[HeavyUser]

For whatever reason my GH has risen from 6*s to 7*s. I have a 20 gallon tank and do 5 gallon water changes every 3 weeks or so.

I was thinking I could mix up a batch of distilled at a GH * of 5, this should lower down the total GH kby 1* leaving me at an optimal level of GH.

My question is will the water change stress the CRS due to the fact its not 5* GH instead of the normal 6* GH I do water changes with?

 

[latchdan]

Do you top off with ro water or tap? You could add ro or distilled and take out some to lower gh. Although, one point isn't much unless your just worried continuous increasing of gh.

 

[HeavyUser]

I think the only reason it has raised slightly is because I bought a handful assassin snails and there a TON of empty snail shells building up in my tank. I'm not too worried about it continuously raising as I have started sucking out the shells as they build up.

Your idea should work, I'll just siphon off a gallon and replace it with some fresh distilled.

 

[Diana]

A change of 1 degree is not much, either up or down, but as suggested above, find out the reason.
If the other parameters are holding just fine with 3 week intervals between water changes, then simply topping off with distilled or RO should help, or small exchanges as you are doing.

A drop from 7 to 5 would be significant. I know some of the more delicate fish would not handle that much change in just one event. Better to split up this much change over 3-4 water changes over a couple of weeks.

GH is a simple math formula. If you remove 50% of the water at 7 dGH and replace it with 5 dGH then the net result will be a full tank of 6 dGH. It might not actually do this for several reasons, such as if the water change was not actually 50%- perhaps you estimated the tank volume incorrectly, or took out too much or too little water. Another reason is the test kit. API for example uses drops and reports the reading in German degrees of hardness, and depends on you to read the color change correctly. Any of these are subject to error.