Post by thread:[ot]: little gyro question

hello, I was thinking of trying to make an electronic level. And I thought the tiny gyros might be used to do this. I don't have one and I was wondering if someone had a recomendation for an inexpensive one that could do the task? Do these things have a built in zero point that you can check against to tell when something is tilting a degree up or down? Or how do they work exactly? Do you turn them on, and then they align themselves to the vertical? or what? Thanks -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Hi! You don't want to use tiny gyros (which are, AFAIK, all rate gyros) because they only output rate of rotation. So, you actually have to integrate their output to get angle. This means that they do not know the difference between level and a 45 deg angle, they can only tell you how much you rotated from a starting angle. In addition, any tiny offset error in their output gets integrated to a continuous drift (rise or fall in output angle). They are best for situations where you need to determine how something is rotating even when it is not an inertial frame (i.e., if it is accelerating). If you simply want to make a level which will sit still on top of something and tell you whether it is tilted, an accelerometer would be a better choice. Accelerometers can measure the force of gravity (indirectly, by actually measuring the force which is preventing the accelerometer from falling) and so if they are tilted, they will register a somewhat reduced value for gravity. You can even get a 2-axis accelerometer (ADXL202 from http://www.analog.com) and then you can read out not only tilt angle but also direction (is it tilted with the left end up or the right end up?). You could actually do this with only a single axis accelerometer, but you can get a more accurate reading by using both accelerometers and I don't think the 202 is any more expensive than its single axis cousins, and it has a PWM output (instead of pure analog) which makes it easier for a PIC to handle. I think Analog Devices even has an app note on using them as level sensors. Note that this only works when the thing is sitting still(or moving at a constant velocity). It is not accurate if you are jerking it around or otherwise accelerating it, since it actually measures applied force, not gravity itself. Sean At 07:00 PM 10/10/01 -0500, you wrote: {Quote hidden}>hello, > >I was thinking of trying to make an electronic level. And I thought the >tiny gyros might be used to do this. > >I don't have one and I was wondering if someone had a recomendation for >an inexpensive one that could do the task? > > >Do these things have a built in zero point that you can check against to >tell when something is tilting a degree up or down? Or how do they work >exactly? Do you turn them on, and then they align themselves to the >vertical? > >or what? > > > >Thanks > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics ---------------------------------------------------- Sign Up for NetZero Platinum Today Only $9.95 per month! http://my.netzero.net/s/signup?r=platinum&refcd=PT97 -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

I'm not a technical bod but I use a gyroscopic sensor for orientation tracking for virtual environments. The thing is that the gyroscope does not have any external reference, used for tracking it is its opening orientation that provides the reference to which all other measurements are made. If one wanted to make a level it is exactly this kind of reference that one would need. You'd need a level to set up your level!! Having said that it might give you an accurate measure of any deviation off of that reference. Having said that in 3d graphics the maths that you need to do that within a 3d environment is not straight forward. If what I've said is wrong I'd be very interested in hearing the reasons why. all the best, Mark On Wed, 10 Oct 2001 19:00:42 -0500 rad0 <spam_OUTrden25TakeThisOuTMINDSPRING.COM> wrote: {Quote hidden}> hello, > > I was thinking of trying to make an electronic level. And I thought the > tiny gyros might be used to do this. > > I don't have one and I was wondering if someone had a recomendation for > an inexpensive one that could do the task? > > > Do these things have a built in zero point that you can check against to > tell when something is tilting a degree up or down? Or how do they work > exactly? Do you turn them on, and then they align themselves to the > vertical? > > or what? > > > > Thanks > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > > ************************************************ Dr Mark Palmer Arts & Humanities Research Board Fellow School of Art & Design College Road Staffordshire University Stoke-on-Trent ST4 2DE tel: +44 (0)1782 294802 (direct) fax: +44 (0)1782 294873 .....M.W.PalmerKILLspam@spam@staffs.ac.uk ************************************************ -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

