After some considerations I decided to make two small modifications. One for the fixed end gripping mentioned above and the other was to make the fixing bolt pockets a little larger than the original ones.
Fixing new vise jaws
Never really milled aluminum on my CNC, so I was a bit pessimistic about the performance and also not wanted to cause any damage, so my idea was to use a BF20 I have access to also. It is not mine, so everything is not set up as I would like to have it set up, but I considered to try out, took off one end of my vise and milled one track on each piece. The results were good, but not good enough, it needed some fixing due to some slight parallelism error. This was best done on my own CNC. At the same time I changed my mind and decided to try out my machine, see what it can do with aluminum, after all, aluminum is not much harder than the Delrin I am normally milling, and my upgraded CNC is now so rigid that it should work.
Milled two new jaw traces at the other end of the vise on my own CNC and was very happy with what I have seen. Excellent finish, very nice and even edges. No lubrication, no cooling, not even air was used, only dry milled. With the right speeds it is not necessary to use anything. The tool or the work piece not even heated up to any sensible temperature and everything worked fine, no melting.
It was a really satisfying feeling to watch the aluminum chip spray flying away from the tool and to see the results when the machine stopped. The pre-milled jaw pairs were also fixed on my CNC to get them as good as possible.
The bolt pockets
The original pockets were too small and only the bolt head would fit in them, no washers. I take this as a design flaw of the manufacturer because it is not a good idea to omit the use of washers, especially not in aluminum, since the bolt eats its way inside the material when tightened hard enough. It is nice to be able to bolt the vise down firmly to the bed, much better than before.
Final steps
I made a vise fixture plate out of 12mm thick aluminum sheet which I will keep it permanently installed on. This gives it much better rigidity and is also easy to remove from the CNC table if I wanted to.
Once everything was as good as I could make it, I assembled and installed the vise, squared the sides on my CNC table and made another final milling run on the jaws to make them absolutely parallel, or at least as good as I could make it. I measured the parallelism error to 0.003mm from one side to the other, over a total width of 140mm, so that is 0.0214mm per meter error, which is good enough for me.
For those who are interested, here is a short video about the modifications I made.
This was the first time I milled aluminum with my upgraded DIY CNC. I am a bit surprised with the results, very satisfying and nice in my opinion. I know some people will disagree with me, but I will continue dry milling aluminum without any lubricant. For now, I see no problems with dry milling. Of course, the speeds must be right, but it seems to work just fine. Perhaps tool ware is much worse, and perhaps the results would be even better with lubricants, so it is a compromise, but at least I don’t have to handle chemicals and inhale the fumes.