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.

TPS40180: Controlling the output via TRK

Part Number: TPS40180
Other Parts Discussed in Thread: TPS40140,

I'm looking for a solution to control the output voltage (or current by design as current source) as simple as possible and in a linear way. The output stage must be parallelized (50A maximum), so the stackability is important and TPS40140 or TPS40180 are ideal an a first view.

The idea was to use the TRK (40140) or SS pin (40180) to control the master via TRK and the slaves have to follow. But is this allowed, what happens after activating of  internal fault circuits at 1.4V and after it, shutting down the TRK pin in a voltage range of 0...0.7V?

I have searched for it, but haven't got a result. I could try it out, but a theoretical approach is needed to be sure.

Thanks in advance...


  • Hi Frank,

    Sorry for the late reply. Just noitice this post.

    When TRK pin is between 0V to 0.7V, the Vout follows TRK voltage with resistor divider ratio.

    When TRK pin is above 0.7V, the internal reference voltage 0.7V will be used for regulation. That means Vout will be regulated at a fixed value.



  • Hello Frank,

    I don't directly work in a group that can support a 50A output current, but I thought I would ask a few questions about your applications request.

    It sounds like you are asking for a circuit that has Constant Voltage/ Constant Current behavior (aka CVCC). When in CC mode you want the upper current limit adjustable from 50A down to some lower value. What is the lowest current you want to set in CC mode?

    What is the voltage level you want to limit to? Does the voltage limit also need to be adjustable or is it fixed?

    Next question pertains to the load itself. What is are the characteristics of the load that it needs a 50A operating current?

    Does the power stage require high-side sensing of the load current or can the load current be measure in the grounded lead of the load?

    --> Low-side current sense is much easier to implement, especially given the high current level of your application request.

    --> High-side current sense is more difficult but certainly appropriate for loads such a batteries that need a CVCC power stage.

    Can your load be subdivided in any way or does the entire 50A go through a single device?

  • Hi Alan, thanks for the response,

    CC would be enough, the load is a single or a string of high power diodes (LED or Laser). That means the voltage would follow the requested forward voltage of the load. The range should be 2V...20V at first. The output current should range between "as low as possible" and 50A, would be depend from the design, esp. the inductance.

    An other feature is low output current ripple, thats why the multi phase design. This would mean: high side current sensing with middle accuracy for current sharing and over current sensing  and  accurate sum current sensing low side. 

    The load could not be divided.

    This is my current level of knowledge to solve such a problem and the target is to find a solution for a multiphase current source with accurate tracking feature (may be externally), active current sharing, neccessary monitoring and protection, stability over all operating points....

    The TPS40140 was from my point of view the nearest, but with some open questions as described before. Alternatives are highly welcome...


  • The output voltage of these two devices is limited to 5.8V because of the 6V Abs Max input voltage rating of the CS1&2 Current Sense and the companion Current Sense RT inputs. Unclear if you can devise an external current sense method to sidestep this limitation of the IC.

    50A x 20V = 1000W output power. This is 10 times larger output power than anything I've ever designed. If you were to manage a design with 90% power conversion efficiency you would still have 100W of heat dissipation in the supply. It will also be difficult to achieve proper stability across such a wide output range.

    Sorry, but this is beyond what I know how to do with devices from my group.

    Do any other TI apps people have a suggestion?

  • Hi Martin, thanks.

    It becomes a bit clearer what I could expect from the 40140 master/slaves configuration and what not:

    - Current sensing of higher CM voltages like 5.8V are done by a voltage divider, this is ok.

    - Over current protection of the slave channels are canceled and must be done externally if needed.

    - Current sharing is active.

    - Current Amplifier is external for my purposes. 

    According 1kW output power class: the problem is clear, you have only one chance, to spread and blow away over a large surface (water cooling is no option here)...