I have a Nitty Gritty 2.5 FI record cleaning machine that I use for my LPs. The cleaning solution is a mixture of distilled water, alcohol and wetting agent. I wasn't going to pay Nitty Gritty $75 per gallon for something consisting of a couple of bucks of ingredients, so I decided to mix my own. Unlike inkjet printers, which are sold at or below cost and the profit recovered by selling ink, Nitty Gritty's cleaning machines aren't exactly cheap, so there's no reason for them to sell the fluid at more than 35 times the cost to produce it. Googling LP record cleaning recipie, I found a lot of ideas. Most were anywhere from 5 to 50 % alcohol, the rest distilled water and a drop or two of liquid hand dishwashing soap. So I tried a few... what follows are my results.

At first I tried a 50/50 mix of 70% isopropyl alcohol and distilled water, which is 35% alcohol. This got the records spanky clean BUT it nearly destroyed my record cleaning machine. The NG 2.5 FI uses epoxy inside to seal the hoses. This high percentage of alcohol liquified the epoxy into goo and the machine started leaking. I had to disassemble the machine and perform surgery with an x-acto knife and epoxy to fix it.

Incidentally, what kind of expoxy? I tried the following, all of which eventually leaked and required a rebuild:
Ace hardware 5-minute epoxy
Ace 60 minute epoxy
Ace marine epoxy
Super seam seal adhesive caulk
High temperature gasket glue for engines

What eventually worked was PC-11 epoxy which comes in miniature paint cans. Perhaps PC-7 would work too. It was next on my list but since PC-11 worked, it didn't have to try it. I chose PC-11 (rightly or wrongly) because it cures underwater which led me to believe that it is designed to be more waterproof for constant immersion applications. The 35% alcohol solution even eats through PC-11 epoxy, but PC-11 holds up longer than any of the above, and holds up indefinitely against solutions with lower alcohol content.

Anyway, back to record cleaning recipies. I knew I needed a lower alcohol mixture. Next, I tried 1 part 70% isopropyl alcohol with 2 parts water and a drop or two of liquid hand dishwashing soap as a wetting agent. This solution is 23% alcohol, gets the records just as spanky clean as the 35% solution, and it doesn't destroy my record cleaning machine.

But, just to be safe, knowing high alcohol content will destroy the machine, I wanted the minimum alcohol content possible. So I tried 1 part 70% isopropyl alcohol to 3 parts water and the same drop or two of liquid soap. This is 18% alcohol and gets the records just as spanky clean as the others. So this is the solution I use.

Here's how I did my computations:

Each unit of 70% isopropyl alcohol is .7A + .3W.
1 part each is: (.7A + .3W) + W = .7A + 1.3W, which is .7 / 2 = 35% alcohol
1 part iso to 2 parts water is: (.7A + 0.3W) + 2W = .7A + 2.3W, which is .7 / 3 = 23% alcohol
1 part iso to 3 parts water is: (.7A + 0.3W) + 3W = .7A + 3.3W, which is .7 / 4 = 18% alcohol

Knowing that most LP cleaning fluids contain a significant amount of alcohol, one may wonder what kind of idiot would design a recording cleaning machine that is destroyed by fluids with high alcohol content. Perhaps it is intentional - use our own overpriced fluid or destroy your machine. Of course I can't know for sure... but other than this, it is a really great record cleaning machine.