Searching \ for '[PIC] ORP problem' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'ORP problem'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] ORP problem'
2010\03\20@221746 by yamanoor sairam

picon face
Hello All,
      I am sorry that I was in a hurry to shoot a mail. I had posted the circuit while writing to the list on a previous occasion. The circuit is available at:

http://mechatronicscraze.wordpress.com/pic-orp/

One of the jpeg image contains the Express PCB file image that I had posted and the other image contains the eagle schematic that I had used for developing this ORP Probe interface. ( I would be happy if someone could offer an opinion on my PCB design skills).

The ORP probe is a glass probe which measures the oxidation reduction potential of a solution based on the potential of the solution. It is commonly used in aquariums by hobbyists and in water treatment plants to measure the dissolved ozone in water.

Now the problem I am facing is that when I provide an input from a DC power supply, my display is providing me the correct voltage with an error of 3 to 4mV. When I dip my probe into a calibration solution of 200mV, it is displaying somewhere close to 203mV.  However, when I remove my probe from the solution, the displayed voltage has to get reduced gradually to zero. This is not happening. It is displaying all values between 0 and 900mV before getting reduced to zero. Some times it is displaying characters like "." "/" etc

I measured the input voltage to the op amp. My multimeter showed 215 mV and the voltage on the input got reduced gradually to zero when I removed the probe from the beaker of water. I measured the opamp output. The opamp output spans between 4.139V and 0V for 0mV to 900mV. My opamp also seems to be sending out the correct voltage.

Could this be a problem with my code?
                 I did the following steps in my code using MikroC complier:
adc_rd = adc_read(0); // read voltage from AN0
t1 = 900-(float)(1.0625*adc_rd);//t1 is declared as float. I am doing this subtraction because my opamp gives an output of 0V for 900mV and 4.139V for 0mV.
t3long = (long)t1long;
//I am proceeding to LCD display section thereafter.

Am I doing the scale conversion correctly or is there anything else that I should do for avoiding my PIC reading the noises.

Sairam



________________________________
From: Russell McMahon <spam_OUTapptechnzTakeThisOuTspamgmail.com>
To: Microcontroller discussion list - Public. <.....piclistKILLspamspam@spam@mit.edu>
Sent: Sat, March 20, 2010 6:08:46 PM
Subject: Re: PIC ORP problem

Sairam.

Congratulations on having got this far with your project.

While there are many things that MAY be causing your problems, and
while we MAY be able to suggest ones which are most likely, it would
be far better if you could provide a circuit diagram and also a
description of any aspects of the circuit which may not be shown well
on the diagram. For example the nature of the probe will affect the
result (and may be what is causing the problem) so knowing about it i
suitable detail will be useful.

The more useful detail that you can provide the more quickly and
easily we are liable to be able to help you.

  regards


             Russell McMahon

2010\03\20@224503 by yamanoor sairam

picon face
Hello Sergio,
    Thanks for the mail.
>For the "junk characters" I would look at the software, in particular the
>routines that convert you binary into displayable ascii. Is a big enough
>buffer available? Can the routines cope with the full range of values you
>might be giving them (can they cope with numbers bigger than 999 for
>example)?

My display cannot exceed 900mV as I am subtracting my ADC values from 900. As mentioned earlier, my opamp sends 0V for a 900mV signal. I am having second thoughts about doubting my code. There seems to be regularity when I provide an input from a DC RPS.

>Regarding the strange random readings, could this be due to the wet probe
>interacting with the oxygen in the air? Or maybe due to movement and a
>faulty connection?

I had this doubt and I tested this by connecting to an oscilloscope. When I rub the probe with my hand, I see that there is an overshoot of voltages.

Sairam



________________________________
From: sergio masci <smplxspamKILLspamallotrope.net>
To: Microcontroller discussion list - Public. <.....piclistKILLspamspam.....mit.edu>
Sent: Sat, March 20, 2010 11:56:57 PM
Subject: Re: PIC ORP problem



On Sat, 20 Mar 2010, yamanoor sairam wrote:

{Quote hidden}

For the "junk characters" I would look at the software, in particular the
routines that convert you binary into displayable ascii. Is a big enough
buffer available? Can the routines cope with the full range of values you
might be giving them (can they cope with numbers bigger than 999 for
example)?

