Wednesday 29 March 2017

What is it?

During this time, I did a lot of research. I spent many, many, hours trying to track down any other examples of my particular machine but only found one - which had been converted back to manual. There are loads of examples of a very similar model (AN-S) but mine is actually an AN-SB which differs in some keys areas. The "ram" doesn't move, it has a proper variable speed motor and VFD rather than a mechanical variator, the table is bigger and the machine in general is bigger and heavier. The family resemblence is very strong but the expoded drawings parts lists are for the AN-S. Bugger. But why did the owner have a parts list for another machine?? It must have been supplied with the original machine in error.

I managed to track down the original owner of this machine. He bought it new in 1984 for about £34k (equivalent to perhaps £100k in todays money and enough for a house in those days), used it for 20 years then put it in storage when he retired. Finally after 10 years he auctioned off his equipment. During its 20 years of operation, he only used neat cutting oil (no water based coolant) and he did mainly precision engineering. It was very well looked after. The guy I bought it off had acquired it at the auction a year earlier but wasn't able to do anything with it. I don't know what he paid for it but I suspect he lost a little in the end.


It's beautifully made and has top-notch components like NSK ball screws, Nachi and NSK bearings and Hitachi motors (spindle and coolant). The accuracy report makes interesting reading. The servo system is DC brushed, with Parker PKS Digiplan digital positioner, servo drives and Heidenhain-compatible encoders. The control software is Micon Numericon 850 (wow) - although it seems to have been quite powerful and capable at the time, it's died and sunk without a trace. Although g-codes were defined before this system was develoed, bizarrely they are different from "normal" g-code. So there was not much point learning them and no chance of finding a suitable CAM post processor to generate them. The program memory is a massive "1/4 megabyte".



No comments:

Post a Comment

Final assembly and test of the spindle nose adaptor - RESULT!!

After the recent distraction caused by the 3D scanner, resurrecting the 3D printer and buggering about with the throttle bodies for my Honda...