A History of Ultrasonic Record Cleaning

@lewm --I don't think there have been any serious scientific studies of damage. I don't think the article was intended to convince otherwise and agree that anecdotal reports are not a complete answer. Mike did mention to Degritter (@120kHz) but not others. 
Could you elaborate on what you mean by "And finally, pure water or what should be the medium?"
My understanding is that ultrasonic cleaning is enhanced by a surfactant. 

My own experience is that I can detect no aural damage, but I only use ultrasonic for limited amount of time (e.g. 5 minute cycle) as part of a larger cleaning regime that includes pre-cleaning and post cleaning rinse using reagent water and vacuum (Monks). 
Fair question though and not one that I think the author intended to resolve. 

@lewm - right now i am not using any surfactant in the US machine since I still am using the KL. But I've had a number of discussions with a manufacturer of ultrasonic equipment for factory lines who, at least in a metal parts (non-LP) context, convinced me of the value of using a surfactant. The biggest issue, of course, is removing it from the record surface. Thus, the pure water rinse. 
My plan when the KL goes is probably to buy the big Elma, given the feature set, plus external filtering and recycling. 

Too low a frequency for too long; if the record doesn't rotate, the cavitation effect focused in one spot could cause damage. I think the literature on ultrasonic cleaning for industrial purposes (non-LP) could be a starting place to look at the risks, which might also include water temperature. I haven't really thought too much about how to damage but it's the flip side of not damaging; lewm raises a fair point about potential damage; we've all engaged in a leap of faith based on manufacturer statements and user anecdotal experience. That's not going to satisfy someone who wants "proof." If I had time, I could ask or explore more; there is a point where experience in cleaning metal parts doesn't translate, but I suppose if it damages metal it is bound to damage vinyl. 

@lewm- Tim, in this article I published, switched from a cheap Chinese unit that crapped out after a year to the P120, see [url]https://thevinylpress.com/timas-diy-rcm-follow-up-2-compelling-changes-improved-results/[/url].
@Terry9-if memory serves is using a Fisher Scientific branded Elma version of the P60, which as Terry could probably explain, is a tight fit for two records, but I gather works well, and is less pricey due to its smaller size/capacity.