0
\$\begingroup\$

I’m trying to integrate a 40+ year old ac motor and remote controller with a device using modern ADC controllers to remotely manage the motor turning process.

The legacy gear is an old CDE Ham-M amateur radio antenna rotator which converts maximum voltage (26 VAC) to a compass heading. The tower mounted rotor includes a 500ohm resistor (wound on a 360 degree platter on the motor axis) which reports the voltage to a meter converted to a compass direction.

The issue at hand is some of the windings on the 500 ohm pot in the remote motor aren’t in perfect condition. When the rotor is turned with the legacy controller, the meter spikes in voltage around some compass headings, which gives the ADC circuit fits. Apparently, the device with the ADC samples so frequently, that when the legacy devices reports momentary voltage spikes (1-2VAC), the ADC device gets so many different values so quickly, that it sends many different relay commands to stop, reverse or advance the motor. The relays in the digital device chatter rapidly, sending commands to turn CW, then CCW and then brake, that finally high current in the motor exceeds the fuse rating. When this occurs, the motor is stalled/paralyzed, often resulting in a blown fuse.

The motor xformer is rated at 26VAC protected by a 3A 3AG (fast blow) fuse.

I’d like to implement a circuit change that will smooth these spikes/variations and reduce the ADC measurements that cause the relays to send CW, CCW and brake commands in rapid succession.

I’m wondering if a decoupling cap across the circuit reporting voltage would smooth the voltage reported to the ADC? If so, what would be a likely uF value? This variability in voltage is for no more than 2-2.5 seconds and less than than 3VAC, until the swiper on the 500 ohm pot reports steady voltages and the ADC is “happy”.

Or do you recommend a different solution to smooth the voltage?

\$\endgroup\$
0
\$\begingroup\$

From your description, the wiper does not have steady contact with the resistive element, thus the reported voltage becomes OPEN CIRCUIT at those moments of loss-of-contact.

Can you insert a RC filter, perhaps 100 ohms and 1,000 uF, which 0.1 second time constant or 1.6 Hertz 3dB point, without making the controller become unstable?

\$\endgroup\$
  • \$\begingroup\$ Sorry for the late response. Your suggestion sounds the most practical and easiest to implement - plain old fashioned physics! Pardon my ignorance for this question..I presume the resistor would be in series with the load and the cap would be across the load? \$\endgroup\$ – Jim McDermott Aug 2 '18 at 1:08
0
\$\begingroup\$

1) Use a serial encoder

2) Use a kalman filter to filter your ADC values.

\$\endgroup\$
  • \$\begingroup\$ Thanks for your ideas. Since I don't have a schematic for the new ADC controller but do for the old analog one, I'll try the RC filter to smooth/dampen the V variation before it gets to the ADC circuits. If that option doesn't work and I get the schematic, I'll follow up with your suggestions. Thanks! \$\endgroup\$ – Jim McDermott Aug 2 '18 at 1:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.