2
\$\begingroup\$

I asked a question the NXP forums here about the TEA2208T (datasheet) which is an active bridge rectifier. I was told this part is a full wave rectifier and not a peak rectifier and therefore a rectified AC voltage would look like this even with a smoothing cap after rectification: "Full wave rectifier"

Where a peak rectifier's output would look like this with a smoothing cap: "peak rectifier"

  1. I cant really find anything on "peak rectifiers", never heard of that term before either. I typically hear what the NXP rep calls a "peak rectifier" as a "full bridge rectifier"

  2. How would the two rectifiers produce a different output even with both having a smoothing cap? My understanding is the active bridge rectifier does the same thing as the diode bridge but at a higher efficiency by essentially replacing the losses from the forward voltage drop of the diode with the Rds_on from the mosfets? I would think no matter how I get to the rectified AC signal, placing a smoothing cap on the end would produce a DC output.

Edit (to respond to Cristobal's answer): So I believe I understand now. Say for the positive half cycle of the ac input, two mosfets will conduct and we can idealize those mosfets as a normal conductor like so

enter image description here

Which is simply (image edited to remove cap value and clarify the voltage source is not vcc)

enter image description here

So the cap just follows the input voltage even as the input voltage drops below it's peak value. Where as in the diode bridge configuration, the cap will charge to the Vpeak value (minus Vf of diode) and once the input drops below Vpeak then the diode stops conducting and the cap is essentially cut off from the input and discharges at the rate of it's RC time constant rather than following the input voltage like in the mosfet bridge.

Edit #2:

Looking at their eval board for this part they show the output waveform of the IC eval board enter image description here

The body diodes do the initial rectification and you can see that the output signal is DC for the first cycle. After that first cycle the mosfets kick in and take over the rectification. And you can see that you no longer get a DC output but just rectified AC.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Yeah, I think you were led astray. The top circuit should similarly filter the ripple, provided it is large enough. \$\endgroup\$ – td127 Sep 18 '20 at 18:47
  • \$\begingroup\$ To your edit: I totally agree with your analysis. But the C1 in your diagram is not the important "Cvcc" from the datasheet nor the final filtered load capacitor. In fact it's not mentioned at all in the datasheet (where did you get 1uF?) It might even be a mistake as it appears to not have any purpose. \$\endgroup\$ – td127 Sep 18 '20 at 22:05
  • \$\begingroup\$ the 1uf value was just in there because i was running a simulation. took it out of the image. Yeah, i agree to the output cap (before the diode-cap combo after the pfc controller) seems to not serve any purpose but was putting it in there just to at least understand what is happening to it in this case. \$\endgroup\$ – Michael Sep 18 '20 at 22:16
5
\$\begingroup\$

The difference here is that a diode allows current to pass in one direction, while the FETs (while on) allow current in either direction. Presumably, NXP designed it to keep the appropriate FETs on at any time one of the input voltages is higher than the other. Since you're conducting through the entire half-cycle, your capacitor is being discharged through the bridge as well as being charged. The FETs will stay on until the input polarity reverses, while the diodes will stop conducting anytime the instantaneous input voltage is lower than the voltage on the cap.

\$\endgroup\$
4
  • \$\begingroup\$ I think this is correct. If you still want use this device to make DC, if you put a diode on the output into a capacitor, then you will get an output similar to a diode bridge ("peak") rectifier \$\endgroup\$ – Pangus Sep 18 '20 at 19:46
  • \$\begingroup\$ I'm dubious of the claim that part of the FETs function is to conduct in the other direction. NXP has designed this for high efficiency: it's nonsensical that they'd add circuitry to actively discharge a cap for which energy has been expended to charge. What the FETs are adding is a low impedance forward path to overcome the diode drop. But maybe I'm completely misinterpreting your answer. \$\endgroup\$ – td127 Sep 18 '20 at 20:44
  • \$\begingroup\$ I believe I understand. Please see edit to my question. Wanted to respond with images \$\endgroup\$ – Michael Sep 18 '20 at 21:52
  • \$\begingroup\$ @td127 It may be that NXP wasn't thinking of charging a cap, but providing near-ideal rectification. A diode (regular or active) on the output could provide peak detection, but this device by itself doesn't seem designed for that. If it were, it could drive the transistors only when the input DC voltage was higher than the output. Michael, did you have any questions about the new content you included? \$\endgroup\$ – Cristobol Polychronopolis Sep 22 '20 at 20:37
2
\$\begingroup\$

The top diagram is a bit misleading: it includes a cap on the output that reasonable adults would assume to be a power supply filter cap.

Upon reading of the TEA2208T datasheet, it appears that cap is not for that purpose - it's probably a small stabilization cap or some such. The TEA2208T is just doing the full wave rectifier function, so its VR output is in fact just that. It's better than a standard diode bridge because in overcomes the forward drop of diodes by using cleverly controlled MOSFETS.

But to get a true filtered output you do need to follow the whole thing with a diode-cap combo as shown in the datasheet:

enter image description here

On power off the MOSFETS are used to discharge surrounding caps, but not during normal operation.

\$\endgroup\$
4
  • \$\begingroup\$ So in terms of efficiency when DC is desired at the output, I guess this half as an improvement as I thought it was over a typical diode bridge. Since the diode-cap combo is needed at the end you're still getting a one diode drop. Still an improvement over two and is still really good for high power applications im sure but was thinking you'd save on two diode drops. \$\endgroup\$ – Michael Sep 18 '20 at 21:57
  • \$\begingroup\$ There's probably something I'm missing about the PFC Control mechanism on the output. That (Power Factor Correction) is something I'm not at all familiar with so am suddenly in over my head. Maybe the cap we've been looking at is really associated with the PFC circuit.. \$\endgroup\$ – td127 Sep 18 '20 at 22:18
  • \$\begingroup\$ just added an edit again. the eval board datasheet for this part is interesting too. most of the schematics in there show a 1uf cap on the output but the BOM doesnt show it and I dont see it on the layout. but the output waveform I put in the edit makes me believe there is. Maybe it's external to the eval board when they did the testing though.. link: nxp.com/docs/en/user-guide/UM11348.pdf \$\endgroup\$ – Michael Sep 18 '20 at 22:33
  • \$\begingroup\$ Interesting! The demo app note goes into great detail about the power lost in that capacitor (due to large AC voltage across it) while acknowledging there is much smaller loss in the diode rectifier (DC with ripple). The benefit appears to be "better THD" (?) I don't get it but as I said, I'm out of my league: I'm only licensed to +48V :) \$\endgroup\$ – td127 Sep 18 '20 at 23:05

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.