servo Vs. linear encoders

[Yovav]

Hi, I hope I am posting in the right forum, if not feel free to correct me.
Anyway, I am building a woodworking CNC and I have a dilemma,
I need to design a control using the PID algorithm, right now I intend on using a brush DC motor with linear encoders on the rails, in theory it will prevent backlash and it easy to program, but I don't really know if that's a good idea, do you have any idea on the subject?
Should I be using servo (brushless DC)?
Will a brush DC will deliver good performance?

Any help will be greatly appreciated

[Al_The_Man]

Brushed servo's have been delivering performance for many years, if good quality motor and the PID loop is sufficiently capable then it should work.
Is there any reason you are using brushed over AC?
But if you are using linear scales to eliminate the effect of backlash then you may want to look at some of the instructional video's on the subject on the Galil Motion site.
In particular the one covering dual encoder feedback.
Al.

[Yovav]

brushless DC are a bit more complex for programming, its allot easier dealing with PWM for the brush DC then the matrix of the brushless DC servo, the Galil web site is great, thanks, but I didn't found the video you mentioned, they mostly deal with servo, did you meant backlash compensation for servo?

[Glacern]

Yovav,

Dual loop compensation is using a combination of a rotary and a linear encoders for more accurate positioning.

The problem with using just a linear encoder is instability. Backlash in the system will cause the motor to constantly overshoot the desired final position, back and forth.

Galil's "Improved Dual Loop" separates the PID. The P and D are applied to the rotary encoder the I is applied to the linear scale.

[Yovav]

excellent, you just solve my problem, I will use servo and linear encoder for the X axis (3m) and Y axis (1.6m), the long axis, and only servo for the Z axis (0.5m), one more thing for now, can you refer me to a dissent engine manufacturer, I found allot of companies in Google but I have no idea how good they are

[Al_The_Man]

This is the web page for dual feedback comp.
http://www.galilmc.com/learning/tuto...sation-methods
There are not that many manuf. now for DC brushed servo's but CMC is one and Aerotech also.
For DC brushless I use Tamagawa or Applied Motion Products.
If you also require the drives, there is Advanced Motion.
Al.

[Yovav]

Thanks Al, Advanced Motion seems very good, especially with there product filtering, and your link defiantly explained allot, and the drive will save me allot of work, I just hope I will stay in budget.
Y

[Yovav]

Hi
I know I am bumping up the thread, and I hope it's all right, I had a few insight about the Q I ask and I finally understood the answers I got, and I didn’t want people to read this thread and might get confused.
For some reason I confused brushless DC with rotary encoders, brushless DC do need encoders for their operation but of course brush DC can also have rotary encoders (I now know I am stating the obvious) .
my system composed of one axis with gear rack and two axis with ball screw, to prevent backlash I intend on using linear encoders for the gear rack axis+ rotary encoder for the axis engine (to compensate for the engine gear and the gear rack), and for the other two axis I will use rotary encoder on the engine + rotary encoder on the ball screw shaft (here I don’t think I need linear encoder since the backlash on the shaft itself is very low, and only need to compensate for the engine gear).
And now I realized I don’t need brushless DC, only brush DC with an amplifier, part of my project is building the control of the machine, so I don’t need a driver just an H-bridge amplifier.
Sorry I didn’t understood your answers earlier, but I guess it's better late than never.
Yovav