PDA

View Full Version : Looking for a way to keep a constant elevation...



deno
08-15-2005, 11:35 AM
angle on a satellite dish. Would anyone have any ideas on what sensor I could use with a stamp and reversing DC motor to maintain the same angle of elevation on a satellite dish regardless of what the base is doing that holds the dish. Elevation angle meaning "up and down", not azmuth.

Ken Gracey (at Parallax), you mentioned that you used some sort of sensor to maintain level flight on your R/C plane...do you think that would work for me.

Deno

Post Edited (deno) : 8/15/2005 4:36:19 AM GMT

Ken Gracey
08-15-2005, 12:36 PM
Deno,

I used the Memsic 2125 accelerometer on the R/C airplane. I've also used it as a slope indicator for avalanche prediction in a backcountry skills course. I'm a bit confused by your explanation of "elevation angle" - do you mean angle?

Andy has written a fantastic set of educational tutorials with this sensor. See http://www.parallax.com/dl/docs/prod/compshop/TiltDispCntrl.pdf and search back in these archives, particularly Stamps in Class and you shall find the following: http://forums.parallax.com/forums/default.aspx?f=6&m=55816 and http://forums.parallax.com/forums/default.aspx?f=6&m=62053.

There's more, too, but this should get you started in a friendly fashion.

deno
08-15-2005, 07:12 PM
Ken, thank you for the info.· I will download and study it.· As you know, to receive data from a satellite, you have to aim the dish at the satellite.· This involves 2 angles.· Azmuth, and elevation.· I have the azmuth figured out, as I have been receiving TV satellite data for years now, with my homebrew Sat Tracker aboard my boat.· Using the V2X compass module and the BS2sx.· This is only for azmuth in calm anchorages as the boat swings on the anchor.·

However, if the boat rolls alittle bit, the dish being fixed in elevation, looses the signal until the boat stops rolling.· I was/am looking for a device that will since this rolling and counter act by driving the dish to remain at a preset angle of elevation.· Most satellite receivers hold/store 2 seconds of video in memory before it is displayed on the TV.· If you can recover the signal before this 2 seconds has expired, you don't loose the TV picture.· So It doesn't have to be very fast acting.

Anyway, thanks again...

Deno

steve_b
08-15-2005, 08:25 PM
If you are rolling along your beam (stem to stern??) I can see this working....but I think the rolling from side to side will require more feedback control.....

Have you looked at gyroscopes....I've seen a few mentioned on here but haven't touched one. Since they use something similar in your 'wobbly' compass to keep it level and readable, maybe using one as a sensor might work in getting the motion you're after....then again, an accelerometer would still probably work (I'm honestly lookin to see how people have done with the gryo's and stamps)...

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve

"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."

its_gotta_hemi
08-18-2005, 08:49 AM
what your trying to do sounds similar to what i will be doing with my autopilot. I havn't yet built it but I'v played with the accelerometer a bit. I have it feeding a VB.NET program that uses graphics to display an artificial horizon (pitch and roll). My intent is to use the same info from the accelerometer to control stepper motors to control ptich and roll on the airplane. By using input from the gps, i can track a flight plan course line. The concept of tracking the flight plan and mainting straight and level flight (correcting turbulance induced flight deviations) is exactly like that described in a series of articals published a few months back in nuts and volts in which they explain and build a PID. Using that as a guide and tying an eccelerometer to a couple steppers or servos or motors, you would be able to keep the dish pointed where ever you want regardless of what the boat is doing..
just my 2 cents
-John

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John Foreman
jwf3@dcx.com

T!
08-18-2005, 11:11 AM
deno said...
angle on a satellite dish. Would anyone have any ideas on what sensor I could use with a stamp and reversing DC motor to maintain the same angle of elevation on a satellite dish regardless of what the base is doing that holds the dish. Elevation angle meaning "up and down", not azmuth.

What is the specific application?· If this is a 18-24 inch Direct TV or DishNet kind of thing it will have different requirements than say a 3 foot C Band dish.· A 24 inch DIrect TV dish will have a beamwidth of something like 2 degrees or so (top of my head here, sorry for any math errors that may show up), requiring a track accuracy of +/- 1 degree or better.· A 3 foot C Band may have a beam width as wide as·6 degrees, and could get by with +/- 3 degrees.

