Jump to content

As of July 17, 2015, the LabJack forums here at forums.labjack.com are shut down. New registrations, topics, and replies are disabled. All forums are in a read-only state for archive purposes.

Please visit our current forums at labjack.com/forums to view and make new posts. To post on the current forums, use your labjack.com login account. Your old LabJack forums login credentials have been retired. There are no longer separate logins for labjack.com and LabJack forums.


Photo

How to use pH probes


  • Please log in to reply
23 replies to this topic

#1 Indy500

Indy500
  • Members
  • 26 posts

Posted 28 November 2009 - 03:18 PM

How to use pH probes

I connected several pH probes to my U6 and I wanted to document what I learned to help others in the future.

Background

The sensitivity of an ideal pH probe at 25C is 59.16 mV per pH unit. An ideal pH probe at 25C will produce 0 V when placed in a solution with a pH of seven. An ideal pH probe at 25C will swing from -7pH x -59.16 mV/pH= +414.12 mV (pH 0 strong acid) to +7 pH x -59.16 mV/pH = -414.12 mV (pH 14 strong base).

The temperature sensitivity of the ideal pH probe increases with temperature. It increases at the rate of 0.198 mV/pH for every C above 25C.

The source impedance of a typical pH probe is very high because the thin glass bulb has a large resistance which is typically in the range of 100 MΩ to 10 GΩ.

The probe

I did extensive research and testing and settled on two different probes: the "American Marine Pinpoint pH Probe", and the "Milwaukee Ma911b/2 Ph Electrode". Both can be easily found on the web for about $38.

Connecting the probe to the U6

First off, since most of my measurements were close to 25C, I ignored the effects of temperature. I only needed a resolution of a tenth of a pH and that was well within the temperature effects I would experience.

The two biggest problems to deal with are the source impedance of the probe and the small voltage range of the probe.

I used a U6 set at +/-1V and a resolution index of 8 which gave me an effective resolution of 24 uV - way more that I needed to measure the pH probe accurately.

I first hooked up the pH probe directly between ground and one of the AINs and started measuring with the pH probe in a calibration solution of 7.00 pH. When I graphed the data, I had an erratic slow sine wave with a period of about two minutes superimposed on my data. That told me right away that I needed a buffer between the pH probe and the U6 because the input impedance of the U6 wasn't high enough for my particular pH probe. I came up with two easy ways to buffer the pH probe.

The quick, easy, and expensive solution is just to buy a pre-assembled buffer made for pH probes. I recommend the PHAMP-1 self-powered pH amplifer for $59. Information is at:
http://builditwithne...nmntl/PHAMP.htm

I was cost constrained, so I decided to build a buffer. I used a Maxim MAX407CPA op amp because it would do unity gain and because you get two op amps in one IC which made it convenient to use on the LabJack LJTick-Proto. You can use any unity gain very high impedance op amp from any manufacturer. You can actually add gain with an op amp and get a higher voltage swing, but that adds resistors to the design. With the U6 I was able to use unity gain and eliminate all other components except for the MAX407. The U6 also provided me the +/- voltages I needed to power the op amp.

The only issue to be concerned about during construction is that with the very high input impedance on the MAX407, you need to use flying lead construction on the two input leads. This technique is described here on page 4.47:
http://www.analog.co...b_Ch4_final.pdf

By the way, Maxim will give you a sample of the MAX407 if you sign up for one on their website.

Calibration

The last thing you need to do is to calibrate your pH probe. Every pH probe is different and will respond slightly differently from an ideal pH probe. Calibration removes these variations and makes your real world probe respond like an ideal probe.

