Anyone familiar with this pH sensor circuit?
Archiver
Posts: 46,084
Just built it on a breadboard but I'm having some trouble understanding some
things.
First, there are two caps that are labelled just ".1". I'm assuming that
means 100uF, would that be correct? If so, I can't seem to find 100uF caps
that don't have a polarity...
The notes say to short the inputs Basically, I'm just running the CA3160 pin
3 to ground through the 10K resistor, is that correct?
It then says to trim the output to 1 volt... Can't figure out what I'm
supposed to trim. The pot is supposed to be trimmed to -1v for the LM358, so
that can't be it. There is a formula for vout (vout=vin(1+R7/R6)-vref but,
R6 is supposed to be either 221K or open and R7 is already specified...
Any hints out there?
things.
First, there are two caps that are labelled just ".1". I'm assuming that
means 100uF, would that be correct? If so, I can't seem to find 100uF caps
that don't have a polarity...
The notes say to short the inputs Basically, I'm just running the CA3160 pin
3 to ground through the 10K resistor, is that correct?
It then says to trim the output to 1 volt... Can't figure out what I'm
supposed to trim. The pot is supposed to be trimmed to -1v for the LM358, so
that can't be it. There is a formula for vout (vout=vin(1+R7/R6)-vref but,
R6 is supposed to be either 221K or open and R7 is already specified...
Any hints out there?
Comments
Original Message
From: "Pat M" <pmeloy@s...>
To: <basicstamps@yahoogroups.com>
Sent: Thursday, January 09, 2003 8:16 PM
Subject: [noparse][[/noparse]basicstamps] Anyone familiar with this pH sensor circuit?
> Just built it on a breadboard but I'm having some trouble understanding
some
> things.
>
> First, there are two caps that are labelled just ".1". I'm assuming that
> means 100uF, would that be correct? If so, I can't seem to find 100uF caps
> that don't have a polarity...
>
> The notes say to short the inputs Basically, I'm just running the CA3160
pin
> 3 to ground through the 10K resistor, is that correct?
>
> It then says to trim the output to 1 volt... Can't figure out what I'm
> supposed to trim. The pot is supposed to be trimmed to -1v for the LM358,
so
> that can't be it. There is a formula for vout (vout=vin(1+R7/R6)-vref but,
> R6 is supposed to be either 221K or open and R7 is already specified...
>
> Any hints out there?
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
The ".1" is likely a .1uf capacitor not 100uf.
The dot within a circle symbol represents a coaxial sheilded cable, like you
would see on a oscilloscope test lead more likely. Usually a 50ohm impedance
coax BNC connector. The outer ring is the sheild on the coax connector going
to ground, the inner dot is the signal wire going to the probe tip.
Pin 3 is the input, the 10k resistor helps to protect the input pin and
reduces larger signal swings from damaging the input pin.
If you short the probe tip (input to pin 3 through the 10k resistor) to
ground, you might have to adjust the values for R6 and R7, to set the output
to 1v.
It looks like you leave the PH probe off and use a wire to short it to
ground.
Depends on the parts tolerances and variations.
Your basically calibrating the unit at that time.
Hope that helps...[noparse]:)[/noparse]
Original Message
From: Pat M [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=dKMqzqP6j_5eCE-oJrr7ip4zv8iqXUO695QAZhdYySuR6bELWOHfmjlWOcQ8lQRKt3GnN314]pmeloy@s...[/url
Sent: Thursday, January 09, 2003 10:24 PM
To: basicstamps@yahoogroups.com
Subject: Re: [noparse][[/noparse]basicstamps] Anyone familiar with this pH sensor circuit?
Doh, forgot the link! http://www.emesys.com/OL2ph.htm
Original Message
From: "Pat M" <pmeloy@s...>
To: <basicstamps@yahoogroups.com>
Sent: Thursday, January 09, 2003 8:16 PM
Subject: [noparse][[/noparse]basicstamps] Anyone familiar with this pH sensor circuit?
> Just built it on a breadboard but I'm having some trouble understanding
some
> things.
>
> First, there are two caps that are labelled just ".1". I'm assuming that
> means 100uF, would that be correct? If so, I can't seem to find 100uF caps
> that don't have a polarity...
>
> The notes say to short the inputs Basically, I'm just running the CA3160
pin
> 3 to ground through the 10K resistor, is that correct?
>
> It then says to trim the output to 1 volt... Can't figure out what I'm
> supposed to trim. The pot is supposed to be trimmed to -1v for the LM358,
so
> that can't be it. There is a formula for vout (vout=vin(1+R7/R6)-vref but,
> R6 is supposed to be either 221K or open and R7 is already specified...
>
> Any hints out there?
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
To UNSUBSCRIBE, just send mail to:
basicstamps-unsubscribe@yahoogroups.com
from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
power and ground on each chip in a circuit to control noise, etc...