If the Memsic 2125 is accurate enough (+/- about·2 degree in the real world, 1 deg at about its best) then it could be used.· A stepper motor and a gear could be used to move the antenna in elevation.· Or an older screw type sat positioner could be used (as was often found on C Band dishes), but normally for·right ascension·motion, not elevation.

I think the real problem is going to be orthogonality, errors in pitch, yaw, and roll all adding up to make the adjustments needed by the system more complex than a simple left-right / up-down.· Have you considered a simple auto-track system?· Something like a monopulse tracker?· Not really all that simple, but it could be a blast to figure out.

T!

knightofoldcode
08-18-2005, 11:13 AM
Deno,

I think the idea is great. I've seen stuff like this done for RV's. My main question though is that it seems like this would be really hard to calibrate. I have a hard enough time getting my house's sat dish to be aimed correctly, without the problems of the sea.

The RV has a signal meter, so it has some sensors to get it in the right area, then it uses the signal strength to "hone in" on the signal. All of the ones that I've seen require about 3-4 minutes to get a picture, I realize that this is setting a signal AFTER a drive, so it has to start new, but I still think it's going to be really hard to get this system working. I love the idea, and would love to see how it's done though, and if you can complete it, more power to you! I certainly don't want to discourage you, it just seems like the sat dishes have to be VERY precisely aligned. I recall first time I set mine up, I didn't have a level pole to begin with, and never got a signal...... After leveling it (other than just by sight) I got the signal within 5 min.

Knight.

T!
08-18-2005, 11:54 AM
Hmmmm...how about this instead of driving elevation.· Something to take care of all errors, pitch, yaw, and roll.

You say you are already correcting in azimuth, and it is functioning for you.· If you built a leveling table, something controlled in two axis to present a level base to your system, this would remove pitch and roll.· Then the elevation could be preset (as it is now on the solid mount) and your existing azimuth axis control would keep the system on target by canceling yaw.

I have done this leveling table concept in small scale using hobby servos before, it was the first project I ever tried with my BS2E and Memsic 2125 combination.



T!

deno
08-18-2005, 08:12 PM
T!...thanks for the reply...in your leveling table, would it hold level when you moved the base around, up and down, etc?

The reason I ask is, on the boat, even a little roll is not noticed by the receiver, using an 18 inch dish. As I said, in an earlier post, if the signal returns back within about 2 seconds, you don't even notice on the TV. Programming is unaffected.

Bigger boats don't roll very fast, (45 feet) so it won't take much of a correction to maintain a level situation. I plan to build a universal joint with a piece of threaded rod thru it coupled to a fast DC reversable motor to act as the linear actuator. Two of these will "screw" the table up and down in both roll and yaw, with a pivot ball centered under the Sat Tracker. The brain will be the Mimsic 2125 and a basic stamp running a routine that will constantally check the 2125 for a correction (if needed) to both actuators. 99% of the time, with the Sat Tracker "hard" mounted on the rail in a level position, in a calm anchorage, you will keep the picture on the TV with no problem.

The problem arises when (sailboat in piticular) you start using water or fuel from an outboard tank and the boat starts to "heel" a little. This forces you to go outside with your nut driver and reposition the Tracker manually as the signal strength as indicated on the meter drops below the minimum level.

Most of the time, you wouldn't even have to turn on the level table, but when you needed leveling, just turn on the stamp and let it run the routine, then turn it off. On flat calm days, I have watch TV while motoring very long distances with the boats auto pilot on and the Sat Tracker on as well, with very little "frezzing" of the picture. It was interesting to note when the auto pilot made a little course correction to the left, the Sat Tracker would make a little azmuth correction to the right and visa-virsa. The V2X compass module from Percision Navigation Instruments does a pretty good job, and is fast enough with the BS2SX (even when averaging 2 readings) to keep the Tracker holding 1 degree of accuracy.

Anyway...T!..getting back to the orginal question, how responsive was that leveling table you put together?

The electronics and related software is always the easiest part of the designs...its the actuators and drive motors, with gears and drives that are the hard part of the design.

Deno

PS...Where did you buy your Mimsic 2125 from?

Paul Baker
08-18-2005, 10:40 PM
I think the problem may be in the way many accelerometers are constructed.

