscle wire. Both can be quite tiny. There are lots of windy days that I'd like to clip one onto my chute, and save a lot of walking (or lost rockets). As I imagine it, you'd set it to an altitude (or just leave it where it was last time, and clip it around your chute. It should turn on automatically when you do that. And be easily recharged (no batteries to find, replace).

dragonsp, you might think about replacing that relay with a simple voltage divider, to step the voltage that comes out of the altimeter's internal switch from 9V down to a value that is compatible with the IO of your microcontroller. For instance, if your microcontroller is 5V tolerant, feed the output of the altimeter into two 50K resistors in a row and then to ground, then tap the center of them into your microcontroller. It will hold your input low when nothing is happening, and instead of 9V when activated, the input to your microcontroller will then be 4.5V. (Or choose resistors for a different top voltage if 5V is too much). Remember, the input of your microcontroller needs almost no current to register a "1". Make sense? Just my two cents...

[dragonsp]

In my humble opinion, this sort of thing is a great idea. Especially if it's small enough for lots of rockets, and easy/reliable enough to use quickly.
I've sketched a mechanical "chute delay" device many times, either with servos or muscle wire. Both can be quite tiny. There are lots of windy days that I'd like to clip one onto my chute, and save a lot of walking (or lost rockets). As I imagine it, you'd set it to an altitude (or just leave it where it was last time, and clip it around your chute. It should turn on automatically when you do that. And be easily recharged (no batteries to find, replace).

dragonsp, you might think about replacing that relay with a simple voltage divider, to step the voltage that comes out of the altimeter's internal switch from 9V down to a value that is compatible with the IO of your microcontroller. For instance, if your microcontroller is 5V tolerant, feed the output of the altimeter into two 50K resistors in a row and then to ground, then tap the center of them into your microcontroller. It will hold your input low when nothing is happening, and instead of 9V when activated, the input to your microcontroller will then be 4.5V. (Or choose resistors for a different top voltage if 5V is too much). Remember, the input of your microcontroller needs almost no current to register a "1". Make sense? Just my two cents...

That sounds like a much better idea. I was a little confused at first about that form of input, so I wanted to use the relay as a switch, especially since there was a schematic on Pololu's website using a switch. Now that I understand everything better, I'll try out this method. In addition, the simpler design gets rid of another point of failure. I've heard mixed reviews about RadioShack's relays, so I'd rather not risk it. Also, getting rid of the relay gets rid of another point of failure in the system.

So, just to clarify, I would connect the Ground and Input Voltage to the altimeter, and in between that circuit, have a three way joint between the two resistors and the signal line to the microcontroller. Is that correct?

Thanks a lot! I wouldn't have made it this far in the design without all of you guys!

dragonsp

[cerving]

That sounds like a much better idea. I was a little confused at first about that form of input, so I wanted to use the relay as a switch, especially since there was a schematic on Pololu's website using a switch. Now that I understand everything better, I'll try out this method. In addition, the simpler design gets rid of another point of failure. I've heard mixed reviews about RadioShack's relays, so I'd rather not risk it. Also, getting rid of the relay gets rid of another point of failure in the system.

So, just to clarify, I would connect the Ground and Input Voltage to the altimeter, and in between that circuit, have a three way joint between the two resistors and the signal line to the microcontroller. Is that correct?

Thanks a lot! I wouldn't have made it this far in the design without all of you guys!

dragonsp

I would strongly recommend against using a relay in any rocketry application, the electromechanical contacts have a tendency to shake loose with G forces and vibration in flight. This is bad, especially for recovery devices. If you are using the relay to turn on a 555 timer or similar circuit to modulate your servo, I would use a transistor switch instead, a 2N2222 NPN or similar should work fine since you're not really drawing much current. Most hobby servos are nominally rated at 5v but they will work fine on a 3.7v Li-Po cell or 3.6v NiMH pack, and if your altimeter will run on that then that's all the power you need. Servos don't have the huge current spike that you see with igniters so it's safe to run on a single battery.

I am currently working on a commerical flight computer that will directly drive a servo as well as an igniter, it is the very late beta stage (about 2 months to market). I have flown it a few dozen times and have successfully done DD's with pyro charges, but I haven't flight tested a servo yet. If you live in the SoCal area and can get to SCRA/ROC events and you would like to beta test it with a servo, send me an email at cris.erving@hotmail.com and we can talk.

[dragonsp]

Well, it has been a while since I started this project, and I have put it on the back-burner until now. I keep on thinking that it is a major possibility that the drogue deployment (motor ejection) could cause a rubber band or similar to come loose. I already have all of the electronics, so I'm thinking about using these mechanics similar to a tether like the Tender Descender. I'm thinking about having the main parachute in a deployment bag, and having a string attached to the drogue to prevent the main from coming out (reading the instructions on the Tender Descender page helps explain this). I plan on having a launch lug sliced in half and a screw or eyebolt that would go through the launch lug. In between the slices of launch lug, the "tether" to the drogue is attached. When the servo turns, it pulls out the screw and releases the tether, and the drogue pulls out the main.

Confusing, huh? When I get it built, I'll show pictures.

I like this idea better because this would allow for dual deployment out of a single compartment, and if done correctly, will have a much lower chance of the main coming loose with all of the jostling during drogue recovery.

Any questions or comments, feel free to post!

Thanks,

dragonsp