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.

LM317: The minimum output current as constant current source

Part Number: LM317

Hello

I'm using LM317DCYR as a constant current source for an amplifier +-12V circuit as below

the current is 1.2/300 = 4ma, it seems fine. But I just read some content which said that the minimun output current should be >10ma, that's mean the R should be <125ohm. 

I also found that some people use 1k resistor to get 1.25ma and still working well.

So....I'd like to know if it's okay for my case?

Thank you!

  • Hi Alex,

    I suppose it depends on your end application and risk tolerance. If it works fine in your application and this isn't a mass production type of project, maybe it's fine. However, if it is a high volume project and you have field failures coming back with this device, we're going to point at the datasheet and say that it says to have a minimum 10mA load. So, my official recommendation for a professional project would be to not do this, but ultimately it will be up to you.

    Regards,

    Nick

  • Hello Nick

    Thanks you for kind advise.

    there is also a minimun load current stated in datasheet 3.5ma while Vi-Vo=40V

    And I just found a topic about this spec

    https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/405442/lm317-minimum-current

    It mentioned that the minimun current will be less for lower Vi-Vo (Vi-Vo=40V plot to 4ma)

    Does it mean that in my case(12V), current above ~2ma should be safe? not sure what does extrapolate curve means?

    Thank you very much.

  • Hi Alex,

    The EC table spec is saying that most devices need only 3.5mA but some devices need up to 10mA. The way to interpret the data in the figure provided by Ron is that the blue data points are real measurements, and the red data points are the same data but scaled such that the worst-case data point (at VIN-VOUT=40V) is at 10mA required minimum current to represent a device that falls at the limit of the worst-case spec. I don't think "extrapolated" is quite the right word for it. So, most devices will be fine with the way you are using it, but there will still be some devices that won't have their minimum load current requirement met. It will probably be a low percentage of devices that fail in this use-case, but not zero, so my official recommendation would not change.

    Regards,

    Nick

  • Hi Nick

    I'm a little bit confused about the devices means, "low percentage of devices that fail in this use-case". Does this devices means LM317?

    Some of LM317 may not met the 3.5ma at Vin-Vout=40V spec, it may need up to 10ma. Am I corrent?

    Thank you!

  • Hi Alex,

    What I mean is that statistically, most devices will be satisfied at the lower current level when process variation is accounted for. Think of the "typical" values as 1 sigma, while the max values are 4 or 6 sigma. What you said is correct.

    Regards,

    Nick

  • Hi Nick

    Thank you very much for your explanation! It helps a lot