It's worth noting that the bucking transformer is only useful if the field leakage is made worse by a transformer starting to saturate. Don't assume that it's a guaranteed fix unless you've tested with reduced voltage first (hence my Variac suggestion - it's not a permanent solution but a very good way to test before mucking about with other transformers).
As I don’t have a Variac, I decided to go for a regular transformer and test it out that way.
The result is quite interesting. It’s hard to tell if it makes a difference for only one of the devices turned on. But if I turn both of them on, this now actually reduces the wobble instead of increasing, as it did before. With both running, it’s basically gone now. So I guess I will keep the transformer and put it into a more permanent form.
Any ideas why running both devices might reduce the wobble compared to just one?
If you have a current probe capture the current waveform with the secondary winding open. This is probably a 50/60Hz thing.
A 50Hz transformer requires more turns than a 60Hz transformer for the same magnetic flux density.
The transformer was probably designed in a 60Hz country.
Unfortunately, I don’t have one. But from my reading (of the previously linked article and some more afterwards) I understood that in this case reducing the voltage should still help. Is that correct?