BS2 Project and ADC''s
Archiver
Posts: 46,084
Thanks for the bone guys. While looking over the datasheets for the LTC1051
show using a LTC1043 Dual Instrumentation Switching Capacitor building
block. While looking very simple to use, some schematics show it while
others do not. Is this thing a must for doing my project?
I'm going to be ordering some parts tonight and play around with it all next
week when it comes in. I'm sure it ought to be fun and I should learn
something out of it all. The 187 looks easy to interface to and the code
examples look pretty easy to follow as well, so I'm sure that won't be much
of a problem.
Again, Thanks for your help.
Richard
Original Message
From: "Tracy Allen" <tracy@e...>
To: <basicstamps@yahoogroups.com>
Sent: Wednesday, October 15, 2003 12:15 AM
Subject: Re: [noparse][[/noparse]basicstamps] BS2 Project and ADC's
> >Okay, let me try a different approach. Sometimes I don't give enough
info
> >and sometimes too much info.
> >I would like to read a sensor with a BS2, I know I will need an ADC,
which
> >needs to be a 12 bit. Can anyone recommend an serial ADC with possibly a
> >4.096 built reference? I've used the 1298 before but I wanted something
a
> >little more precise and since the 1298 uses the Vcc as it's reference I
> >wanted to avoid it (Unless someone has a great method to make it very
> >accurate and repeatable).
>
> Do you need just one channel? You may need more sooner or later. I
> like the TLC2543, which has 11 input channels.
> <http://www.emesystems.com/OWL2face.htm#TLC2543> It needs an external
> reference. Another series with good Stamp documentation is the
> single channel MAX187 (with 4.096 internal reference) or MAX189
> (external reference) or 8 channel MAX186 (4.096 internal) or MAX188
> (external). One disadvantage of an internal reference is that they
> are usually not very accurate (i.e. +/- 2%) , and you can't adjust
> them with an external potentiometer. You end up having to do
> corrections in software.
>
>
> >
> >The next part does not pertain to a stamp, but does have something to do
> >with my project, if you can't help, at least try and point me in the
right
> >direction.
> >
> >The max my sensor puts out is 400mv, so I figured I would use an
> >instrumentation amp to increase this signal by approx. 10 x with a small
> >adjustment for calibration purposes. This would give me the full range of
my
> >ADC and the best resolution, I think. The amp needs to be a single
> >voltage like + 5vdc and Grd. I was looking at the AD626 and I several
> >LTC1052's but I never used them before. Can someone tell me if I'm on
the
> >right path or if they have another item which is easy to work with and is
> >fairly accurate?
>
> I use the LTC1051 (dual CAZ op amp) a lot when DC accuracy is
> paramount--input offsets of less than 10 microvolts. I think the
> LTC1052 is similar. Don't expect it to swing above 3.5 volts when it
> has a 5 volt supply. Maybe you can power it from the higher
> unregulated voltage. Or limit your amplification to x5 instead of
> x10, and use a 2.5 volt reference on your ADC. Or just accept 11
> bits resolution. I don't know of any rail to rail op-amps that are
> terrifically accurate, but they would certainly work if you want to
> adjust both the input offset and the scale factor with pots. Is the
> Oxygen sensor really that accurate and stable? The op amp circuit
> will be really simple, especially if you use the CAZ op amp. The
> classic non-inverting amplifier.
>
>
>
> >
> >I don't mind learning by hard knocks, but I rather not do it a dozen
times
> >either.
> >Throw me a bone please....
> >Richard
>
>
> Don't worry, you still get the hard knocks. The bone falling may hit
> you on the head. A corollary of Murphy's law.
>
> -- Tracy
>
>
> >
> >
> >
Original Message
> >From: "Richard Skinner" <rwskinner@w...>
> >To: <basicstamps@yahoogroups.com>
> >Sent: Monday, October 13, 2003 1:04 PM
> >Subject: [noparse][[/noparse]basicstamps] Sensor project
> >
> >
> >> I working on a small project which requires me to read a sensor which
> >> outputs from 0 mv to 400mv +/- 40 mv full scale. It's almost linear
but
> >> does have a couple of items that could use some error correction. I
have
> >> several questions regarding the electronics end and some on the math
end.
> >> The sensor is an Oxygen sensor that reads from 0% to 25.5% o2. At 21%
o2
> >it
> >> reads approx. 400mv.
> >>
> >> I don't know much about OP Amps but willing to learn. I figured I
would
> >use
> >> an 8 bit ADC with a 4.096 reference, then use an OP Amp with a Gain of
10.
> >> I thought about using a pot to slightly adjust the gain in order to
> >> calibrate the sensor. The sensor always has zero output when it senses
> >zero
> >> input, so I was only going to adjust the high end. Would this be the
> >proper
> >> procedure?
> >>
> >> Next, the sensor exhibits a Temp Coefficient of 0.2% signal / Degree C
so
> >I
> >> may want to perform a temp. correction for the temp at the sensor. The
> >temp
> >> would range from about Ambient (32F to 100F) to about 150 Degree F when
> >> operating in it's environment. It's calibrated in an Ambient
condition.
> >I
> >> was leaning towards using an Thermistor to measure the temp during
> >> calibration, then correcting the displayed units for the operating
temps.
