LM338 Regulator - Voltage And Current Adjustment.

crutschow

Joined Mar 14, 2008
34,283
..................
a 12v zener is 20 cents.
You are so smart, design the circuit without the 12v zener.
You're such a sweet talker. :p

But a 12V zener has a poorer temperature coefficient than the regulator's built in reference.

And I don't have to design the circuit.
It's shown in post #8. All you have to do is add the pot across R1.
 

hp1729

Joined Nov 23, 2015
2,304
Yes they are. But I m trying to understand stuff, with the minor knowledge i got. An answer like "no the author is not wrong" it would be perfect for me.

No I wont. I will built one. Tomorrow i ll start experimenting on the circuits i ll choose tonight.

Thanks anyway.
Re: "Tomorrow i ll start experimenting on the circuits"
Good attitude in my opinion. The higher objective is the learning. Anybody can buy a power supply if that were the only objective.
With some hands-on time you get a better understanding of why circuits are built the way they are.
And yes, you have a problem in the current limiting section of your schematic. Way too much resistance.
 

AnalogKid

Joined Aug 1, 2013
10,987
Regarding the circuit in post #14, the TS did not express himself clearly. He is correct that the majority of the load current does not go through the current limit adjustment pot, and that is a good thing.

The circuit uses the 317 as a smart pass transistor, a pseudo-emitter-follower. When operating normally, the 317 output always is 1.25 V greater than the adjust pin. So it acts sorta kinda like an NPN darlington emitter follower - unity voltage gain and lotsa current gain, but with -1.25 Vbe (negative Vbe means the emitter is above the base).

For the input voltages shown on the schematic, the zener diode current is between 6.4 mA and 12.8 mA. An LM317 with a conventional voltage adjustment circuit has a typical adjust pin current of only 50 uA, a tighter reference tolerance, and a significantly lower temperature coefficient. So while the circuit is an unusual application of the LM317, it is not at all clear if it is better than the standard circuit.

Colin, as always, your responses make the rest of Australia proud.

I think he is the inventor of the universe..
He certainly does.

ak
 
Last edited:

crutschow

Joined Mar 14, 2008
34,283
Regarding the circuit in post #14, the TS did not express himself clearly. He is correct that the majority of the load current does not go through the current limit adjustment pot, and that is a good thing...................
Yes, that's true. I misread his answer. :oops:
Actually the pot could have a higher resistance so that essentially no load current goes through it since all it has to do is provide a small current to the transistor base.
 

crutschow

Joined Mar 14, 2008
34,283
Below is the LTspice simulation of a current-limit circuit with an LT1084 regulator (similar to an LM338 but with a lower dropout voltage). It uses a comparator to sense the shunt resistor voltage which allows the use of a smaller shunt resistor, giving a current limit range of a few mA to the maximum current limit of the regulator (5A in this case) with one resistor of a moderate power rating.
U5 generates the stable reference voltage for the current limit.
C3 provides negative feedback compensation to prevent oscillations in the comparator output when it is limiting.

The simulations shows results for current limit pot settings (U2) of 0%, 25%, 50%, 75%, and 100%, over a load resistance range of 25Ω to 2Ω (top trace).

Note that since the minimum output voltage of these types of regulators is 1.25V so you can't limit the current below that voltage output (e.g. a short circuit output).
Edit: If you need controllable current limiting for a short circuit, then the circuit would need to be modified with a power MOSFET in series with the regulator input.

upload_2016-4-6_10-49-59.png
 

Attachments

Last edited:

Thread Starter

vol_

Joined Dec 2, 2015
93
Lacking 0.56R resistor and other components, after all i build the circuit in the attachment. Everything seems to go good, except that my minimum voltage is about 7V, while expecting smthing like 1.3V..

If anyone has an easy clue about this, tell me.. For the moment the regulator is working

Crutschow thanks for the last post, my next regulator will be something like it. After I study power regulators..

thanks everyone!
 

Attachments

crutschow

Joined Mar 14, 2008
34,283
The voltage across the pot at zero setting should be near 0V.
If it's not then you either have a faulty pot, as Colin stated, or you have some sort of miswire.
 

Thread Starter

vol_

Joined Dec 2, 2015
93
LM338 Voltage Regulator - Conclusions (?)

The project I've been trying to built its done. After all, I found out that:

1) I got a counterfeit LM338

2) The AC/DC transformer used has 38V output (no load), while rated for 24V.

3)When i fed the (probably counterfeit) LM338 with 38Vdc the minimum voltage output was about 7.8V. When i supplied the input of the LM338 with 9V the minimum voltage output was 1.3V. I was using the exactly same circuit (on strip board).

4) When i supplied a TDA7052 guitar amp with 8.80V through the regulator circuit of the LM338, the sound of the guitar on the speaker was oscillating heavily and with strong sustain, while when supplied through a battery was not. I dont know what this means..

I think that with a genuine lm338 this would n't have happen. But i dont know..

If its easy to someone to understand something by these conclusions and comment, i would be grateful.

Salute!
 
Top