**********
> > First, there are two caps that are labelled just ".1". I'm assuming that
> > means 100uF, would that be correct? If so, I can't seem to find 100uF
caps
> > that don't have a polarity...
>things.
>
>First, there are two caps that are labelled just ".1". I'm assuming that
>means 100uF, would that be correct? If so, I can't seem to find 100uF caps
>that don't have a polarity...
>link! http://www.emesys.com/OL2ph.htm
They are 0.1 uF. Sprinkled like salt--they are a common electronic
part. One in this circuit is a power supply bypass. Another filters
out noise that may be present on the pH signal.
>
>The notes say to short the inputs Basically, I'm just running the CA3160 pin
>3 to ground through the 10K resistor, is that correct?
That's right, shorting the input simulates an ideal pH probe in a
neutral (pH 7) solution. Both the input and the output of the first
stage (x2) amplifier should be zero volts. Then the offset can be
adjusted on the second stage.
>
>It then says to trim the output to 1 volt... Can't figure out what I'm
>supposed to trim. The pot is supposed to be trimmed to -1v for the LM358, so
>that can't be it. There is a formula for vout (vout=vin(1+R7/R6)-vref but,
>R6 is supposed to be either 221K or open and R7 is already specified...
It says the reference voltage is "~-1 volt", and the squiggly line is
supposed to mean approximately -1 volt. The inverting amplifier
(LM358) should turn that into +1 volt at its output. However, there
are offsets in the amplifier itself, in both stages. You can use the
pot to trim the output offset to exactly one volt. (In which case
the Vref then will not be exactly -1 volt.) Sorry for the confusion.
>
>Any hints out there?
The main calibration is meant to take place in the Stamp software.
pH probes vary from one to the next one and also they change their
calibration as they age. So it is necessary to re-calibrate from
time to time. How often depends on the construction of the probe,
how "dirty" the solution is, and what accuracy is required.
Usually there are two steps to the calibration. You could have a
Stamp set up with a temperature sensor measuring the temperature of
your reference solutions. It also has to measure the temperature of
your working solution for best accuracy.
1) Put the sensor in a neutral (pH 7) butter solution and press the
"Standardize" button
2) Put the sensor in a pH4 (or pH10) solution and press the
"Calibrate" button.
pH4 for acid, pH10 for alkaline. Or some other reference near
the working pH.
The Stamp takes the raw data from those two measurements, and the
temperature reading, and calculates the slope and offset of the
individual pH sensor at that point in time. When moved to the
working solution, it uses the calibration data and the temperature to
calculate the ongoing pH.
You can use the circuit without doing any calibration, and assuming
that your pH probe has the ideal Nernst response. You will see a
good response to changing pH. But to get good quantitative accuracy,
you do have to do standardization and calibration with buffer
solutions. You can purchase buffers solutions at chemical, or
photographic or even some aquarium supply stores.
-- regards,
Tracy Allen
electronically monitored ecosystems
mailto:tracy@e...
http://www.emesystems.com
> They are 0.1 uF. Sprinkled like salt--they are a common electronic
> part. One in this circuit is a power supply bypass. Another
filters
> out noise that may be present on the pH signal.
Ah ok, I have tons of these, specifically for that purpose. Just
never saw then named that way in all my 3 months of electronics
experience (tongue firmly in cheek).
> That's right, shorting the input simulates an ideal pH probe in a
> neutral (pH 7) solution. Both the input and the output of the
first
> stage (x2) amplifier should be zero volts. Then the offset can be
> adjusted on the second stage.
Ok, so at least I an follow some instructions [noparse]:D[/noparse] Not getting zero
volts on the CA3160 output though... With R6 open I get 4.5v and with
221K there I get -4.5 volts (its a moving target though). I'll rip
these all apart and re-build, I've got something wrong that I can't
see.
> It says the reference voltage is "~-1 volt", and the squiggly line
is
> supposed to mean approximately -1 volt. The inverting amplifier
> (LM358) should turn that into +1 volt at its output. However,
there
> are offsets in the amplifier itself, in both stages. You can use
the
> pot to trim the output offset to exactly one volt. (In which case
> the Vref then will not be exactly -1 volt.) Sorry for the
confusion.
Ah ok, wondered what that struck-through diamond was supposed to mean!
> The main calibration is meant to take place in the Stamp software.
> pH probes vary from one to the next one and also they change their
> calibration as they age. So it is necessary to re-calibrate from
> time to time. How often depends on the construction of the probe,
> how "dirty" the solution is, and what accuracy is required.
>
> Usually there are two steps to the calibration. You could have a
> Stamp set up with a temperature sensor measuring the temperature of
> your reference solutions. It also has to measure the temperature
of
> your working solution for best accuracy.
