• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 983 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

UCC256404: UCC256404 Won't start up

Peter Baxter
Intellectual 390 points
Part Number: UCC256404
Other Parts Discussed in Thread: UCC25640EVM-020, UCC256304, UCC28056, PMP

I've been having a lot of trouble getting the UCC256404 to start up and run.

Not being able to get a first QR SMPS version to work, I built a second, very standard, 24V 2A version, with the desire to replicate one of the reference designs (I couldn't copy it - as there was no information on the transformer type and turns). So I used the UCC256404 Excel spreadsheet calculator values.

To remove additional issues, I am powering the PFC and QR IC's VCC from an external 15V isolated supply (via a diode). I have shorted BW (pin-8) to GND and removed the diode between the Bias Winding and VCC.
 
The PFC is giving 388V. It appears fine.

My transformer uses "off-the-shelf" EPCOS parts from Mouser. ETD-44 core set with a 0.75mm gap Ferrite and 45 turns (Litz) of primary, giving Lm of 530uH. I've been experimenting with secondary windings and am currently on 5-turns (Litz) which just reaches the 24V output. I initially had 6-turns and found the output went above 24V to about 30V (the opto feedback appears to be ignored).

I have observed inefficiencies in the secondary coupling that is requiring another turn or two.

I observed the opto feedback switching hard high then low - but not having any impact upon the output. (With the UCC256404 removed) I tested the opto circuit successfully activating when the secondary is around 24V.

Adding a 1k load (24mA) to the 24V output results in a much faster output cap discharge rate versus no-load.

I have re-tested the values of all components used. The ICs are all new (not re-used).


I am getting 1 second start-up pulsing.

Considering the main "variable" might be my transformer, I consider ISNS and VCR to be the only values I could work on (change the capacitor value?).

Do you have any ideas on why it won't start?

I've notice that this is the latest evolution in the QR series. Is there a document describing how to get them going?

Attached should be some files

U80495_24V_V1.pdf

  • 0
    TI__Guru* 86266 points

    Hello,

    I reviewed your schematics and the UCC256404 and the connections to the opto an LLC are not in your attached schematic.  Could you include them and repost?

    Regards,

  • 0 in reply to Mike O'
    Intellectual 390 points

    Hopefully this zip file comes through. It includes a 3 page schematic, images and the Excel file.

    Thanks

    PeterU80495_24V_V1.zip

  • 0 in reply to Peter Baxter
    TI__Guru* 86266 points

    Hello Peter,

    Thanks for the schematic and the design tool.  I am in the process of reviewing them presently.  I notice in your original post you had shorted the BW pin to ground.

    Regards,

  • 0 in reply to Mike O'
    TI__Guru* 86266 points

    Hello,

    1. I reviewed the excel tool and the calculated Np/Ns was 8.125 and  your design had chosen 9.  To get the proper output voltage you would generally round down to the lowest whole turn.  I would consider making this 8.
    2. I reviewed your schematic and see that you are using a full wave rectifier on the output with a split secondary.  This is not conventional but you should be able to get it to work.
    3. The only thing that I saw on the schematic that the opto feedback did not look correct.  We have a evaluation module (UCC25640EVM-020) that shows an example of 365V to 410V to 12V 180W output.  The schematic in figure 3 shows how the opto should be connected to the UCC256404. Use this as a reference when correction your schematic.  The following link will get you to the Users Guide for the evaluation module. 
    4. I review your layout and have the following comments.
      1. I can't see where the UCC256404 gnd connects to the power ground.  It should be the C152 and C153 return.  If that pin is missing it could contribute to the misbehavior.
      2. C230 VCC high frequency capacitor should be as close to the UCC256404 VCC and GND pins as possible.  If you use a double sided board you mount the capacitor across the two pins underneath the IC.
      3. The data sheet has layout recommendations and  an example as well that could help you improve your layout.

    5. It might be worth while for you to order a couple of the UCC25640EVM-020 to evaluate the LLC design as well as one for you to modify for your specific power needs.
    6. It might be worth while reviewing the following two application notes as well.  The first one gives practical design guidelines, the second discusses how to select primary magnetizing inductance to achieve ZVS.

    Regards,

  • 0 in reply to Mike O'
    Intellectual 390 points

    Hi Mike,

    Thanks for looking at my problem and your comments.

    1) Good to know. I'll change the winding ratio.

