Making an electrode for a conductivity meter

[Wayne]

Hi

I am trying to make a conductivity meter and would like to know if any body
can advise on how to make the electrodes? Do I make them from paladium or
can I use something like alluminum?
Also, would I get an improved conductivity value using an AC signal?

Cheers

Wayne

 

[Guy Macon]

n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

--
Guy Macon <http://www.guymacon.com/>

 

[John Larkin]

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon
<http://www.guymacon.com/> wrote:

n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

Wouldn't a potted torroidal inductor make an interesting conductivity
sensor? No contacts to crud up.

John

 

[Rich Grise]

On Wed, 29 Dec 2004 12:11:36 +0000, Wayne top-posted:

Thanks Bill
Do you know where I could buy a few of these 1cm^2 platinum
plates. My research is taking on a great deal of equipment and I will be
modifying and customising the equipment that is why I can not just buy one.

You might try some of these links:

http://www.google.com/search?q=platinum+electrodes&btnG=Google+Search

Good Luck!
Rich

 

[Rich Grise]

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon wrote:

n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

It's probably harder than gold. It's more expensive, so it's more
impressive. ;-) And Bill Sloman mentioned "Platinum black", which
apparently can easily be plated onto a Pt substrate to increase the
surface area.

And who ever heard of "gold electrodes?" ;-P

Cheers!
Rich

 

[n2mp]

Hello,

Try Goodfellow. They should have this.

Best regards.


--
Enlevez ".nospam" de mon adresse e-mail pour me répondre.
--------------------------------------------------------------------------------------------------------
Remove ".nospam" from my email address to reply me.

 

[colin beeforth]

On Tue, 28 Dec 2004 19:57:29 +0000, Wayne wrote:

Hi

I am trying to make a conductivity meter and would like to know if any body
can advise on how to make the electrodes? Do I make them from paladium or
can I use something like alluminum?
Also, would I get an improved conductivity value using an AC signal?

Cheers

Wayne

Hi Wayne,

The company I work for sells conductivity meters, and while I don't know
much about the principles I do know that they use high density carbon
electrodes (graphite?) and use AC drive to measure the conductivity.

Try searching Google for conductivity electrodes.

Cheers, Colin

 

[n2mp]

Hello,

1- It exist conductivity meters in the market. You can just buy one of them
: you'll save your time and money.
2- If you really want to make one by yourself for pratice, then :
* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded
in insulator and separated from each other by 1cm.
* measuring signal is AC to avoid problems linked with mass transfer
limitations when a DC current is applied that will add overvoltages that
will distort the measurement.
* platinum is used in standard cells, despite its high cost, because it has
very low charge transfer resistance. Charge transfer resistance adds to the
ohmic resistance you want to measure and thus distort the result.
* aluminum is covered by a layer of Al2O3 that will distort your
measurement, because the electrons will need to pass this layer.

Best regards.


--
Enlevez ".nospam" de mon adresse e-mail pour me répondre.
--------------------------------------------------------------------------------------------------------
Remove ".nospam" from my email address to reply me.

 

n2mp wrote:

Hello,

1- It exist conductivity meters in the market. You can just buy one
of them
: you'll save your time and money.

Very true.

2- If you really want to make one by yourself for pratice, then :
* conductivity cells usually consist of two 1cm2 plates of platinized

platinum correctly embedded
in insulator and separated from each other by 1cm.

This is a "standard" conductivity cell - you can make the plates bigger
and closer together if you want to see a lower resistance when
measuring the conductivity of a particular solution. Making them
smaller and further apart lets you see a higher resistance, but you end
up averaging the conductivity over a range of paths through the liquid,
and the maths can get complicated.

* measuring signal is AC to avoid problems linked with mass transfer
limitations when a DC current is applied that will add overvoltages
that
will distort the measurement.

You really have to measure conductivity with an AC signal - you get
additional voltage drops at the electrode surfaces in addition to the
resistive losses through the bulk of the liquid, and you have to
minimise these additional losses if your measurement is going to say
anything useful about the conductivity of the solution. Even with an AC
signal, you've always got a capacitative impedance at the electrodes in
series with the resistance of the solution - a higher frequencies this
can be just the "double layer capacitance" but at lower frequencies you
can also see the Warburg impedance which looks like a frequency
dependent capacitance in series with a frequency dependent resistance -
see

http://www.gamry.com/App_Notes/EIS_Primer/EIS_Primer.pdf

* platinum is used in standard cells, despite its high cost, because
it has
very low charge transfer resistance. Charge transfer resistance adds
to the
ohmic resistance you want to measure and thus distort the result.
* aluminum is covered by a layer of Al2O3 that will distort your
measurement, because the electrons will need to pass this layer.

Platinum is used in standard cells because it doesn't corrode, and
because it is easy to electroplate a layer of "platinum black" onto a
platinum surface, which vastly increases the effective surface area and
reduces the voltage drops at the electrode surfaces in the same
proportion. Carbon is often used in big cells where cost is a problem,
but it is much more difficult to work with.