Regarding the strange random readings, could this be due to the wet probe
interacting with the oxygen in the air? Or maybe due to movement and a
faulty connection?

Regards
Sergio Masci

2010\03\21@034859 by Russell McMahon

face picon face
Sairam,

pH probe impedances are in the 10's to 100's of megohms, maybe 1000
megohms, so signals induced into the probe leads cab be expected to
need considering.  would expect the result might easily be much as you
describe.

It depends in part on how your probe behaves when it only has target
solution on random parts of its exterior. If probe impedance rises
when it is not fully immersed then induced signals in the probe leads
may be significant. Placing a meter at the input is quite likely to
make a difference. A 10 Mohn input resistance meter could swamp your
probe and even a 100 Mohm input R meter may cause major problems.

Measuring at Vsense (IC3 pin1 ) is safer. An oscilloscope on Vsense
should show you if the signal is stable.

Op amp input bias current is 20 pA so may not be an issue - although
with a 1000 MOhn probe it would be noticeable.
eg 20 E-12 x E9 ohm = 20 mV.
So if the probe resistance varies greatly from immersed to 'in air'
then it could cause problems.

Having the display show "/" indicates software problems unless you are
applying "/" volts at the time :-)




           Russell McMahon

2010\03\21@041628 by sergio masci

flavicon
face


On Sat, 20 Mar 2010, yamanoor sairam wrote:

> Could this be a problem with my code?
>                   I did the following steps in my code using MikroC complier:
> adc_rd = adc_read(0); // read voltage from AN0
> t1 = 900-(float)(1.0625*adc_rd);//t1 is declared as float. I am doing this subtraction because my opamp gives an output of 0V for 900mV and 4.139V for 0mV.
> t3long = (long)t1long;
> //I am proceeding to LCD display section thereafter.

Should the above read

t1 = 900 - (float)(adc_rd) * 1.0625;
t3long = (long)t1;

Regards
Sergio Masci

2010\03\21@222941 by yamanoor sairam

picon face
Hello Mr. Russell,
              Thanks for your mail. I shall check the op-amp output and get back to you.

Having the display show "/" indicates software problems unless you are
applying "/" volts at the time :-)

Before getting reduced to zero, my voltage values are shooting upto those values which is not possiblei.e While getting reduced to 0 from 225mV, I am reading values as high as 300 or 400mV. This cannot happen when my opamp input behavior is normal. Can I attribute coding errors for the same?

Sairam


________________________________
From: Russell McMahon <EraseMEapptechnzspam_OUTspamTakeThisOuTgmail.com>
To: Microcontroller discussion list - Public. <piclistspamspam_OUTMIT.EDU>
Sent: Sun, March 21, 2010 1:18:38 PM
Subject: Re: [PIC] ORP problem

Sairam,

pH probe impedances are in the 10's to 100's of megohms, maybe 1000
megohms, so signals induced into the probe leads cab be expected to
need considering.  would expect the result might easily be much as you
describe.

It depends in part on how your probe behaves when it only has target
solution on random parts of its exterior. If probe impedance rises
when it is not fully immersed then induced signals in the probe leads
may be significant. Placing a meter at the input is quite likely to
make a difference. A 10 Mohn input resistance meter could swamp your
probe and even a 100 Mohm input R meter may cause major problems.

Measuring at Vsense (IC3 pin1 ) is safer. An oscilloscope on Vsense
should show you if the signal is stable.

Op amp input bias current is 20 pA so may not be an issue - although
with a 1000 MOhn probe it would be noticeable.
eg 20 E-12 x E9 ohm = 20 mV.
So if the probe resistance varies greatly from immersed to 'in air'
then it could cause problems.

Having the display show "/" indicates software problems unless you are
applying "/" volts at the time :-)




          Russell McMahon

2010\03\21@224930 by yamanoor sairam

picon face
Hello Sergio,
       Thanks for your mail.

