• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 62 subscribers
  • Views 266 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.

LMZ14203H: Need an efficient voltage clamp

Stuart Warshawsky
Prodigy 10 points
Part Number: LMZ14203H
Other Parts Discussed in Thread: LM22670, LMZ14203, LMR38025-Q1, LM70840, LM76003-Q1, LMR36015S

Tool/software:

I am using the LMZ14203H where the input is 24V - 36V and the output is 5V. this part has successfully passed our radiation testing.

I had a requirement that needed to switch a load with a 22 V - 36V, 2A load and was going to use the TPS2663x which met this need and has also been successfully tested under radiation. However, the requirement changed where the input voltage is still 24 V - 36V but the load being switched has increased to 2.5A max and the output voltage is now 22V - 34V. So with the input battery voltage at 34 - 36V, I need to limit it to 34V max.

One thought was to use a DC-DC converter and set its output at 23V. This would work but the load would draw more current operating at this lower voltage.

Another thought was what if the above DC-DC converter was set to 34V, when the input dropped to below 34V, would the converter keep the FET on most of the time and more or less just follow the input voltage.

A third option would be to use another FET to switch between regulated output and below the regulated out, pass the input directly to the load.

This design is space limited.

I was considering the LMZ14203H to use as a load switch/regulator. I did not see on the datasheet that the output was limited to 30 V, but that is what Webench indicates. Is this really the case?

I was looking at the LM22670 family as it will handle the 34V output but this will require radiation testing.

Does anyone know if there is a regulator that would just pass the input voltage to the output when the input is below the output set point?

Suggestions?

  • 0
    TI__Mastermind 35210 points

    Hi Stuart, being that this is a COT based converter, I don't forsee you having an issue operating at 34V.

    There maybe a concern associated with the internally, emulated voltage ramp not having sufficient amplitude at the high duty cycle.

    If that is the case, we could try operating at a lower switching frequency, but, you may have a risk of hitting current limit. I believe the low inductor ripple would help.

    If you have the EVM for LMZ14203, I would say it is worth investigating if the device can meet your requirements.

    You need to be very certain that your input is regulated enough that you will not exceed ABSMAX on Vin,

    I am not aware of an integrated solution with a bypass switch.

    Perhaps, if you want to go that route, if you want bypass the input to output you can have a load switch which would be activated by a comparator output with Vout. You could OR the output with Vout. 

  • 0 in reply to Marshall_Beck
    Prodigy 10 points

    Thank you for your quick response. I do have some follow questions:

    1. In the email I received, it said you didn't see an issue operating at 30V but in the forum it lists 34V which is what I would want. I want to verify that it will work at 34V over temperature?

    2. Since I could not find an upper limit in the datasheet, how high a voltage would you expect the device output could regulate to? I don't want to risk being on the cusp of failure. Failure of this device could potentially result in a failed mission.

    3. I have a few EVMs on order and will experiment. Do you believe if I am able to get satisfactory results after tweaking the EVM, that it will be repeatable on hundreds of units? I would assume that to be the case as long as I am not exceeding some parameter internally that I'm not aware of.

    4. Can Webench simulate this scenario? I tried with mixed results.

    5. Since we are running off a battery (there's no chance of exceeding Vin), the input voltage and FB pin will continue to drop during the mission. Does it matter how low FB goes below 0.92V? Does the ramp rate change with different FB values?

    6. Do you know what kind of duty cycle I should be able to achieve?

    7. I considered something similar to your circuit but am space constrained. There are 2 of these circuits on a very small board with other components.

    8. Any thoughts on a different device (LM22670 for example) that might work better?

    Thanks for your assistance.

  • 0 in reply to Stuart Warshawsky
    TI__Mastermind 35210 points

    Hi Stuart, you are welcome.

    1. I revised my post. I meant to say 34V
    2. We don't have a spec' the concern isn't necessarily having an electrical overstress risk, but, achieving stability. I understand the concern and I definitely don't want to steer you in the wrong direction
    3. I am hopefully you we have strong confidence of a robust design by evaluating on the EVM. I can provide suggested changes. The concern would be if Vin were to reduce and how it would effect the stability. If Vin were to come very close to Vout, then, the Vout ripple may become large and switching frequency may no longer be constant.
    4. I can take a look and follow up
    5. Related to point 3. I don't understand how FB pin will drop in voltage?
    6. The device can operate up to almost 100% duty cycle, but as the duty cycle becomes very large, their maybe irregular switching and increased output ripple
    7. ok
    8. For this high vout application, if you are considering a current mode control buck, you may face issues with trying to stabilize the converter when the converter has fixed internal compensation. In addition, with the high duty cycle and using a power module with a fixed inductor and current mode control, you run into the problem of slope compensation issues. A COT based device such as avoids those two pit falls, but, there must be sufficient vin to vout differential for the given switching frequency and selected inductor to ensure that the switching can be consistent.

    With higher output voltage you can decrease the load, which is good based on our product offerings.

    I think if possible (based on your input supply), it is best to keep the vin to vout differential as big as possible

    Here is a few converter options that I think would be a good fit, have some risk that due to stability, but would enable the smallest size:

    LMR38025-Q1

    LM70840

    LM76003-Q1

    LMR36015S