Could ultrasonic cleaning damage your LP? Yes, if you over do it. So could any other record cleaning system, if you overdo it. Damage could come from exposing the vinyl surface to either excessive cavitation (bubble bursting) energy or to excessive heat. Also, you want to avoid excessive heat when working with isopropyl alcohol. Let me describe how I have kept within safe operating boundaries.
Higher ultrasonic transducer frequency increases the number of bubbles and, correspondingly, decreases the energy released by each bubble. That is a good thing! Lower individual bubble energy lowers the risk of eroding the surface of the vinyl. Moreover, smaller bubbles can fit into smaller crevices – the vinyl surface is cleaned more gently and more thoroughly as the transducer frequency rises.
Most ultrasonic LP record cleaning systems use a 40 kHz transducer. Cost rises as transducer frequency rises. If you want safer, go higher in frequency and in cost. For instance, the individual bubble cavitation energy from a 40 kHz transducer is about 3x greater than that of a 60 kHz transducer and 10x greater than that of an 80 kHz transducer.
The magic upper bound for heat appears to me to be 95 degrees Fahrenheit (35 degrees Celsius) for three reasons. First, heat influences but does not determine the efficiency of the ultrasonic cleaner. A 60 kHz machine cleans LPs very nicely at temperatures well below 95 degrees Fahrenheit. Second, the vinyl surface is not troubled at or below this temperature. Remember that the temperature within the fluid bath is not the same everywhere because the transducer itself is causing the temperature to rise unevenly as it broadcasts into the tank. I have seen an instance in which the surface of the record ABOVE the water bath line was affected when the bath temperature approached 110 degrees Fahrenheit. I have seen no adverse effects when the bath temperature is at or below 95 degrees Fahrenheit. Third, the flash point of a 10% solution of isopropyl alcohol is about 110 degrees Fahrenheit. The cleaning bath solution ought to be a 3% to 4% isopropyl concentration. In sum, an upper bound of 95 degrees Fahrenheit gives a deeply respectful margin of error.
http://sanfranciscoaudiophilesociety.com/forums/topic/observations-about-cleaning-lp-records-ultrasonically/