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.

UCC3895: the different feedback circuit between UCC3895EVM-001 and PMP9622

Part Number: UCC3895
Other Parts Discussed in Thread: PMP9622, , TL431, PMP

Hi Team 

My customer is curious about the different feedback circuit (EA+ EA- EAO) between PMP9622 and UCC3895EVM-001

why we designed two types of feedback circuit, any different and which one is better?  




EA- and EAO are connecting each other,  

  • Hi Red

    Interesting -
    The PMP9622 design is a CV/CI circuit (Constant Voltage/Constant Current) - it uses some op-amps on the secondary circuit as error amplifiers and this is then transferred onto the primary through an optocoupler. The opto output is fed into the error amplifier of the UCC3895 through R11. The UCC3895 error amplifier (EAN, EAP, EAOUT) is configured as a unity gain inverter.

    The UCC3895 EVM has a CV output and uses a TL431 as the error amplifier - again, the error signal is transferred to the primary through an opto-coupler. The UC3895 error amplifier is configured as a voltage follower - EAN connected to EAO and the signal is fed onto EAP.

    In both the EVM and the PMP9622 designs if Vout increases, the error amplifier will sink more current in the optocoupler which increases the current in the phototransistor. The difference is that the increased current in the PMP9622 increases the feedback voltage but in the EVM the increased phototransistor current reduces the feedback voltage. The PMP9622 photo transistor pulls FB up, the EVM phototransistor pulls FB down.

    The different FB signal polarities in the two designs require that the error amplifier on the UCC3895 either invert the FB signal (PMP9622) or simply passes it through (voltage follower (EVM)the EVM inverts the feedback signal.

    Which is better ? There is little difference between the two approaches. It might be that the EVM requires the opto to operate closer to saturation at no-load (it has to pull the EAOUT pin below 500mV to trip the no-load comparator) but most optocouplers can do this without difficulty (Vcesat typ 200mV on the H11A87 used on the PMP - the CNY17 DS I looked at doesn't give a VceSat so that would be a question to the opto mfr).

    I think it comes down to different designers, at different times making different decisions - both of which are workable solutions to the problem.