LM317 CCS - Any limitations?

[Archwn]

Just my purely curiousity and would worth make my understanding of LM317 CCS more clear. Hope somebody could point me out.

My question will be, if LM317 forms a CCS of 1.5A, then i put 1 meg resistor to its downstream to the ground, will I get 1.5A through this 1Meg resistor? I don't think it will work but having limited knowledge in electronics make me seek for the explanation.

Many thanks,

AK

 

[AndrewT]

Hi,
the 317 adjusts it's throughput to ensure ~1.25V between it's output and adjust pins.
Placing a resistor between these pins, as shown in the datasheet, allows the regulator to behave as a CCS.
Insert a 10r resistor and the 1.25V will give a 125mA CCS.
Now measure or calculate your worst case maximum input voltage to the CCS and use this to determine the maximum voltage drop across the CCS. then check how big the heatsink needs to be.

If you need 1.5A CCS then the resistor will need to be ~830milliohms.
The 317 won't last very long.

 

[Gordy]

Power = Current (squared) x Resistance
so 1.5 Amps of current through 1000000 Ohms of resistance would produce 2,250,000 Watts of power.

Voltage = Power / Current
so 2,250,000 Watts of power divided by 1.5 Amps of current requires 1,500,000 Volts.

The 317 is good, but that may be pushing things a little!

 

[Circlotron]

If you need 1.5A CCS then the resistor will need to be ~830milliohms. The 317 won't last very long.

If the '317 has enough heatsinking it should be just fine, in my experience at least. Bear in mind it will dissipate 1.5 watts for every volt dropped across it, so you would probably have the '317 tab straight to the metal of the heatsink (with heatsink paste) rather than use an insulating washer with it's poorer thermal conductivity.

 

[Circlotron]

Voltage = Power / Current
so 2,250,000 Watts of power divided by 1.5 Amps of current requires 1,500,000 Volts. The 317 is good, but that may be pushing things a little!

Technically at least, if you had a DC supply of 1,500,000 volts for the resistor and 5-37 volts for the LM317, so a total of 1,500,005 to 1,500,037 volts then it should work alright... Just make sure I am not there when you turn it on!

 

[Archwn]

Hi AndrewT,

Thanks as usual. I mean after I form a 1.5A of CCS by placing your 0.83R between it's output and adjust pins, I should get 1.5A out of that. Then I place 1Meg resistor from this point to ground. I'm sure the current wouldn't be that much but cant really explain what will really happen.

Hi Gordy,

I think you are right. Let's say if I have only 10volts before the 317, I forms 1.5A CCS as per AndrewT's, then I place 1Meg in downstream, what's actually happens? Will the current only be at 7.5(roughly)/1,000,000 Ampere? Then will there be a problem in CCS loop as they set out to be 1.5A at first?

Hi Circlotron,

I got your point. 317 is a floating type and my example here is quite a bit exagerated though.

 

[AndrewT]

I mean after I form a 1.5A of CCS by placing your 0.83R between it's output and adjust pins, I should get 1.5A out of that. Then I place 1Meg resistor from this point to ground. I'm sure the current wouldn't be that much but cant really explain what will really happen.

Hi,
a large value resistor fed from a high voltage operates as a current source. Small variations in the load make almost no difference to the CCS current.
To avoid the need for these very high voltages and large resistors, we instead use solid state CCS circuits, that mimic this behaviour from more usable voltages.
If you fit a 317 CCS and use it to feed a high value resistor then the resistor will dominate the CCS behaviour.
Only when the current in the resistor has increased sufficiently to activate the 317 CCS action will the 317 become the controlling element.
To pass 1.5A through 1M will require 1.5MV. for all supply voltages below 1,499,999V the resistor controls the current. when the voltage increases to 1,500,000 the 317 starts to take over control.
at 1,500,003 the 317 has 3V across it and is starting to work well as the CCS controlling element. But, it is now dissipating about 2.625W [(3-1.25)*1.5]. it will need a heatsink. Increase the supply voltage another few volts to 1,500,007 and dissipation has gone to >8.6W. You are close to blowing up the317 for a 0.0003% increase in voltage.
Forget it.

 

[Speedskater]

In AudioXpress magazine Walt Jung wrote a 2 part article on current regulators.

http://www.audioxpress.com/magsdirx/ax/addenda/index.htm

4/07
Sources 101: Audio Current Regulator Tests for High Performance.
Part 1: Basics of Operation
Walt Jung
Jung2778.pdf

5/07
Sources 101: Audio Current Regulator Tests for High Performance.
Part 2: Precise High Current/Voltage Operation
Walt Jung
Jung2779.pdf

 

[Archwn]

Hi AndrewT,

This is what I 'm looking for. The basic behind this thing. Thanks for the explaination. I really appreciated your help.

My example here is too extreme. Just to clear up my head how to use 317 as a CCS properly.

Hi Kevin,

Thanks for the links. Definately will have a look in to it.

Cheers everyone.

AK