LM317 as Precision Current Limiter

Not the first time TI has made a mistake in their datasheets. Should be 1.25.

The Vref in the LM317 isn't exactly 1.25V among all devices, and the device is comparatively noisy. 12.5 ohms should get you close to 100mA. Just drop a 1 ohm from the adjust pin to ground and measure the voltage drop.
 
There is no such thing as "exactly" 100mA.

The tolerance or accuracy required must be specified to be meaningful. Given that the internal reference in an LM317 has a tolerance.

You will either need to select your resistor by measurement of the actual current, or provide for some means of adjustment.

Be aware that at 100mA, the power dissipation could cause significant heating and change in current value.
 
Thanks for your reply
Can you dig into this?
My measurement now is limited to a cheap and bulky multimeter, I have my small lab dismantled for moving. This is why I need an exact 100ma, to calibrate some other measurement tools, at least for the moment.

Trim the value of resistor R1 until the voltage drop across R2 is 100mV

You might want to read this series of articles by Walt Jung on current sources:
Sources 101: Audio Current Regulator Tests for High Performance - Full Article | audioXpress

I'm open to any other way to get 100ma +-1% of current with my actual tools. I choose LM317 beause was there, but I could get other parts.
If you want 1% precision, you have a small issue in metrology -- the Vref as mentioned is not precisely 1.25V, easily varies +/-3%, the 1 Ohm resistor requires precision of better than 1% (parallel 10, 10 Ohm 1% resistors and you will get very close) and lastly the precision of your DMM. If you can get your hands on a Fluke 177 (accuracy 0.01%) or similar you'll be good to go.
 

Attachments

  • LM317_CurrentSource.png
    LM317_CurrentSource.png
    6.8 KB · Views: 219
Last edited:
Hi,
Thanks for the reply.
I think I'll wait for my workbench to be ready again to go for the 1%.


I calculated 12, 12.5, and 13 Ohms and the output should range from 96 to 104ma. With 12.5Ohm and the TI formula I get 96ma. +-4% should be good to go for the moment.


Thanks :grouphug:
 
It seeems to me it is handy use 13 Ohm (or more) resistor, measure real current, and add parallel resistor (1k..3k ?) to this 13 Ohm, to get exact current you want.
I'm not sure about LM317 temperature stability, so you may will have a liitle worse than 1%, if 317 will became too hot. But if it is steady state, you can set current in hot state, so it will be in your 1%.