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.

  • TI Thinks Resolved

TPS82130: DC/DC converter choice TPS82130 or TPS62130?

Part Number: TPS82130

For a customer of us we are designing a new PCB where we need to generate +5V DC out of +12V for supply of max 6x USB2.0, 2x touch panel (LCD) and for internal logic 

For USB we need max 3A en for 2 displays max 1.9A and for the remaining logic at least 2.5A

Therefore I was thinking of using your TPS82130 converter with integrated inductor. However Webench simulation shows a body temperature of max 105 degrees Celsius at max load which is too much for us (we have no forced cooling).

So I have some options:

- Limiting the max current for the TPS82130 converters to 1.5A . This reduces the body temperature to 59 degrees max which looks acceptable. This means that I need more TPS82130 devices.

- Using TPS62130 with external inductor. Webench shows a temperature of max 69 degrees at 3A load. Still quite high but I think acceptable because the 3A is a worst case situation ( I think that a customer in practive will not use 6 USB ports at max +5Vusb load).

What would you suggest here? The TPS62130 is cheaper but I need an external inductor so the total needed PCB space will be more but possibly I need less converters because the heat is spread between the converter and the inductor. In your datasheet a  COILCRAFT XFL4020-222MEB is suggested. I found a much cheaper 2.2uH inductor with similar specs at Murata but it has a bigger size . Is it possible to use this one or do you suggest the inductor as mentioned in your datasheet?

Best Regards


Chris van der Aar

Sr Hardware Engineer

NTS Systems Development

Eindhoven The Netherlands
.

  • Hi Chris,

    Yes, the TPS82130 has thermal limits as explained in the D/S. TPS62130 is better in this respect, due to not having the inductor's power loss and the better thermal package.

    Yes, you can many different inductors. The D/S has some recommended ones and this one looks ok as well.

    Chris Glaser

    Texas Instruments

    ________________________________

    TPS62147 – New 17Vin, 2A buck with forced-PWM option for low output ripple.
    1% accuracy over temperature, 0.8V to 12Vout, pin-compatible to 4A (TPS62136), for industrial applications.

  • In reply to Chris Glaser:

    Hi Chris,

    Because I only need fixed +5V I could choose the TPS62133 I just found in the datasheet so I don't need the resistors for voltage setting. However T.I also has a TPS62135 and I don't see much price difference between these devices. Could you explain the differences between these devices and when I have a max load of 3A would you suggest a TPS62133 or TPS62135 (both +5V out) especially regarding temperature at full load?
    And about the conductor. Do you suggest a full shielded version here or can a semi-shielded version used here? These semi shielded inductors looks much cheaper compared with full shielded but I can imagine that the EMI radiation will increase with semi shielded?

    Regards
    Chris van der Aar
  • In reply to Chris van der Aar:

    Yes, TPS62135 and TPS62136 are newer devices, which are smaller, support higher output currents, are more accurate, support a wider Vout range, and support a forced PWM mode. TPS62136 will be more efficient due to its lower switching frequency.

    However, due to its smaller package, TPS62135/6 will be hotter than TPS62130/3. If this is the primary concern, then use TPS62133.

    In case it's of interest, here is a USB port reference design that I just did: www.ti.com/.../tida-01567

    On the inductor, a shielded one is better for EMI and noise. But if you are very cost constrained, you might choose an inductor which has both shielded and unshielded options in the same footprint.

    Chris Glaser

    Texas Instruments

    ________________________________

    TPS62147 – New 17Vin, 2A buck with forced-PWM option for low output ripple.
    1% accuracy over temperature, 0.8V to 12Vout, pin-compatible to 4A (TPS62136), for industrial applications.

  • In reply to Chris Glaser:

    Ok thanks. I think I willl go for the TPS62133. Webench shows a max body temperature around 70 degrees at 3A load but this load is worst case in our application so in practice it will be lower.

    I found a shielded inductor with a rather small size (max 6.3x6.3mm) and low cost and can handle max 7.8A and 19mOhm max so I think this one is suitable for this application. For the capacitors I will select the TDK types (X7R) as suggested in your application (and Webench).

    https://nl.mouser.com/ProductDetail/Murata-Electronics/1264EY-2R2N=P3

  • In reply to Chris van der Aar:

    Just one question concerning the TPS62133. What would you suggest for the FSW pin for minimal dissipation at the highest load? At this moment I have connected this pin with Vout so the switching frequency should be set at 1.25MHz. So the same as in the Webench simulation circuit.
  • In reply to Chris van der Aar:

    Yes, this gives the lower operating frequency which is more efficient and therefore has a lower power loss.

    Chris Glaser

    Texas Instruments

    ________________________________

    TPS62147 – New 17Vin, 2A buck with forced-PWM option for low output ripple.
    1% accuracy over temperature, 0.8V to 12Vout, pin-compatible to 4A (TPS62136), for industrial applications.

  • In reply to Chris Glaser:

    Hi Chris,

    In the first place a very happy newyear!
    One question concerning the TPS62133 . I have added these in my schematics where I use 3 of them for different supplies.
    Do we need to take special precautions in case of short-circuit at the connector?
    E.g. like adding polyfuses in series with the output pin? Or will the device switch off or reduce current flow in case of short-circuit?

    Regards

    Chris van der Aar
    NTS Systems Development
    Eindhoven, The Netherlands.
  • In reply to Chris van der Aar:

    Hi Chris,

    Happy New Year!

    The TPS62133 is a fixed 5Vout device. Do you have 3 locations where 5Vout is needed?

    Section 8.4.4 explains the current limit behavior of the IC. In general, it will protect itself but you need to consider if your system needs further protection for it (current rating of wires, for example).

    Chris Glaser

    Texas Instruments

    ________________________________

    TPS62147 – New 17Vin, 2A buck with forced-PWM option for low output ripple.
    1% accuracy over temperature, 0.8V to 12Vout, pin-compatible to 4A (TPS62136), for industrial applications.

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.