> >>
> >> Any input would be greatly appreciated.
> >> Richard
> >>
> >>
show using a LTC1043 Dual Instrumentation Switching Capacitor building
block. While looking very simple to use, some schematics show it while
others do not. Is this thing a must for doing my project?
I'm going to be ordering some parts tonight and play around with it all next
week when it comes in. I'm sure it ought to be fun and I should learn
something out of it all. The 187 looks easy to interface to and the code
examples look pretty easy to follow as well, so I'm sure that won't be much
of a problem.
Again, Thanks for your help.
Richard
Original Message
From: "Tracy Allen" <tracy@e...>
To: <basicstamps@yahoogroups.com>
Sent: Wednesday, October 15, 2003 12:15 AM
Subject: Re: [noparse][[/noparse]basicstamps] BS2 Project and ADC's
> >Okay, let me try a different approach. Sometimes I don't give enough
info
> >and sometimes too much info.
> >I would like to read a sensor with a BS2, I know I will need an ADC,
which
> >needs to be a 12 bit. Can anyone recommend an serial ADC with possibly a
> >4.096 built reference? I've used the 1298 before but I wanted something
a
> >little more precise and since the 1298 uses the Vcc as it's reference I
> >wanted to avoid it (Unless someone has a great method to make it very
> >accurate and repeatable).
>
> Do you need just one channel? You may need more sooner or later. I
> like the TLC2543, which has 11 input channels.
> <http://www.emesystems.com/OWL2face.htm#TLC2543> It needs an external
> reference. Another series with good Stamp documentation is the
> single channel MAX187 (with 4.096 internal reference) or MAX189
> (external reference) or 8 channel MAX186 (4.096 internal) or MAX188
> (external). One disadvantage of an internal reference is that they
> are usually not very accurate (i.e. +/- 2%) , and you can't adjust
> them with an external potentiometer. You end up having to do
> corrections in software.
>
>
> >
> >The next part does not pertain to a stamp, but does have something to do
> >with my project, if you can't help, at least try and point me in the
right
> >direction.
> >
> >The max my sensor puts out is 400mv, so I figured I would use an
> >instrumentation amp to increase this signal by approx. 10 x with a small
> >adjustment for calibration purposes. This would give me the full range of
my
> >ADC and the best resolution, I think. The amp needs to be a single
> >voltage like + 5vdc and Grd. I was looking at the AD626 and I several
> >LTC1052's but I never used them before. Can someone tell me if I'm on
the
> >right path or if they have another item which is easy to work with and is
> >fairly accurate?
>
> I use the LTC1051 (dual CAZ op amp) a lot when DC accuracy is
> paramount--input offsets of less than 10 microvolts. I think the
> LTC1052 is similar. Don't expect it to swing above 3.5 volts when it
> has a 5 volt supply. Maybe you can power it from the higher
> unregulated voltage. Or limit your amplification to x5 instead of
> x10, and use a 2.5 volt reference on your ADC. Or just accept 11
> bits resolution. I don't know of any rail to rail op-amps that are
> terrifically accurate, but they would certainly work if you want to
> adjust both the input offset and the scale factor with pots. Is the
> Oxygen sensor really that accurate and stable? The op amp circuit
> will be really simple, especially if you use the CAZ op amp. The
> classic non-inverting amplifier.
>
>
>
> >
> >I don't mind learning by hard knocks, but I rather not do it a dozen
times
> >either.
> >Throw me a bone please....
> >Richard
>
>
> Don't worry, you still get the hard knocks. The bone falling may hit
> you on the head. A corollary of Murphy's law.
>
> -- Tracy
>
>
> >
> >
> >
Original Message
> >From: "Richard Skinner" <rwskinner@w...>
> >To: <basicstamps@yahoogroups.com>
> >Sent: Monday, October 13, 2003 1:04 PM
> >Subject: [noparse][[/noparse]basicstamps] Sensor project
> >
> >
> >> I working on a small project which requires me to read a sensor which
> >> outputs from 0 mv to 400mv +/- 40 mv full scale. It's almost linear
but
> >> does have a couple of items that could use some error correction. I
have
> >> several questions regarding the electronics end and some on the math
end.
> >> The sensor is an Oxygen sensor that reads from 0% to 25.5% o2. At 21%
o2
> >it
> >> reads approx. 400mv.
> >>
> >> I don't know much about OP Amps but willing to learn. I figured I
would
> >use
> >> an 8 bit ADC with a 4.096 reference, then use an OP Amp with a Gain of
10.
> >> I thought about using a pot to slightly adjust the gain in order to
> >> calibrate the sensor. The sensor always has zero output when it senses
> >zero
> >> input, so I was only going to adjust the high end. Would this be the
> >proper
> >> procedure?
> >>
> >> Next, the sensor exhibits a Temp Coefficient of 0.2% signal / Degree C
so
> >I
> >> may want to perform a temp. correction for the temp at the sensor. The
> >temp
> >> would range from about Ambient (32F to 100F) to about 150 Degree F when
> >> operating in it's environment. It's calibrated in an Ambient
condition.
> >I
> >> was leaning towards using an Thermistor to measure the temp during
> >> calibration, then correcting the displayed units for the operating
temps.
> >>
> >> Any input would be greatly appreciated.
> >> Richard
> >>
> >>