stupid lm317 driver :(

[TTerbo]

i made an LM317 driver.
it uses an lm317 200 ohm pot between the two left pins a capacitor.
when setting the current i can easily get it to 120 then it will make a big jump to 180 (if im lucky and it doesnt go straight over 200) then it goes of the scale on my DMM (goes up to 200mA then 10A )
anyone know why it jumps like this?
cheers
[EDIT] this hapens while adjusting the pot not just randomly

 

[Things]

It sounds like your pot is reaching the end of it's run, and pretty much giving the LM317 zero resistance. What you need to do is calculate the absolute max you want your driver to output, then use a resistor in series with the potentiometer.

This is helpful if you don't know the maths: LM317 Current Calculator - Electric Circuit

 

[lasersbee]

Maybe your pot has already been damaged and the resistive
surface is no longer linear..
Or you don't have enough voltage supplying the Driver...
Or maybe your load is changing...

We have no clue as to how your LM317 is set up....


Jerry

 

[Seoul_lasers]

Yet again it could be a jumpy meter.....? Does the LM317 chip get hot when you get those strange readings?

 

[Bionic-Badger]

Remember that for most of those LM317 designs, if you want higher currents the amount of resistance for the pot is extremely small; you're looking at 1-5 ohms of your pot being used to regulate the current, which is a really small part of the whole range of your pot. You might want to use alternative methods such as the pot being in parallel with another resistance, and using multi-turn pots so you have more precision.

For your multimeter, the amperage meter is probably not very good. Cheap meters usually go up to 200mA, and then have some other setting for higher currents. Even expensive meters may not have great range or accuracy.

A better, and safer method of measuring your current is to construct a dummy load, i.e. a high wattage resistor of low known resistance (e.g. 1W, 1 ohm resistor), and maybe some rectifier diodes (e.g. 1N400n diodes) in series with it to further drop the voltage. Then you measure the voltage across that resistor to get your current using Ohm's Law (Voltage = Current x Resistance). That way you're not putting your meter in series with the driver (it can blow your fuse!), and you can measure any current you want via the voltage metering function, which has a better range than the current metering in general.

 

[lasersbee]

^^^ You are absolutely right... he has no limiting resistor in series
with his pot and his pot value is a bit too high...


@TTerbo...
How about posting a drawing of exactly what you built..


Jerry

 

[TTerbo]

working on the pic....
yeah i have a testload lol should start using it
will i be able to test current with the test load if my DMM doesnt have a fuse in it
it blew and i have to get a new one

 

[lasersbee]

It seems that you need to do a little reading on how to use
a Test Load... The info is all over the Forum...

You set your DMM to low volts and measure the Voltage
across the Test Load 1 Ohm resistor...
1Volt=1 Amp or 1mV=1mA...


Jerry

 

[TTerbo]

i know how to use it.....i just lost it.
i usually just hook the wires up to the DMM and measure the current.

 

[moonreft]

Sounds like a bad pot. Hook it up to your DMM and test its resistance. Turn it very slow, if it jumps at all it is bad.

Might want to use a smaller value pot also.

 

[TTerbo]

ok
Eudaimonium helped me fix it.
a few resistors (3x 1R) in series with the Pot.
cheers

 

[millirad]

Awesome! I think the LM317 is a decent way to power LD's. Their only drawback is that they should be heat sinked at high current flows. Or you can get the LM317K and use a mica spacer witih a finned aluminum heat sink for even better temp control.

 

[PlasmaCube]

Awesome! I think the LM317 is a decent way to power LD's. Their only drawback is that they should be heat sinked at high current flows. Or you can get the LM317K and use a mica spacer witih a finned aluminum heat sink for even better temp control.

Linear regulators like the LM317 have the main problem of inefficiency and, thus, heating. They act as dynamic variable resistors, so the energy that they are controlling gets converted to heat and lost away. In contrast, switching regulators modulate the current extremely quickly between full on and full off, to "simulate" an impeded circuit, and because of this they do not create as much heat. This also ends up being much better as far as battery life goes.

Also, the LM317 is only a buck regulator - it can only lower voltage, so if you need voltages that are higher than your supply, then you'll have to use something else.