Calibration solutions of 7.00 pH and 4.01 pH are standard and easy to get. Basically, you put the probe in each solution and record the voltage. This gives you two points on a line and with that you can calculate the slope and intercept of the line:

slope = (7.00 - 4.01) / (VoltageAt7.00 - VoltageAt4.01)
intercept = 4.01 - (slope * VoltageAt4.01)

Once you have the slope and intercept, you can calculate the pH from any InputVoltage:

pH = (slope * InputVoltage) + intercept

I hope this helps someone in the future that needs to use pH probes with the LabJack!
-David Spoelstra

#2 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 14 December 2010 - 11:31 AM

We have used the EA-BTA electrode amplifier from Vernier for $40:

http://www.vernier.c...bes/ea-bta.html

Ours is a little plastic box, so it looks like they changed the packaging, but otherwise I think it is the same. We cut off the RJ45 connector and pigtailed 3 wires for VS (5 volt supply), ground, and signal output.

#3 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 22 August 2012 - 01:00 PM

Pinouts for the Vernier EA-BTA here:

http://www.vernier.c...sensor-pinouts/

Black = GND
Orange = +5V Input
Red = Signal Output

#4 AJ123

AJ123
  • Members
  • 3 posts

Posted 04 September 2012 - 09:53 AM

I have a U3-HV and I want to be able to log pH and DO with it, can I just follow what is posted here for the U6 and have the same results or is it something different entirely?

#5 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 04 September 2012 - 01:47 PM

Pretty much the same info. With the U3-HV, the decision on whether to connect to a high-voltage or low-voltage analog input would depend on what signal range you want to handle.

Note that the EA-BTA amp ties the shield of the BNC (connected to reference electrode) to ground, so if you use 2 electrodes in the same liquid, you are effectively shorting the 2 reference electrodes together, and will have problems.

One solution to this is to use 1 U3 (and 1 EA-BTA) for each probe, connected as described above, and put an isolator on each U3 to keep the probes isolated from each other. Here is a $38 isolator for the U3:

http://microcontroll...roducts_id=3769

A possible solution using a single U3 and single EA-BTA, is to have the U3 control mechanical relays to connect a different probe to the EA-BTA ... one at a time.

#6 AJ123

AJ123
  • Members
  • 3 posts

Posted 06 September 2012 - 07:40 AM

I am looking at only having one probe in a solution at a time and looking at controlling pH in the range of 4.5 to 10. I am not sure what that is in mV but that is what I am looking at doing. In that case would i still need the extra voltage isolator?

#7 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 06 September 2012 - 07:48 AM

If you only have 1 probe in the solution at a time, then no you don't need to worry about isolation.

#8 AJ123

AJ123
  • Members
  • 3 posts

Posted 13 September 2012 - 08:15 AM

Pretty much the same info. With the U3-HV, the decision on whether to connect to a high-voltage or low-voltage analog input would depend on what signal range you want to handle.


I am not sure which I need to use high-voltage or low-voltage, I will be looking at pH's all across the board as low as 1 and as high as 12. I am not sure what that is in mV. I am also looking to get the highest level of accuracy that I would be able to obtain.

Also I bought the EA-BTA from and am wiring it up and would just like to confirm that the black wire goes to GND, orange wire goes to VS and the red wire will go to which ever AIN I am looking to use( in this case AIN0)

#9 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 13 September 2012 - 10:08 AM

Your connections between the EA-BTA and U6 are correct. Seems a little strange, but their pinouts page confirms that orange is power is red is signal. A standard pH measurement electrode provides a raw output of +/-420mV compared to the reference electrode. The EA-BTA amp inputs accept -450mV to +1100mV, so it is good with the standard pH input signal. The EA-BTA has a gain of x2.2 and and offset of 1.20V. That means the +/-420mV raw signal from a pH probe should get converted to to 0.28V to 2.12V by the EA-BTA. That 0.28-2.12V output from the EA-BTA is perfect for a low-voltage channel on the U3, so if you were using a U3 you would use FIO4 perhaps. On the U6, just use any AINx channel with the default +/-10V (BIP10V) range and you will get excellent results.

