• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 61 subscribers
  • Views 526 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PMP6712

Taiwan
Prodigy 70 points
Part Number: UCC28950
Other Parts Discussed in Thread: TL431, , UCC21521, PMP6712, UCC27714

Hello Leonardo
We happen to have a development case like the PMP-6712 architecture. The input battery voltage is 48-60V, the output voltage is 325Vdc (Robot application), the output power is 1600W, and the output voltage of PMP-6712 is much higher. Leonardo:
1 How to change the feedback circuit? Can you support the solution?
2 Does the PMP-6712 have a demo board?
3 Is it possible to provide a circuit description for the PMP-6712?
4 If we try and mass produce according to the PMP-6712 circuit architecture, what advice do you have?
    Thank you

  • 0
    TI__Mastermind 25570 points
    Hello

    1/ The TL431 used in the feedback circuit will not support a 365V output so it needs to be supplied from an auxiliary - fixed rail - perhaps at about 12V. This rail could be derived from a small extra winding on the transformer secondary. The main output voltage is sensed via a resistor divider in the usual way. This arrangement changes the feedback dynamics - see - wenku.baidu.com/.../2235433d5727a5e9856a617e for details. The UCC28950 is optimised to be used on the secondary side of the isolation barrier - this eliminates the optocoupler in the feedback path and the issues about the voltage rating of the TL431. It can also give you faster transient response. You will have to design a small auxiliary PSU - using a push pull for example to generate the 12V needed to power the UCC28950.
    TI manufacturers isolated gate drivers which can take the OUTx signals from the controller and transfer them across the isolation barrier to drive the primary side MOSFETs - UCC21521.

    2/ We don't stock demo boards for our reference designs - normally only one or two are built for use in our Lab. Full design information is available on the PMP6712 page at http://www.ti.com/tool/PMP6712

    3/ I'm afraid we don't have circuit descriptions for this design. I would suggest that you consider replacing the SRs by diodes (SiC) to simplify control. The efficiency penalty is relatively low at such high voltages.

    4/ The single most important thing is to be very careful with the PCB layout to minimise stray inductance in the switching loops and stray capacitance between nodes with high dv/dt and feedback lines into the controller. Please note that the reference design is intended as a design guide - it was not developed as a mass production design.

    Overall - I would consider moving the UCC28950 onto the secondary side to ease the feedback task and to allow improved transient performance. The PMP6712 was developed before we had the UCC21521 isolated gate drive product available and this is one reason the designer chose to put the controller on the primary side. I would also consider replacing the SRs with diodes and also look at a centre tapped secondary rather than a single winding secondary with a bridge rectifier.

    The UCC27714 EVM may be of interest - it uses the UCC28950 as a secondary side controller - an ISO7240 isolator and a UCC27714 high/lo side driver. Neither the isolator nor the driver were available when the PMP6712 was designed.

    Regards
    Colin
  • 0 in reply to Colin Gillmor
    TI__Mastermind 25570 points

    Please  post here if you have any further questions.

    Regards

    Colin

  • 0 in reply to Colin Gillmor
    Prodigy 70 points
    Dear Sir
    Regarding the PMP-6712 circuit, we hope to use this architecture to set up the output 1600Watts 330V robot power supply. Since this situation is high voltage and low current, your recommended feedback scheme uses TL431 in the feedback circuit that does not support 330V output, so it needs to be provided from the auxiliary fixed rail - probably about 12V. This rail can come from a small extra winding on the transformer secondary. The main output voltage is detected in a conventional manner by a resistor divider. If we then provide another protection circuit (OVP.OTP.OLP.(OP-AMP)) for this power supply (12V), is it feasible?
    We also use diodes (SiC) instead of SR to simplify control.
    Do you have any other suggestions?
    Now TI no longer offers free samples? Is it convenient for us to buy samples there?
    Thank you

    Regards
    Dick Liu
  • 0 in reply to Taiwan
    TI__Mastermind 25570 points

    Hello Dick

    SiC diode rectification at 300V makes a lot of sense - the current is low enough that the power lost in the Diodes will not have a significant affect on the PSU efficiency.

    Your point about the voltage on the TL431 is correct and the solution is to power the TL431 from a bias winding or else an extra winding on the transformer secondary. The output voltage is sensed through a resistor potential divider in the normal way.

    One point to consider is that the feedback frequency response changes if the optocoupler anode voltage is constant instead of being proportional to Vout. This is well described by Ridley in the article at https://www.researchgate.net/publication/280308828_Designing_with_the_TL431_-_the_first_complete_analysis     

    Please take care that if you use a regulated bias supply then the optocoupler anode voltage will be constant. if you use an unregulated bias supply which is proportional to Vout then the optocoupler anode voltage will also be proportional to Vout. Given the likely effects of transformer winding resistances and the associated cross regulation between Vbias and Vout I would think that using a regulated Vbias would be a better option.

    You can reduce the voltage stress on the diodes by using a single winding secondary and a full bridge rectifier - This will also reduce the power dissipation in the transformer secondary. The tradeoff is that this puts two diode drops in the rectification path rather than just one.

    Regards

    Colin