• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 232 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.

TPS23730: TIDA-050045 - substitute parts

David Miller36
Prodigy 110 points
Part Number: TPS23730
Other Parts Discussed in Thread: TIDA-050045, CSD18503Q5A, UCC2897A

Hi,

I'm designing a product based on the TIDA-050045 reference design. One of the synchronous rectifier transistors is NRND (BSC105N10LSF), pretty much all FETs are very hard to obtain now, 52 week lead times, etc. I'm searching for alternate parts that we can use. Some parameters are obvious, Vdsmax, Vgsmax, Idmax, Rdson. What about Ciss, Coss, Qg, Qgd, Qoss, rise and fall times, Qrr, trr? Is there a rule of thumb for choosing them, eg, lower C, Q, rise and fall times are better and/or acceptable?  How much increase in charge, capacitance, rise and fall times can be accepted? Any particular traps in choosing MOSFETs for this design? 

Thanks,

David

  • 0
    TI__Expert 3700 points

    Hi David,

    Great question. Are you using Q8 in your design? If so, the Q8 FET can be used to replace Q6.

    To answer your question directly, there are three main considerations when choosing this FET:

    • Voltage and current rating
      • Typically you will want to use the same voltage and current rating FET. However, the Q6 FET voltage rating is actually overkill in this design and a lower rated voltage FET can be used such as 40V. As an alternative for Q6, Q8 (CSD18503Q5A) can be used to replace Q6.
    • Rdson 
      • A lower Rdson will have lower conduction losses so better efficiency. 
    • Gate charge
      • In ACF designs the MOSFETs need to switch very fast and slow FETs are not recommend. 
      • There is a higher risk of shoot through if higher gate charge device is used because this slows down the on/off time of the FET. 

    Please let me know if you have any questions about this. Thanks! 

    Best Regards,

    Brett Colteaux

  • 0 in reply to Brett Colteaux
    Prodigy 110 points

    Thanks Brett! This is very helpful!

    I don't understand what you mean by "slow Rdson FETs".

    The CSD1503Q5A unfortunately is unavailable and 40 weeks lead time, so that has to be substituted as well.

    The CSD1504Q5A IS available, but it has ~ 50% higher Rdson. It also has ~ 50% lower Qg. Does the lower Qg cancel the higher Rdson so that it is acceptable? It also has 20% lower Id continuous, but it is still 50% above the 10A output current.

    Are the considerations similar for choosing Q11 (the main switch) and Q10 (the clamp)? While there is some stock on them, the lead times are also long and I need multiple sources.

    Best wishes,

    David

  • 0 in reply to David Miller36
    TI__Expert 3700 points

    Hi David, 

    Apologies about the confusion, I edited my comment above to make it more clear. 

    Higher gate charge and higher Rdson than what is used in the design would both impact the efficiency of the design negatively.

    I don't think they will cancel each other out but if I were to choose between two similar FETs: one with lower gate charge and one with lower rdson, I would choose the one with lower gate charge based on my experiences in order to improve efficiency. But I don't see an issue with using CSD1504Q5A in this design.

    For Q11 and Q10 the same considerations apply as in my post above. In addition I recommend to pay special attention to the PFET current rating. Choose one that is the same or higher because the peak current can be high during shutdown. 

    I hope this helps. 

    Best Regards, 

    Brett Colteaux

  • 0 in reply to Brett Colteaux
    Prodigy 110 points

    Thanks! Again, very helpful.

    Two further questions, the first regarding Q6 and Q8. You said Q8 can be used to replace Q6. The specs of Q6 regarding Rdson, Qg, and even Idmax are much looser than Q8. Can Q6 be used in place of Q8 (the reverse direction of what you said)? In other words are the requirements for both sychronous converters similar, and can the specs of Q6 regarding Rdson, Qg, and Idmax be considered an outer bound for the specs of Q8 as well?

    Second question is, can you refer me to an app note that will help me understand the operation of active clamp forward converters? I have found several articles, but they are all pretty basic and are silent on how to choose components.

    Best wishes,

    David

  • 0 in reply to David Miller36
    TI__Expert 3700 points

    Hi David,

    In general the specs for the series FET and the parallel FET will be different. Typically the series FET voltage rating will be higher than the parallel FET voltage rating for ACF. So the voltage on the series FET should be higher and you could use the same FET part number on the parallel FET as the series FET. 

    As FYI, the designer for this reference design was trying to minimize the amount of changes from the EVM so he only changed the series FET but we are working on getting this updated on TI.com. 

     

    For ACF resources, here is an app note I recommend which discusses fundamentals of ACF power stage:

     

     

    Let me know if you have any other questions. Thanks!

     

    Best Regards,

    Brett Colteaux