• Join
  • Sign In with my.TI Login
Texas Instruments
  • Products
  • Applications
  • Tools & Software
  • Support & Community
  • Sample & Buy
  • About TI
Sample & Purchase Cart Sample & Purchase Cart
  • Search
  • Advanced
TI E2E™ Community
  • Support Forums
  • Blogs
  • Groups
  • Videos
  • 简体中文
  • More ...
TI Home » TI E2E Community » Support Forums » Power Management » AC/DC and Isolated DC/DC Power » AC/DC and Isolated DC/DC Power Forum » ucc28070
Share
AC/DC and Isolated DC/DC Power
  • Forum
Options
  • Subscribe via RSS

Forums

ucc28070

This question is answered
Arthur Padget
Posted by Arthur Padget
on Mar 13 2012 09:39 AM
Prodigy100 points
We are currently working on a design using your UCC28070 chip.  The board we are using is similar to the TI PR779/PR780 EVM boards, but we added some things to make it work for our project.  We made the schematic in such a way that we could bypass our additions, leaving us with an equivalent to the PR779/PR780 EVM boards. 
We started out using the board configured for our changes, with values calculated for our project.  The result of the testing was that we were losing a FET, and blowing the input fuse.  We then decided to disable phase B by removing the mosfets.  With phase B disabled we were able to regulate to the voltage we had planned for under light loads (300mA @ 390V).  However if we switched over and used phase B and disabled phase A we the mosfet would fail and blow the input fuse.
We were concerned that we may have calculated the values wrong, so we grabbed a second PCB and used the values from the EVM board.  We populated both phases of the driver, and turned on the board.  We experienced the same results, dead FET and a blown fuse.  We then disabled phase B again, and again we had were able to get the board to regulate to the correct voltage.
We have checked our schematic over and over again, and feel that we have the same connections as the EVM board.  We also think that it is unlikely that the PCB has an error on it as it is only a 2-layer board and visually appears to have good alignment.  Is anyone aware of any conditions that result in a phase not working, while the other seems fine?  We also didn't know if there was some function of the part that only used one phase but not the other?  Specifically we weren't sure if RSYNTH was connected to both phases in the same way.  Any help would be appreciated.
UCC28070
Report Abuse
  • Reply
You have posted to a forum that requires a moderator to approve posts before they are publicly available.
All Replies
  • Mike 1414007
    Posted by Mike 1414007
    on Mar 13 2012 10:26 AM
    Verified Answer
    Verified by Mike 1414007
    Mastermind18045 points

    PR779/PR780 was a reference design that systems engineering used to evaluate the UCC28070.  There is a 300W EVM that you can order and evaluate http://www.ti.com/litv/pdf/sluu312b.  It also has a proven layout where you can download the Gerber files if you like.  http://www.ti.com/litv/zip/sluc111  http://www.ti.com/litv/zip/sluc110

    If phase A is working and phase B will cause the FET to fail.  It could be something wrong with phase B's current sensing network.  The current amplifier feedback might not be seeing a current sense signal.  If that is the case the peak current limit for that FET would not work also.  This would cause the FET to blow up.  Note this is only one scenery to what could be causing the FET to fail.

    You might try evaluating your design by limiting the input current of your AC source so the fuse does not blow.  In this way you can evaluate the current sense signals to make sure everything is O.K.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Arthur Padget
    Posted by Arthur Padget
    on Mar 14 2012 13:08 PM
    Prodigy100 points
    Mike,
     
    We limited the input current to our circuit and we were able to turn the board on without incident.  However with the current limited the input voltage dipped so low that it was not able to regulate (presumably because phase B is trying to draw too much power).  Both phase A and phase B went to a maximum duty cycle and the input voltage became non-sinusoidal.  This made it hard to trigger on either phase, but from what we could see it looked like both CSA and CSB were ramping up in the same way, and appeared to be working.  When it is regulating (with just phase A enabled), we are able trigger easily and see the signal ramp up then go to 0V during each switching period as expected.
    We have gone back through our board and values.  We made sure that there were not any soldering errors or incorrect parts on the phase B.  We have done more testing with only phase A enabled, which seems to work great by itself.  We wondered if there was anything else that could be the problem besides the current sense feedback or the current amplifier compensation since we have been over these circuits very carefully.  Specifically we were wondering if the R_Synth, R_IMO, and R_PkLmt blocks treat both phases exactly the same, or do they use only one phase and extrapolate to the other phase.  We're unclear on how the R_Synth is able to synthesis two current waveforms with only one resistor.

