I am currently developing an app for mixing (Android), and I plan to implement support for weight.
However, this is (or can be) a tedious process. If I want to keep calculations super-accurate, I need to maintain a database for flavor gravities. Also, these data may not always be accessible.
One option would be to just specify base type (PG, ethanol etc), and then use that as a gravity indicator for the flavor. This will not be 100%, but maybe 90% accurate in most cases.
But, the way I see it: The main point of doing this, is to be able to reproduce the product exactly next time you mix. Therefore, "molecular accuracy" is not necessarily that important.
Also, the difference would probably no be big enough to do a blind test.
Of course, this is a nice excuse to justify laziness :D However, I will try to maintain a list of flavors and add specified gravity if available.
What's your thoughts on this ?
There is no point in trying to measure specific gravity of every flavour on the market, because the manufacturers change their ingredients from time to time, based on costs, availability and performance. You can expect a given brand to produce a similar quality of end product over the long haul, but that isn't to say their ingredients or their proportions will be carved in stone.
The best way to do this is to use some default value such as the weight of PG, and allow users to change it if they wish.
The standard methods seems to be using the weight of PG for flavorings, because as you said, it creates reproducible results and avoids the tedious process of hunting down and entering the specific gravities of hundreds of flavors.
You may consider adding an option for non-PG based flavorings, allowing users to use VG's weight or enter a specific grams/ml.
I was playing with the idea of starting a similar project but with an eye towards compatability with the recipe calculator from e-cigarette-forum. I really like that calculator and it's quite fully featured, I just wish it had a mobile interface, at least a way to view recipes on a mobile device without taking screenshots and throwing them in Dropbox.
Yes, I will definitely let the user specify base type (PG/alcohol/water) for each flavor. I will also supply a list of the most common flavors and their base type.
In addition, I could let the user specify a more accurate gravity if they wanted to, and use the base type gravity as default.
I measured most of my flavourings then put the results into my calculator but occasionally i notice that some aren't quite as accurate as i'd like. But i wasn't too worried as i had the approach you're talking about - it doesn't matter so long as you can reproduce the same ratios.
Tbh though, if someone did all the work of properly measuring gravities i would rather use that data, so at least i know it's accurate as it can be.
So, yes, i would happily trade my laziness for yours and know all the gravities are correct. :)
I work in a lab and have access to both accurate glassware and 0.1mg analytical balances. I already have my small library of flavors that I'd love to do this with (more so I could figure out how many drops per mL individually), but I don't like wasting stuff just for my own curiosity. I could probably give up a mL or so of each flavor for you over some weekend. I'm concerned about viscosity though in a small pipette like this one. I'll keep it in mind.
Sounds good and really excited that you are working on this. I agree that your workaround when the data is not available is a good place holder and should be accurate enough.
One thing I can't stand about certain calculators out there is the inconsistency when it comes to picking flavors. I realize the importance of allowing the user to input their own text when inputting a flavor, but over time a list for say strawberry type ripe by tfa becomes seventeen different iterations (I.e. strawberry ripe tfa, strawberry ripe TPA, ripe strawberry the perfumer's apprentice, strawberry (ripe), etc.). If you could find a way around that and had a decent database of recipes and a lively user base I think it could be amazing.