I have a nice sensor calibration kit at work .We run 4 regular
swimming pools, 5 hot tubs, a wave pool and a toddler's wading pool,
all with automatic controls (2xStrantrol, 3xAccutrol, and 5 x
Honeywell knock-together). I have lots of experience with pH and ORP
sensors from the user side of things and several spare sensors to
test things with [noparse]:D[/noparse]
So, if I set up the 3160 and *don't* hook its output (pin 6) to the
other op-amp, I should get zero volts with the input shorted,
correct? I'm thinking I should just do one op amp at a time and get
everything working correctly there before trying to get the next one
set up.
Oh and my 7660 isn't supplying -5.0, its being drawn down to -4.6
once the 3160 and LM385 are hooked up (vcc is about +4.93 at this
time). Would the differential screw up the outputs?
And thanks for being here Tracy!
>
>> They are 0.1 uF. Sprinkled like salt--they are a common electronic
>> part. One in this circuit is a power supply bypass. Another
>filters
>> out noise that may be present on the pH signal.
>
>Ah ok, I have tons of these, specifically for that purpose. Just
>never saw then named that way in all my 3 months of electronics
>experience (tongue firmly in cheek).
chuckle. I usually prefer to use the 0.1 notation. But you can be
99.99% sure that .1 does not have units of Farads! That would be
100000 microfarads. I think it is safe to say that capacitance
values are usually in uF, unless stated otherwise.
>
>> That's right, shorting the input simulates an ideal pH probe in a
>> neutral (pH 7) solution. Both the input and the output of the
>first
>> stage (x2) amplifier should be zero volts. Then the offset can be
>> adjusted on the second stage.
>
>Ok, so at least I an follow some instructions [noparse]:D[/noparse] Not getting zero
>volts on the CA3160 output though... With R6 open I get 4.5v and with
>221K there I get -4.5 volts (its a moving target though). I'll rip
>these all apart and re-build, I've got something wrong that I can't
>see.
There must be something wrong in the wiring. Maybe R7, the feedback
resistor, is not connected. The input (pin 3 of the CA3160 via the
10k resistor) does have to be connected to ground. Keep the lead
lengths short, particularly the connection from the 2.2 megaohm
resistor to pin 2. That point is very suseptable to noise pickup.
If you wee a funny voltage, also put your meter on AC measurement,
and that way you can see if there is an oscillation. (There should
not be more than a few millivolts of AC).
>
>> It says the reference voltage is "~-1 volt", and the squiggly line
>is
>> supposed to mean approximately -1 volt. The inverting amplifier
>> (LM358) should turn that into +1 volt at its output. However,
>there
>> are offsets in the amplifier itself, in both stages. You can use
>the
>> pot to trim the output offset to exactly one volt. (In which case
>> the Vref then will not be exactly -1 volt.) Sorry for the
>confusion.
>
>Ah ok, wondered what that struck-through diamond was supposed to mean!
Macintosh Geneva font!
>
>> The main calibration is meant to take place in the Stamp software.
>> pH probes vary from one to the next one and also they change their
>> calibration as they age. So it is necessary to re-calibrate from
>> time to time. How often depends on the construction of the probe,
>> how "dirty" the solution is, and what accuracy is required.
>>
>> Usually there are two steps to the calibration. You could have a
>> Stamp set up with a temperature sensor measuring the temperature of
>> your reference solutions. It also has to measure the temperature
>of
> > your working solution for best accuracy.
>
>I have a nice sensor calibration kit at work .We run 4 regular
>swimming pools, 5 hot tubs, a wave pool and a toddler's wading pool,
>all with automatic controls (2xStrantrol, 3xAccutrol, and 5 x
>Honeywell knock-together). I have lots of experience with pH and ORP
>sensors from the user side of things and several spare sensors to
>test things with [noparse]:D[/noparse]
Ahh, you're swimming ahead of the pack!
>
>So, if I set up the 3160 and *don't* hook its output (pin 6) to the
>other op-amp, I should get zero volts with the input shorted,
>correct?
Correct, with input connected to ground. Whatever voltage is on the
input wrt ground should be transfered directly to the output, times
two with the R6=221k in place. e.g., 0 volts in, 0 volts out... 1
in, 2 out... -1 in, -2 out.
>I'm thinking I should just do one op amp at a time and get
>everything working correctly there before trying to get the next one
>set up.
Always a good idea (at least if it doesn't work!)
>
>Oh and my 7660 isn't supplying -5.0, its being drawn down to -4.6
>once the 3160 and LM385 are hooked up (vcc is about +4.93 at this
>time). Would the differential screw up the outputs?
That is to be expected. The 7660 is not regulated.
>
>And thanks for being here Tracy!
>
You're welcome, and good luck with the project! In looking back at
the page I found a couple of typos, so I uploaded it again to
http://www.emesys.com/OL2ph.htm. When I get a chance, I'll try to
locate the push-button auto-calibration program and add it to that
page.
-- Tracy