Should the above read
> Could this be a problem with my code?
>                  I did the following steps in my code using MikroC complier:
> adc_rd = adc_read(0); // read voltage from AN0
> t1 = 900-(float)(1.0625*adc_rd);//t1 is declared as float. I am doing this subtraction because my opamp gives an output of 0V for 900mV and 4.139V for 0mV.
> t3long = (long)t1long;
> //I am proceeding to LCD display section thereafter.

Should the above read

t1 = 900 - (float)(adc_rd) * 1.0625;
t3long = (long)t1;


The second line was a typo. Am I wrong in the first line?

Sairam



________________________________
From: sergio masci <@spam@smplxKILLspamspamallotrope.net>
To: Microcontroller discussion list - Public. <KILLspampiclistKILLspamspammit.edu>
Sent: Sun, March 21, 2010 5:35:22 PM
Subject: Re: [PIC] ORP problem



On Sat, 20 Mar 2010, yamanoor sairam wrote:

> Could this be a problem with my code?
>                   I did the following steps in my code using MikroC complier:
> adc_rd = adc_read(0); // read voltage from AN0
> t1 = 900-(float)(1.0625*adc_rd);//t1 is declared as float. I am doing this subtraction because my opamp gives an output of 0V for 900mV and 4.139V for 0mV.
> t3long = (long)t1long;
> //I am proceeding to LCD display section thereafter.

Should the above read

t1 = 900 - (float)(adc_rd) * 1.0625;
t3long = (long)t1;

Regards
Sergio Masci

2010\03\22@085507 by Russell McMahon

face picon face
> Before getting reduced to zero, my voltage values are shooting upto those values which is not possible

If it is not possible, then is there any way for it to NOT be a code
problem? :-)

(Actually, yes, but unlikely - eg the PIC could be behaving badly  eg
as may happen if you bias its pins badly as per prior discussions).

Looking at the voltages being measured with an oscilloscope (as
already suggested) is a very easy way to see if what is being reported
is anything like what is happening.

But, if your display shows strange characters or if the circuit does
not reach the voltages shown by the uP then the Holmes'  law of
necessary reality* [tm]  says that your code or your processor must be
malfunctioning.


              R

* After you have eliminated the impossible, what remains, no matter
how improbable, must be reality.

(The trick is in determining the bounds of "impossible").








i.e While getting reduced to 0 from 225mV, I am reading values as high
as 300 or 400mV. This cannot happen when my opamp input behavior is
normal. Can I attribute coding errors for the same?
{Quote hidden}

>

2010\03\22@101945 by sergio masci

flavicon
face


On Sun, 21 Mar 2010, yamanoor sairam wrote:

{Quote hidden}

Think about it. If you NEED to convert "1.0625*adc_rd" to a float AFTER
the multiplication, then you are going to have problems.

You shouldn't "need" to convert adc_rd to a float because the compiler
should do that automatically for you before the multiplication because you
are multiplying an integer with a float. The problem is I don't know how
well behaved the MikroC complier is (I have not used it) when it comes to
promoting integers to floats in conjunction with floating point constants.
I've seen lots of issues in the past with other compilers so I would play
safe and rewrite it as I've shown (and even then I wouldn't like to
guarentee it).

I haven't done the math but you should check yours (that adc_rd * 1.0625
cannot produce a result greater than 900) otherwise you'll get negative
numbers.

Also something else that comes to mind, is adc_rd a 16 bit integer
variable or an 8 bit integer variable and does adc_read(0) return a 16 or
8 bit value? If adc_read(0) returns a 16 bit value and adc_rd is an 8 bit
integer, you could be seeing a high value (above 255) that is dropping to
zero and you are seeing it jump about because you are losing the high
bits.

e.g.

