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.

TPS92512EVM-001: Search

Part Number: TPS92512EVM-001
Other Parts Discussed in Thread: TPS92512, TPS92512HV, PMP6024, UCC28811

The TPS92512EVM-001 has Vout=5-25V while the input VIN =12VDC-48VDC, my question is:

How to define the max VOUT and Why it has to be just 25V, can't it be greater? 

In my design I have a power source with constant voltage Vin= 60V/1.5A and need a wide range VOUT= 20-50 V, ILED (max) = 1A . I am considering to use the TPS92512HV instead for the TPS92512. I plan to modify the TPS92512EVM-001 to achieve this output specification. I have tried the Webench simulation for the upper limit of VOUT (50V) and lower limit on 21V have got a suggested designs. Obviously the designs give different component values. I am not sure how to achieve the full VOUT range 20-50V. I will also need to control the LED with PWM and it has to be between 0-100% duty cycle.  

Is this right way to go? Is there any calculation spreadsheet with the related equations available for customer usage?

  • Hello Conny,

    The design equations are in the datasheet and pretty easy to follow. But this EVM is pretty flexible since it is already populated with high voltage components. The 25V output limitation is only because C5 is a 25V capacitor. So if you change that to a 50V capacitor you can use the EVM as-is for your requirements. However, at the higher input voltage the AC losses will be greater, so I would recommend a lower switching frequency. If I wanted to modify the EVM to fit your specs I would do the following:

    -Change C5 to a 50V capacitor

    -Change L1 to 47uH

    -Change the switching frequency to around 350kHz

    -Change the UVLO level to around 55V (ideally it is above the highest output voltage)

    -Change R8 to 0.3ohm

    That would give you a design that works well.

    Regards,

    Clint

  • Hi Cliton!
    Thanks for the advise! I will try that. The resistor
    Another question regarding the IADJ pin.
    In case 1 analog surrent adjustment not use: I assume that the IADJ pin will be connected through pull-up resistor 10M to VIN and have VADJ on 1.8V, the ISENSE pin has therefore 300mA. And by change R8 to 0.3 ohm the current limit will be 1A.

    In case 2 this pin use to set max ILED. When a good number of LED current limitation steps between (250mA- 1000mA ) are required I consider to use external voltage source (reference voltage on 1.8V) to supply a resistor divider an connect to a DIP switch.
    I consider to adjust the switch frequency by connect the RT/CLK to the PWM output of the MCU. With that I hope to achieve a better performance of the driver circuit. Do you think this is right way to design a selectable LED current limitation. The R8, L1 and C5 are defined for highest level 1A.

    Br
    Conny
  • Hello Conny,

    Case 1 is correct. That uses the internal 1.8V clamp to set 300mV on ISENSE.

    Case 2 you can use the same circuit and externally apply 1.8V to IADJ for 1A and 450mV to IADJ to get 250mA. Using the synchronization function to set the switching frequency is fine, but I'm not sure what performance benefit it would give you. It's really only useful if you need to have a specific switching frequency or if you have multiple circuits that need to have the exact same switching frequency.

    Regards,

    Clint

  • Hi Clinton,

    I think to apply the external voltages to control the max. current output levels is clear. Another way to define the output current which I have experienced is adjusting the PWM range but by applying different voltage to the IADJ input the PWM range will be the same for all output levels.

    I have not any required specific switching frequency, just concern about the wide range output levels but as you say there is not any benefit with different switching frequencies. 

    So what is your recommendation about to define a fixed switching frequency for this application while the Vin is around 60V and Vout = 20-50V and Iout = 250mA- 1A? Previously you had mentioned that I should keep the switching frequency around 350 kHz, to keep the AC losses low due to the higher input. To my understanding the lower switch frequency the bigger output inductor is required but the higher switching frequency the lower supply efficiency. I am not sure how to find the best balance, but if the efficiency can be around 95% it would be accepted. Is there any guideline on how to define this parameter?

    When trying the Webench simulation on the above mentioned output current levels I have seen different values on R1 and R2 (voltage divider to the input UVLO) with the same input Vin max/min. According to the equations for calculating R1 and R2 it should not be different, do you have any clue on that?  

    Another question about the input UVLO setting TYP value = 11.5V, is that correct? The Maximum rating in the datasheet SLVSCT1B states that it is 5V. I may 

    Br,

    Conny

  • Hello Conny,

    I am not sure exactly what you are trying to achieve with your dimming, but you can use analog alone, PWM alone, or both at the same time.

    AC losses do get exponentially higher (given the same Fsw) with higher input voltage. Due to that 95% efficiency is very difficult for any regulator to achieve. But to get even close you may need to go down to 100kHz to 200kHz. This will require a higher inductor value and also a larger inductor if you want to keep the DCR low for improved efficiency. Even at 350kHz you might still get close to 90%, but 95% may be hard. But if that is your main spec then go with the lower range.

    Webench usually uses the output voltage to spec the UVLO setting, making sure there is some margin above Vout before switching starts. That may be the reason for different values, but webench does have some oddities and is only a good starting point usually. The 11.5V spec in the user's guide is the VIN at which the UVLO pin itself crosses the 1.22V threshold.

    For your application I would set UVLO at around 55V as I mentioned. Try R1 = 1.5k and R2 = 66.5k.

    Regards,

    Clint

  • Hi Clint,
    Thanks for the advises, what I am trying to do is using both analog and PWM dimming at the same time. I did some tests on the EVM board and it works well.
    I have considered to use this device, the TPS92512HV together with the UCC28811DR and found that TI have a reference design very close to my application, the PMP6024. I am looking for the Design Spread sheet for the UCC28811DR for to be able to design for instant the switch transformator to the AC/DC part.

    Do you think it is possible to get som design documentations, calculations and simulation data on the PMP6024 ?

    Appreciate your great support!

    Br,
    Conny
  • Hello Conny,

    I am going to close this one out as I see you have another post regarding this. The folks that respond to that one are the correct people to support the UCC28811.

    Regards,

    Clint

  • Ok Clint, so far there is no answer on that post.
  • Hello Conny. It looks as though somebody responded. Sorry for the delay, I do not know the person handling that.