    You indicated that the PR779/PR780 was not a good choice for a starting schematic.  Since we are designing a 1.5kW design the 1.2kW reference seemed like a better starting point than the 300W reference.  Is there a problem with the PR779/PR780 design?  The only significant difference that we could find was that the 300W termination network for the current sense transformers used a resistor and capacitor instead of a diode and a zener, and the 300W design has a light-load biasing circuit so that the gate drive biases the current sense signal.
    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Arthur Padget
    Posted by Arthur Padget
    on Mar 19 2012 09:26 AM
    Prodigy100 points

    Mike,

    I thought I would post a follow up on our continued testing.  We were able to get phase A working well with a new set of calculated values, however we are still trying to figure out what is wrong with phase B.  With only Phase A enabled we are able to regulate to the desired voltage and took the opportunity to take some measurements.  While the current sense signal did a great job of mimicking the AC wave form, we noticed that our meter was showing a PF of .54.  After using our oscilloscope to monitor input Voltage and current sense we noticed that there was a phase shift between Current and Voltage.  We did this testing at a fairly light load, is the result of this test due to the light load or maybe the fact that only one  phase is enabled?

    We look forward to hearing your thoughts on Phase B, and our questions concerning Power Factor.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Mike 1414007
    Posted by Mike 1414007
    on Mar 19 2012 10:33 AM
    Verified Answer
    Verified by Mike 1414007
    Mastermind18045 points

    The definition of power factor is real input power divided by RMS output power.  At lighter loads there will be a phase shift between input current and input voltage.  This will drive down the power factor.    It should not matter if you have one phase or two for a given power level. 

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Arthur Padget
    Posted by Arthur Padget
    on Mar 22 2012 09:23 AM
    Prodigy100 points

    Mike,

    We were able to find the problem with phase B. The Gate drive signals from the UCC28070 were running under one of the high current traces on phase B.  We filtered that line and after that we have been able to run with both phases.  

    While attempting to get both phases working we had calculated values that would result in peak current limit at a very small output level.  Now that both phases are running we changed our values back to our intended values.  After the switch we noticed that there is no longer a phase shift between current and voltage.  However we did notice that the input current is no longer sinusoidal at any load.  We were wondering if any of the following changes adversely effect power factor.

    Going to two Phases.

    Peak current limit was raised to 3.75V.

    Switching frequency went from 125kHz to 60kHz.

    Termination values for the current sense transformers were changed.

    Voltage compensation network VAO was chagned.

    Both Current compensation networks were changed COA and COB.

    Imo resistance was changed.

    We noticed that the 3.18 volt over voltage protection limit was being reached on our Vsense pin.  We increased the capacitor on that filter in order to filter unwanted noise, however our filter RC constant is much longer than recommended in the Data sheet.  We used a 3300pF capacitor and a 1.1megaohm resistor making for a time constant of 3.63 milliseconds, compared to the recommended 100 microseconds.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Mike 1414007
    Posted by Mike 1414007
    on Mar 23 2012 08:33 AM
    Mastermind18045 points

     

    Look at the voltage across your IMO resistor.  It needs to be a rectified sine wave.  If it is not more than likely you have too much filtering at the VINAC pin.  It might be worthwhile for you to order one of the 300W EVM to evaluate.  You can look different points and compare them to your design.  Points of interest to you might studying the Rimo CSA/B outputs and the voltage amplifier output.

    By heavily filtering Vsense you might be masking another problem.  You might have too much output ripple voltage and that is what is triggering the over voltage protection.

    Answers to your questions below

    Going to two phases

    >If the load is the same PF for one or two phases should be very similar.

    Peak current limit was raised to 3.75V.

    >Using a larger current sense signal will reduce crossover distortion caused by current amplifier offsets.  This will lower THD.  This could improve power factor.

    Switching frequency went from 125kHz to 60kHz.

    >Switching frequency should not have an effect on power factor.  However, the inductance needs to be designed for the specific frequency in question.

    Voltage compensation network VAO was changed.

