First time seeing this post Stevohdee. I'm frustrated by the extreme runout on ER collets, and it does not matter what size they are. Cheap is cheap. And they may advertise 0.005mm runout, but it always seems it's way beyond 0.03mm. Some sellers are 'more' honest, and claim 0.015mm runout, but they are still greater then those numbers. Worse is the chucks, where they are not even ground with a concentric offset to the axis of the tool, but at an angle so the further from the chuck the greater the work piece or tool runout. I check the runout at the outer lip, and inner edge of the 8 degree conical ground surface, and I've not see the same runout to find just a slight offset from the axis of the final grind process.

I've had much better luck with R8 and 5C collets then my ER40 and ER50 collets.
I wonder how they could be held to repair them. Would sliding them onto a mandrel (custom to each size) and grind the exterior 8 degree taper be the best way to fix them? That would be hope-ing that the friction of the un-clamped collet would be enough to overcome grinding forces. And would require using another collet to hold the mandrel, and just how concentric that collet is. The mandrel could be ground in the 5C collet (thinking of a spin indexer on a surface grinder), to correct for it's runout. But what a job that would be to fix a cheap set. I'll have to see if there's a YouTube video of someone doing this.