Sorry to knock the dust off this one, but I had an idea that I wonder if anyone has ever tried in these situations.
Since the drops are simply reacting with the nitrites (or ammonia or any other compound measured in ppm) in the test sample, the sample size itself is irrelevant to the actual reaction.
If the readings are too high with 8 drops of the test solutions, it seems like you can adjust the sensitivity of the results by cutting the number of drops in 1/2. Or...use the same amount of drops but with a 10ml sample instead of 5ml. In either case, the reading will be 1/2 of what they would normally be so all you have to do is double the value of the results to know the actual value of the sample.
Does this make sense?
I've done this when measuring chlorine and TA in swimming pools when the normal readings are off the charts and repeatedly come up with accurate results when compared against wide range tests performed by the pool supply stores.
I've never tried it on a pH reading, but since a pH reading cannot logically show more than 14, if you cut the number of drops in 1/2 (or double the sample size), it seems like you'll never logically get a reading higher than 7. Seems to me like it should work. I hesitate only because of the 0-14 pH scale range that doesn't directly correlate to parts of something.