This is meant for the knowledge base and just a contribution to the Murphy user base. I am not sure where to post it

Problem statement: It will be seen that if you drive a proportional valve via JUST PWM from the PDM that there is a severe feedback and interference to the PWM driver and the response is not linear and "jump"s in steps. The same response is typically seen even if you use a external motor driver ( as to be expected of course).

Our answer was found to be a "low pass" filter using the inductance of the coil(manufacturer specific), a capacitor and resistor. Just google it ( Low pass filter), there are many sites that will give you values to use.

we found it better to use the 1000 hz PWM and not the 100Hz pins. It should not matter if you use highside or H bridge, but understand the coil does not care what direction the current flows!

The trick is to realize the resistor must pass full current. In our case, trial and error we used a 3.3ohm ( 25w!!) resistor and 4.7MFarad capacitor and got a decent ( almost linear) response to PWM control over the full range. The resistor must be rated for full continuous current!

the alternative is to actually measure the current flowing (using the parameter from the pdm that is available), but we found it only was available in 125mA steps which is less than 8 steps over the whole range of our valve and not nearly practical or accurate enough. Ideally, if the PDM gave feedback in much smaller steps, you could just measure the current and control it that way. This method ( measuring current) is the way most hydraulic companies controllers work

NormanButchGrant

Thanks for sharing this,