I know the TA's are rated well but I haven't had them apart. I'd like to see the two machines next to each other both with the same wave, EN, EP, etc and hear the difference, maybe I could give you a better idea of what's going on. Now that I think of it, it could be how the magnetics are would and the gaps, etc. that each company uses, could you be hearing this I wonder?
Mark:
I'm not sure what you're saying, a true square wave is easy to generate, hence the "mod square wave" inverters that flooded the alternate energy market before "pure sine" inverter technology caught up 8 or 10 years ago.
Switching transients, even 'N' type high side charge pump driven MOSFETs (Metal Oxide Silicon Field Effect Transistors) are a product of rise time, capacitance loading (the Miller Effect, etc.) new MOSFETs are fast and relatively cheap with slew rates in the 100 and 1000's of V/uS. RDSon values are in the mOHMs so I can't see where this would be something even a circuit designer with minimal experience in magnets would have a problem.
The noise difference is probably not saying much just the different approaches to generating the final wave, and there can be many ways to get to that end. I'd say the thing that separates machines is core materials, winding philosophies, winding material purity, temperature tolerances of the winding material used and how overbuilt the circuits are. Some of those things denigrate gradually and may even take a few years to show up full bore, that's one reason why I bias toward machines like Miller and Lincoln. Just me probably, to each his own, and I don't fault anyone's machine choice.
Rob