       900 = 0x384 -> 0x84 = 132
       800 = 0x320 -> 0x20 = 32
       700 = 0x2bc -> 0xbc = 188
       600 = 0x258 -> 0x58 = 88
       500 = 0x1f4 -> 0xf4 = 244
       400 = 0x190 -> 0x90 = 144
       300 = 0x12c -> 0x2c = 44
       200 = 0xc8  -> 0xc8 = 200

Regards
Sergio Masci

2010\03\22@141354 by Michael Rigby-Jones

flavicon
face


> -----Original Message-----
> From: TakeThisOuTpiclist-bouncesEraseMEspamspam_OUTmit.edu [RemoveMEpiclist-bouncesspamTakeThisOuTmit.edu] On
Behalf
> Of yamanoor sairam
> Sent: 21 March 2010 02:18
> To: Microcontroller discussion list - Public.
> Subject: [PIC] ORP problem
>
> Hello All,
>        I am sorry that I was in a hurry to shoot a mail. I had posted
the
> circuit while writing to the list on a previous occasion. The circuit
is
> available at:
>
> mechatronicscraze.wordpress.com/pic-orp/
>
> One of the jpeg image contains the Express PCB file image that I had
> posted and the other image contains the eagle schematic that I had
used
> for developing this ORP Probe interface. ( I would be happy if someone
> could offer an opinion on my PCB design skills).
>
> The ORP probe is a glass probe which measures the oxidation reduction
> potential of a solution based on the potential of the solution. It is
> commonly used in aquariums by hobbyists and in water treatment plants
to
> measure the dissolved ozone in water.
>
> Now the problem I am facing is that when I provide an input from a DC
> power supply, my display is providing me the correct voltage with an
error
> of 3 to 4mV. When I dip my probe into a calibration solution of 200mV,
it
> is displaying somewhere close to 203mV.  However, when I remove my
probe
> from the solution, the displayed voltage has to get reduced gradually
to
> zero. This is not happening. It is displaying all values between 0 and
> 900mV before getting reduced to zero. Some times it is displaying
> characters like "." "/" etc

Have you checked the output of your high impedance buffer op-amp (IC3A)
with an oscilloscope?  It's quite possible it could be oscillating
around a mean DC value.  I notice you don't have any components to limit
the bandwidth of this stage, and with the very high input impedance it
would take only a tiny amount of coupling (a few pF) from the output to
the non-inverting input to become unstable.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2010\03\22@154633 by Marcel Duchamp

picon face
On 3/20/2010 7:17 PM, yamanoor sairam wrote:
>
> Now the problem I am facing is that when I provide an input from a DC
> power supply, my display is providing me the correct voltage with an
> error of 3 to 4mV. When I dip my probe into a calibration solution of
> 200mV, it is displaying somewhere close to 203mV.  However, when I
> remove my probe from the solution, the displayed voltage has to get
> reduced gradually to zero. This is not happening. It is displaying
> all values between 0 and 900mV before getting reduced to zero. Some
> times it is displaying characters like "." "/" etc
>
> I measured the input voltage to the op amp. My multimeter showed 215
> mV and the voltage on the input got reduced gradually to zero when I
> removed the probe from the beaker of water. I measured the opamp
> output. The opamp output spans between 4.139V and 0V for 0mV to
> 900mV. My opamp also seems to be sending out the correct voltage.
>
> Could this be a problem with my code?

As others have pointed out, this is normal operation for an ORP probe.
It is extremely high impedance when out of the solution.  It then is
susceptible to picking up electrical noise (line frequency, etc.) or
simply floating to one rail or the other.  When you connect your DVM to
the input, you are pulling the input to ground.  But with no DVM on the
input, the circuit will produce indeterminate signals.

2010\03\22@225613 by yamanoor sairam

picon face
Hello Mr. Russell,
                  Thanks for your mail.
But, if your display shows strange characters or if the circuit does
not reach the voltages shown by the uP then the Holmes'  law of
necessary reality* [tm]  says that your code or your processor must be
malfunctioning.

You were right on the malfunctioning. As Mr. Segio pointed out, I corrected the error in my code.

To eliminate frequent jumping in my display readings, I learnt that I must sample the ADC readings and take an average of the same. Hence, I averaged 256 samples with 1ms delay between them. I see that my display is stable now.

I dont get any junk characters on my display since I have controlled the code section. However, when I lift my probe out of the water, I see that my amplifier is amplifying all noise signals and sending them to the PIC. It takes a lot of time for the PIC to read '0000'. I have added a 0.1uF capacitor and a 10uF capacitor between the PIC input and opamp output. Since I have a 1K capacitor, could this act as a RC filter combination?

