LM 317 R2 ?

[petemoore]

 LM 317 adjustable regulator shows a variable resistor between the adjustment pin and ground.
What would be a good fixed value adj. resistor for minimum voltage drop ?
All wired up working and humming like it's not even there...using a 5k drops voltage about 2volts less than input.
Wired earlier with a 470ohm variable resistor adjusts voltage but only up to 4.7volts max.
Input voltage is 9.1vdc from a linear full wave rectifier/big capacitor..
I couldn't view any recommended values for R2 adj. pot, but vaguely remember it being a 5k I used with success in the past.

[theehman]

http://www.qsl.net/dl5dbm/Projekte/Netzteil/Regulated%20Power%20Supply,%20Variable.htm

[davent]

For the LM317 to regulate you need  an input voltage at least a few volts greater then the desired output voltage. So with a 9v input and R1=200r, R2 would need to be 920r for regulation to 7v.

R1=200r, R2=5k, output voltage would be 32.5v if the input voltage was high enough above the regulator dropout.

http://www.reuk.co.uk/LM317-Voltage-Calculator.htm

[R.G.]

What he said, but more:
The LM317 needs 2V more than it's output voltage to regulate, which is another way of saying that the output voltage can can't be higher than 2V less than the input voltage. So with 9V in, it can't put out more than around 7V regulated, but can regulate down to less than 7V from a 9V supply.

The three pins of the LM317 are input, output, and adjust. The inner workings let current through from the input pin to the output pin to make the output pin be 1.25V higher than the adjust pin.

So if you ground the adjust pin and put a 240R resistor from output to adjust pin, the output voltage is 1.25V.

If you make the resistance between ground and the adjust pin be more than zero, then the 317 STILL makes the output be 1.25V higher than the adjust pin. But the resistor between adjust pin and output still sees 1.25V, so a current of 1.25V/R flows. If you use the suggested 240R resistor there, a current of 1.25V/240R = 5.2ma. The adjust pin is designed not to have much current at all in or out of it. The same current flows in the resistor from output to adjust pin as flows from adjust pin to ground. You program the current through the resistors by changing the top resistor, and the total output voltage by changing the bottom resistor.  The LM317 likes to see resistors near 240R for programming its voltage, so that's always a good starting point.

So a resistor R between the adjust pin and ground sees that same 5.2ma flowing, and the output voltage then becomes 5.2ma*R_{(to ground)} + 1.25V. If the top resistor is 240R and the bottom resistor is 1K, then the output voltage is 5.2ma*1240R = 6.448V nominally. The maximum output of about 7V with a 9V input is reached with a resistor from adjust to ground of 1105 ohms and a programming resistor of 240R.

The actual output voltage is Vout = (1.25V/Radj)(Radj+Rgnd) = 1.25V* (1+Rgnd/Radj)

[tubegeek]

the output voltage can be higher than 2V less than the input voltage.

... the output voltage*** can't *** be higher than...

[R.G.]

the output voltage can be higher than 2V less than the input voltage.

... the output voltage*** can't *** be higher than...

Oh, sure, insist that I think as slowly as I type...   ]

Corrected.

[Jdansti]

There's always a first time for everything...

[tubegeek]

He thinks, I proofread. Works for me.

[petemoore]

 Yupp, it makes variable output voltage but I haven't been able to make it de-rippulate, there is still strong hum coming through.
The 'top' resistor between adjust and output is 220r, and 1k from adjust to ground, input voltage is 9.1vdc, and output is 6.4vdc still humming strong.
I must have somehow miswired the circuit. I'll try again.

[R.G.]

Yupp, it makes variable output voltage but I haven't been able to make it de-rippulate, there is still strong hum coming through.
The 'top' resistor between adjust and output is 220r, and 1k from adjust to ground, input voltage is 9.1vdc, and output is 6.4vdc still humming strong.
I must have somehow miswired the circuit. I'll try again.

Three terminal regulators like the 78xx series and 317-style can suppress ripple by about 40db - about 100 to 1 - without special work on the regulator circuits and surroundings. When the regulator saturates, this suppression drops to zero, and the output follows the input, about 1.5-2V down from the input. So if you want a regulated voltage of 7V, your input must never drop below 9V, even for an instant, or you'll hear it as hum. It's OK for the ripple to be a volt or two coming in, but the lowest valley of the input voltage still has to be more than the regulator minimum dropout voltage. So you can filter 12V with 1V p-p ripple to 9V and get quiet DC out. You can filter 14V with 3V of ripple and get quiet DC. But at 11V in, you can't have any ripple that doesn't go through to the output because there's just enough input voltage to barely let the regulator work for a 9V output.

How well filtered is your input voltage ( which is equivalent to saying "what's the minimum instantaneous voltage on your input DC, not the meter DC reading?")

[petemoore]

 Tried a new board..
100uf electrolytic + .1uf between input +/-.
258ohm between adj and output.
1k from adjust and gnd.
100uf between output and gnd.
BRAAAAAAPPPP
very strong power supply input AC tone signal.
Nasty little buggar...ordered a power supply.
Might get around to pinching the circuit board between bootheel and concrete !
Nice clean job on the board build, 9.1vdc input, 6.77dc and ac output..tone is fantastically loud this time.

[R.G.]

Have you read power supplies basics at geofex?

100uF first filter sounds very, very small unless the current being pulled is small.

Or you may have a shorted rectifier, or solder problems. There are many other possibilities.