Re: [CAD_CAM_EDM_DRO] Power supplys, encoders, servos, techometers and screws

info.host@... wrote:

> I have a few questions I'd like to ask.
>
> 1.) Can any of you recommend a place for regulated power supplies for quite large servos? 2-3n/m constant and 6-7n/m peak. Infact, thinking about it, if servos this size where to all (4) run at peak I could use a few arc welders! I found the peak of servos this size (5.3 amps at 320vdc) to be 1696 watts, that's only a few hundred off my welder and that's for ONE. I guess you have to be careful with the wiring? :)

Why do you want regulated power supplies? UN-regulated supplies should work fine!
You will likely have to build your own. Marlin P. Jones and Burden's Surplus Center
are places that come to mind.

> 2.) How do encoders 'work out' (Visualise) the movement of the actual rotor? I've heard the encoders measuring lines, does that mean the encoder sees a load of lines moving out from the inside out making segments (Like a spiders web without the spiral) that it detects in? What exactly does quadrant encoder mean? Just that it measures quadrants made by the lines? I've heard of three different types of encoder, I don't remember their names though.

Shaft encoders generally have a wheel with a bunch of radial stripes, and 2 photocells spaced
1/4 of the line pitch apart. (This is an oversimplification, they use 2 gratings that match the
spacing of the wheel's grating when the stripes get small.) The lines do not 'move out from the
inside', they are a ring of radial stripes. There are also resolvers, inductosyn encoders
in both linear and rotary, magnetic encoders, Newall steel ball encoders, magnetostrictive
linear encoders, laser interferometers and the list goes on.

> 3.) This will sound stupid probably but are all higher torque servo's higher voltages? The higher torque ones I've found at Pacific Scientific use 320vdc at a few amps when under peak load. But that doesn't make sense, don't the Gecko's control up to 15 amps at 80vdc? If they can control these currents does that mean somewhere sells high amperage, low voltage, high torque motors or is it just a huge safety gap? I don't even want to ask how much a controller from Pacific Scientific would cost.

Generally, going to highr voltage is cheaper than going to massive currents. The wiring
itself becomes a problem over about 25 A. Pac Sci isn't that expensive. It is the Allen-Bradley,
Siemens, Fanuc, etc. that are REALLY expensive.

> 4.) I understand the encoders are the feedback from the servos to make the system closed and to keep the system aware of errors but tachometers? Is that to take some pressure off the controls, to have a physical piece of hardware measure the rotor's speed rather than to have the software count the pulses from the encoder?

In older systems, the encoders had neither the resolution or the bandwidth (time response)
to keep the system smooth. A DC tach coupled to analog processing electronics in
the servo amp did that job. In old systems with light bulbs in the encoders, it was desirable
to have another channel to measure motion. If the bulb went out (about once a month on
some systems) the tach could be compared to the encoder. If the tach showed faster
movement than the encoder, it would estop the system.

> 5.) I spoke to someone who distributes for AHHA! who told me the servo's in their Bridgeport deal where selected because of the screws pitch, that being .200". Do most Bridgeport style mills use this pitch or are there lots? Surely it also has something to do with the weight of the table that's being moved on that pitch?

Yes, most US machine tools of the size of a Bridgeport series I use a 5 TPI screw. Some
larger machines use 4 or even 2.5 TPI. The original screws in a manual Bridgeport are 5 TPI.

Jon