Need help with replacing sampled H2O.

[arking_mark]

I would like to use my 4th doser pump to replace saltwater used by the KDH.

How do I figure out the amount to replace?
The sample size is set to 65ml
The sample tube volume is 33ml
The measured waste is ~85ml
The sample pump draws 110ml and then returns unused portion
The reagent pump seems to use about ~1ml

By my calculations h2o taken from tank should be 110 - 33 = 77. However, based on waste quality it should be 85-1 = 84ml.

What am I missing?

Also, can the doser pull from ~50ft away?

[Thilo]

Before each measurement the measuring cell is rinsed with 10 ml sample water. So the total amount per measurement is sample volume + 10ml.
To the waste water, the dosed reagent from the rinsing process and the measurement is added. Further deviations are caused by inaccurate calibration and sample tube length.

The doser can dose over several feet. How far exactly also depends on the height. It would be best if you tried it out.

[arking_mark]

I'm not sure your # is correct. I'll recalibrate the pumps, but I'm pretty sure I have them calibrated well.

Based on your answer, the total water draw is sample size plus 10ml, which would put me at 65 + 10 or 75ml.

This is very close to the actual pump draw which is 110ml minus the tube volume of 33ml which is 77ml. (Close is not good for such an expensive unit)

The fact that there is ~85ml of actual waste water produced is way more troubling to me. ~1ml of reagent is used which means that it actually pulled ~84ml of sample water.

I'll recalibrate and re-measure everything to be certain its set up correctly.

There is one additional thing I noticed. In my setup, the filter hangs at an angle and never completely fills with water. If it volume is completely included, could that account for the extra water. Does the filter need to be completely vertical? If so which way?

Sent from my SM-G988U1 using Tapatalk

[Thilo]

Yes, that might be the problem. It is best if the tube + filter are installed continuously upwards.
You can also measure the sample volume by filling the tube (and filter) completely and then emptying it into a measuring cup to measure the volume.

[arking_mark]

I'll fix the filter so its vertical. I'll post my results.

Sent from my SM-G988U1 using Tapatalk

[arking_mark]

Ok. So here is what I did and my results.

1. I re-ran and re-measured the sample tubing to ensure the filter is verticle. This resulted in 102 inches of tubing. The calculated sample tubing volume with filter is is 40ml.
2. I then just ran the test, which resulted in an initial sample draw of 52ml followed by a waste of 8ml
3. Then the sample pump did a 65ml draw (= to my set sample size) followed by a waste of 75
4. The reagant draw was 1ml

Expected waste would be 12ml + 65ml + 1ml = 78ml actual was 5ml greater than this.

Since these numbers don't add up, I pulled the sample tube off of the KH Director and did a measured dose of 40ml (the calculated volume) and got 5ml dosed into a graduated cylinder. I'm assuming that the 5ml extra is because that calculation is wrong. To be sure the pump was calibrated correctely, I did anoter dose of 12ml which it hit exaclty, so that doesn't seem to be the issue.

This 5ml is exactly the missing volume of waste.

So it would seem that the device pulls 12ml for the initial clean and 65 for the sample volume.

I assume that I just need to correct the sample tube volume to 35ml to get a more precise KH measurement and the expected 77ml.

Is this correct?

[arking_mark]

So after making the above adjustments, we were still off by 3ml. Expected waste was 79ml and actual waste was 82ml.

So I filled the sample tube and then completely emptied it which resulted in 38ml. I then recalibrated the pump and it went from 37.5ml to 39ml...and reran the test. I'm not sure why the pump calibration resulted in a different number...maybe it's wearing in as I've only had it a week.

Test results coming shortly....

[arking_mark]

...and finally I got expected waste of 77ml plus ~1-2ml Reagent!! Running it a 3rd time to confirm.

[arking_mark]

...and 3 waste readings that are perfect.

One other note, with all these tests, the KH only varied from 6.8 to 7.2.

[Thilo]

Okay, great. Thank you for sharing your findings and solution.