Forum Navigation
Forum breadcrumbs - You are here:ForumGeneral: IntroductionsGadgeteer
You need to log in to create posts and topics.



My name is David Pollack  I am English, after my wife  died in the UK I moved to China in March 2012, married a Chinese girl and now  live permanently in China.  When I left the UK I had to sell all my tools and machines (mainly woodwork) . but I also  a had a Sieg SC3 Milling machine and lathe. When I moved to China I had to sell all my gear in the UK at a huge loss  and after a year in  China I moved the  into a 4 bed apartment  where I used the second bedroom and an internal balcony for my workshop/study. Purchasing the Sieg lathe and milling machine for the second time in China was a lot cheaper than in the UK, over the last 5 years I have bought lots of tools/accessories and a floor standing pillar drill with 1500 watt motor ( a basic Chinese model but good value for money, I will add a VFD and X-Y table later for more precise drilling). While I was working in a large multinational  company research lab  in UK  where I ran the instrument lab (we made non standard instrumentation for the scientists ) I came across a White paper in  one of the scientific journals about the possibility of multi stepping a stepper motor, there were no circuits or plans just speculations really. At the same time the scientist running the microscopy lab was having problems moving the target in the scanning electron microscope at 100,000 x mag. even one division on the x-y metric micrometers would move the target completely out of view. I found an electronic engineering company in the UK that made high  quality very accurate 1.8 degree stepper motors (this was 1978 very few motors of this type then). Long story short, after lots of talks with the company engineers and understanding the motors I designed and made a circuit that micro stepped the standard motor 1.8 degree steps into 128 steps. Then one of my tool makers made a zero backlash coupling to connect one motor to each micrometer of the X-Y axes on the SEM stage .  This worked very well after the SEM stage reached  a stable temperature. In 1980 we were all made redundant  and I found a job with  Analog Devices for two years , the pressing problem at the time I was there was to get their new prototype 16 bit  A to D to work reliably.For the next 10 years I did something completely different although   I did get involved in getting a mad computer scientist to program a paperwork system into a PC based system. After that I worked for a large high street computer company as a PC technician for 6 years (W95 and W98) then I took early retirement at 63.

My big problem now is the lack of space and to keep the noise down as I live on the 14th floor of an apartment block, I don't have welding facilities as the wiring in the apartments is 4^2mm ( I specified this when the apartment was being built). In retrospect I should have specified at least 1 circuit with 60 amp capacity, I don't know what wiring system is used in residential property in China (220VAC 50HZ) At present I have a small light weight 30/40 Chinese milling machineI bought 4 years ago it had it's own built in controller with integrated VFD and post software. The unit lasted for about 6 months before it died,  however the mechanicals were quite good and it came with a 800 watt 400hz motor. I fitted the next size up nema 17 steppers and the last version of Mach3 and connected a 1.5 kw VFD , it works well and is surpprizingly accurate. I use it for PCB's (after I got my head round the appropriate software)  and later to engrave panels  My  other cnc  machine ( my folly) is a 800x420 mm x 165mm with 2.2Kw 400Hz spindle and 2.2Kw VFD The X axis fitted with 30mm  C5 ground  ball screw with double ballnut Y: C5 25mm ground ball screw and double ball nut Z: C5 20mm ground ball screw single ball nut. This is my second set of ball screws the first set was C7 rolled and single ball nuts all round. I had the base and some of the othe parts made by a Chinese prototyping company in mainly 6061 T3 20mm thick milled flat and with a matrix of M10 threaded blind holes in the  X-Y table. I found the ball screws  both C7 and C5 ground or rolled to be very in accurate over their full length so I bought a three axis DRO and 3 glass scales and temporally fixed them to the machine. I blueprinted each axis every mm and recorded the figures in a spread sheet, I did 2 runs for each axis and averaged the results while trying to keep the temperature as stable as possible(after AC stabilized) .25-27 degrees C.  Using the Mach3 calibration function(It really sucks!!!) I applied the appropriate correct for each mm When I plotted the raw data for each ball screw the graph was like a dog's hind leg. The spec. for the glass scales: resolution +/-  1 micron  accuracy +/- 5 micron ? For the future I want to add flood cooling a coolant tray under the bed with a filtered drain tap down to a recirculating tank and a full enclosure to the machine. Before I buy the Masso G3 (cost 670 pounds sterling including relay board and pendant) I want to make sure it  will do what I want and that there are no nasty surprises after I buy it. By the middle of next year I am looking at extending the cut footprint to 1000mm x 600 mm with my existing base, adding 1.5Kw servos for all axes and for the spindle also possibly ATC . I have already bought the 4 servos and drivers + spindle and 4 bt30  Er32 tool holders. Since the servos are a lot more powerful and heavy than the previous setup I will have to beef up the whole machine to make it more rigid (don't know how I will do that yet)!! I also use Gwizard Calculator and Gwizard Editor as I have not got the experience or training to estimate these things myself.