Post by thread:[EE]: Navigation sensors

I'm working on a sensor pack which will measure azimuth, inclination and roll. Currently I'm involved in literature review and parts specification with a leaning towards Honeywell's magneto-resistive magnetometers and an Analog accelerometer. Both Honeywell and Analog provide excellent application notes and support, but the input of any board members would also be appreciated, specifically direction to further online research sources. Thanks, Shawn -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

You can also buy expensive mechanical pendulums ($100 for one) that output a standard gray code (like a rotary encoder) with an index pulse. http://www.usdigital.com/products/t2/ http://www.usdigital.com/products/a2t/ Then there are capacitave sensors. http://www.maxim-ic.com/appnotes.cfm/appnote_number/198 For my most recent design I'm using an ADXL213, which should be good to 1/10 of a degree through its range, but with very slow response time (1Hz bandwidth). Once you start needing faster updates the resolution drops quickly. There are companies that have done all the hard work (temp compensation, sensor testing, etc) and have drop in modules to do what you want for $90. -Adam Shawn Mulligan wrote: {Quote hidden}>I'm working on a sensor pack which will measure azimuth, inclination and >roll. Currently I'm involved in literature review and parts specification >with a leaning towards Honeywell's magneto-resistive magnetometers and an >Analog accelerometer. Both Honeywell and Analog provide excellent >application notes and support, but the input of any board members would also >be appreciated, specifically direction to further online research sources. > >Thanks, Shawn > >-- >http://www.piclist.com hint: To leave the PICList >.....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu > > >. > > > -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu

Originally posted by Adam: For my most recent design I'm using an ADXL213, which should be good to 1/10 of a degree through its range, but with very slow response time (1Hz bandwidth). Once you start needing faster updates the resolution drops quickly. Thanks for the input, Adam. I've sampled the Maxim part and I'm also using the ADXL213 in a current design. It's an accurate, easy to incorporate little accelerometer. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

Memsic (http://www.memsic.com) makes some single and dual axis accelerometers. My coworker has their 2-axis demo board ($25) sitting on his desk - pretty cool. I think one limitation of accelerometers in a 3-D environment is that centrifugal force can play tricks on them. For example, in a coordinated turn in an airplane, you're pulling 1G straight down the yaw axis - your roll-axis sensor wouldn't register the roll I don' think. I want to say that most manufacturers of 3-axis navigation devices combine gyro inclinometers, accelerometers and solid-state compasses with an alogorithm whose name escapes me - to compute very accurate pitch/yaw/roll angles. Crossbow (http://www.xbow.com) is one. -matt redmond {Quote hidden}> I'm working on a sensor pack which will measure azimuth, inclination and > roll. Currently I'm involved in literature review and parts specification > with a leaning towards Honeywell's magneto-resistive magnetometers and an > Analog accelerometer. Both Honeywell and Analog provide excellent > application notes and support, but the input of any board members would also > be appreciated, specifically direction to further online research sources. > > Thanks, Shawn > > -- > http://www.piclist.com hint: To leave the PICList > EraseMEpiclist-unsubscribe-requestspam_OUTTakeThisOuTmitvma.mit.edu > -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu

Matt Redmond writes: Memsic (http://www.memsic.com) makes some single and dual axis accelerometers. My coworker has their 2-axis demo board ($25) sitting on his desk - pretty cool. I think one limitation of accelerometers in a 3-D environment is that centrifugal force can play tricks on them. For example, in a coordinated turn in an airplane, you're pulling 1G straight down the yaw axis - your roll-axis sensor wouldn't register the roll I don' think. Thanks. Yes, the Memsic evaluation boards are very well priced and cool. I'm using these devices in a directional drilling application, and surveys are only taken when the sensor is stationary, so centrifugal forces won't be an issue, but I'll follow that train for future reference. -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-requestKILLspammitvma.mit.edu

>I think one limitation of accelerometers in a 3-D environment is that >centrifugal force can play tricks on them. For example, in a coordinated >turn in an airplane, you're pulling 1G straight down the yaw axis - your >roll-axis sensor wouldn't register the roll I don' think. These exists no such thing as a coordinated 1g turn. The sum of the acceleration vectors for any kind of travel excepting straight line is significantly different from 1g. Peter -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

----- Original Message ----- From: "Peter L. Peres" <KILLspamplpKILLspamACTCOM.CO.IL> To: <RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU> Sent: Tuesday, August 24, 2004 9:35 PM Subject: Re: [EE]: Navigation sensors {Quote hidden}> >I think one limitation of accelerometers in a 3-D environment is that > >centrifugal force can play tricks on them. For example, in a coordinated > >turn in an airplane, you're pulling 1G straight down the yaw axis - your > >roll-axis sensor wouldn't register the roll I don' think. > > These exists no such thing as a coordinated 1g turn. The sum of the > acceleration vectors for any kind of travel excepting straight line is > significantly different from 1g. > > Peter > No, a descending coordinated turn in an aircraft can feel just like 1g straight and level flight. I realize its possible to tell the difference over time using precession effects, but instantaneously, the acceleration vector is simply down. Thats why pilots use gyros for flying in clouds. Regards, Bob Monsen -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Tue, 24 Aug 2004 17:20:32 -0700, Robert Monsen wrote: > ----- Original Message ----- > From: "Peter L. Peres" <spamBeGoneplpspamBeGoneACTCOM.CO.IL> >...< > > These exists no such thing as a coordinated 1g turn. The sum of the > > acceleration vectors for any kind of travel excepting straight line is > > significantly different from 1g. > > No, a descending coordinated turn in an aircraft can feel just like 1g > straight and level flight. I realize its possible to tell the difference > over time using precession effects, but instantaneously, the acceleration > vector is simply down. > > Thats why pilots use gyros for flying in clouds. Indeed, and why the life-expectancy of an inexperienced pilot entering IMC unexpectedly is very few minutes! Whoever coined the phrase "flying by the seat of your pants" was utterly wrong - if you rely on the "feel" of things when you can't see the horizon or the ground, you're in real trouble. Happy Landings... Howard Winter St.Albans, England -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>> These exists no such thing as a coordinated 1g turn. The sum of the >> acceleration vectors for any kind of travel excepting straight line is >> significantly different from 1g. >> >> Peter >> > >No, a descending coordinated turn in an aircraft can feel just like 1g >straight and level flight. I realize its possible to tell the difference >over time using precession effects, but instantaneously, the acceleration >vector is simply down. You are right. Peter -- 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