    2) The Full Wave Rectifier was a modification to simplify things - to try to get the SMPS functioning.

    3) The opto circuit was based on the UCC256304 Eval board schematic. The UCC256404 Eval hadn't been released when I designed this circuit.

    4) The grounds do connect. The bypass Cap is next to the IC.

    5) I've ordered an Eval board and initially want to swap my transformer for its transformer.

    Last week, after your response, I changed all values on my current PCB to the Eval boards values (made it a 12V design - so I could use the exact same topology and values). I changed the transformer secondary turns as suggested. Powered it up and every second, it dragged down the PFC DC output voltage. I just have to put it aside, do something else, then come back to it later.

    So I'll work with the Eval Board from now on.

    6) I discovered that the "new-ish" UCC256404, is the latest evolution of a design family (UUCC25630x...). Many documents created for early versions (UCC25630x) are valid for later versions. It might be worth adding any relevant documents (such as slua836a UCC25630x Practical Design Guidelines) to the UCC256404 product page. I wouldn't have though of looking for it otherwise.   

    New questions


    7) On UCC25640EVM-020 Evaluation Module (SLUUBX3B) Figure 4 page 9, what is the bridge rectifier for (self-discharge)? My design is a typical combination  UCC28056 and UCC256404, so I'm not sure whether that circuit is necessary.

    8) The PMP21160 336-W Auxless AC/DC Power Supply Reference Design is good example of what I want to do (combined PFC and QR), but it is possibly a superseded design (18 month old). Is it a complete product ready design? Or more a proof of concept? As in, does it have the Active x-capacitor discharge function implement or does it require additional parts.

    9) The PMP21160 design has a MOSFET switch on the PFC bulk capacitor that never gets explained. Is that circuitry required for a compliant design?   

    10) When I download the UCC256404 EVM Gerber files tidcfi8.zip, I get a warning "trojan" message (Avast) and it closes down. Same with CAD files and probably some others.  

    11) On the UCC256404 Excel Spreadsheet (sluc675a), select the SCHEMATIC tab at the bottom of the page. I've noticed that the CSS field C47, has question marks in it. Not sure whether that is on purpose.

    Cheers,

    Thanks

    Peter

  • 0 in reply to Peter Baxter
    TI__Guru* 86266 points

    7) On UCC25640EVM-020 Evaluation Module (SLUUBX3B) Figure 4 page 9, what is the bridge rectifier for (self-discharge)? My design is a typical combination  UCC28056 and UCC256404, so I'm not sure whether that circuit is necessary.

    >This circuitry is just show how the Xcap discharge circuitry works.

    >You can use diode D8, D9 and R26, R27 and R28 to discharge the XCAP on the UCC28056 design before the bridge rectifier.

    8) The PMP21160 336-W Auxless AC/DC Power Supply Reference Design is good example of what I want to do (combined PFC and QR), but it is possibly a superseded design (18 month old). Is it a complete product ready design? Or more a proof of concept? As in, does it have the Active x-capacitor discharge function implement or does it require additional parts.
    >The PMP design are design for real applications to give the power supply designer are starting point.

    >The Gerber files can be downloaded and modified for the end users specific application.

    >They are designed for 25 deg C, they will need to be modified for the customers specific ambient temperature

    >Reliability testing and production testing needs to be completed by the end user.


    9) The PMP21160 design has a MOSFET switch on the PFC bulk capacitor that never gets explained. Is that circuitry required for a compliant design?   

    > Q100 switches in and out the current inrush limiting resistor RT100


    10) When I download the UCC256404EVM Gerber files tidcfi8.zip, I get a warning "trojan" message (Avast) and it closes down. Same with CAD files and probably some others.  

    > I was able to open the Zip file with Microsoft explorer with no issues.  This may have to do with any security software that is running on your PC.

    11) On the UCC256404 Excel Spreadsheet (sluc675a), select the SCHEMATIC tab at the bottom of the page. I've noticed that the CSS field C47, has question marks in it. Not sure whether that is on purpose.

    > I noticed the same thing.  I will report it too the applications engineer that developed the tool.

    >I did notice that actual soft start capacitor is calculated in the Design and Input and Calculations.  That seems to be working O.K. so you can still select the CSS capacitor.

    Regards,

  • 0 in reply to Mike O'
    Intellectual 390 points

    Thanks for all of that.