The cheapest variety has a hollow chamber and in the center there is a small heating element, located around the perimeter are temperature sensors. When the unit undergoes acceleration the little bubble of heated air drifts in the direction opposite of the acceleration which is detected by the temperature sensors. This scheme while inexpensive has serious limitations on the responsiveness of the signal in respect to the acceleration (the bubble drift time, the temp sensor aquisition delay etc.).

There are MEMS based accelerometers which have a faster response time but are more expensive and the application they are most used in (airbag deployement sensors) only need a single axis and operate off a threshold scheme (either the bag should be deployed or not) so I don't know if there are MEMS accelerators which have the features nessesary to be used in your application.

Another alternative is to use a gyroscope which can be found in small packages designed for use in R/C helicopters, these provide immeadiate feedback but cost between $100 and $200.

Finally you can use a tilt sensor, like this one (http://www.goldmine-elec-products.com/prodinfo.asp?number=G15429&variation=&aitem=8&mitem=11). Read the spec sheet (http://www.spectronsensors.com/datasheets/SDS-117-1104.pdf)·to determine if the tilt sensor is suitable for your project.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·1+1=10

Post Edited (Paul Baker) : 8/18/2005 3:44:13 PM GMT

steve_b
08-18-2005, 11:32 PM
ISn't there any low profile sat antenna's out there?· Something obviously powered for increased gain, but that is basically a helical type antenna (similar to external GPS antennas)...



I know it's no fun to do it this way....

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve

"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."

Larry~
08-18-2005, 11:45 PM
Satellite TV for boaters

» Satellite TV for boaters (http://www.kaptainsatellite.com/blog/archives/2005/03/02/satellite-tv-for-boaters/) Filed under: News (http://www.kaptainsatellite.com/blog/archives/category/news/)· Posted By: Tomas @ 4:02 pm 3/2/2005

Satellite TV (http://www.kaptainsatellite.com/) is now available even for boaters far on the sea. Raymarine announced in February its satellite television system that will bring high quality satellite signal to boaters while at sea. It shall be available to European boaters in summer 2005 and in spring 2005 for North America.

The Raymarine 45 STV has the ability to search for nearest satellite instantaneously and to maintain the most reliable connection while the boat is in the motion. It is housed in a white dome and measures 45 cm to complement all boats easily. It is also compatible with all digital video broadcast satellite TV services, so that each customer has all the freedom and flexibility to choose his or her own most suitable provider.


<!-- -->This off a web search<!-- You can start editing here. -->

kelvin james
08-20-2005, 01:01 AM
This is a simple idea that may work, and i will try to explain it the best i can. You could use a rotary encoder as a level sensor to control the motor. You add a weight to the shaft of the encoder, they are free-wheeling (no indents), so the shaft moves very easily without any friction. The weight would hold the shaft plumb with gravity as a level reference. As the boat tilts the encoder shaft will turn giving a pulsed output to the stamp, which then in turn would activate the motor to compensate for the tilt. You would probably not need a high resolution encoder, a 128 would likely be overkill, and the response time from the stamp should be good enough, as you said you had a 2 second buffer in the receiver. The software would be set up so the sensitivity could be controlled, like only have the motor move when it reaches a certain angle, otherwise it would be trying to correct itself with any movement. Encoders are fairly easy to use with a stamp, use little power, and will probably last forever with that type of operation.

kelvin

Jonathan
08-20-2005, 02:14 AM
Kelvin,

Seems to me that your idea would require some heavy damping not to over correct. Maybe a NIB magnet and a large chunk of copper? This is what I do on my seismic sensor.

Jonathan

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.madlabs.info (http://www.madlabs.info) - Home of the Hydrogen Fuel Cell Robot

metron9
08-20-2005, 02:25 AM
Hmmm Would you not need 2 encoders at right angles?

Larry~
08-20-2005, 04:15 AM
I would like to think outloud here

if the antenna were mounted on something that would float
and then placed into a fluid contained in a (bucket) would the antenna
seek level on its own.

Bruce Bates
08-20-2005, 04:28 AM
Hi Larry -

Yes it would, but why bother. The liquid is somewhat superfluous to the solution. A two-axis gimbal mounting would provide nearly the same thing and no liquid mess to contend with, if the action were too fast or too far.

