Why are records still warped?

If you really want to know:

An inherent outgrowth of the polymerization process is that each individual polymer chain has a different molecular weight (chain length). When the PVC pucks are formed (before the record stamping) they are created from small plastic pellets which are melted with heat to form the disk.

When the plastic melts it expands. When the plastic re-solidifies it shrinks an-isotropiclally (different amounts in different directions) due to the viscosity variations caused by the molecular weight (polymer chain legth) differences. This variation in shrink by direction causes warp. Some of the warp occurs initially but some will occur over time as the record sees elevated temperatures that cause the internal stresses (caused by the variation in shrink in different directions) to relieve themselves.

Even though the record business is run on antiquated equipment, all of the above challenges are faced by the highest precision plastics companies in the world. While improvements could be made through more modern equipment, longer cycle times, etc., no one will ever be able to make a perfectly flat record that stays flat it's entire life until constant molecular weights are achieved in the initial polymerization process.

I did not mean to imply that very flat records aren't possible, in fact, by far the majority of records produced now and back in the day are flat. I was trying to explain the challenges involved. Molecular weight distribution at the end of the bell curve create the problem. As said, longer cycles and modern machines will improve the odds. In addition, using well maintained older machines and processing techniques that minimize oxidation and hydrolysis (which break molecular chains further exacerbating molecular weight variation) can improve the likelihood of a precision outcome. Finally, a 100% QC inspection process will obviously eliminate those warped records which occur when initially produced (but those aren't the problem)

I am certain some record suppliers do a better job than others but zero defect production is much more difficult than Bdp24 implies. Some records still warp even when left in the press long enough to come out flat and when packaged and stored properly . A small percentage will warp down the road because the molecules in that particular shot of material happened to all have lengths at one end of the distribution range (be it long or short) which completely changes the rheology of the flow of the material which changes the internal stresses. They go into a sleeve perfectly flat and change in the days and weeks to come.

This is an entirely different challenge than faced by manufacturers in the metals businesses (machining, casting, edm, etc.).

06-02-15: Lowrider57

The record labels and the pressing plants all say that demand is so high for manufacturing vinyl that their orders are backed up. There was a thread on Steve Hoffman, I believe, where somebody stated that the cooling stage had been shortened as one way to speed up the process. I'm sure there are other areas where vinyl production specs don't measure up to the original days of record cutting. Especially quality control as previously mentioned.

Lowrider, I am certain your are correct. Bad manufacturing practices just make all the problems I have discussed even worse.

06-03-15: Moonglum
Given your level of insight, perhaps you are the man to ask about MRA?

I am not familiar with MRA's quality issues but what you describe (if I had to guess) is probably more related to bad practices.

Not necessarily. There are many variables involved. The wider wall stock will reduce the pressures required to make the puck which can result in lower internal stresses so it helps in that regard but the more plastic the more shrink which as you state could make things worse. I have no direct experience with how these to countervailing influences effect records.

Complex finite analysis models are used to predict these things but they are only so accurate. There is no substitute for prototyping the mold and making actual parts to get the best result.