The original post was actually concerned with how to make a decent electronic level indicator. If the level indicator is used in the normal way that a bubble level indicator is used (i.e. to measure inclination of a stationary object, such as a table top or brick wall), then you can make a quite simple one using a potentiometer. You start with a flat base. This becomes the part that will rest on the surface whose inclination you want to measure. At a right angle to the base you attach an upright that is several inches tall. Mounted to this upright is a standard linear pot. Something in the range from 1k to 50k would be fine. It is not important what the pot's value is, but you want a pot that has low friction at the shaft. The body of the pot is firmly fixed to the upright. Rotate the pot to the approximate center of rotation. To the pot's shaft attach a stiff metal rod that has a reasonable sized weight attached to the far end of the rod. Connect one side of the pot to ground, the other side to a voltage source, such as +5 v. Output is measured between the ground and the wiper of the pot. Let's assume you are using a PIC to measure the inclination angle. Place assembly on a known flat surface. Push "level" button. PIC reads voltage and stores it as zero reference. Incline unit some predefined calibration angle, such as 30 degrees. Push the "calibrate 30" button. PIC measures voltage, finds difference between level and 30 degree values, divides by 30 and comes up with the delta value per degree of inclination. Now, as the device is placed at various degrees of inclination, the PIC reads the current voltage, subtracts it from the reference voltage, determines +/- , if it is negative, then a minus sign is displayed and the value is converted to its positive equivalent. The current delta value is divided by the delta value per degree and then this value is displayed on the readout in degrees. Instead of a PIC you could use a couple of op amps and a few trim pots to adjust zero point and circuit gain so that the output voltage would be scaled to something like .1 volt per degree. You could drive an analog or digital voltmeter direct with this arrangement. Fr. Tom McGahee -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

rad0, I think that you are mixing two things here. I think that the little gyros you mention are actually Tokin and Murata pizoceramic rate gyros. These only indicate change of radial speed (angular acceleration) and have no zero or verticality indication. For an inclinometer try an inclinometer (or clinometer) sensor. These are mass produced for the automotive industry (alarms). You can also use a monolithic 2-axis accelerometer (ADXL...) for the same thing. You can also diy using a pendulum or half wheel in various combinations. Peter -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Hi Peter, I agree with your conclusions, but just one correction: the piezoelectric rate gyros indicate angular speed directly, not angular acceleration. In other words, you only have to integrate once to get position. Because of the need for integration and a starting reference, they are not suitable for a level sensing application, but they aren't quite as bad as, say, using accelerometers to get position, where you have to integrate twice. Sean At 01:12 AM 10/12/01 +0200, you wrote: {Quote hidden}>rad0, I think that you are mixing two things here. I think that the little >gyros you mention are actually Tokin and Murata pizoceramic rate gyros. >These only indicate change of radial speed (angular acceleration) and have >no zero or verticality indication. > >For an inclinometer try an inclinometer (or clinometer) sensor. These are >mass produced for the automotive industry (alarms). You can also use a >monolithic 2-axis accelerometer (ADXL...) for the same thing. You can also >diy using a pendulum or half wheel in various combinations. > >Peter > >-- >http://www.piclist.com hint: The PICList is archived three different >ways. See http://www.piclist.com/#archives for details. ---------------------------------------------------- Sign Up for NetZero Platinum Today Only $9.95 per month! http://my.netzero.net/s/signup?r=platinum&refcd=PT97 -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

