The trim outputs of the autopilot panel are just two wires for each axis and the states are either OFF, +/- or -/+ depending on which way you're driving the servo.
Just to clarify, if the SV-AP-PANEL isn't driving the motor, both outputs for that axis are pulled to ground. That's the "OFF" state.
There are lots of inexpensive motor controllers out there [...] You can get those for under $20 from Amazon [...]
Indeed, many of the electronic hobbyist modules available online are impressive, at least in theory. The problem is that a very high percentage of the modules from China that are sold on Amazon, eBay, AliExpress, etc. -- especially those that advertise using a specific part number from a well-known manufacturer -- are made with counterfeit ICs that often don't meet the real manufacturer's specifications.
As an example, take the ubiquitous motor drive modules claiming to use two Infineon BTS7960 half-bridge ICs. Here's an
eBay seller in Texas advertising them for $8.50 delivered. There's no way that guy can import, package and ship a module using two genuine Infineon ICs that cost ~$3.60 ea. in 1,000 quantity for $8.50.
As you pointed out, even Amazon is in on the counterfeit game. Their marketplace listings are overflowing with the same module, starting for as little as
$6.33 ea. in packs of three (helpfully mislabeled as stepper motor drivers), stocked and shipped by Amazon for free Prime delivery. They're apparently blissfully unaware of the Chinese IP theft that they're facilitating on their platform.
Also, the BTS7960 has been obsolete for well over a year and legitimate component suppliers have zero stock. How is China producing thousands of those modules using two ICs per board?
Anyway, my point is that I prefer to source components from legitimate suppliers so that I know what I'm getting, especially when it's going into my airplane.
Hypothetically, one could see how a person might use one of these controllers in a trim system compatible with the Dynon SV-AP-PANEL trim MOTOR outputs and preserve both speed and direction. You'd need one set of these for each trim axis though you only need one controller board. Connect the "AC" pins of a bridge rectifier to the motor output of the the SV-AP-PANEL. Connect the "DC" pins of the bridge rectifier to the PWM input of the motor controller through an optocoupler to isolate the grounds. Connect a 2nd optocoupler from the motor output of the SV-AP-PANEL should only conduct in one direction. Add a filter circuit to integrate the PWM pulses to give a continuous 0 in one direction (negative pulses that the optocoupler doesn't conduct for) and a continuous 1 in the other. Connect the output of the controller to the trim motor and it does both speed and direction control from the Dynon SV-AP-PANEL.
I'm afraid that I don't follow your circuit description; I don't understand the need for a bridge rectifier, dual optocouplers, isolated grounds or an integrating filter.
Feeding the motor outputs from the SV-AP-PANEL through a bridge rectifier would eliminate direction information. Regardless of which "AC" input to the bridge was high, only its "DC+" output would ever go high. Feeding that into a motor driver would result in unidirectional movement regardless of trim commands.
Putting a PWM signal through an integrating filter will produce a variable analog DC voltage, not a discreet 0 or 1. The PWM signal is what controls motor speed; once you filter it away, you've lost speed control (or to be pedantic, you've lost speed control at full motor torque).
An optocoupler doesn't provide an isolated ground; it isolates a signal. In a single-supply DC system you would need an isolated DC-DC converter to achieve isolated grounds.
The only interface that might be required between the SV-AP-PANEL outputs and a motor driver is level translation to satisfy the driver's input voltage limit. Each of the two motor outputs from the SV-AP-PANEL will connect to the input of one-half of a full-bridge driver, and each half-bridge responds to a high signal on its discreet input. Without a common ground reference this won't work.
An oscilloscope or frequency counter will give you the PWM rate.
Indeed they will, but unfortunately it's not quite that easy. The SV-AP-PANEL will only produce a PWM speed control output when it's powered by and communicating on the SkyView Network. If that connection is lost --
and if its alternate power pin (D15 pin 9) is supplied from elsewhere -- the SV-AP-PANEL reverts to a simple full-speed directional control. Thus, I would have to mock up the entire system in order to make the SV-AP-PANEL produce a PWM output that I could measure.
This is why I'm asking if Dynon can simply confirm that the PWM frequency is below 25kHz: because I can't easily measure it until my plane is nearly complete.