#10 alkaram

alkaram
  • Members
  • 18 posts

Posted 15 May 2014 - 07:59 AM

I'm trying to monitor pH and CO2 within a solution at the same time using electrodes with the U6. I had a question about the response below: 

 

Pretty much the same info. With the U3-HV, the decision on whether to connect to a high-voltage or low-voltage analog input would depend on what signal range you want to handle.

Note that the EA-BTA amp ties the shield of the BNC (connected to reference electrode) to ground, so if you use 2 electrodes in the same liquid, you are effectively shorting the 2 reference electrodes together, and will have problems.

One solution to this is to use 1 U3 (and 1 EA-BTA) for each probe, connected as described above, and put an isolator on each U3 to keep the probes isolated from each other. Here is a $38 isolator for the U3:

http://microcontroll...roducts_id=3769

A possible solution using a single U3 and single EA-BTA, is to have the U3 control mechanical relays to connect a different probe to the EA-BTA ... one at a time.

 

Does this mean that I can't connect the two probes to two individual EA-BTAs and then connect both of these to the same U6? 



#11 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 15 May 2014 - 10:39 AM

The EA-BTA amplifiers will short the reference electrodes together, but they likely need to be at slightly different voltages, so you might have problems.  To avoid that possibility, you can connect 1 EA-BTA normally, but for the other you need to do 2 things:

 

1.  Power the amp (+Orange/-Black) from an isolated 5V supply rather than VS from the LabJack.  Any AC/DC wall-wart will be isolated, so you can use any extra USB charger you might have around.

 

2.  Connect the signal (+Red/-Black) to the U6 in a way that maintains some isolation from GND:

 

    2A.  I suspect it will work if you connect +Red/-Black differentially to AIN0/AIN1, and then add a resistor from AIN1 to GND.  You want a big enough resistor to maintain some isolation from GND, but small enough to hold the signals in the proper common-mode range.  With the U6, 1M will probably do the trick, but my guess at a range would be something from 10k to 1M.  When you try a 10k or 1M, look at 2 things:  does the signal voltage stay the same whether you have just this probe connected or both probes connected, and does a single-ended measurement of AIN1 return something reasonably close to 0 (+/-1V is probably good ... 10V is pretty bad).

 

    2B.  Put an analog isolator between the signal (+Red/-Black) and the U6.  One option is a 5B31 module from Analog Devices, for which you might want the 5B03 or 5B04 mounting board.



#12 alkaram

alkaram
  • Members
  • 18 posts

Posted 18 June 2014 - 07:30 AM

Hello, I have a few more questions about this setup.

 

1) Will shorting the electrodes together harm the sensors? Or will it just disrupt the readings?

2) An earlier post mentioned pigtailing the wires after cutting off the RJ45 connector. Is pigtailing primarily used here for extending the wire length? Or is any other special configuration involved?

3) In the above post, how do you safely split the black wire to connect it both to the power supply and LabJack without exposing the metal? Would pigtailing also be used in this instance? 

 

THANK YOU! 



#13 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 18 June 2014 - 08:20 AM

1.  No harm.  Just disrupted readings.

 

2.  In post #2 we say that we cut the RJ45 off the EA-BTA and pigtailed the wires.  We just did this so we had bare wires we could connect to the screw-terminals on the LabJack.

 

3.  I assume you are talking about doing steps 1 & 2A from post #11.  That means you want 3 things connected together:  Black wire from the EA-BTA, negative of the external power supply, and AIN1.  You can do that however makes sense in your setup, but one convenient way would be to clamp the black wire from the EA-BTA and the negative from the power supply both into AIN1 (along with one end of the resistor to GND).  If you are not clear, draw a diagram in Google Drawings and post a link here so we can look at it.



#14 alkaram

alkaram
  • Members
  • 18 posts

Posted 20 June 2014 - 11:52 AM

Thanks! That makes sense. And just to be sure I am understanding everything correctly, is the drawing linked below correct?

 

http://imgur.com/IQiZHZL 



#15 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 20 June 2014 - 12:23 PM

That link does not work for me.



