New pH meter doesn't seem accurate

[h22lude]

I'm using a MW-102 for the first time. I calibrated it using 7.01 and 4.01. I rinse with RO water. I took a mash sample at 15 minutes and chilled it. My first reading was 4.90. I recalibrated and got the same thing. I let it sit during my mash, recalibrated then tested it again and got 5.42. Rinsed the probe and measured it again and got 5.53. I put it in the 7.01 again and it measured fine.

Why am I getting random readings for my wort?

 

[ajdelange]

Make sure, after a cal, that if you put it in 4 buffer it stays at a reading of near 4.00 for at least half an hour and better yet an hour. This is the stability test described in the sticky.

pH meters are very sensitive instruments and it doesn't take much to throw one well off. Such things as loose wires, dirty connector contacts and cold solder joints can all do that as can manufacturing problems with the electrode.

pH can also vary a lot within a mash before it is thoroughly mixed expecially if you have added acid or sauermalz. Mash pH also varies over time.

 

[stz]

I've no idea about your particular meter, but we've got quite a good hannah at work. It wants to be calibrated across 3 reference solutions at worst, 5 is good, but it would like up to 10 and it is noticeably less accurate with 3. Our read also changes, especially if you swirl the sample, mostly with temperature, but also as 'things' happen during the mash.

While ours might fluctuate, it is less than your reading, maybe like between +/- x.x5 so 5.39 to 5.44. We test 10 minutes in at the same temperature for consistency and make acid or bicarbonate additions at that point. It can take a while to stabilise, around 25 seconds or so though we do swirl the sample while it is doing its thing. It has an hourglass and eventually it'll beep and show a tick to indicate it is happy.

I've basically resigned myself that even with a nicer meter it is still just a tool, while suited to the purpose allowing us to take a reference at the same time and adjust following a set process the consistency is more important than slavishly following the meter. I wouldn't want to trust it with 1/100th, we shoot for > 5.3 < 5.5 and adjust as little as possible to get within the range.

 

[ajdelange]

I've no idea about your particular meter, but we've got quite a good hannah at work. It wants to be calibrated across 3 reference solutions at worst, 5 is good, but it would like up to 10 and it is noticeably less accurate with 3.

Something wrong here. IF you calibrate with the 3 standard buffers (4, 7, 10) the 10 buffer reading should have no effect on readings between 4 and 7 as the meter constructs a two parameter (slope and offset for 4-7 and slope and offset for 7 -10) model between the pairs and then uses the parameters which matches the sample to compute pH from the mV and temperature reading.

It is possible to get acuracies slightly better than the buffer accuracy halfway between a pair of buffers as long as the meter reads mV to ±0.1 and temperature to ±0.5 °C. If you can't do this with a two buffer cal (assuming that you are measuring at mash pH) something is wrong.

I wouldn't want to trust it with 1/100th, we shoot for > 5.3 < 5.5 and adjust as little as possible to get within the range.

A brand new Hanna HALO right out of the box exhibited rms accuracy of less than 0.01 in the stability test I ran against it (buffers ±0.01). I did not claim better than the buffers based on this test but rather on mathematical analysis. The test just confirms that analysis.

 

[stz]

A brand new Hanna HALO right out of the box exhibited rms accuracy of less than 0.01 in the stability test I ran against it (buffers ±0.01). I did not claim better than the buffers based on this test but rather on mathematical analysis. The test just confirms that analysis.

It is a Hanna Edge. What you say about two point calibration makes sense, though I didn't know it is calibrating itself by only the two points within the range sampled. I figured it'd incorporate the 10 into the curve somewhere, but that makes sense. It doesn't consider itself calibrated unless we calibrate with 3 points and it is pretty good at nagging us.

We only use it with 4, 7 and 10 for calibration and while I now understand what it is likely doing under the hood better based on what you say, it did seem more accurate when we went to 5 points. Wishful thinking perhaps! We didn't continue with 5 points because we eventually ran out of everything except the common buffer solutions. It claims ±0.2 mV, ±0.01 pH, ±0.002 pH. Getting it working as a basic meter instead of trying to connect to a PC etc was maybe the most difficult part of setting it up!

 

[h22lude]

Thank you both. I will calibrate it again later on and see how long it stays close to 4.01 after an hour. It did seem to be more accurate as the brew day went on. Maybe it needed some time because it was new.

 

[rayg]

pH meters are designed to measure pH in pure or at worst relatively pure solutions that are homogeneous at equilibrium. Trying to get an accurate reading from a non homogeneous solution which also has chemical reactions taking place is a waste of time. Even if there were no reactions going on, you would still have to stir the entire mash to make it homogeneous. Your best bet for an accurate reading is to wait till it's complete and measure the wort, or run small test batches and draw off wort after 5, 10, 30 min intervals. You might also want to rethink why you need the pH to begin with, as more than likely it is not necessary.

Ray

 

[ajdelange]

pH meters are designed to measure pH in pure or at worst relatively pure solutions that are homogeneous at equilibrium.

Where did you ever get that idea? pH meters are frequently used to measure the pH of things like meat and cheese.

Trying to get an accurate reading from a non homogeneous solution which also has chemical reactions taking place is a waste of time.

Not at all. pH measurement can be and is used to determine the progress of reactions. A good example might be the progress of the lime softening reaction.

Even if there were no reactions going on, you would still have to stir the entire mash to make it homogeneous.

Well you are not completely off base. Clearly you want mash to be as homogenouts as you can get it for most meaningful pH readings.

Your best bet for an accurate reading is to wait till it's complete and measure the wort, or run small test batches and draw off wort after 5, 10, 30 min intervals.

Most people stir and measure at five minute intervals or so up to 30 min. Things should be reasonably stable at 20 - 30 minutes though pH is still approaching its final value assymptotically.

You might also want to rethink why you need the pH to begin with, as more than likely it is not necessary.

His thinking recognizes that the difference between mash in which pH is not monitored and mash in which it is is the difference between beer which is OK and beer which elicits an 'Ahh' from the drinker. Multitudes of people on this site and others have discovered this.

 

[ClaudiusB]

It is a Hanna Edge. It doesn't consider itself calibrated unless we calibrate with 3 points and it is pretty good at nagging us.

You maybe able to tell your meter to ignore the third point. I have a Hach sension pH meter and do it all the time. When asked for the third solution I press the return button to tell the meter not required.