• State TI Thinks Resolved
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 48 views
  • Users 0 members are here
Support feedback
Options
Options
Related

UCC5881-Q1: Compatibility of UCC5881 in 3000VDC systems

Prathik P
Prodigy 30 points
Part Number: UCC5881-Q1
Other Parts Discussed in Thread: UCC21520

Tool/software:

I am working on a 3kVDC DC-DC converter. I want to know what the compatibility of the UCC5881 driver is for such high voltage systems. I noticed in the datasheet that it has a lower continuous withstand voltage rating of 1000Vrms, compared to 1500Vrms in the UCC5871 IC. Is low continuous withstand voltage rating of the 588x IC a concern for high voltage systems? Do you have any driver recommendations for such high-voltage systems.

  • 0
    Prodigy 30 points

    Also, when is the expected EoL (end of life) for the UCC587x ICs?

  • 0 in reply to Prathik P
    TI__Mastermind 25700 points

    Hi Prathik,

    To reach higher working isolations, you would most likely need a multi-level topology or two series isolation barriers.

    I strongly recommend using the UCC5881-Q1 over the UCC587x-Q1 devices as the 5881 includes significant feature and performance improvements.

    I don't believe that we have an EoL spec for this device. The 5871 uses the same isolation technology as the UCC21520, which has a 29 year projected lifetime, so it may be safe to assume that it is the same if not similar.

    Regards,

    Max Verboncoeur

  • 0 in reply to Max Verboncoeur
    Prodigy 30 points

    Hi Max,

    Let's put the topology (2L or multilevel) aside for a minute. The system design is such that certain insulation faults must be withstood indefinitely, which means that the ICs and boards between the HV and LV boards must be able to withstand 3kVDC indefinitely. 

    Keeping in this mind, we are working on a two-stage isolation strategy with digital isolators + isolated gate drivers. In order to be safe, each isolation barrier must be able to continuously withstand at least 50% (but preferably 70%) of the total bus voltage.

    Based on this, do you still think it is safe to use the newer UCC588x IC? The UCC5871 has reinforced isolation that can continuously withstand 2kVDC, which meets our design requirements, but the 5880 and 5881 both are only rated for 1.4kVDC, which falls short of our requirement.

    Let me know your thoughts on this.

    Regards,

    Prathik

  • 0 in reply to Prathik P
    TI__Mastermind 25700 points

    Hi Prathik,

    I think should be fine to use the UCC588x parts with slightly higher working voltage. If you refer to the TDDB curve on the datasheet, the primary effect of increasing the working voltage is decreasing the insulation lifetime. Luckily, the 588x devices have a very long insulation lifetime at the spec'ed voltage, so there is some room work with.

    The other tradeoff from operating past our spec'ed value is that we can't guarantee performance since we haven't performed full performance characerization at higher working voltages. The device should operate normally at a higher working voltage, just with a shortened expected lifetime.

    If you're interested in digging deeper into TI's isolation technologies and learning more about these specs, I recommend this white paper on how we define and test our isolation technologies: High-voltage reinforced isolation: Definitions and test methodologies.

    Regards,

    Max Verboncoeur

  • 0 in reply to Prathik P
    TI__Mastermind 25700 points

    Hi Prathik,

    The figure is located in the UCC5881 datasheet, and I believe is going to be added to the 5880 datasheet in the next revision. The figure is applicable to both devices as they use the same isolation technology.

    Also, the full UCC5880 datasheet is NDA restricted. Please refrain from posting images of the full datasheet or discussing exact performance/specs on the public forum. I'm deleting your response with the images.

    Regards,

    Max Verboncoeur