PH Problems

A

Thread Starter

Alex

I currently have a situation where there are two PH probes in a waste water tank. The first probe measures the PH of the liquid within the tank, while the second is used to record the PH of the liquid leaving the tank. When both of the probes are removed from the tank it is easy to calibrate them using buffer solutions of PH 4, 7, and 10, but when both probes are placed back into the tank, thay will both read the same PH, even though it is way off. They will read a PH of 11, but the PH will actually be 5 on a hand held calibrated PH meter. At first I thought this was a simple problem, the PH probes are connected to a device that converts the signal to 4-20ma so the PLC can read it. I thought for sure that this device was wired incorrectly, or not properly grounded. Well here I am three days later, still looking for answers. Any suggestions you can give me will be greatly appreciated. The tank is all plastic, with all plastic piping. I have heard theories of transient voltage within the liquid, how would I eliminate this? Is there a specific probe I should be using?
 
D

david mertens

Be sure to have both your reference probes properly earthed to a clean earth with minimal resistance. As the signal you are measuring is a
milivolt signal, even the smallest equalisation current will render the measurement useless. Also provide an earthing for your product as it
can build up static electricity in a plastic vessel.
 
B

Bob Peterson

> > Erroneous readings from PH probes in a plastic tank...
>
>
> Be sure to have both your reference probes properly earthed to a clean
> earth with minimal resistance. As the signal you are measuring is a
> milivolt signal, even the smallest equalisation current will render the
> measurement useless. Also provide an earthing for your product as it can
> build up static electricity in a plastic vessel.

I second this suggestion. This is really critical with pH readings. The best solution is probably to screw the pH cell into a metallic sample cell, and make sure you ground the sample cell. Use a small sample pump to pump liquid out of the tank and through the sample cell back into the plastic tank. Bond the metallic sample lines to the sample cell.

Another problem you may be having is that the cell may be drying out. Once it drys out it may take a while before it re-hydrates and you get a
reasonable reading. Some cells don't ever really recover from being dried out. The best answer is never let them out of the liquid. The pumped sample line is a good solution for this problem as well.

Bob Peterson
 
C

Curt Wuollet

Sounds like you might want to try a guarded probe with the guard either driven or grounded depending on the circuitry. Does putting just one of the probes in the tank work OK? This would be the way to diagnose it they were interacting or if some outside factor is affecting them. If each works but both togeher fail the guard idea might help. If they don't work individually then you probably do have stray current from someplace.else. The idea behind the shield is to shunt the current to ground or provide a voltage close to you measurment voltage to reduce leakage. With everything plastic the only return path for stray current is through your sensor.




Regards

cww
 
W
You are undoubtedly picking up transient voltages within the liquid. Try separately grounding each electrode. In addition, you might try using completely separate E/I converters, with isolated retransmitters for each sensor loop. In all likelihood, the E/I converters are not isolated and you are running a ground loop through the PLC.

It sounds like you are using a sensor-E/I converter-PLC system. Have you considered actually using one of the many excellent pH controller/transmitters on the market? They are relatively inexpensive and there are over a dozen manufacturers including Horiba, Rosemount, E+H, GLI, Quantim, ABB, GFSignet, and a host of others (don't stone me if I didn't mention your name, folks, there are a lot of you). One of the benefits these devices give you is on board calibration, diagnostics and retransmission capabilities so you don't have to program them into a PLC.

Walt Boyes

---------------------------------------------
Walt Boyes -- MarketingPractice Consultants
[email protected]
21118 SE 278th Place - Maple Valley, WA 98038
253-709-5046 cell 425-432-8262 home office
fax:801-749-7142 ICQ: 59435534

"Strategic marketing, sales and electronic
business consulting for the small and medium-sized
enterprise: http://www.waltboyes.com"
---------------------------------------------
 
R

Richard King

Hi Alex,
I would try to simulate the problem by running a conductor from the buffer solution to the process, if the reading goes bad then you will know what the problem is.

On the other hand, the probe and instrument should be galvanicly isolated, so transient voltages should not be a problem (if you have grounded the instrument then that might be your problem).

Another reason for a false pH reading could be placing the probe in a fast flowing fluid (eg. probe to close to the inlet). Some probes, by design, would give better readings in fast flowing fluids.

(Is your process reacting with the solution in your reference electrode ?)

If the tank is plastic then earth the process somehow, the instrument will not like high static voltages.

I hope this helps,
Richard King
 
V

Vijay V Harne

I would suggest you to check the distance of cable between electronics and pH probe. It should not cross the recommended length. Also check whether the power signal (110VAC etc...) running parallel to the this cable. The error may cause due to Electro Magnetic interfearance. We had similar problem once.

Best luck.

Vijay
 
A

Al Pawlowski

My experience is that stray currents cause erratic readings rather than a constant, set one. You say that the units calibrate correctly out of tank. If you are disconnecting their outputs from the PLC to do that, maybe you
have encountered the old "trying to drive mulitple common-connected inputs with non-isolated transmitter outputs" trap. This can be solved by isolating, or "floating" (when possible), the transmitters or isolating the
PLC inputs.
 
Alex,

Maybe you have a "ground-loop" problem. This is a somewhat common problem with pH probes. Talk with the manufacturer of the meter about this possibility.

D.Best

===============================================
 
Well if there is one thing I have learned in instrumentation, it's start with the basics...
for example.. check 4-20 output at device then check at plc input. Also check scaling at PLC.
 
Top