As you had suggested, I am going to gain access to an oscilloscope and check the op-amp signals.

Sairam




________________________________
From: Russell McMahon <apptechnzEraseMEspam.....gmail.com>
To: Microcontroller discussion list - Public. <EraseMEpiclistspamMIT.EDU>
Sent: Mon, March 22, 2010 6:24:46 PM
Subject: Re: [PIC] ORP problem

> Before getting reduced to zero, my voltage values are shooting upto those values which is not possible

If it is not possible, then is there any way for it to NOT be a code
problem? :-)

(Actually, yes, but unlikely - eg the PIC could be behaving badly  eg
as may happen if you bias its pins badly as per prior discussions).

Looking at the voltages being measured with an oscilloscope (as
already suggested) is a very easy way to see if what is being reported
is anything like what is happening.

But, if your display shows strange characters or if the circuit does
not reach the voltages shown by the uP then the Holmes'  law of
necessary reality* [tm]  says that your code or your processor must be
malfunctioning.


              R

* After you have eliminated the impossible, what remains, no matter
how improbable, must be reality.

(The trick is in determining the bounds of "impossible").








i.e While getting reduced to 0 from 225mV, I am reading values as high
as 300 or 400mV. This cannot happen when my opamp input behavior is
normal. Can I attribute coding errors for the same?
{Quote hidden}

> -

2010\03\23@001717 by yamanoor sairam

picon face
Hello Mr. Sergio,
                   Thanks for your mail
I haven't done the math but you should check yours (that adc_rd * 1.0625
cannot produce a result greater than 900) otherwise you'll get negative
numbers.

You were right. My program generated negative numbers and I have controlled my program.

Also something else that comes to mind, is adc_rd a 16 bit integer
variable or an 8 bit integer variable and does adc_read(0) return a 16 or
8 bit value

adc_rd is an unsigned integer. Since it is a 16 bit integer, I think the adc value is well within the limits. I am worried about the linearity of the displayed voltage now. The method adopted keeps me worried whether I am able to adopt a linear scale. I tried testing it by providing a known voltage from my DC RPS. My display seems to read the values closer to the known voltage. How may I ensure it?

Sairam


________________________________
From: sergio masci <RemoveMEsmplxTakeThisOuTspamspamallotrope.net>
To: Microcontroller discussion list - Public. <EraseMEpiclistspamspamspamBeGonemit.edu>
Sent: Mon, March 22, 2010 11:38:51 PM
Subject: Re: [PIC] ORP problem



On Sun, 21 Mar 2010, yamanoor sairam wrote:

{Quote hidden}

Think about it. If you NEED to convert "1.0625*adc_rd" to a float AFTER
the multiplication, then you are going to have problems.

You shouldn't "need" to convert adc_rd to a float because the compiler
should do that automatically for you before the multiplication because you
are multiplying an integer with a float. The problem is I don't know how
well behaved the MikroC complier is (I have not used it) when it comes to
promoting integers to floats in conjunction with floating point constants.
I've seen lots of issues in the past with other compilers so I would play
safe and rewrite it as I've shown (and even then I wouldn't like to
guarentee it).

I haven't done the math but you should check yours (that adc_rd * 1.0625
cannot produce a result greater than 900) otherwise you'll get negative
numbers.

Also something else that comes to mind, is adc_rd a 16 bit integer
variable or an 8 bit integer variable and does adc_read(0) return a 16 or
8 bit value? If adc_read(0) returns a 16 bit value and adc_rd is an 8 bit
integer, you could be seeing a high value (above 255) that is dropping to
zero and you are seeing it jump about because you are losing the high
bits.

e.g.

   900 = 0x384 -> 0x84 = 132
   800 = 0x320 -> 0x20 = 32
   700 = 0x2bc -> 0xbc = 188
   600 = 0x258 -> 0x58 = 88
   500 = 0x1f4 -> 0xf4 = 244
   400 = 0x190 -> 0x90 = 144
   300 = 0x12c -> 0x2c = 44
   200 = 0xc8  -> 0xc8 = 200

Regards
Sergio Masci

2010\03\23@033626 by yamanoor sairam

picon face
Hello Mr. Mike,

Thanks for your mail.
Have you checked the output of your high impedance buffer op-amp (IC3A)
with an oscilloscope?  It's quite possible it could be oscillating
around a mean DC value.  I notice you don't have any components to limit
the bandwidth of this stage, and with the very high input impedance it
would take only a tiny amount of coupling (a few pF) from the output to
the non-inverting input to become unstable.

