I see that Randy has started a new thread to pull out the fuel mix issue from the FP 40 Mods thread. So at the risk of offending someone by dropping my prior post into this one.... Randy has obviously provided the answer. But perhaps my example will help illustrate what we are both saying.
This came up (again) when Gil asked for a tutorial on why measuring by weight was different than measuring by volume. Here was my answer from the other thread.
Here is the refresher course--
The density of the three ingredients are all different. If they were the same, then there would be no difference between a percent by volume mix and a percent by weight mix.
The densities are as follows: Nitromethane 1.14 gm/cc; Castor 0.961 gm/cc; methanol 0.79 gm/cc
For reference, the density of water at 25C is 1.00 gm/cc
We assume that the ingredients are not miscible, thereby screwing up volume-after-mix observations or calculations. (edit--I meant to say that the three constituents, when combined, do not sum to a lesser [or greater] volume than the original parts. I can't remember the correct term for that. Chemistry was a long time ago....)
Example
For a 10N, 22C, 68Al fuel mix by volume, the mix by weight would be 13.22N, 24.51C, 62.28Al. So if you are accustomed to thinking in volume measurements what has been working well for you, you can see that you would have to specify a higher oil content percentage if it were communicated on the basis of weight. Else, you would be a few percent short.
Conversely, if you used your "normal" percentages (volume) but your fuel mixer assumes you meant by weight, because that may be how he does it, then...
A 10N, 22C, 68Al fuel mix by weight, results in a 7.45N, 19.44C, 73.11Al mix by volume
Your engine would not run exactly the same. As you run a lower and lower oil and nitro percentage, this unintended "reduction" might be critical and cause damage.
Note that the alcohol constituent is by far the least expensive of the three. A cynic might say that a fuel mixer would make his product based on weight, but be vague about the basis, because it lets him put a percent for nitro and oil on the bottle that make you feel good, but reduces his costs significantly.
Note that it does not matter how the manufacturer is blending; in other words whether he uses a flowmeter or a scale or a whatever (as long as it is accurate) but we need to know the basis for the percentage given. I think all of us using the fuel assume that the percentage is by volume. But if the peerson blending does not follow this convention, and it isn't stated on the bottle, then we are not going to be happy with the results.
I thought that I could get to the bottom of this by looking up the Manufacturer's Safety Data Sheets (MSDS) which are required for chemical products. Unfortunately, the ones I have looked up did not give the basis of the percentage which surprised me, and further that most are also written giving a huge range in percentage of the ingredients. So they are not useful for our purpose here.
That leaves you with the same evaporative measurement testing that has been described in other posts. Unless you are doing large samples, or have very accurate measuring capabilities, you probably are not learning the true composition of your fuel.
Hope this helps,
Dave