Log in

View Full Version : Another reason not to trust test kits



matt
Thu, 11th Sep 2003, 08:44 PM
I've been tweaking my reactor for the past several weeks, trying to get the effluent ph lower and steady. The reason for this is that although my tank looks very healthy, and alk is a steady 9.0 dkh, my calcum has been reading anywhere from 360-400 with the "kwik kit" test that I got from Premium Aquatics when they were out of Salifert. I couldn't understand why my calcium wasn't higher, especially because my reactor (one that I made) holds a full gallon of ARM, and the korallin reactor "rated" to 300 gallons holds less than half that. Anyhow, I finally got a salifert Ca test kit in and it reads 460+ PPM. Soooo....

The good news is that my reactor, which is much smaller than the one I brought to the meeting, is obviously having no problem with my 100 gal sps tank. The bad news, once again, I've been trying to deal with a non-existent problem due to a faulty test kit reading. This happened to me with ph until I got a probe, and alk (with the aquarium systems kit) until I got the salifert carbonate hardness kit. I sure hope the salifert ones are accurate!

This means I can run a slightly higher reactor ph (6.8) and still get plenty of Ca in my system.

Richard
Thu, 11th Sep 2003, 10:05 PM
I have to agree with Joshua on the red sea test kits. A rep gave me a kit to try at the store and I really don't like it. The alkalinity test gives a reading of Low, Medium, High...how crappy is that. I've had good luck with kits from Salifert, Aquarium Sytems, and even the Hagen kits are pretty accurate although their ph kit is hard to read unless the lighting is just right. I also bought an expensive Ca kit from Lammonte and do not like it.

matt
Fri, 12th Sep 2003, 01:07 AM
No, I mean don't put too much stock in a test kit reading if it seems at odds with what you can observe in the animals, unless you can confirm it with a different, higher grade test. Of course, this takes experience and familarity with a given set of animals.