KH Director reading too high - tried a lot of the known fixes

Hi
I've recently setup a KH Director and it is still reporting high dKH. Reports around 2-3 dKH higher than actual value. Here is what I have tried, some of it several times

1. Measured and remeasured all the lines
2. Calculated volume by disconnecting and dispensing into a 50ml graduated cylinder - twice - this was within 2ml of measured/ calculated volume by length
3. Re-calibrated pH probe 3 times - all readings after calibration have been within .2 of 4 and 7
4. Re-calibrated dosing heads twice - each calibration within 0.5ml of each other
5. Checked for air bubbles, multiple flushing and venting reagent
6. I ordered some GHL KH reference fluid and its ran three times 10.2, 10.1 and 10
7. Hanna checker gives me a dKH of 8 (tank water)
8. KHD gives me a consistent 10.3
9. Saltwater is Tropic Marin Pro made up to std strength i.e. 1kg/30litres which should be about 7-8 dKH

I really don't know what else to check, the only thing which is odd is the waste is being reported much higher than is being produced - waste on dosing pump 3 is showing as 1439ml in the container at most its 950ml - this has been consistently high

Thanks

Consistency is the key ... my KHD records between 9.5 and 10 testing twice a day. Recent Triton N-Doc test shows 9.8.

I could never get Hanna tester to agree so stopped using it.

If the KHD is accurate versus the ref fluid then I would be happy

That's the problem, it isnt accurate versus a brand new GHL KH reference fluid, I've now done 4 tests with the reference fluid 10.2, 10.1, 10 and 10.5

Quote:

---

Originally Posted by Row1974

That's the problem, it isnt accurate versus a brand new GHL KH reference fluid, I've now done 4 tests with the reference fluid 10.2, 10.1, 10 and 10.5

---

That's 10, 10, 10 and 10

Quote:

---

Originally Posted by Bob_The_Reefer

That's 10, 10, 10 and 10

---

Yes but it should be 7.5, 7.5, 7.5 and 7.5 (+/- 0.2) - its consistent but it isnt within the tolerance of the reference solution so why would it be consistently high

What sample volume are you using?

Quote:

---

Originally Posted by Bob_The_Reefer

What sample volume are you using?

---

80ml

So you've consumed 320ml of reference solution?

Quote:

---

Originally Posted by Bob_The_Reefer

So you've consumed 320ml of reference solution?

---

looking at the bottle yes, its almost half gone

Then I can only suggest raising a ticket to verify the probe is not faulty .... sorry

BTW my waste output is much less than reported too ... to answer the final question in your original post

ok thanks for your help, someone on reef2reef suggested weighing the fluids rather than using a graduated measuring cylinder - going to try that first

I found a standard graduated cylinder accurate enough to get whole number accurate but it would be interesting to know

Quote:

---

Originally Posted by Bob_The_Reefer

I found a standard graduated cylinder accurate enough to get whole number accurate but it would be interesting to know

---

I've ordered some scales accurate to 0.1ml so I'll post my findings here. I did a rough test with some kitchen food scales and it showed the reagent amount in ml when calibrating the doser to be 4ml (+/- 0.5ml as scales aren't accurate) versus 6.5ml using a graduated measuring cylinder. Scales should arrive tomorrow.

My KHD is 0.7 out.

Here are some Tips to setup

• Put the filter at the end of the sample line in my sump.

• Sealed the line connection on the water filter with a small zip tie.
• I ended up calibrating the pump speeds 4X and took the average and used that for the measurement.
• I moved the water sample intake to a different part of the sump and shortened the line.
• Instead of measuring the sample water tube with a ruler, I filled it with water and then measured the exact ml of water that came out. Definitely a different measurement than measuring the length of line with a ruler.
• Switched to a sample size of 100mL instead of 80mL.


Sent from my iPad using Tapatalk

I used the digital scales accurate to 0.1ml - reagent was off by 2.1ml - around 30%, sample was off by 1ml - around 2% - with the weighed calibration, results for reference fluid were 7.4, 7.5 and 7.2 so its clear the documented process has limitations and perhaps should be reviewed

Vinny posted this

Quote:

---

Originally Posted by Vinny

Hello Oscar,

The correct tolerance when using the KHD and Reference solution is +/- 0.4° dkH (0.2 from KHD and 0.2 from Reference solution).
Given your results, you are within the tolerance range.

If you would like to get the KHD result a bit closer to the 7.5 mark, check again your tube lengths and calibrate your pump heads. When calibrating pump heads, you can use a digital scale to get the most precise measurement; 1g = 1mL

You can also verify your water sample tube volume by temporarily disconnecting the tube going into the KHD and manually filling both ends of the tube. Empty this tube into a digital scale or measuring cup and type-in the amount into the tube volume field in the KHD settings page.

---

In addition to putting the filter at the end of the sample tube I found tiewrapping the filter in a 25mm pvc tee helped the consistency (and kept the filter where I wanted it)

Quote:

---

Originally Posted by Bob_The_Reefer

Vinny posted this

In addition to putting the filter at the end of the sample tube I found tiewrapping the filter in a 25mm pvc tee helped the consistency (and kept the filter where I wanted it)

---

Thanks, I'm happy with it now I've determined the calibration volumes by weight and results are as expected - less happy that I had to purchase microscales and I've used almost an entire bottle of reference fluid trying to work out what was wrong - in my opinion the documentation should reflect the importance of the calibration volume particularly the reagent and that the volume should be determined by weight

I also ordered a set of coffee scales to improve the calibration process for the KHD and the dosers .... I guess it might also come in handy once my IonD arrives