HELP !!! My regulators are oscillating

[youyoung21147]

Hello,

I have built two different regulators, one based on LM317/LM337 and the second one is a discrete one.

The schematic for the first one can be found here (first schematic)

http://www.tnt-audio.com/clinica/regulators2_impedance1_e.html

and the second one here (two transistor shunt regulator)

http://www.tnt-audio.com/clinica/regulators_noise4_e.html

R1,R2,R4,R10 = 160ohms
R3=1,62k
R5=24 ohms
D2=16V
C1,C2 = 220µF
Q1=BC550C
Q2=BD140

I have added C3 = 22pF between Q1's collector and ground for stability.



Both oscillate very badly : they do have the correct DC voltage at the output, but my multimeter also tells me the output shows up to 38V AC ripple !!!

They are connected to an unregulated 35V supply.

What I don't understand is the behavior of the discrete reg : when I put a 160Ohms resistor across the output, the voltage drops from 17,8V to 12,3V... And the output shows 25V AC ripple on the multimeter...


Can someone help me ??

 

[Conrad Hoffman]

Build the first one just as they've shown it, but be sure to put an input bypass capacitor within an inch or less of the regulator. I prefer something like a 47uF electrolytic and maybe a .01 disk, but lots of things will work.

 

[AndrewT]

Hi,
two completely different regulators and yet both oscillate.

I wonder if you have provided a DC supply?

 

[youyoung21147]

Conrad : I cannot use the same values as they did, I don't have the resistors...

AndrewT : the unreg PSU is taken from my power amplifier, and it works fine. The AC voltmeter shows no sign of AC on the rails...
I have even used the PSU from another amplifier and the result is the same.


Things get stranger : the Zener reg oscillates no more, but still has this unacceptable voltage dropout. Do you think I should have used something else than the BD140 transistor ?

I have also rebuilt 2 regs around the LM317/337, and both work fine so I don't understand why the first one oscillates while it it is wired the same (same layout, same component values...)

Really strange...

 

[sidiy]

Hi,
Werner posts sometimes here, in this forum. He may be able to help you.

Good luck!

 

[youyoung21147]

Oh my GOD !!!!!!!

I've been fooled by this [censored !!] multimeter !!!!!!!!!!! Holy ****, I've spent one day finding what was wrong, in fact this is what happens :

When used in AC mode, one connection polarity displays a value when connected to a DC source, while the other does not. That explains why I found some oscillation sometimes, and sometimes not.

I got sure of this measuring a 9V battery : one polarity shows 20V AC !!

I'm really angry, it made me loose some precious time to finish my system 🙁


It still doesn't explain the Zener reg behavior, but I think I'll leave it for later 😉