• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 63 subscribers
  • Views 1020 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM2675 as inverting regulator.

martin b
Intellectual 560 points
Other Parts Discussed in Thread: LM2595, LM2675, TPS5402, TPS54233, TPS5420, LM2650, LM2672

I'm currently using an LM2595(DDPAK/TO-263) as a split rail design with a coupled inductor, and this has worked well so far, but its huge (Application Note 1118.pdf). I need something more compact but because I'm still rolling my own boards I cant use anything with a power pad, its just too difficult to install with just an iron.

Can you recommend some regulators that will allow me to do this. I understand the Simple Switchers are perfect for this arrangement.

I have also noticed the LM2675 comes in an SOIC package, will this replace the LM2595(DDPAK/TO-263) successfully, are there any pros and cons that will mess up my day. Also I'm not sure if this is the correct forum for the next question, but can I use any of these devices listed below in a similar way to what I have described above.

TPS5402
TPS54233
TPS5420
LM2650

Cheers

Martin.

  • 0
    TI__Genius 15810 points
    Hello,
    Looks like there isn't a simple answer. The SOIC package will be different from the DDPAK. There is not thermal pad in the SOIC package. That also seems to be the case for most of the other TPS parts you have listed. But otherwise the LM2675 theoretically should work alright. But the LM2675 is a voltage mode control device and it could get a bit challenging to stabilize this circuit. The TPS5402 is a current mode part available in SOIC package. It looks more suitable, but it has external compensation. If you do not have the expertise you will need to seek support from the folks in the "Non Isolated DCDC" forum. Hope this helps.

    Regards,
    Akshay
  • 0 in reply to Akshay Mehta
    Intellectual 560 points

    Hi Akshay.

    Can you give me some more details on what would be the challenges to stabilize the LM2675 circuit? if it was to present any issues what might they be?

    Martin

  • 0 in reply to martin b
    TI__Genius 15810 points
    The LM2675 is a voltage mode control device. The resulting con is that the inductor and output capacitor double pole sits at frequencies below the crossover and causes the phase margin to drop significantly. The current mode control devices do not have this problem, because the internal current sense loop breaks this double pole interaction and results in two real poles (the low frequency capacitor pole and the high frequency inductor sampling pole). The situation then gets compounded by the fact that you desire to use this in a inverting application. In the inverting application there is a right half plane (RHP) zero. This element behaves as a zero in the magnitude plot, but causes phase decrease in the phase plot.

    The RHP zero can't be easily compensated and usually we get around the problem by designing the crossover to be lower than the RHPZ frequency. To compensate the double pole you'd have to either add feedforward capacitors in parallel with the upper feedback resistor and/or use aluminium electrolytic capacitors at the output that have relatively higher ESR. All this while you would have to make sure you do not cause the crossover frequency to go too near the RHPZ frequency. Hope this helps.

    Regards,
    Akshay
  • 0 in reply to Akshay Mehta
    Intellectual 560 points
    Thanks Akshay, helps a lot.
  • 0 in reply to martin b
    Intellectual 560 points
    One other thing Akshay, in relation to Application Note 1118.pdf, how small an ESR can I use and still be stable, can I use parallel X7R ceramics to replace the output capacitors C2 and C3? ie: 2 x 22uF on each output. Also the same for the input capacitor C1.

    Martin.
  • 0 in reply to martin b
    Intellectual 560 points

    bump

    Akshay, in addition to the above question, are you able to tell what the startup current draw would be for the LM2675 in the split power configuration, as I noticed in the circuit I'm currently using (the LM2595 in split power configuration), its high on initial power up.

    Martin.

  • 0 in reply to martin b
    TI__Genius 15810 points
    Hello Martin,
    The LM2595 and the LM2675 most likely have very short softstart times. This would mean that the device attempts to bring the VOUT up pretty quickly. This means the caps at the output will pull current in short time to get up to the applied voltage. This translates to large initial startup current. If you could consider, LM2672 has a softstart pin and comes in a DIP package.

    Regards,
    Akshay