ac vs dc servo motors


Thread Starter

still smiling

Hi, we are designing a new machine. We have previously always used DC servos, but now we have a clean(ish) drawing sheet. I have read the blurb but since no-one ever mentions the downside risks I am having some difficulty deciding which way to go.

The application is vertical motion lifting a 200kg mass (maximum speed 1.0m/s and max acc 0.5m/s2) We need resonably accuracte positon contol -- 0.1mm. The machine could be used 24/7 so reliability is an issue.

I would appreciate any thoughts. Thanks

Thomas B. Bullock

Most new applications today are going AC brushless which is an AC synchronous motor.
It is more reliable (no brushes), smaller, more efficient, and has better resolution/accuracy. It is more costly, but the differential between AC and DC is getting less each year.
Is your vertical load counterbalanced?
If not and you keep the servo loop closed, the motor will sit with a torque and current sufficient to hold the load up from gravity.
This will cause I squared R loss and motor heating whether it is AC or DC.
Tom Bullock
Hi, I would think that you are likely using a screw or some other mechanism for moving this weight. Things like screw-pitch and gears will affect positional accuracy, when combined with your servo motor's resolution. If the servo will have to be at stop with a holding current near it's maximum, consider getting a motor/drive that is larger. Servos today generally have excellent resolution, repeatability and holding characteristics. I think using a servo is a good idea for all of the options it gives you. I did a project sounding very similar to your's a year ago with excellent results. I would definitely consider a servo motor with a brake, considering the the possible constant pull from gravity on the motor.

John Franks (still smiling)

Hi thanks for the responses.

The load is not counterbalanced. It adds unnecessary complication and makes achieving the resolution more demanding.

We are using a lead screw to drive the mass. During a cycle the load will be at rest for possibly up to 2-3 minutes. A brake is essential for safety, if the power fails the mass must not fall. However, can we apply the brake in process and retain the table position? Or should we rely on the motor to hold it in position against gravity?
If you're starting with a clean sheet, maybe your application (depending on details) is suitable for linear brushless servomotor technology? LM
technology eliminates all the rotary to linear mechanics and can be extremely repeatable, accurate and reliable. Most standard brushless
commutated servodrives can be adapted to control LMs. One hitch is that there is no such thing as an off-the-shelf LM [safety] brake.
Another thing to evaluate is the type of AC drive you use for the brushless motor. Brushless PM motors can be driven with sophisticated digital
amplifiers that develop the AC sinusoidal commutation and are driven by standard +/-10v analog controller signals. The other way to go is with "dumb AC amplifier" which receives two phases of the sinusoidal signal from the
controller and allows the PID loop to be handled in the controller as well. The latter is a substantially lower cost approach. Some controllers can provide this output from one channel, others require signals from two

Bob Close
Executive V.P.
Precision MicroControl Corp
TELE 760-930-0101
FAX 760-930-0222
[email protected]
Power blocks are available from several vendors where they supply only the transistors and heat sinks.
All logic, loop closures, and compensations are done within the main controller. On the surface, this looks less expensive, but the additional application effort will more than offset the lower hardware costs.
OEMs who use high volumes can justify the one-time
engineering costs and spread it over a large number of units, but the small users can't.
Todays drives also have much software in them which complicates using the main controller. The drive manufacturer wants to keep control of his auto-tuning, his diagnostics, his torque loop and various other things which make separating the software less desirable. My observation is that the market has opted for more intelligent drives, not dumber ones. This seems to be true for many peripherals as well.
Tom Bullock
The sine drives of today come in two categories: those that are full digital dives that want to close the PID and do all that grand tuning, and these are great for distributed control systems, but if you have coordinated axis that require fine control and tuning then there are the other type power blocks that depend upon relatively easy to use controllers that provide tuning
scopes, which are easier to use than the primitive systems on the Drives.
With the controller you also have excellent trajectory planning in high speed RISC processors, right on the control board. These reduce processing and communications time over using the smart drive and a computer. The
controller may also have a DSP right on the axis module that has very fast update rates for the PID loop... PMC's controller runs about 125
microseconds, Delta Tau's may be faster. As for the power block drives... they're much lower cost. The programming is more powerful in the Controller than the smart drive... Today you program in Labview or C++, there are libraries for almost anything... Most controllers have FPGAs on each axis as well, allowing the user to set up notch filters.

I'm really not certain why anyone wants to put a controller in each of their amplifier boxes, unless distributed control is necessary or they just have one axis to drive... the difference in price is about 3 to 1....

Bob Close
Executive V.P.
Precision MicroControl Corp
TELE 760-930-0101
FAX 760-930-0222
[email protected]
Yes, the brake is essential for safety.
Is the lead screw a ball screw or acme screw?
The acme screw is much harder to back-drive
and may hold the load with little help from
the motor.
When sizing the motor, you should do a torque
profile of your motions and add in the steady state torque required to counterbalance the weight. Compute the rms of this profile and compare it to the torque rating of the motor.
Using the brake whenever the axis will be still
for 2 to 3 minutes will certainly reduce the duty
cycle as long as the control allows for an easy
reclosing of the loop.
Tom Bullock
If the motors are properly sized holding the mass at rest should prove little difficulty. Ideally run them through a gearbox of at least 16 to 1 and it will probably be the case that the load will not drop even upon loss of power. The brakes are still a good idea for safety's sake, but you will then be able to use smaller (cheaper) motors to drive the load.

Davis Gentry
Applications Engineer
Delta Tau Data Systems