Please check over my LM317 power supply design!

[niflheimer]

With the right filtering caps it actually performs on par with some cheap bench supplies.

I have the same two stage supply working at this moment , with a power NPN across the potentiometer for current control. While you lose about .7V output ( I suppose because of the voltage drop in the diode stage of the transistor ) , the result is smooth and the potentiometer (or in my case a trim pot , was too lazy to drive to the store for another wirewound) is not heating up.

I'v actually been surprised with the this tiny supply. Sure , the current regulation is less than stellar ( it will vary by a few mA - typically 2-3 mA at full load) , and the voltage will drop while in current limit , the voltage supply is quite stable.

My fluke 87 is showing a constant 10.000 if set to 10v with a light load , and it will vary by a few milivolts on a higher load.

[microbug]

OK, having read through the recent posts I have decided to use Paul Price's design. Thanks for your advice!

Edit: On Paul Price's design it says to use a 5V 1W zener diode, but I can only find 5v1. Would a 5.1v diode be OK instead of a 5v one?

[Paul Price]

Yes, any 1 watt rated  5V to 8.2V zener will do nicely.

C4757p:  You say I still end up equating constant current with current limiting...I don't think so.


For instance a 555 timer, a 7805 regulator and the LM317 in this case all have current limiting. The current limiting is not fixed to an exact nor linear current limit value and current limiting varies without user control.

From my experience what I would call a  constant current supply, when we are talking about lab power supply equipment, is usually firstly a constant voltage regulated power supply that switches over to constant current, not current-limited operation.

The word constant means just that, it doesn't vary over the output voltage compliance range of the power supply...current limiting does not require that the current remains constant.

[westfw]

We might just be facing a wording, semantics problem here

I think perhaps so.

Although, I think it's also true that the usual two-317 CCCV supply does do the CC first, and it gets away with it because the current consumed by the voltage regulation stage is small and nearly constant; it doesn't matter whether you regulate the input current (as the circuit does) or the output current (as you'd really want to do.)

[Paul Price]

Westfw: says,"it doesn't really matter..."

Yes it does, that's what this OP's circuit  inquiry is all about, getting a circuit to work to providing a constant current.

[ali6x944]

http://www.onsemi.com/pub_link/Collateral/LM317-D.PDF
in page 8 see there is a circuit there, the circuit is simple and easy but it uses FETs and negative voltages which is quit annoying, but u can still use a simple MC34063 in a inverter config can give u negative voltages, but the FETs ..... well i still haven't find a way around them....

[SeanB]

FET's are easy to get rid of, just use a small constant current source driven from the negative rail, and use that. A simple red LED, run at around 1mA, to drive the bases of 2 single NPN transistors set up as current sources.  Emitter resistors of around 10R will give roughly 9mA of  constant current to emulate those 2 constant current sources. Use a common 100V rated signal transistor in both locations, and thermally couple them as well. 100V transistor only needed for the first, but easier to match, and the current sources only need to be able to sink the regulator quiescent current with zero load.

Your negative supply though has to be able to sink around 30mA at -5V, but does not need really good regulation, as it has those clamping diodes providing a reference. Makes the supply design easier.

[Audioguru]

A lot of replies for such a simple LM317 circuit.
You never said the input voltage or the maximum output voltage you want. Your first circuit with the wrong 560 ohms and 10k pot would have a max output of about 23.6V if the input is at least 26V.
Then the 100 ohm output resistor will fry with 5.6W. Reducing the values of the 560 ohm resistor and pot would eliminate the 100 ohm resistor and its heat.

[Zero999]

The main problem with cascading an LM317 constant current source and an LM317 voltage regulator is, the drop out voltage will be huge. The LM317 needs at least 2.5V to regulate properly (3V is ideal). The current source needs an additional 1.25V of headroom on top of the 2.5V, giving 3.75V. The total drop-out voltage, at the maximum current setting will be 6.25V.

[ZeTeX]

Can you afford a little bit more voltage loss? This version doesn't require as much voltage overhead, so you can get your "negative voltage" with a little trickery: put two diodes in the return path so that the output is shifted up 1.4V or so. That way, the original "ground" is 1.4V less than your new ground. For everything that comes after this part, consider the new low output to be ground and forget about the original one.

Don't let the way I drew it confuse you - R3's not part of the circuit, it's just a load to test it with. The two wires coming out to the right are the output.

In theory, Q2 and Q3 should be matched in gain, threshold and temperature; in practice, this isn't a precision circuit, just make sure they're the same part number and you'll be fine.

Just so you know, you have LT317 in LTspice.

[basinstreetdesign]

Has anyone here yet mentioned that the schematic of the OP has output ground NOT connected to GND?  As it is the output voltage will be at constant 1.25Vdc.  I just thought I'd throw that in...

[not1xor1]

You can get any positive power supply to generate a negative bias voltage with a 555 timer. See attached. It generates about  minus two volts.

if you have a transformer there is no need to add another IC and more high frequency noise, you just need a couple of diodes and a couple of capactitors to get a further negative (or positive rail)

in the example below you can see how it is easy to get both double Vout and -Vout when you need just few mAs or few tenths of mA...

you need just to add a resistor and a zener diode to the negative rail to adjust the voltage to the needed value