You are right about the DC voltage oscillating around the mean value. I am yet to gain access to an oscilloscope. I checked the output of my opamp using a multimeter. The opamp remained at 3.3V when I removed the probe out of water instead of increasing to 4.139V. It took a long time before increasing to the desired value. How am I supposed to prevent amplification of all noises at this stage?

Sairam



________________________________
From: Michael Rigby-Jones <RemoveMEMichael.Rigby-JonesKILLspamspamoclaro.com>
To: Microcontroller discussion list - Public. <piclistSTOPspamspamspam_OUTmit.edu>
Sent: Mon, March 22, 2010 11:43:50 PM
Subject: RE: [PIC] ORP problem



> {Original Message removed}

2010\03\23@034716 by yamanoor sairam

picon face
Hello Mr. Marcel,

Thanks for your mail.


As others have pointed out, this is normal operation for an ORP probe.
It is extremely high impedance when out of the solution.  It then is
susceptible to picking up electrical noise (line frequency, etc.) or
simply floating to one rail or the other.  When you connect your DVM to
the input, you are pulling the input to ground.  But with no DVM on the
input, the circuit will produce indeterminate signals.


In order to avoid noises, I have used an RC filter combination at the PIC input & Opamp output. I used a 0.1UF & 10uF capacitor in parallel. I am also averaging 256 samples before displaying it. My display has stabilised now and I have also removed all junk characters on my display.

When I remove my probe from the BNC connector, I am able to see that the displayed value is dropping to zero immediately. I shall check the oscilloscope output and get back to you. I am currently searching for a possible solution for controlling the erratic behavior when the probe is removed out of water but still connected to the PIC.

Sairam

________________________________
From: Marcel Duchamp <spamBeGonemarcel.duchampSTOPspamspamEraseMEsbcglobal.net>
To: Microcontroller discussion list - Public. <KILLspampiclistspamBeGonespammit.edu>
Sent: Tue, March 23, 2010 1:16:30 AM
Subject: Re: [PIC] ORP problem

On 3/20/2010 7:17 PM, yamanoor sairam wrote:
{Quote hidden}

As others have pointed out, this is normal operation for an ORP probe.
It is extremely high impedance when out of the solution.  It then is
susceptible to picking up electrical noise (line frequency, etc.) or
simply floating to one rail or the other.  When you connect your DVM to
the input, you are pulling the input to ground.  But with no DVM on the
input, the circuit will produce indeterminate signals.

2010\03\23@053358 by Russell McMahon

face picon face

CCs- Widlar at end. Thoughts on shield driving?

> You are right about the DC voltage oscillating around the mean value. I am
yet to gain access to an oscilloscope. I checked the output of my opamp
using a multimeter. The opamp remained at 3.3V when I removed the probe out
of water instead of increasing to 4.139V. It took a long time before
increasing to the desired value. How am I supposed to prevent amplification
of all noises at this stage?

--

MANY possibilities, all with compromises.
Looking how commercial units do it would probably be informative.

Adding a resistor from 1st amplifier (probe) input to ground will start in
the right direction, at the cost of loading the probe signal. Try 10 to 100
megohm or so and see what it does.

A SMALL capacitor at probe input will slow things down - maybe too much.

Better shielding (or shielding) of the probe cable may help.
Using a shield driver may help. This is more usually a driving end solution
but may be useful.

Fig 7 / Fig 8 here give the idea.
Note the writers name. A past superhero of the art* (died at age 54).

Note shield grounding may or may not be best solution.

Your 2nd stage amp is a low pass filter. You COULD lower its frequency, but
in normal use that may be undesirable.

