I agree if it is running an R/C clock network. It might be used for a VCO (Voltage Controlled Oscillator).
But I am thinking he installed a 500K trim pot and turned it down quite a bit.
I am having a hard time understanding the 240K (early run) vs 130K (later run) aspect. It makes me think they really had to tune the boards to various batches of TI chips. I know we noticed subtle differences from game to game when they were brand new but never anything too far out of pitch, I seem to recall it being mentioned on RGP years ago that the chips currently on the market could be rejects that failed to meet TI's specs at the time. So, before discarding a newly purchased 5200 that seems dead, I would certainly try the trim pot thing to see if it will respond to other R9 values.
I fired up my spare FG S&T in my Centaur this morning. With the 8048 date code and 130K R9, it sounded just like Flash should. I just couldn't tell if there were sounds missing (still missing the AD558) by just batting the ball around a bit.