3
\$\begingroup\$

I'm trying to understand the PoE reference designs contained in the datasheet and the evaluation board user's guide for the TI TPS23756.

The major difference between the "typical application" and the actual design they ship seems to be the role of the second transformer.

The schematic in the datasheet makes sense to me -- the control voltage for the chopper is phase-shifted, so it is synchronous to the current on the secondary windings on the main transformer, so using that signal for switching will give a DC signal.

The evaluation board instead builds a perfect diode to create a half wave rectifier, which is also a valid approach of course, but they still have a second transformer hanging off the gate voltage, and feed its secondary side to the current sense input on the PoE controller.

  • Is this really proportional to the current on the primary side of the main transformer?
  • Can I combine both designs, use a current sensing resistor on the primary side and a perfect diode on the secondary, and leave out the second transformer completely?
  • Am I missing something?
\$\endgroup\$
  • \$\begingroup\$ Just to make sure we're on the same page. Datasheet: which figure? Eval board has three transformers T1, T2, T3; which one are you referring to as "second" ? \$\endgroup\$ – Nick Alexeev Dec 23 '15 at 5:07
  • \$\begingroup\$ In the datasheet, figure 27 in the "Application and Implementation" chapter is what I'd think of as their recommendation. In the eval board, T1/T2 are alternatives, so T3 is the second transformer in this case. \$\endgroup\$ – Simon Richter Dec 23 '15 at 10:37
1
\$\begingroup\$

enter image description here

T3 is used to check the maximum current flowing into the mosfet drivers - it's a current limiting device. If you look in the normal data sheet it replaces the component below: -

enter image description here

Can I combine both designs, use a current sensing resistor on the primary side and a perfect diode on the secondary, and leave out the second transformer completely?

There's no connection between current monitoring on the primary (via T2, a current transformer) and secondary side rectification.

Is this really proportional to the current on the primary side of the main transformer?

It's a 1:100 CT so it's going to work fine even though the burden (R12 @ 11 ohms) is after the rectifer D9.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Ah, I see, so these have nothing to do with each other, and the current monitor with the transformer avoids the inline resistor. \$\endgroup\$ – Simon Richter Dec 23 '15 at 11:50
  • \$\begingroup\$ I reckon so but I'd be keen on avoiding T3 and using a resistor. I have a suspicion that this can be done but unfortunately xmas shopping is going to rule the next few hours. However, I'd want to determine max current flow through the primary and work out why the chip needs that level of amplification using the 100 to 1 step up. \$\endgroup\$ – Andy aka Dec 23 '15 at 12:00

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.