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.

LM2842 does not regulate when not loaded

Other Parts Discussed in Thread: LM2842


We are using the LM2842YQMK/NOPB in one of our projects.

It is used to supply a power amplifier at certain times. When used, it has to output 13.5V@300mA. When not used, there is virtually no power consumption on its output apart from the voltage setting bridge.

We tested about 300 boards, measuring the output voltage when loaded, and also when not loaded. We had no problems up to now. But on the last lot of boards, we discovered that on some boards, the LM2842 was not able to regulate when not loaded.

In the following capture of the output voltage, we see that every 150ms, we switch off our power amplifier for 14ms. The consequence is that the LM2842 is no more able to regulate on the defective boards, up to the point where we enable the power amplifier again.

This behavior could not be a problem if the settling time of the LM2842 was very short. But as can be seen below, it takes 800µs before to reach the target voltage. This results in the loss of some modulated data bits coming out of our power amplifier.

I noticed that the addition of a resistive load of 22kΩ solves the problem. This adds less than one mA of load, but is enough to remove the regulation loss.

Unfortunately the data sheet does mention any issue like this, and therefore I am not sure that the addition of a 22kΩ load will solve the problem in any case.

Therefore my question is : what is the minimum load in order to ensure regulation in load dump situation in any case ?

Here is the WeBench schematic that we use:

NOTA: By the way, I saw no mean to define the minimum load within WeBench. It seems to be forced to 10mA by default.

  • As a follow up of this issue, we found out that

    • Our input voltage is higher when the circuit is loaded, due to the design of this input voltage. Therefore the (Vout - Vin) difference is small when not loaded, and larger when loaded.

    • When (Vout - Vin) is small the Toff time is too small in the worst case of the LM2842 data sheet, as pointed out when using TI Power Stage Designer

    This closes this post.

  • Aprocha,

    Thank you for following up on this post. We have many product experts that monitor these forums regularly, so typically you will receive a response quickly. I apologize that you did not get a response initially, but please continue to post and add valuable content to our forums. 

    Best,
    Michael