Multiplex Signal Conditioner

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
Ian Robinson
Posts: 3
Joined: Wed Apr 09, 2003 1:01 am
Location: UK
Contact:

Multiplex Signal Conditioner

Post by Ian Robinson » Wed Apr 09, 2003 11:09 pm

I have searched everywhere for explicit circuits for a multi-channel, signal conditioner for use with strain gauges, without success. It needs to be able to interface with a computer.<p>The Maxim website has details of various IC's that appear suitable [MAX 1452, MAX 408] but though extremley detailed information is given, some aspects are outside my experience/abilities, making me doubt that I could complete a project based around them.<p>
Can anyone suggest where I might find alternative circuits or point me in the direction of a textbook that might help? <p>Thankyou
Ian Robinson

Dimbulb
Posts: 324
Joined: Sun Oct 13, 2002 1:01 am

Re: Multiplex Signal Conditioner

Post by Dimbulb » Thu Apr 10, 2003 12:58 am

A max1457 a battery and ground plane intensive, shielded and organised for parts layout together with it's A/D counterpart.<p>a clean reference voltage is stepped to different voltages constant as current can be. Each output being affected by it's strain gauge resistance where tiny flows of current give evidence of forces against squeezing the geometry of the mico electrical pathways of current. Reducing coefficients of circuit error would require use of precision components specific by type and yet scattered in an ocean of semiconductor descriptions and lengthy discourses only to be purchased by the giants of production.

russlk
Posts: 563
Joined: Thu Dec 06, 2001 1:01 am
Location: New Hampshire
Contact:

Re: Multiplex Signal Conditioner

Post by russlk » Thu Apr 10, 2003 3:51 pm

A digital voltmeter with PC interface is the simplist way to interface. You will need a preamp to raise the signal to 200mv and a multichannel analog switch to select the strain guage.

Ian Robinson
Posts: 3
Joined: Wed Apr 09, 2003 1:01 am
Location: UK
Contact:

Re: Multiplex Signal Conditioner

Post by Ian Robinson » Tue Apr 15, 2003 11:33 pm

Thanks for the replies.<p>Ian
Ian Robinson

Will
Posts: 310
Joined: Tue Feb 11, 2003 1:01 am
Location: Katy Texas
Contact:

Re: Multiplex Signal Conditioner

Post by Will » Wed Apr 16, 2003 12:21 pm

Ian,
Because of the relative insensitivity of strain gauges they are usually installed/configured in what are called `Wheatstone Bridges ' (After the inventor, your UK antecedent Sir William Wheatstone) You need to look these up ina text book because I can't draw circuit diagrams on this medium. The purpose of a Wheatstone Bridge is generally to measure resistance by comparing a variable resistance (the strain gauge) to other fixed resistances by way of voltage differences produced. If you connected together four resistances in such a way as to form a square then applied a voltage to two diagonally opposite corners of the square, then, if you apply a voltmeter to the other two diagonally opposite corners of the square the voltage thus measured would change according to the changes in resistance of the one in which you were interested (The strain gauge) Before you can configure such a circuit you have to be aware of the standing resistance of the strain gauge and by how much it will change under the circumstances you are studying. Since any current drawn by the voltmeter or measuring device will affect the quality of the reading then high inpout impedance measuring devices are necessary so you would probably connect a one to one buffer op amp directly to the bridge corners and then follow that up with an op amp scaled to produce the desired output. Feed multiples of such circuits as desired into a multiplexed Analog to Digital converter and lead the digital output into your PC. I think Linear ? Devices makes an a/d converter with a serial output that you could probably connect directly into the RS 232 port of PC. Hope this helps Will
BB

Will
Posts: 310
Joined: Tue Feb 11, 2003 1:01 am
Location: Katy Texas
Contact:

Re: Multiplex Signal Conditioner

Post by Will » Wed Apr 16, 2003 12:23 pm

Ian,
Because of the relative insensitivity of strain gauges they are usually installed/configured in what are called `Wheatstone Bridges ' (After the inventor, your UK antecedent Sir William Wheatstone) You need to look these up ina text book because I can't draw circuit diagrams on this medium. The purpose of a Wheatstone Bridge is generally to measure resistance by comparing a variable resistance (the strain gauge) to other fixed resistances by way of voltage differences produced. If you connected together four resistances in such a way as to form a square then applied a voltage to two diagonally opposite corners of the square, then, if you apply a voltmeter to the other two diagonally opposite corners of the square the voltage thus measured would change according to the changes in resistance of the one in which you were interested (The strain gauge) Before you can configure such a circuit you have to be aware of the standing resistance of the strain gauge and by how much it will change under the circumstances you are studying. Since any current drawn by the voltmeter or measuring device will affect the quality of the reading then high inpout impedance measuring devices are necessary so you would probably connect a one to one buffer op amp directly to the bridge corners and then follow that up with an op amp scaled to produce the desired output. Feed multiples of such circuits as desired into a multiplexed Analog to Digital converter and lead the digital output into your PC. I think Linear ? Devices makes an a/d converter with a serial output that you could probably connect directly into the RS 232 port of PC. Hope this helps Will
BB

Chris Foley
Posts: 146
Joined: Mon Jan 13, 2003 1:01 am
Location: Chicago IL
Contact:

Re: Multiplex Signal Conditioner

Post by Chris Foley » Wed Apr 16, 2003 8:05 pm

Hi, Ian. Will's right. Strain gauges can be a bear to interface. You have a number of separate problems. Strain/resistance is kind of non-linear. You need a tempco compensation for the metal you're placing it on You have to worry about self-heating changing the reading. Microvolts count big time, even if you apply the maximum voltage possible across the strain gauge. Not to mention that, if you don't affix them to the metal properly, they won't work at all. Oh, yes -- overloads put them out of calibration. All of this tends to discourage a quick hack here.
Having said that, there are a number of ways to go with your project. A "store bought" solution usually has a laser-trimmed version of Will's Wheatstone bridge built in, which linearizes the strain gauge and provides temperature compensation. There is usually a built-in precision voltage source to provide bridge excitation, and a buffer amplifier to give a mV output. Easy, straightforward, and fairly precise. If this is an educational project, you will probably have to do these things yourself, and learn through reading and trial/error. A good first place to look for information is the Omega Pressure catalog, which is available through Omega.com, and there is usually a somewhat dated edition in most college libraries. Their reference section in the back of the book has a good intro to the strain gauge concept, as well as a summary of most of the basic physical equations you will need to apply the strain gauge in your application. If you purchase some of their stuff, you can also call their applications hotline, and actually talk to an engineer who might be able to help. If you're planning on a home-brew, Analog Devices is the first source when looking for information on the electronics of interfacing transducers. Go to
http://www.analog.com/
and punch in "strain gage" on their site search. There are plenty of good ideas there. AD also published a book around 1980 called "Transducer Interfacing Handbook". If you can find it, there's a whole section on strain gauges.<p>I'm not too enthusiastic about the multiplexing issue for a home brew job, unless you just want to know if the elephant in the pink tutu is dancing on the beam. Measuring micro-ohms through a multiplexer is something that, to get accuracy/precision, you need to be very aware of a lot of very small details. Analog Devices references one article from Electronic Design which, though it refers primarily to analog multiplexing of RTDs, has a lot of applicability. It might be best, if this is a real project (meaning you've got a budget, and can make some real decisions about things) to go with the "store-bought" strain gauge setups, and then take Russ' idea and multiplex the voltage outputs to a 5-1/2 digit multimeter. If you want to roll your own, do the signal conditioning at the strain gauge, and then multiplex the voltage signal as above. As always when measuring the vanishingly small, beware of ground loops. Good luck (you'll need it, but luck is the residue of hard work). Happy hunting.

Ian Robinson
Posts: 3
Joined: Wed Apr 09, 2003 1:01 am
Location: UK
Contact:

Re: Multiplex Signal Conditioner

Post by Ian Robinson » Thu Apr 24, 2003 11:16 pm

THANKS Will & Chris for your posts which, only now have I had the opportunity to read. <p>Since my original post, I have been looking at the Anderson Loop as a possible solution to some of the problems you mention but of the circuit I can find, there are missing details which I am slowly working my way through. Whatever else, accurate strain measurement isn't straightforward.<p>Regards
Ian Robinson

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 15 guests