Yes, that is correct - I brain-farted on that one. Coordination just keeps the centrifugal force / gravity vectors such that you are being pulled straight down into your seat rather than slipping sideways in it. -matt {Quote hidden}> > From: "Peter L. Peres" <TakeThisOuTplpEraseMEspam_OUTACTCOM.CO.IL> > Date: 2004/08/25 Wed AM 04:35:33 GMT > To: RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU > Subject: Re: [EE]: Navigation sensors > > >I think one limitation of accelerometers in a 3-D environment is that > >centrifugal force can play tricks on them. For example, in a coordinated > >turn in an airplane, you're pulling 1G straight down the yaw axis - your > >roll-axis sensor wouldn't register the roll I don' think. > > These exists no such thing as a coordinated 1g turn. The sum of the > acceleration vectors for any kind of travel excepting straight line is > significantly different from 1g. > > Peter > > -- > http://www.piclist.com hint: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > -- 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

----- Original Message ----- From: "Matt Redmond" <mdredmondEraseME.....CHARTER.NET> To: <EraseMEPICLISTMITVMA.MIT.EDU> Sent: Wednesday, August 25, 2004 2:04 PM Subject: Re: [EE]: Navigation sensors > Yes, that is correct - I brain-farted on that one. Coordination just keeps the centrifugal force / gravity vectors such that you are being pulled straight down into your seat rather than slipping sideways in it. {Quote hidden}> > > From: "Peter L. Peres" <RemoveMEplpEraseMEEraseMEACTCOM.CO.IL> > > Date: 2004/08/25 Wed AM 04:35:33 GMT > > To: RemoveMEPICLISTspam_OUTKILLspamMITVMA.MIT.EDU > > Subject: Re: [EE]: Navigation sensors > > > > > > These exists no such thing as a coordinated 1g turn. The sum of the > > acceleration vectors for any kind of travel excepting straight line is > > significantly different from 1g. > > Both are correct. What will be seen in typical cases is something slightly over 1g for a non-radical coordinated turn. Someone made a comment about a coordinated descending turn as being able to produce a 1g turn. This is only true if the downward speed increases constantly (i.e. you accelerate downward). Of course you can't continue to do this for very long until something very bad happens ;() This is all nit picking though. Regards, Gordon Williams -- 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

----- Original Message ----- From: "Gordon Williams" <RemoveMEg_willTakeThisOuTspamCYBERUS.CA> To: <EraseMEPICLISTspamspamBeGoneMITVMA.MIT.EDU> Sent: Wednesday, August 25, 2004 11:53 AM Subject: Re: [EE]: Navigation sensors > ----- Original Message ----- > From: "Matt Redmond" <RemoveMEmdredmondKILLspamCHARTER.NET> > To: <PICLISTSTOPspamspam_OUTMITVMA.MIT.EDU> > Sent: Wednesday, August 25, 2004 2:04 PM > Subject: Re: [EE]: Navigation sensors > > > > Yes, that is correct - I brain-farted on that one. Coordination just > keeps the centrifugal force / gravity vectors such that you are being pulled {Quote hidden}> straight down into your seat rather than slipping sideways in it. > > > > > From: "Peter L. Peres" <spamBeGoneplpSTOPspamEraseMEACTCOM.CO.IL> > > > Date: 2004/08/25 Wed AM 04:35:33 GMT > > > To: KILLspamPICLISTspamBeGoneMITVMA.MIT.EDU > > > Subject: Re: [EE]: Navigation sensors > > > > > > > > > These exists no such thing as a coordinated 1g turn. The sum of the > > > acceleration vectors for any kind of travel excepting straight line is > > > significantly different from 1g. > > > > > Both are correct. What will be seen in typical cases is something slightly > over 1g for a non-radical coordinated turn. > > Someone made a comment about a coordinated descending turn as being able to > produce a 1g turn. This is only true if the downward speed increases > constantly (i.e. you accelerate downward). Of course you can't continue to > do this for very long until something very bad happens ;() > Pilots call it a graveyard spiral... VFR only pilots who venture into clouds can end up in one of these fairly easily. The end is when they notice the altitude decreasing quickly, and attempt to pull the nose up with the elevator. Unfortunately, the lift vector has to pull 5.8 Gs in order to keep level, because of the unnoticed 80 degree bank... this will occasionally pull off the tail, causing the pilot to have a bad day. {Quote hidden}> This is all nit picking though. > > Regards, > > Gordon Williams > > -- > 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 > -- 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