pH-Measurement

[G-rooon]

Hello,

i'm a german student, using LabVIEW now, but at one point got stuck.

So, i'm using a PCI 6024e Card and a BNC 2120. LabVIEW Version, not sure, quite old, around 6.0 or 6.1 I guess.

Furthermore i got a pH-Sensor with an BNC connector-plug.

When i measure the voltage over that sensor, I get a relative high voltage and the voltage is permanent falling after some minutes.

Got someone an idea why the voltage isn't constant? Any idea what to check/improve?

I would also appreciate any advices of what hardware I could buy to get this project started.

Thanks

Regardful

G-rooon

[Albert.Geven]

The problem is the bnc plug. What comes out of it and how did you connect it to the iocard?
If you measure with a standard multimeter, do you see the same effect?
The input impedance of the iocardis very high but that does not explain this behaviour.

[G-rooon]

The card is direcly connected to the BNC connection block.

The Analog-Input's of this connection block are the counterparts of that BNC plug, so that you can directly connect that pH-Sensor with that block.

The measurement with an DMM gives me a value of ca. 50mV. With that NI Hardware i get around 2V.

The only explanation i got, was that the impedance of that connection block is by far not that high as the impedance of an DMM.

greetings

[Albert.Geven]

If the parallel resistor on the block are switched off the impedance is higher than most dmm's.

And that proves the 2V to 50mV difference, but joking aside, I don't know what the output really is so is 50mV a correct value for the pH that you are measuring?

[Dennis_Knutson]

Did you configure the acquisition for rse or differential?

[Stephan_D]

Hey G-rooon,

according to my information, the input impedance of some ph-probe are high (in the range of 600 MOhm according to google). The PCI-6024e is designed for low impedance signal sources. You can find this information in the manuel of the PCI-6024e:

E Series User Manual - 370503k.pdf
http://www.ni.com/pdf/manuals/370503k.pdf

Page 43

Use Low Impedance Sources
To ensure fast settling times, your signal sources should have an impedances of <1 kΩ. The settling time specifications for your device
assume a 1 kΩ source. Large source impedances increase the settling timeof the PGIA, and so decrease theaccuracy at fast scanning rates.

You have to take into account, that you sampling rate is very imitated (=slow). This is may not the main problem of your measurement, but a fact which is maybe important, too.

Regards, Stephan

[Intaris]

What kind of pH sensor do you have?

There are the old analogue types which can have impedances of up to 1 Gigaohm (increasing exponentially at lower temperatures) and there are the modern pre-amplified versions with a few kiloOhm impedance.

I would claim that you will need some kind of signal conditioning for the former whereas the latter should work OK.

On to the next point.  What solution was your pH electrode placed into for your measurement?  You are aware that pH measurements in air will drift unpredictably?

Full disclosure: I developed pH electrodes for 5 years.

[G-rooon]

Frist of all, this is the pH-Sensor -->pico DrDAQ® pH-sensor DD011  ( http://www.conrad.com/ce/en/product/128670/pico-DrDAQ-pH-sensor-DD011?ref=searchDetail )

Secondly, I want a pH-measurement of saline water.

My presumtion was that there is a problem between the high impedanz of the sensor and the and the low impedanz of the PCI-Card, too.

But this does not explain the decline value ...

However thanks for all the support I got here so far!

[Intaris]

OK, so we have a conventional pH electrode.  Input Impedance will be (based on experience) around 200-400 Megaohms.

If you measure with a dMM and you're getting 50mV, this corresponds to approximately pH 6, which given your extra information of "Saline water" seems about correct.

2V would mean a pH value of 40 which is physically impossible.

As such we can safely conclude that the measurement values you are receiving over the NI hardware are definitely wrong.  Please check the differential mode and input impedance of your NI card.  If needed, there are some relatively cheap voltage followers on the market which can handle up to 1GigaOhm impedance (which your sensor will exceed at approximately 10°C, a bit above freezing point.

I think the declining value is simply a result of an incorrect measurement set-up and cannot be attributed to the sensor in itself (unless it's broken of course but then it wouldn't work with the DMM either).

[Intaris]

On a side note, measurement of pH in saline solutions is not as trivial as it sounds.

--Warning--

Information based on years of pH electrode research and develpoment follows

--Warning--

A major problem with "low cost" pH electrodes is that their glass membranes are not optimised for such environments and their reference electrode liquid junction (diaphragm) are (to be polite) sub-optimal.

Traditional wisdom is that the higher the resistance of the solution, or the higher the resistance of the electrode the harder it was to measure and get stable readings.  The latter point relates to the sub-optimal equipment people tended to use about 30 years ago.  With modern electronics, a pH electrode with 1 Gigaohm can be measured without any great problems these days.  The membrane resistance as a measure of reaponse time / accuracy is no longer any where near as important a factor as it used to be.

The other point is the resistance of the solution.  Solutions with low conductivity have high resistance.  If a low conductivity solution would guarantee fast and accurate measurements, then pH measurements in saline would be easy.  They are not.  Chances are ou will see very similar behaviours between measuring in pure water and measuring in 3M KCl.  Although the reference electrode has less to do in a 3M KCl solution than in pure water (in pure water even small disturbances in the reference electrode can lead to large measurement errors, in 3M KCl much less so) the glass membrane still has a lot to do.  This is because it is not actually the resistance / conductivity / ionic strength of the solution which guarantees accurate or fast measurements but rather the buffer capacity.  There is a small link between solution ressitance and buffer capacity in that a solution with good buffer capacity will have a certain conductivity but the reverse is not true.  The buffer capacity of pure water and of 3M KCl is almost the same (near zero).  We see this manifest itself around pH 7 where both H+ and OH- ions are present in equal quantities.  If you look at the combined concentration of H+ and OH- ions across the pH range we see that it reaches it's lowest value at pH 7 (total concentration is 2*10^-7M).

What this means for the electrode is that dynamic processes at the surface of the pH electrode become the limiting factor as the ability of the solution to provide ions to the pH membrane is extremely limited.  Only pH electrodes specifically designed for this area of measurement will have a chance of giving an accurate reading.

So, that's my rant finished.  I spent years developing electrodes for exactly this purpose and it's amazing how little other manufacturers actually understand the processes going on during such a measurement.  If you want a really good electrode for this kind of measurement, either the GK201C from Radiometer or the Aquatrode plus from Metrohm are good choices.  These will set you back more than your electrode from Conrad but will (assuming you can solve your electronics problem) give far better results.  without knowing the electrode you have purchased, my experience would tend to suggest that you'll have an error of approximately 1-2 pH units in saline if the real pH is anywhere near pH 7.  The Radiometer will reach an error of approx. 0.5 pH units whereas the aquatrode plus should get to within 0.1-0.2 pH units irrespective of the actual pH value.

There's a great ISO norm calles ISO 10523 which outlines some very weakly buffer pH solutions which are a good test for your electrode.  They test exactly this aspect of performance whereas standard buffer solutions have such a high buffer capacity that they work fine with ANY electrode.  My favourite is 0.05mM NaH2CO3 which is left to equilibrate with atmospheric CO2.  It reaches a pH of 7.005 with an extremely low buffer capacity.  It's great for separating the men from the boys in this regard.