Stainless steel is just as bad as aluminium - it is "stainless" because
the surface is protected by a thin layer of Cr2O3 and I've had
conductivity measurements wrecked by the additional resistance of this
layer (which changes with the chemical nature of the solution being
measured). My bosses were not pleased with having to pay for platinum
rather than stainless steel, but should have been more upset about ever
having wasted time on stainless steel electrodes in the first place.
-------
Bill Sloman, Nijmegen

 

[Wayne]

Thanks Bill
Do you know where I could buy a few of these 1cm^2 platinum
plates. My research is taking on a great deal of equipment and I will be
modifying and customising the equipment that is why I can not just buy one.

Cheers

Wayne

<bill.sloman@ieee.org> wrote in message
news:1104321738.687048.116070@f14g2000cwb.googlegroups.com...

n2mp wrote:
Hello,

1- It exist conductivity meters in the market. You can just buy one
of them
: you'll save your time and money.

Very true.

2- If you really want to make one by yourself for pratice, then :
* conductivity cells usually consist of two 1cm2 plates of platinized

platinum correctly embedded
in insulator and separated from each other by 1cm.

This is a "standard" conductivity cell - you can make the plates bigger
and closer together if you want to see a lower resistance when
measuring the conductivity of a particular solution. Making them
smaller and further apart lets you see a higher resistance, but you end
up averaging the conductivity over a range of paths through the liquid,
and the maths can get complicated.

* measuring signal is AC to avoid problems linked with mass transfer
limitations when a DC current is applied that will add overvoltages
that
will distort the measurement.

You really have to measure conductivity with an AC signal - you get
additional voltage drops at the electrode surfaces in addition to the
resistive losses through the bulk of the liquid, and you have to
minimise these additional losses if your measurement is going to say
anything useful about the conductivity of the solution. Even with an AC
signal, you've always got a capacitative impedance at the electrodes in
series with the resistance of the solution - a higher frequencies this
can be just the "double layer capacitance" but at lower frequencies you
can also see the Warburg impedance which looks like a frequency
dependent capacitance in series with a frequency dependent resistance -
see

http://www.gamry.com/App_Notes/EIS_Primer/EIS_Primer.pdf

* platinum is used in standard cells, despite its high cost, because
it has
very low charge transfer resistance. Charge transfer resistance adds
to the
ohmic resistance you want to measure and thus distort the result.
* aluminum is covered by a layer of Al2O3 that will distort your
measurement, because the electrons will need to pass this layer.

Platinum is used in standard cells because it doesn't corrode, and
because it is easy to electroplate a layer of "platinum black" onto a
platinum surface, which vastly increases the effective surface area and
reduces the voltage drops at the electrode surfaces in the same
proportion. Carbon is often used in big cells where cost is a problem,
but it is much more difficult to work with.

Stainless steel is just as bad as aluminium - it is "stainless" because
the surface is protected by a thin layer of Cr2O3 and I've had
conductivity measurements wrecked by the additional resistance of this
layer (which changes with the chemical nature of the solution being
measured). My bosses were not pleased with having to pay for platinum
rather than stainless steel, but should have been more upset about ever
having wasted time on stainless steel electrodes in the first place.
-------
Bill Sloman, Nijmegen

 

[John Larkin]

On Wed, 29 Dec 2004 15:46:37 -0500, John Popelish <jpopelish@rica.net>
wrote:

John Larkin wrote:

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon
http://www.guymacon.com/> wrote:


n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

Wouldn't a potted torroidal inductor make an interesting conductivity
sensor? No contacts to crud up.

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.

That's bizarre. Why not a single torroid?

John

 

[John Popelish]

John Larkin wrote:

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon
http://www.guymacon.com/> wrote:


n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

Wouldn't a potted torroidal inductor make an interesting conductivity
sensor? No contacts to crud up.

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.


--
John Popelish

 

[John Popelish]

John Larkin wrote:

On Wed, 29 Dec 2004 15:46:37 -0500, John Popelish <jpopelish@rica.net
wrote:

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.

That's bizarre. Why not a single torroid?

I think the two core approach makes it easier to separate the small
current transformer currents from the large excitation current.

http://vfaut.axelero.net/honeywell/5000tc.pdf

--
John Popelish

 

[John Larkin]

On Wed, 29 Dec 2004 15:59:36 -0500, John Popelish <jpopelish@rica.net>
wrote:

John Larkin wrote:

On Wed, 29 Dec 2004 15:46:37 -0500, John Popelish <jpopelish@rica.net
wrote:

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.

That's bizarre. Why not a single torroid?

I think the two core approach makes it easier to separate the small
current transformer currents from the large excitation current.

http://vfaut.axelero.net/honeywell/5000tc.pdf

OK, that makes sense for low-conductivity liquids.

