|
|
Page 2 of 5
1 2 3 4 5
A really big powersupply.
18:28 23/11/2004 |
| Since I decided to use rather beefy servos on my mill I needed a quite high voltage, high-power, power-supply. Even on Ebay a 60V
350W power-supply is damned expensive, and usually tends to be switched mode (i.e. similar to a computer power-supply) which is
both overkill, and in some ways problematic (switched mode power supplies can induce noise!). So the only real solution was to
build my own. Actually this is nowhere near as hard as it sounds as we're building a linear power supply and not a switched mode one. There are 3 components to a linear power-supply: a transformer (coil) to step down the voltage from mains volts to a bit under
the voltage you want (1.42 times less if you want to know), a bridge-rectifier (a small diode bridge) to convert the AC-current
into a pulsed DC one, and finally a big capacitor to smooth the pulsing voltage into a nice DC stream. Did I say big capacitor?
Oh yeah! I mean big. If that cap was a beer-can it'd hold about 3 cans of beer. It has actual M5 bolt-threads in the terminals - you actually screw bolts into the cap to hold on your connecting wires. |
| I "borrowed" an old external disk case from work to house the power-supply in, which looks pretty good. Unfortunately the
massive cap didn't fit in the case - so I decided to put the capacitor in the driver/electronics case instead and leave the coil
and it's magnetic field away from my electronics. |
| The rest was just a matter of soldering, and some crimping really. I used crimping (and spade connectors) wherever possible as
this makes disconnecting easy in the case a components fails. Finally I added a power LED to the front (with a big resisor as
the small ones would get too hot from the high voltage and cook) and was done.
The performance of the power-supply when
I ran it was stellar: my strips lights actually DIM when you turn it on! The cap is amazing too... too good in some ways in fact. Servos only draw power when they move (unlike steppers) so I had the problem that even after turning off power I could still
control the mill with CNC! In fact, I could actually RAPID for several seconds! Sounds good? Well, it is.. except when say, you
wanted an Emergency Stop.. ("My kitten is getting sucked in!") I had to solve this problem by using a relay to cut power to the capacitor and then short a heavy
resistor over it (acting as a heating element) to cut the power. Now it shuts down in about a 1/4 second. But I have to replace the
relay about every 6 months. |
|
Mounting servos to leadscrews
21:01 25/11/2004 |
When I first started, I wanted to get something running before I potentially ruined my machine by installing ballscrews, so
I decided to begin by motorizing the leadscrews that came with the mill. I'd also then be able to get an idea of how accurate the
machine really was before I invested the effort in overhauling it. The easiest option I could see was replacing the graduated handwheels (you can see one in the first pic - it's right between the the handle and the bearing block) with a timing belt pulley.
I ordered some from RS components and damn were they expensive! Nearly 50 GBP for all you see there - AND I managed to mess up
calculating the belt size initially too. Can you see where I skipped a multiply by 2?
I trusted my little Sherline mill and rotary table better than my (at that time poor) lathe to bore out the pulleys to the correct
size and remove the lip on them. Finally I drilled a small 2.5mm hole through, and tapped it so I could put a grub screw in to
tighten it to the handwheel shaft.
I then machined a little groove in the handwheel's "spiggot" so that the grub screw would have something to hold to, and with that
the bored out pulleys easily slid on, but could be nicely tightened up to hold well. The result was that the pulley took the place
of the graduating wheel (as you can see in the last pic) but no other life-threatening modifications needed to be made to the mill.
|
|
Page 2 of 5
1 2 3 4 5
|
|
|