0
\$\begingroup\$

I'm attempting to design a PoE PD for fairly high-power applications; chiefly, powering/driving stepper motors. Currently, both motors and drivers run at 24V and will consume about 30W per PD. At that kind of power, regulation gets a little expensive, particularly when wanting isolated regulation.

My question is: if I do something like the below (which is heavily simplified), replacing the motor drivers (I think the motors can handle the higher voltage already) with higher-powered versions and tapping off the unregulated PoE voltage, will I be violating any important standard, safety or otherwise, that would present a hazard or cause problems if I wanted to get the device certified for commercial sale? Note the Tx/Rx pairs are device-side and already isolated, everything network-side is just the one line on the left.

Application will be a microcontroller and one or two minor ICs

\$\endgroup\$

2 Answers 2

0
\$\begingroup\$

The problem is that POE isn’t unregulated anywhere.

As to the safety factor, that probably moot, because a standard Poe supply is going to shut the powered device down if it pulls too much current. Also, who is going to want to buy or use something that isn’t Poe compliant?

As to the physical aspect of it, it isn’t realistic to do this. 30w/48v is 0.625a. While I don’t know your application exactly, I believe this would take a carefully crafted power supply and motor control to work around the limitations.

You might check out some of the Poe options from LT/analog devices for greater than 30w options.

\$\endgroup\$
1
  • \$\begingroup\$ IEEE802.3bt specifies power up to 30W, and we've found parts/a PSE that can supply it, so that's not a concern. We could even go to 60W, but we don't need that much. \$\endgroup\$ Nov 26, 2019 at 3:51
0
\$\begingroup\$

Afaik, isolation in PoE is not required, but a strong preference. The power is already 57-52VDC. The issue is whether the stepper motors consume current lower than the overcurrent threshold at all times. So the current spec must not be a mean value, I would measure it. PoE PSE chips are notoriously sensitive to that, in bad designs they trip from interference. The long, relatively highZ cable from the PSE to PD might also be an issue there. You might be drawing a current under the threshold, but when you consider resonances it might still be instantaneously high and trip the PSE.

I would give it a go, though. It might be just hard, but what is not?

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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