I started making my third batch of weights this week. I took some pictures so I could show more of the process instead of just the finished product.
First, I had to clean out the molds, since they had just been used. I simply scrubbed with water and IPA (not the beer, but still alcoholic), and they looked much better afterwards:

Before cleaning 
After cleaning

Next, I hot glued the metal ring into the mold. The ring is what forms the hole that lets you slide the weight on and off of the barbell. It comes out with the concrete at the end and stays in permanently- the reason for the hot glue is to stop concrete from seeping in and blocking up the hole.
Once the ring was sealed on, it was time to get mixing. Since I had two molds to experiment with, and since I was dissatisfied with the quality of the weights I had previously made, I decided to try a new mixing strategy with each mold. With the larger mold, I would use precisely the water:concrete ratio as recommended on the concrete bag. No matter how it looked, I was going to follow that ratio exactly. With the smaller mold, I would eyeball the ratio as I did with previous batches, but with the new knowledge that I needed a much less soup-y texture in order to make my weights less prone to cracking.
Here is the mix for the larger mold, for which I followed the instructions precisely:

50lb bag of concrete added 
Precisely measured water poured on 
Water and concrete mixed
Following the instructions resulted in what seemed to me a very dry, rocky mix. I was skeptical that it would be wet enough, but looking at how my first two batches turned out, what did I know? After mixing in the fiberglass reinforcement provided by Stix and Stone, I poured the mix into the mold until getting the right weight:

The final mix 
For a 45lb plate- as the water evaporates and the concrete cures, you can expect to lose 1-2lbs
I am very doubtful that what you see above is going to somehow cure into a nice smooth surface, but I will find out for sure when I get to school on Monday.
I followed the same process for the smaller mold, but rather than using the prescribed amount of water, I guessed. I was aiming for a texture that lay somewhere in between the soup of my first two batches and the gravelly texture of the mold I had just finished. A big mistake I made with the first batches was that I dumped large quantities of water at once to speed up the mixing process- not good! So, an important change I made this time around is that I added water MUCH more slowly than I did before, so that I could control the texture more precisely. Here’s the resulting pour:

Much wetter than the first mix 
For a 10lb plate- I am not sure if this will end up being too heavy
I have a much better feeling about this one than I do with the first. I think the wetter mixture will result in a much smoother surface. Have I finally cracked (no pun intended) the formula for making the perfect weight plate? Or will the quest continue? I will find out on Monday… stay tuned!