    >This does have an impact on PF and THD.  If the 100/120Hz ripple is not attenuated this can result in third harmonic distortion which will lower power factor and increase THD.  Generaly the voltage loop should cross over at 10 to 20 Hz.  The voltage amplifier 100/120Hz ripple should be less than 50mV.

    Both Current compensation networks were changed COA and COB.

    >If the current loop is not stable the CS signals will not track the voltage at IMO.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Arthur Padget
    Posted by Arthur Padget
    on Mar 23 2012 09:30 AM
    Prodigy100 points

    Mike,

    Thank you this post was very helpful, I will check our board based on your advise.  

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Arthur Padget
    Posted by Arthur Padget
    on Mar 26 2012 13:53 PM
    Prodigy100 points

    Mike,

    We have made a lot of progress on this board, we are now trying to improve our power factor over the output range.  We are happy with the power factor at full load, and we understand that power factor will fall off as load decreases.  We were wondering if you had any tips to slow down the power factor roll off.  

    Also in a previous post we discussed the phase shift between current and voltage at light loads.  Is this phase shift caused by the input capacitance(2.2uf film capacitors) becoming a higher and higher percentage of the load?  Do you have any tips for reducing the magnitude of the phase shift?

    Thanks for all of your help.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Evan Shultz
    Posted by Evan Shultz
    on Jun 20 2012 16:42 PM
    Prodigy65 points

    MikeO

     Look at the voltage across your IMO resistor.  It needs to be a rectified sine wave.  If it is not more than likely you have too much filtering at the VINAC pin.  It might be worthwhile for you to order one of the 300W EVM to evaluate.  You can look different points and compare them to your design.  Points of interest to you might studying the Rimo CSA/B outputs and the voltage amplifier output.

    Hi Mike,

    What's the best method to measure across RIMO? It's very sensitive. Attaching a probe makes the supply chirp and make other funny noises. While it doesn't get damaged, the supply is unable to start.

    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
  • Shishir Kulkarni
    Posted by Shishir Kulkarni
    on Sep 06 2012 13:49 PM
    Prodigy30 points

    Sir,

    The problems stated above are same to mine too, i.e. I have blown a FET and the fuse too. In the above discussions you have mentioned, that the problem was solved, but not how it was solved. Kindly provide the details as to what were the changes you made after the FET blowing in Phase B started till you got both phases working and till achieving desired 0.9 p.f. I am building a bridgeless converter as given in SLUA517, but at reduced switching frequency. The above fault occurred at no-load. Please provide me a solution ASAP. Thank you.

    boost converter AC-DC Bridgeless UC28070
    Report Abuse
    • Reply
    You have posted to a forum that requires a moderator to approve posts before they are publicly available.
TI E2E™ Community
  • Support Forums
  • Blogs
  • Videos
  • Groups
  • Site Support & Feedback
  • Settings
TI E2E™ Community Groups
  • TI University Program
  • Make the Switch
  • Microcontroller Projects
  • Motor Drive & Control
Other Communities
  • Deyisupport
  • Designsomething.org
  • beagleboard.org
  • TI on Element 14
  • TI on TechXchangeSM
Other Technical & Support Resources
  • WEBENCH® Design Center
  • Product Information Centers
  • Technical Documents
  • TI Design Network
  • TI Technical Articles
  • TI Training

All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.

Content on this site may contain or be subject to specific guidelines or limitations on use. All postings and use of the content on this site are subject to the Terms of Use of the site; third parties using this content agree to abide by any limitations or guidelines and to comply with the Terms of Use of this site. TI, its suppliers and providers of content reserve the right to make corrections, deletions, modifications, enhancements, improvements and other changes to the content and materials, its products, programs and services at any time or to move or discontinue any content, products, programs, or services without notice.

Follow Us Texas Instruments on Facebook Texas Instruments on Twitter Texas Instruments on LinkedIn Texas Instruments on Google+
TI Worldwide | Contact Us | my.TI Login | Site Map | Corporate Citizenship | mobile m.ti.com (Mobile Version)

TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs and
embedded processors, along with software, tools and the industry’s largest sales/support staff.

© Copyright 1995-2013 Texas Instruments Incorporated. All rights reserved.
Trademarks | Privacy Policy | Terms of Use