#16 alkaram

alkaram
  • Members
  • 18 posts

Posted 25 June 2014 - 01:59 PM

 Sorry about that.

 

This link should work: http://imgur.com/Fogl11d



#17 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 25 June 2014 - 02:36 PM

Looks good.



#18 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 01 July 2014 - 10:46 AM

Omega is selling amps for $69 or $79:

 

http://www.omega.com/pptst/PHTX21.html

 

The battery-powered PHTX-21 should be useful for applications with multiple probes, as you don't need an isolated power supply for each probe.  You would just connect the +/- output to a differential analog input with a resistor from the negative to GND.  See post #11 above.



#19 alkaram

alkaram
  • Members
  • 18 posts

Posted 16 March 2015 - 10:38 AM

Hello,

 

I'm back with another question regarding pH sensor setup.  I purchased the omega amp for connecting the pH probe. It worked great! But now that we have our whole setup within a small incubator, I think there is electrical noise coming from a high-power LED lamp that is located close to the probe (i.e. the probe is now constantly bouncing back and forth by about .05 pH units).  Would the best way to alleviate this problem be a low-pass RC filter? Or would there be a better option? 

 

Thank you again for all your help.



#20 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 16 March 2015 - 10:45 AM

If you turn off the lamp does the extra noise go away?

 

Can you determine the frequency of the noise?



#21 alkaram

alkaram
  • Members
  • 18 posts

Posted 16 March 2015 - 05:54 PM

I noticed more noise with the light on (or so I thought).  The main lamp is a high power LED lamp so I think its creating noise when generating DC current, much like wallwarts do?  We also have fluorescent lights in the background that could be causing other noise.

 

After doing a FFT of signal at 1200 Hz, it looks like there is a lot of background noise regardless of whether the light is on or off. However, I noticed the bouncing in the pH value more with the light on.  Here is a link to the FFT breakdown of the signal with the light on and off. 

 

 

So perhaps a filter would be beneficial for this sensor regardless of whether the main light is on or off? 



#22 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 17 March 2015 - 07:02 AM

I can see that your noise is 60 Hz (plus harmonics).  Using a U6-Pro or T7-Pro at ResolutionIndex=12 is a quick way to get rid of 60 Hz noise.

 

Do you have 2 probes in your system as described earlier?  Please provide a wiring diagram showing all connections between the LabJack, sensors, amplifiers, ground and power supplies.

 

Do both probes have the same noise, or just 1?

 

If you remove 1 probe, does the other still have the same noise?



#23 alkaram

alkaram
  • Members
  • 18 posts

Posted 17 March 2015 - 09:01 AM

Unfortunately, we just have the U6 device currently.  

 

We don't have the two probes set up together in this setup, yet (trying to troubleshoot this issue first). But each probe has their own isolated power supply from the omega pre-amp. I know the pH probe is very sensitive to any electrical noise. For example,  opening the solenoid valve makes the pH probe go especially crazy. As a result I tested the CO2 probe. I didn't notice as much of an impact with this probe because the electrode is shielded within the metal cap sleeve, which contains the CO2 membrane.  We minimized the solenoid interference issue by placing the solenoid outside of the incubator and wrapping it in foil.  

 

As a fair warning, there is a lot going on in our setup. We are still in the testing phase, so our setup/wiring is not finalized.  Here is a very rough diagram of the wiring. I haven't had time to draw out a nice schematic since we are still figuring out the best setup.  Most of the setup is inside an incubator (which you can think of exactly like a refrigerator with shelves). Hopefully the diagram can give you an idea of our setup.  I apologize because I know it's not easy to follow all the wiring.  On that note, do you have any suggestions for open-source programs to draw these type of wiring schematics?  



#24 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 19 March 2015 - 08:52 AM

Google Drawings is useful for wiring diagrams.  You can share it via link, or can share it with [email protected]

 

https://support.goog...er/177123?hl=en




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users