Photometers

[combo]

Ok, in full tank shot topic is a war, but lets get back to the subject.

I also made my own photometer, using one very good photodiode (Infrared Rejection Filter Planar Photodiode), operational amplifier and chemicals from Salifert.

While I was using zeolites I also tested phosphates.

- regarding resolution: 0,01V because it is not calibrated yet (estimation -200 usable steps)

- regarding accuracy: missing calibration

But, what I learned up to now?

1. Clean cuvettes are the most IMPORTANT (need special chemical for cleaning, otherwise thin film inside gives wrong results, washing with RO water is not enough=

2. CLEAN cuvettes on outside side is also essential

3. It takes 13 minutes that sample stabilizes

4. Comparable I achieved PO4 levels to the Aleksander Girz or Corallen Zuht shop, but that not enough. We can observe colourful SPS tanks with 10 ppm nitrates and low phosphates.

If you are not aware of rules 1. -3., you can get very wrong result - up to 300%.

Calibration is also very tricky and accurate process. I have a bottle of standard solution from Hatch 3.00 +- 0.03 mg/l. I need very good RO water, and then I can calibrate my photometer.

[lduncan]

Some other sources of error can include:

Water turbidity.

Electrical noise within the measurement circuit. LNA instrument amplifiers take a bit of skill and money to build.

Errors in calibration fluid accuracy.

Most other error sources can be minimised by calibration.

Layton

[wasp]

Wow Combo, that's awesome!!

You are a switched on dude!

Layton mentioned water turbidity - this is just a complicated way of saying discoloration of the tank water. In other words, the sample of water you test, may have some color in it, already, before you test.

To get around this problem, if using a Hanna colorimeter, the procedure is that the sample is "zeroed" ( Calibrated ) to the water sample, before the reagent is added. Then the colour change is accurately measured. Does your instrument have the capability to do this?

What is your work? How do you have the skills to make this?

[wasp]

Another questio also, you say RO is insufficient to clean the inside of the curvette. What would you suggest?

( To Combo, not lduncan )

[combo]

your instrument have the capability to do this?

What is your work? How do you have the skills to make this?

It does not. Instead of that I am calculating coefficient between tank water measurement and phosphate test measurement.

You do not need complex engineering skills. This photometer is just a little bit better then the most of schematics you can find on the internet.

here you can see light degradation in the sample:

[combo]

Another questio also, you say RO is insufficient to clean the inside of the curvette. What would you suggest?

( To Combo, not lduncan )

My co-reefer friend gave me one strong chemical normally used in public restaurants to automatically wash dishes. That is so strong, that I destroyed part of my floor. In six months this chemical dissolved plastic bottle and dissolved leather suitcase. But I do not know the name right now.

For certain Salifert chemicals leaves thin transparent film inside curette and that gives wrong results.

[lduncan]

Wow Combo, that's awesome!!

You are a switched on dude!

Layton mentioned water turbidity - this is just a complicated way of saying discoloration of the tank water. In other words, the sample of water you test, may have some color in it, already, before you test.

To get around this problem, if using a Hanna colorimeter, the procedure is that the sample is "zeroed" ( Calibrated ) to the water sample, before the reagent is added. Then the colour change is accurately measured. Does your instrument have the capability to do this?

Turbidity is turbidity. It's not a complicated way of saying discolouration. It refers to the opacity not the colour.

Calibration may help, but the turbidity has the potential to be affected by the addition of the reagents used in the test. Meaning true calibration may not be possible for that error source.

Layton

[wasp]

Thanks Combo!

Please keep us updated on your interesting project.