A2M-80SV02430 servo and S2-SVD15 driver

Recently picked up this servo and driver off Amazon because it was suspiciously cheap at $106 for the set.

Sounds too good to be true but oddly enough it arrived and works.

Here's some notes:

• Running at 110V unloaded it's able to get up to about 3.5-3.6kRPM.
• Running at 110V I was able to confirm, with my limited instruments, that it could produce ~6Nm peak torque. It may, with better testing and its full voltage, be able to reach its advertised peak torque of 7.2Nm.
• There is no serial port, even if the manual/product page says there is. Not sure if the firmware would support it if you populate the empty headers.
• It appears to use a single-turn absolute position encoder. Likely it communicates with the driver via some sort of serial method instead of the usual quadrature.
• The encoder/position data is not available as an output on the parallel port.
• The parellel port IO is rated for 12-24v. However, using 5v for the enable signal appears to work fine. Wired for single-ended quadrature inputs, I found I needed about 11.6v for it to work reliably. If I configured it with a fake differential signal (VCC and ground are different, right?) it seems to work fine at 5v.
• Opening it up, it appears to have an internal brake resistor.
• Not sure if there's any real difference between the A2M-80SV02430 servo and a regular 80ST-M02430 servo beyond the encoder.
• It was cheaper with the holding brake, so my servo has a "Z" at the end of the model number. Pump ~24v into the holding brake connection to release the holding brake. (Background: The holding brake helps keep the servo from turning when it's unpowered/e-stopped. The servo is actually strong enough to overcome it. The holding brake is different from the brake resistor which is used to throw away power when the servo is electrically decelerating.)
• To get wires into the black terminal thing, jam a screwdriver into the *side* and insert the wire into the round hole. I think you're supposed to be able to put a screwdriver in the top rectangular slots, but I wasn't able to get it to release that way.
• All the parameter numbers I've looked at in the manual match up with AASD-15A parameters so far.

@Thanos @RaaR: Hopefully that helps provide some more data to this conversation.

@Thanos: Do you see any reason it wouldn't work with your controller? Is it an issue if the servo driver doesn't send live position feedback? Namely, does sensorless homing (hard stop torque limit) still work?

I am confused. First you say

• The encoder/position data is not available as an output on the parallel port.
• 
  Later you say you use the quadrature encoder on the parallel port. This is a key point for me since I haven’t seen one this cheap with encoder outputs. Can you clarify?

Also do you have e an actual manual?

The current position of the drive isn't shared as a parallel port output, however, you can tell the drive where you want it to go via parallel port input (quadrature, S+D, or direction pulse). They're two different sets of pins.

Check out the attachments RaaR shared on this post.

I was contemplating these same servo kits on Amazon. Looking at the pinout of Thanos' controller, there doesn't appear to be any connections to the Encoder Signal Output of the servo driver, so the lack of those outputs shouldn't be an issue for compatibility. One minor concern is... do these drivers have the high frequency "whine"? I look forward to any more info regarding your experience with these. Thank you.

Yes, and an ugly EMI output, too.

Yes, as you can see, this is a low-cost servo system. It uses a magnetic encoder. The 80ST-02430 model you mentioned employs an optical encoder. While they share some compatibility in specifications and offer a degree of functional replaceability, optical encoders generally deliver superior performance overall. I hope this information is helpful.

In general, I agree. But you have to be careful, there are many different types of encoders, not only optical and magnetic. The older optical encoders are incremental with different tracks for position (high resolution) and commutation (hall sensor replacement) information. Some older models have a relatively low resolution of 1000CPR. For simulators accuracy doesn't matter much but you can hear stepping noises at lower speeds. Another drawback is that they need cables and connectors with high wire/pin count (usually 14 or 15).

Newer optical encoders have 17 or more bits resolution and only need 4 wires because they transmit the (absolute) position information over a digital interface.

Then there are cheap magnetic encoders. They work like the hall-sensors in joysticks. You can recognize them by the magnet glued onto the end of the motor shaft and a bare PCB mouted over it. As they work like a compass they are sensitive to external magnetic fields. So if you hold a strong magnet close to the motor (or a cable with high current) the position of the servo can drift.

The high quality magnetic encoders use a magnet wheel with multiple pole pairs. Their sensors work differentially and are (almost) immune to external fields. Permanent magnets don't cause position drift but high frequency EMI from unshielded cables can cause a bit of noise.

I have tested cheap to medium priced drives ($100 to $300 per axis) from many different manufacturers with all of the encoder types mentioned above. The cheap ones have some disadvantages but the all run quite well and are well suited for simulation. Position drift as high as several degrees would be a show stopper for CNC applications but is not noticable for simulators.

So my advice would be: Don't worry much about the encoder. Rather watch out for this:
* Is there a manual available in English that explains parameter setup, wiring and tuning
* Is the pinout of the command input compatible to your controller
* If you're in the USA: does the servo run with 110V? Some don't like it
* Does it make the high pitch whisteling sound (~10kHz PWM) or is it quiet?