Shielding the Compass
SRLM
Posts: 5,045
Does anybody have a recomendation on how the shield the Parallax compass? I am constructing a robot with a compass, and discovered that when the compass was on the board the numbers were pretty much the same all the way around the circle. So I moved the compass off the breadboard using a couple of extension cables, but now I am alerted. I would like to completely isolate it from the electrical noise, but still have it close to the electronics (due to space constraints). Any suggestions?
Comments
More precautions don't necessarily improve anything and they sometimes make things worse. At the very least, they will give a false sense of security.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Stephen
Once you get a good chassis configuration you still may have trouble with indoors operation due to metal table legs and other such hazards if the bot wanders closed to them.
Of course you can turn a disadvantage into an advantage and use it as ferrous metal detector. In that context, each table leg might be mapped as a landmark within the boundaries of the room. You just end up with a bit more complex map.
Adjustment always have had to be made for compasses. On ships, they would 'degauss' the entire hull. And magnetic north is often not the same as true north.
So, the devil is in the details. A really good application software might overcome quite a lot of these problems. There is a good reason that military prefers brass buttons. Steel buttons and zippers throw off compass bearings for both navigation and guns. So do 'dog tags'.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's sunny and warm here. It is always sunny and warm here.... (unless a typhoon blows through).
Tropically, G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Kramer) : 8/14/2008 3:07:45 PM GMT
Have you considered how does a steel hulled ship manages to use a compass?
Rather than move the compass from the ship's bridge to some perfect location, they have adjustible iron balls around the compass that offset the imbalances in the magnetic field.
If the batteries and motor are centered perfectly below the compass's center, they don't effect the readings.
If you can find the major magnetic axis of the robot [noparse][[/noparse]by using any compass], you only have to add two offsetting magnetic masses that are perpendicular to the major axis and cross at the center of the compass.
That is how ships have done it for ages. You see the compass mounted in a housing with two large iron balls perpendicular to the ship's main axis. It would help to keep the robot 'magnetically symmetrical'.
Degaussing was a later addition because radar, sonar, and all the other additional electronic gadgets began to affect the compass.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
It's sunny and warm here. It is always sunny and warm here.... (unless a typhoon blows through).
Tropically, G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Perhaps use homemade electro-magnets that you can turn on when you want a compass reading? You could adjust their windings to
get the proper "balance" for the compass.
I don't have a compass myself, but I'm curious... is the problem the reading doesn't change at all, or that there's no stable point of reference?
If its the former, I'm stumped. If its the latter, then maybe you can always make sure the robot is pointing north when you turn it on, then let it
use that as a point of reference.
Edit: I'm SO sorry for bringing back a dead thread, I got here through links and didn't notice the date of the last post... Sorry guys [noparse]:([/noparse]
I Think a change in building materials would be the easiest way out. The use of Aluminum ,Brass, and stainless steels will help alot. If tinsel strength is important Stainless steel is the way to go, But it needs to be 308 or higher in grade. were 316 SS is most desirable for this project. 316 SS has no "FE" or Iron in it. it is nonferrous and is very strong. But it is also expensive.
I have used mid steel to shield motors and transformers from radiating a magnetic field.
Sometimes the Mild steel will have a slite magnetism to it. this can be removed with a heat treatment.(The use of a torch, and red heat)
I'm just looking at the project from a deferent angle!; And using techniques that have worked for Me in the Past, like "ZERO SPEED" sensors, speed detection with hall effect senors, Generator shaft VIB. inside a Highly fluctuating magnetic field,3Ph on some large Mega watt Generators.
I've seen these new phones do a lot with a compass and some other sensors, I would look in to what Their doing for ferrous noise
_______________$WMc%___________
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Truth is out there············································ BoogerWoods, FL. USA
Your problem is not a shielding problem -- it's clearly a stray-field problem.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
Unfortunately, I fried my compass last night when I breadboarded it to work on some code. A big puff of smoke, and then silence (maybe a few ***). Anyway, I found that I had hooked the power up backwards. [noparse]:([/noparse]
It's almost embarrassing how long this project has been going on. I though the thread was safely buried away, but I guess not quite. Well, college is a killer of the hobby dream. I try to sneak in projects when I can, but it gets busy. Oh well...