John

 

[John Woodgate]

I read in sci.electronics.design that John Larkin <jjlarkin@highSNIPland
THIStechPLEASEnology.com> wrote (in <ok56t0hb8n0a6kdvlu4hqt39vof9bd80os@
4ax.com&gt about 'Making an electrode for a conductivity meter', on Wed,
29 Dec 2004:

On Wed, 29 Dec 2004 15:46:37 -0500, John Popelish <jpopelish@rica.net
wrote:

John Larkin wrote:

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon
http://www.guymacon.com/> wrote:


n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

Wouldn't a potted torroidal inductor make an interesting conductivity
sensor? No contacts to crud up.

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.

That's bizarre. Why not a single torroid?

That was my initial reaction. But calculating the conductivity from the

resistive component of the inductor impedance is not at all simple.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[John Larkin]

On Wed, 29 Dec 2004 21:02:52 +0000, John Woodgate
<jmw@jmwa.demon.contraspam.yuk> wrote:

I read in sci.electronics.design that John Larkin <jjlarkin@highSNIPland
THIStechPLEASEnology.com> wrote (in <ok56t0hb8n0a6kdvlu4hqt39vof9bd80os@
4ax.com&gt about 'Making an electrode for a conductivity meter', on Wed,
29 Dec 2004:
On Wed, 29 Dec 2004 15:46:37 -0500, John Popelish <jpopelish@rica.net
wrote:

John Larkin wrote:

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon
http://www.guymacon.com/> wrote:


n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter electrodes.
Do you have any information as to why Platinum would be prefered?

Wouldn't a potted torroidal inductor make an interesting conductivity
sensor? No contacts to crud up.

It certainly would. And they are a commercial product. They contain
two toroidal cores, axially aligned. The winding on one produces an
EMF in the liquid, driving a current that passes through the hole.
The winding on the second one acts as a current transformer, measuring
that current.

That's bizarre. Why not a single torroid?

That was my initial reaction. But calculating the conductivity from the
resistive component of the inductor impedance is not at all simple.

Well, some experimental calibration would be in order. But the loss
associated with a resistive shorted turn should pretty well scale to
conductivity, after subtracting core loss.

I'd think that the single torroid would work pretty well for
reasonably conductive stuff. I'll try it maybe.

John

 

Wayne wrote:

Thanks Bill
Do you know where I could buy a few of these 1cm^2
platinum
plates. My research is taking on a great deal of equipment and I
will be
modifying and customising the equipment that is why I can not just
buy one.

In the U.K. we got platinum and similar stuff from Goodfellow Metals

http://www.goodfellow.com/csp/active/gfHome.csp

but I think you can also get it from jewellery suppliers. For platinum
black you need hexachloroplatinic acid (H2PtCl6) which we got from our
regular chemical supply house (it's poisonous and corrosive ...).

You ought to be able to get platinum plated onto printed circuit board
type materials - which include Teflon/PTFE and alumina - if you chase
around a bit. My bosses weren't prepared to mess around with the
mechanical design of our conductivity sensor to take advantage of this,
so I've not tried this out in practice.

The better solution to measuring solution conductivity is to skip the
electrodes completely, and use the inductive technique (which we've
discussed on s.e.d. within the last few years). You need a rather
bigger volume of solution than you can get away with in a 1cc
conductivity cell, and it doesn't seem to be too good below the
milliSiemen/cm level, but you lose all the voltage drops at the
elelctrode surfaces.

--------
Bill Sloman, Nijmegen

 

Rich Grise wrote:

On Wed, 29 Dec 2004 18:22:54 +0000, Guy Macon wrote:


n2mp wrote:

* conductivity cells usually consist of two 1cm2 plates of
platinized
platinum correctly embedded

I have always used pure Gold plating for conductivity meter
electrodes.
Do you have any information as to why Platinum would be prefered?

It's probably harder than gold. It's more expensive, so it's more
impressive. ;-) And Bill Sloman mentioned "Platinum black", which
apparently can easily be plated onto a Pt substrate to increase the
surface area.

And who ever heard of "gold electrodes?" ;-P

22-carat gold ought not to tarnish, but it contains enough copper and
silver for a tarnish layer to form if there is a lot of sulphide
around.

24-carat - 0.9999 pure gold - won't tarnish, but it is very soft, and
probably as hard to get hold of as platinum.
-------
Bill Sloman, Nijmegen

 

Only if you do it that way - if you put just enough current through the
second coil to precisely cancel the induced voltage, the inductor
impedance would fall out. Nulling techniques are always nice.

And don't forget that the windings on both inductor cores have got to
be non-progressive, otherwise you get an effective single turn around
each of the toroidal cores, which rather messes up the concept.
----------
Bill Sloman, Nijmegen

 

In fact the commercial inductive sensors don't seem to be too good
below about 1mS/cm so they wouldn't be too good with tap water. which
isn't my idea of a low-conductivity liquid.
-----------
Bill Sloman, Nijmegen