It should be extremely easy to get software to do what you want. Probe
output behaviour should be quite different in and out of solution so you
should be able to detect the "anomalous" behaviour and adjust what is
displayed.
OOS maybe ? :-) - out of solution?

If desperate (and it shouldn't come to this) you could consider putting an
op amp buffer in the probe head, so that the cable pickup is irrelevant. (4
wires trivial, 3 wires easy, 2 wires - extra points :-)).


Russell

______________________________

*
Robert J. Widlar

(November 30, 1937 - February 27, 1991) was a pioneer in the design of
integrated circuits. He made his fame with Fairchild Semiconductor in the
1960s. Many of his designs were the first of their kind, or became industry
standard parts. These include the ìA702, which was the first IC opamp, and
the ìA723 voltage regulator. Widlar designed the LM101, the parent of (and
nearly identical to) the ubiquitous LM741.

He also invented circuit topologies, including the Widlar current source and
bandgap voltage reference, that are used in almost any practical analog IC.

CCt not in front of me but if 1st stage had unity gain then a filtered
version of the IC output may make a good shield drive.
  http://www.intersil.com/data/an/an1308.pdf

1973. May give some useful ideas. No promises.
http://www.freepatentsonline.com/3720879.pdf

2010\03\23@054202 by sergio masci

flavicon
face


On Mon, 22 Mar 2010, yamanoor sairam wrote:

> I am worried about the linearity of
> the displayed voltage now. The method adopted keeps me worried whether I
> am able to adopt a linear scale. I tried testing it by providing a known
> voltage from my DC RPS. My display seems to read the values closer to
> the known voltage. How may I ensure it?

I don't understand what you mean here.

Regards
Sergio Masci

2010\03\23@055220 by sergio masci

flavicon
face


On Tue, 23 Mar 2010, yamanoor sairam wrote:

> You are right about the DC voltage oscillating around the mean value. I
> am yet to gain access to an oscilloscope. I checked the output of my
> opamp using a multimeter. The opamp remained at 3.3V when I removed the
> probe out of water instead of increasing to 4.139V. It took a long time
> before increasing to the desired value. How am I supposed to prevent
> amplification of all noises at this stage?

This would be easy in software. You know what to expect when the probe is
in the solution (how the output of the probe should vary over time) and
you know how the probe IS behaving when it is taken out of the solution
(you are getting random noise). So write a bit of code that can tell the
difference and only display a reading when this "detector" code tells you
the probe is probably in the solution. Otherwise you could just display
"???" in place of the reading.

Just think about how you would tell the difference between random and
non random readings.

Regards
Sergio Masci

2010\03\24@000317 by Gerhard Fiedler

picon face
yamanoor sairam wrote:

>> As others have pointed out, this is normal operation for an ORP
>> probe. It is extremely high impedance when out of the solution.  It
>> then is susceptible to picking up electrical noise (line frequency,
>> etc.) or simply floating to one rail or the other.  When you connect
>> your DVM to the input, you are pulling the input to ground.  But
>> with no DVM on the input, the circuit will produce indeterminate
>> signals.
>
> In order to avoid noises, I have used an RC filter combination at the
> PIC input & Opamp output. I used a 0.1UF & 10uF capacitor in
> parallel. I am also averaging 256 samples before displaying it. My
> display has stabilised now and I have also removed all junk
> characters on my display.
>
> When I remove my probe from the BNC connector, I am able to see that
> the displayed value is dropping to zero immediately. I shall check
> the oscilloscope output and get back to you. I am currently searching
> for a possible solution for controlling the erratic behavior when the
> probe is removed out of water but still connected to the PIC.

What may work, depending on the impedances involved, is to inject a low
current at the input. The current should be so small that it doesn't
affect the measurement significantly when the probe is in the solution
(low impedance) but does bring the opamp output to one of the rails (the
lower is probably best) when the probe is out of the solution (high
impedance).

Something similar is used for break detection with thermocouples.

If it doesn't work directly (that is, if the impedance difference is not
so big that the necessary current wouldn't affect the measurement, but
not so small that it couldn't be measured), you can switch the current
on and off periodically and use it to measure the impedance, and discard
the measurements while the impedance is indicating that the probe is not
in the solution.

Gerhard

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...