I stuck some 5670's in my Collins 26U once, and they seemed to work a bit, but the metering was totally off. I considered going down the path of the conversion ala Manley, but never did. Shortly after that Manley admited the 5670 wasn't so great after all.
6ES8 is a proven quantity in the Langevin Leveline compressor.
There's a Gates limiter that uses PP 6J7's for gain reduction, but in combination with a lightbulb based variable resistor bridge shunt. The bulb shunt and the entrance to the variable mu sidechain are the same spot, so there's interaction and the points are probably set so the bulbs keep the vari-mu chain from bottoming out. I haven't measured the range of the vari-mu section, but it has to be relatively small. The vari-mu section acts as an initial brake before the limiting action of the bulbs kicks in. Of course it's not much of a stretch to convert that tube position to 1612's if you wanted to do so.
6ES8 is a proven quantity in the Langevin Leveline compressor.
There's a Gates limiter that uses PP 6J7's for gain reduction, but in combination with a lightbulb based variable resistor bridge shunt. The bulb shunt and the entrance to the variable mu sidechain are the same spot, so there's interaction and the points are probably set so the bulbs keep the vari-mu chain from bottoming out. I haven't measured the range of the vari-mu section, but it has to be relatively small. The vari-mu section acts as an initial brake before the limiting action of the bulbs kicks in. Of course it's not much of a stretch to convert that tube position to 1612's if you wanted to do so.