Regards,

Bruce Bates

Post Edited (Bruce Bates) : 8/19/2005 9:29:59 PM GMT

Larry~
08-20-2005, 04:59 AM
This was a different way of thinking about the problem. Most gimbil mounts are not self leveling and
very spendy. you could use oil as the media, or liquid would not have to be used and a leveling gimbil
mount would be great. I was thinking of the way a sail boat stays upright and picturing a mount like this
even if it was held by springs.

Larry~
08-20-2005, 05:49 AM
http://www.wkarc.org/Research/ARCH/PrecisionAg/OWB.htm

Self-leveling Antenna Mount

We designed, built, and field-tested a simple, self-leveling device (gimbal) to serve as the RTK antenna mount The gimbal consisted of two perpendicular, sleeve bearing-supported, axes mounted in a light–weight, triangular steel chassis with three legs and handles. We chose three legs because it was more stable on uneven ground. Handles on the chassis made it easy for two persons to move and position the gimbal pointer. The entire unit weighs approximately 25 kg . It is 70cm tall and the triangular chassis is 100 cm x 100 cm x 80 cm. One person could easily operate the device, if one or two wheels were added.

The antenna was attached to a counter-weighted (plumb bob), 16 mm all-thread rod that threaded through an aluminum block at the center of the gimbal mechanism. The RTK receiver, radio modem, battery, cables, and data collector were mounted strategically on the chassis for optimum balance and ease of use. The chassis was moved until the pointer below the gimbal was directly above a specific location, then the RTK “fixed” position solution was acquired. Using the gimbal-based unit, we conducted two unique comparisons of data sets on multiple dates:

Larry~
08-20-2005, 05:53 AM
http://nsidc.org/data/docs/noaa/g00791_iabp/mechanical.html


another good device for buoy antenna mount

T!
08-20-2005, 06:42 AM
deno said...
T!...thanks for the reply...in your leveling table, would it hold level when you moved the base around, up and down, etc?


..........

Anyway...T!..getting back to the orginal question, how responsive was that leveling table you put together?

The electronics and related software is always the easiest part of the designs...its the actuators and drive motors, with gears and drives that are the hard part of the design.

Deno

PS...Where did you buy your Mimsic 2125 from?
Deno,

Yes, the two axis table would hold level as the base was moved and twisted.· How responsive the table was depended on the servos used.· I tried two different sets, one set older and slower, and one set newer, faster and higher torque.· Both responded well but the slower set took about 1 sec to correct an instantly input·30 degree angle on the base.· The newer set was much faster.· 30 degrees was about the mechanical limit of the linkages I used on that particular table.

Naturally, hobby servos will not work for your application, but I suppose you could use linear actuators instead.· Probably a ball screw type would be the ticket.· The drive motors really do not need to have encoders or position feedback.· The Memsic 2125 is the feedback, the drives just need to be reversable.· You can filter the response rate in your software if you need to.

I got the 2125 at parallax.com.

T!

deno
08-20-2005, 11:16 AM
T!..thanks for your reply...I don't think I will need to correct more then 5 to 10 degrees, if that much.· Like I said, there is quite alot of give when it comes to being off on the pointing of the dish.· Especially, in the United States.·It does get more touchy in the Bahamas because of the spot beaming that Directv uses to keep the signal in the States, but I can still get it with an 18 inch dish. Along with the rest of the Bahamian citizens.·(They have US sponcers) ·I have been able to receive the sat signal and view TV with a signal strength down to 68% on the signal meter.

I have ordered my 2125 and will start to mess around with it tomorrow.· I generally use DC gear motors, with the L293E H bridge as the driver buffered from the stamp with optoisolators.·(Learning the hard way, if a L293E goes bad from abuse, without a optoisolator, it will put 12 volts right back into the stamp I/O...$50 down the drain)·They do take alot of abuse, and it is easy to heat sink them with a piece of thin copper soldered across the top and down to the 4 center pins on both sides. (these 8 pins are ground anyway.) But the opto saves the stamp everytime.

Surplus center has some nice linear actuators which were designed for electric seat positioners on autos.· 12 volt.· Anyway, that is the plan.· It's all just a hobby anyway.

Again, thanks to all for you knowledge base...

Deno