316 Stainless Steel is (typically) about 38% iron.· It is an austenitic stainless steel and is therefore essentially nonmagnetic, unlike martensitic stainless steels which are·magnetic.· Most stainless steels that you will encounter are austenitic, however.
Heating to the Curie temperature (which varies with various steels) will make steel nonmagnetic; but when it cools it becomes magnetic again, often acquiring weak permanent magnetism from the Earth's magnetic field.· That's the usual source of the magnetic field in welded steel ships, by the way -- the hull's bulk magnetic field depends on the direction they were pointed on the building ways.
Even if you completely degauss a steel object, it can acquire a permanent field again merely by being shocked, say with a hammer, when immersed in a magnetic field such as that of our planet.·
Placing the motors (or other magnets) directly beneath the compass will not eliminate their effect.· Don't believe it?· Just visualize the magnetic field surrounding the magnet or motor -- the lines of force go around in circles, right through the compass.· Still don't believe it?· Put a little magnet down there and twist it around.· The compass will move around too.
Soft-iron masses beside the compass on the binnacle are often necessary on steel ships.· The determination of how large they should be, and where placed, is very complex, a job for specialists.· Their adjustment is extremely critical, also a job for experts.
But how much accuracy do you need?· Surely a toy robot will be OK if it's within, say, ten degrees!· You want it to be able to go north without wandering south.· You're not trying to get to Hawaii from Marcus Island.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
while You are correct about the heat treat method I recommended for iron materials, It is seldom that the Magnetic fields return when cooled down.But some times they do.
Remember this is on a hobby level
Degaussing would be an ideal way to deal with ferro noise, But it is not easily implemented
The 38% FE or iron that you speak of is for 304 SS and not 316 SS
One of My mentors worked for a company called Stellite. This company invented Stainless Steel.
This company was bought by Union Carbide a long time ago.
I haven't herd from You in awhile Mr.Hayes, I'm Glad Your Back
_____________$WMc%____
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Truth is out there············································ BoogerWoods, FL. USA
Post Edited ($WMc%) : 2/1/2009 8:48:10 PM GMT
http://www.sandmeyersteel.com/316-316L.html#2
or any of several other places.· When listing the ingredients in steels, usually the iron is not listed.· Perhaps that's the source of your confusion.· It is assumed that, after you add up the listed percentages of other elements, all the rest is iron.· This is not always explained, for it is assumed anyone would know it -- but it's explicit in the link above, and in many other places.
Steel always implies iron as a basis, with other elements alloyed.· Anything that lacks iron isn't steel, by definition.
Degaussing is actually very easy.· I use an old·Radio Shack tape eraser that looks kind of like a clothing iron, and it does very well.· Takes only a minute or two, including digging out the tool and putting it away.
Cheers!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
It is the carbon content that makes it STEEL, not the IRON.
Take a look at www.ASTM.org This is what the NRC uses for material standards. (Nuclear Regulatory Commission)
The Tape Degausser is a great Idea!
_________________$WMc%_________
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Truth is out there············································ BoogerWoods, FL. USA
Post Edited ($WMc%) : 2/1/2009 10:37:05 PM GMT
However, if you choose to continue to believe that 316 stainless contains no iron, I suppose it's a harmless delusion. Enjoy.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
I'm attempting to use the HM55B with the Propeller Control Board mounted on a QuadRover and I'm getting an enormous amount of magnetic interference from the engine as its drive shaft rotates. Running the engine doesn't seem to help the situation, I had thought that with the magnetic field constantly in flux the compass might be able to read the background field though the engine's field. I can't imagine that Parallax didn't encounter this problem when designing the prop control board, there has to be a fix. I'm not sure where to go from here and neither is our QuadRover.
Any ideas?
