Another way is to find out what your experimenting on is soluble or insoluble in different things. Find one that dissolves everything but what you want left behind , and there might be a gem or two staring you in the eyes. Twinkling.
This is the method I used to remove primarily water soluble contaminants.
Like dredd09 said in his post the idea is to find what your compound is soluble and insoluble in.
In our case we are talking testosterone which is soluble in a solvent but is insoluble in water.
The solvent in our grear is BA but I chose ethanol (everclear) 95% idealy you would use usp grade 100% or 200 proof available online for $60 a pint plus $25 hazmat fee min order of 2.
The other ingredient is chilled distilled water.
I found 7ml of ethanol per gram of testosterone to do the trick to disolve it.
Some mild heat may be required as well 180F or so.
Chill in fridge or freezer 10ml of distilled water per gram while disolving test.
Once disolved and all crystals are gone leaving a clear liquid allow to cool somewhat then
using a 20ml pin or larger draw up the chilled distilled water.
Then slowly, very slowly dribble the water into the flask along the inside edge.
You should see white crystals from as the water comes into contact with you solute.
The testosterone is falling out of the solution or precipitating out of the solute.
Any water soluble substance will remain in the liquid.
Continue to slowly add water till the test no longer precipitates out of solute.
Using a coffee filter and another beaker filter the clumpy white powder out of the liquid.
Then using more distilled water rinse the crystals a few more times to get any of the ethanol off.
Allow to dry, for test esters with a higher melting point this can be done with heat as long as it is well below the melting point.
My results so far have been a 65% yield on some test E and a 95% yield on some test p
Obviously there was additional water soluble products in the test E and a little bit in the test p
Now there can still be contaminants in them that have the same properties of testosterone
soluble in solvent and not in water.
Both products tested very well after using LM and the TestE very well after brewing, I have yet to brew the test P.
Tip: weigh everything before hand and write it down. beakers, filter, stir stick, rubberbands or paperclips used to hold the filter.
This will make calculating yield much easier, for instance if you know the weight of you filter then you can leave powder in filter and deduct its weight, because you will lose some in the filter.
If anyone else has something to add or notice an error please post up, I am an amature and enjoy learning and sharing what I have found.