{Quote hidden}> -----Original Message----- > From: pic microcontroller discussion list > [.....PICLISTKILLspam.....MITVMA.MIT.EDU]On Behalf Of Sean H. Breheny > Sent: Thursday, October 11, 2001 6:11 AM > To: EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU > Subject: Re: [ot]: little gyro question > > > Hi! > > You don't want to use tiny gyros (which are, AFAIK, all rate > gyros) because > they only output rate of rotation. So, you actually have to > integrate their > output to get angle. This means that they do not know the difference > between level and a 45 deg angle, they can only tell you how much you > rotated from a starting angle. In addition, any tiny offset error in their > output gets integrated to a continuous drift (rise or fall in output > angle). They are best for situations where you need to determine how > something is rotating even when it is not an inertial frame > (i.e., if it is > accelerating). > > If you simply want to make a level which will sit still on top of > something > and tell you whether it is tilted, an accelerometer would be a better > choice. Accelerometers can measure the force of gravity (indirectly, by > actually measuring the force which is preventing the accelerometer from > falling) and so if they are tilted, they will register a somewhat reduced > value for gravity. You can even get a 2-axis accelerometer (ADXL202 from > http://www.analog.com) and then you can read out not only tilt angle but also > direction (is it tilted with the left end up or the right end up?). You > could actually do this with only a single axis accelerometer, but you can > get a more accurate reading by using both accelerometers and I don't think > the 202 is any more expensive than its single axis cousins, and it has a > PWM output (instead of pure analog) which makes it easier for a PIC to > handle. I think Analog Devices even has an app note on using them as level > sensors. Note that this only works when the thing is sitting still(or > moving at a constant velocity). It is not accurate if you are jerking it > around or otherwise accelerating it, since it actually measures applied > force, not gravity itself. Yeah, I had seen a circuit in Elektor Electronics a couple of years back (1998 ?) Which did this using a single Analog Accelerometer and two led bars. It was titled "Electronic Spirit Level". It did'nt use any Micros. Perhaps you could find it on the elektor website. Jeethu Rao -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

A non-electric analog method: Find a string and a weight, something pointy is good. Hold the loose end at the top of the post, and line it up with the weighted string. Works on most planets. with a decent gravitational field. alice -- _______________________________________________ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup Have you downloaded the latest calling software from Net2Phone? Click here to get it now! http://www.net2phone.com/cgi-bin/adforward.cgi?p_key=NH211JK&url=http://commcenter.net2phone.com/ -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> A non-electric analog method: > > Find a string and a weight, something pointy is good. > Hold the loose end at the top of the post, and line it > up with the weighted string. Works on most planets. > with a decent gravitational field. > > alice Hello Alice, Your answer is correct, if considering the planet a perfect spherical shape. Are the Earth planet a good spherical shape? The planet (as almost others around) is a little bit "fatty" at the Equator, so, what happens at 19 degrees south or north? the pointy weight will be pointing to the gravitational field center, but the vertical line will not be exactly at 90 degrees relative to the horizontal surface... a little bit off, right? The buildings will not fall down, but the grounds would seems to be descending when walking toward the poles, isn't it? The american expression "up north" seems a little bit wrong. Walking from the North Pole or South Pole towards the Equator, it is always UPWARD the terrain, so, americans should say UP-SOUTH or DOWN-NORTH instead... messing, eh? :) ---------------------------------------------------------------------------- ---- Wagner Lipnharski - Director - KILLspamwagnerKILLspamustr.net UST Research Inc. - Orlando - FL - http://www.ustr.net -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> > A non-electric analog method: > > > > Find a string and a weight, something pointy is > > Hello Alice, > > Your answer is correct, if considering the planet a perfect spherical shape. > Are the Earth planet a good spherical shape? The planet (as almost others > around) is a little bit "fatty" at the Equator, so, what happens at 19 > degrees south or north? the pointy weight will be pointing to the > gravitational field center, but the vertical line will not be exactly at 90 > degrees relative to the horizontal surface... a little bit off, right? The > buildings will not fall down, but the grounds would seems to be descending > when walking toward the poles, isn't it? Well, it's actually even worse than you describe. Any tall mass will deflect the plumb bob. The Brits in the 18th century had a heck of a time measuring the height of Mt Everest, because its mass deflected the plumb bob of the surveyor, and the vertical angle to the peak was always off. That, and the mountain has a big 'root' beneath it, that adds even more mass off at an angle. In fact, that's the way they proved that most mountains Have roots by measuring local deflections in the gravitational field. I suppose we even need to take the thickness of the sring into account, and use monofilament instead of hairy old twine for more accuracy. welcome back alice -- _______________________________________________ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup Have you downloaded the latest calling software from Net2Phone? Click here to get it now! http://www.net2phone.com/cgi-bin/link.cgi?157 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body