Sean KD0EHH
A magnet (or motor, or magnetized battery case) placed directly beneath the compass, or directly above it, or anywhere else near enough,·can affect it strongly.· If you don't believe this, experiment with a compass and a small magnet.· You will find its effect extremely strong, even when the magnet is directly above or below.·
Radar, sonar, etc. do not affect a compass at all, although the cables carrying power to them might affect it.· Degaussing is necessary because of the unavoidable magnetic field of the steel hull.· A steel object (such as a ship) under construction will acquire and retain a magnetic field induced by the Earth's magnetic field.· Its direction depends on where the hull was built, and on the orientation of the building ways (or graving dock, if it's built in one).
Degaussing coils are used almost exclusively on warships.· Their job is to buck the ship's magnetic field so that the resultant field near the ship (especially below it and, in the case of a submarine, above it) is cancelled as nearly as possible.· This is not done for the sake of the compass, which can be adequately compensated without degaussing coils.· The sole reason for degaussing coils on a ship is to prevent activation of magnetic detectors in mines and other ordnance (and in the case of submarines, magnetic detection by ASW vessels and aircraft).
Unlike degaussing coils we might use to demagnetize small objects (such as tape heads or screwdrivers), the degaussing coils on a ship do not demagnetize the hull.· Instead, they are constantly energized with DC in order·to cancel (imperfectly) the external field of the hull.· This is expensive and seldom worthwhile except on warships.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
The problem is not correctly identified.
The problem is not that the compass is unshielded.· Shielding the compass will prevent it from working.
The problem is not that the motors and other magnetic field producers are unshielded.· Shielding them might solve the real problem, but may not be the best way to solve the real problem, which is --
The problem is that the magnetic fields from the motors, relays, etc. are getting into the compass.· That may not seem like much of an insight, but I think no one has seen it before.
Now then:· without shielding the compass, how do we keep these fields from reaching it?
(1)· We could shield each motor, each relay, etc., to confine its field· This would take a lot of shielding.· Mumetal would work, or soft iron (not steel) sheet.· Either adds expense, complication,·and weight.
(2)· We could, instead, enclose all that stuff in a single shield.· This would be even more expensive, make the thing hard to work on, and add even more weight.
(3)· We could seek to control those fields by steering them away from the compass.· This, I think,·is the correct tack.
Some time ago, I designed and built a telephone-call screener that controls who can ring my telephones.· This wonderful blessing works by using reed relays to connect telephone instruments to the Telco line when the call is welcome, and disconnect them when it is not.· The PC board that does this has five reed relays, side by side.· See accompanying photo (dusty, isn't it?· It's been sitting out in open air).
When first tested, the relays interfered with each other.· Every relay worked when no other relay was actuated, but when two or more were actuated the external field from one relay interfered with the function of any next-adjacent relay.· Oops.
The solution was to create·low-reluctance paths for the magnetic fields.· I did this by cementing a piece of soft iron (snipped from a PC expansion slot cover) over the whole row of relays.· The magnetic field from each relay follows a path through this piece of iron and doesn't go through the much higher-reluctance path that takes it through a lot of air space and through the adjacent relay.· It works.· If you look closely at the photo, you can see this little piece of soft iron in the middle of the board.· Under it are five reed relays that no longer interact.· Simple and cheap, and no redesign needed.
You can do the same thing with your motors and relays.· It takes only a narrow strip of soft iron around the waist of each motor to steer its external field.· For most relays, the strip of iron should be around each relay end-to-end.· The idea is to go from pole to pole of the coils that create the fields.· The effect is to short out the external magnetic field so that it doesn't desolate the entire countryside.· That countryside includes your compass!· It will keep the field away from your compass, and you don't have to shield the whole device.· A narrow strip, similar in cross-section to a popsicle stick, ought to work well -- subject to experiment.
Incidentally, it may also make the relays and motors work a little better.
No guarantees -- this is something to try, not anything I'd bet the farm on -- but it's cheap to try and it is very likely to work well enough to enable you to compensate that compass normally and make your day.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
Post Edited (Carl Hayes) : 4/1/2009 7:07:21 AM GMT