analog output question

[Jim]

I have an Omega OS102 ir thermometer that I am using in a project to record
temps continuously. My data logger has 0-5 volt inputs at 12 bit
resolution.

The Omega unit has an analog output of 1 mv per degree C. I am doing all my
recording in the 5-37 degree C range, and it would be very useful if I was
recording 10mv per degree instead of 1mv.

Is there a way to do this conversion with a simple op amp? What's the best
approach to this?

Thanks

 

[[email protected]]

I have an Omega OS102 ir thermometer that I am using in a project to record
temps continuously.  My data logger has 0-5 volt inputs at 12 bit
resolution.

The Omega unit has an analog output of 1 mv per degree C.  I am doing all my
recording in the 5-37 degree C range, and it would be very useful if I was
recording 10mv per degree instead of 1mv.

Is there a way to do this conversion with a simple op amp?  What's the best
approach to this?

Your signal ranges from 5mV to 37mV.

A simple follower-with-gain curcuit will do what you want.

http://www.physics.brocku.ca/Courses/3P92/lab-manual/392/node16.html

The National Semiconductor LF353 is probably not the amplifier you'd
want to use

http://www.national.com/ds/LF/LF353.pdf

since the input offset voltage is typically around 5mV with a worst
case of 13mV over temperature, and it would require positive and
negative voltage supplies in your application.

I'd be more inclined to use the LT1006

http://www.linear.com/pc/downloadDocument.do?navId=H0,C1,C1154,C1009,C1021,P1211,D1414

which is a single supply op amp with a 50uV input offest
specification.

In principle you could probably use a gain of 100 in your application,
but you might need to do some heavy low-pass filtering if you did want
to run with that kind of gain.

You can build a Sallen and Keyes low pass filter around a single
follower with gain, but few text-books analyse the circuit operation
at high gain.

http://electronicdesign.com/Articles/Index.cfm?ArticleID=7486&pg=2

The two amplifier version of the circuit - as discussed in Williams
and Taylor's "Electronic Filter Design Handbook"

http://www.edacafe.com/books/McGraw_Hill/Electronic_Filter_Design_Handbook/1715-ElecFilter_FM.pdf

- allows you to choose more or less arbitrary values for the two
capacitors involved, which makes the design a lot more practicable. E-
mail me - my e-mail address is real - if you want more detail. The
dual version of the LT1006 - the LT1013 - would work in such a
circuit.

 

[Winfield]

Your signal ranges from 5mV to 37mV.

A simple follower-with-gain curcuit will do what you want.
http://www.physics.brocku.ca/Courses/3P92/lab-manual/392/node16.html

The National Semiconductor LF353 is probably not the amplifier you'd
want to use http://www.national.com/ds/LF/LF353.pdf

since the input offset voltage is typically around 5mV with a worst
case of 13mV over temperature, and it would require positive and
negative voltage supplies in your application.

I'd be more inclined to use the LT1006
http://www.linear.com/pc/downloadDocument.do?navId=H0,C1,C1154,C1009,...
which is a single supply op amp with a 50uV input offest
specification.

In principle you could probably use a gain of 100 in your application,
but you might need to do some heavy low-pass filtering if you did want
to run with that kind of gain.

You can build a Sallen and Keyes low pass filter around a single
follower with gain, but few text-books analyse the circuit operation
at high gain.

http://electronicdesign.com/Articles/Index.cfm?ArticleID=7486&pg=2

The two amplifier version of the circuit - as discussed in Williams
and Taylor's "Electronic Filter Design Handbook"

http://www.edacafe.com/books/McGraw_Hill/Electronic_Filter_Design_Han...

- allows you to choose more or less arbitrary values for the two
capacitors involved, which makes the design a lot more practicable. E-
mail me - my e-mail address is real - if you want more detail. The
dual version of the LT1006 - the LT1013 - would work in such a
circuit.

Hey, Bill, time's ripe for an ASCII schematic!

 

[Fred Bloggs]

Your signal ranges from 5mV to 37mV.

A simple follower-with-gain curcuit will do what you want.

http://www.physics.brocku.ca/Courses/3P92/lab-manual/392/node16.html

The National Semiconductor LF353 is probably not the amplifier you'd
want to use

http://www.national.com/ds/LF/LF353.pdf

since the input offset voltage is typically around 5mV with a worst
case of 13mV over temperature, and it would require positive and
negative voltage supplies in your application.

I'd be more inclined to use the LT1006

http://www.linear.com/pc/downloadDocument.do?navId=H0,C1,C1154,C1009,C1021,P1211,D1414

which is a single supply op amp with a 50uV input offest
specification.

In principle you could probably use a gain of 100 in your application,
but you might need to do some heavy low-pass filtering if you did want
to run with that kind of gain.

You can build a Sallen and Keyes low pass filter around a single
follower with gain, but few text-books analyse the circuit operation
at high gain.

http://electronicdesign.com/Articles/Index.cfm?ArticleID=7486&pg=2

The two amplifier version of the circuit - as discussed in Williams
and Taylor's "Electronic Filter Design Handbook"

http://www.edacafe.com/books/McGraw_Hill/Electronic_Filter_Design_Handbook/1715-ElecFilter_FM.pdf

- allows you to choose more or less arbitrary values for the two
capacitors involved, which makes the design a lot more practicable. E-
mail me - my e-mail address is real - if you want more detail. The
dual version of the LT1006 - the LT1013 - would work in such a
circuit.

There is also this incredibly precision conditioner for which 12-bits is
a snap:
http://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1154,C1009,C1045,P1678
....and available for buy off their website.

 

[MooseFET]

On Jan 22, 5:01 pm, [email protected] wrote:
[....]

You can build a Sallen and Keyes low pass filter around a single
follower with gain, but few text-books analyse the circuit operation
at high gain.

Add a capacitor across the feedback resistor and you can make the
filter better and the capacitor values easy to get ones.

From memory:

------------------------------------
! !
=== 1000p !
100K ! 100K !
---/\/\----+----/\/\----+-------!+\ !
! ! >------+-----+---Out
0.1 === --!-/ !
! ! !
GND +----/\/\----+ 22K
! !
GND---/\/\---------+-----!!----- 0.1
220

This gets quickly to 10 bits, (IIRC) 50mS and reduces the 180Hz by
(IIRC) 30dB.