Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum

Message Sort: Post Order or Asylum Reverse Threaded

LM317 Regulator

71.81.113.220
Posted on January 1, 2023 at 18:55:16
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007




I'm testing a breadboarded 12 VDC power supply using a LM317 regulator, which I built a long time ago, but only used once.

As shown above, the (13,600 uF cap) filtered input DC is 19.26 VDC. My meter is a true RMS type, and the AC at the LM317 V in pin checks 4 to 5 mV.

The output voltage checks 12.22 VDC (which is what I have it set for), with the AC on the output measuring between 27 to 35 mV.

I've bypassed R1 with an optional 10 uF cap, which is supposed to increase ripple rejection from 65dB to 80 dB. I'm at a loss as to why I apparently have more ripple on the output of the regulator than I do on the input. Could the LM317 be defective, or am I measuring the wrong thing? (I don't own a scope)

These measurements are taken with no load, if that makes any difference.

Thanks.

 

Hide full thread outline!
    ...
RE: LM317 Regulator, posted on January 1, 2023 at 23:24:16
Triode_Kingdom
Audiophile

Posts: 10049
Location: Central Texas
Joined: September 24, 2006
 The data sheet states that the minimum load current to maintain regulation is 3.5mA typical and 10mA maximum. Maybe adding a 390Ω resistor across the output will help.

 

RE: LM317 Regulator, posted on January 2, 2023 at 03:18:56
RayP
Audiophile

Posts: 727
Location: Maryland
Joined: June 30, 2005
 If you would like a long read, you could wade through a thread on DIYAUDIO about a noise canceler for an LM317 regulator.

ray

 

minimum load current, posted on January 2, 2023 at 05:21:22
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
 I could possibly see if the ripple wasn't reduced by the unloaded regulator, but the fact that it seems to actually increase the AC component over what it's being fed is what has me scratching my head. I wasn't sure if it's a defective LM317, my meter, or something else which I might have overlooked.

I'll try adding a resistor across the output and see what happens. Thanks TK.

Update: I added 390 ohms across the output, and the AC component still checks about five times what's reading on the regulator input.

 

RE: LM317 Regulator, posted on January 2, 2023 at 05:35:23
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
 My question isn't so much as to how I could improve LM317 performance, but, as stated in my reply to Triode Kingdom below, to try and understand why I seem to be getting more AC ripple on the output of the regulator than I'm feeding in.

My plan was to see if this supply offered any audible improvement over a switch mode wall wart on a device drawing about 200 mA. If it did, then I would build a supply using a better regulator than the LM317.

Thanks for the link, Ray. I'll check it out.

 

RE: minimum load current, posted on January 2, 2023 at 10:50:18
Triode_Kingdom
Audiophile

Posts: 10049
Location: Central Texas
Joined: September 24, 2006
 In that case, I would replace the LM317. The output noise spec is well under 1mV, and ripple rejection should be at least 66dB (0.0005 X input ripple). If the replacement does the same, I would suspect the meter is picking up something it shouldn't.

Regarding your answer above to Ray, it's not possible to stereotype wall wart switchers in this way. Some generate quite a bit of trash, while others are relatively clean. An AC voltmeter really isn't the best tool for separating the wheat from the chaff, either with the switchers or the LM317. You're handicapped by not being able to identify the frequency of the unwanted noise.

 

"replace the LM317", posted on January 2, 2023 at 14:15:16
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
 I'll try that first, and see if anything changes. I agree with you that using a meter isn't the best way to go. But even while suspecting that it may not be giving me accurate hard number readings, I thought that at least the ripple on the output would measure lower (or even the same as) the ripple on the regulator input.

My reply to Ray wasn't intended to be a wholesale condemnation of all SMPS. In fact, the maker of the Geshelli Labs DAC this supply is intended for said in a youtube interview that they had to work hard to find a switcher which would pass EU regulations, so I'm assuming the one supplied with my DAC performs well. He also said that he didn't think a linear supply would be any better, but implied to those who wanted to use one to go ahead.



 

Layout?, posted on January 3, 2023 at 10:03:37
gusser
Audiophile

Posts: 3649
Location: So. California
Joined: September 6, 2006
 You mention it's on a breadboard? If it's just tacked together with long lead lengths, noise pickup is a strong possibility. If so, try rebuilding on a perf board or some other tight layout technique.

 

Have you tried a, posted on January 3, 2023 at 11:16:04
Cougar
Audiophile

Posts: 4593
Location: SoCal
Joined: June 25, 2001
 12V Regulator to see if you get the same result/issue?

 

Poor choice of words on my part, posted on January 6, 2023 at 15:10:55
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
 The circuit itself is hardwired on perfboard. I wrote "breadboard" because the perfboard is just laying in a wooden box, with the filament transformer screwed onto the box. I should have written "cobbled together".

I omitted the recommended 0.1 uF caps at the input and output of the regulator, so I will add them to see if oscillation might be my issue.

 

I have not, posted on January 6, 2023 at 15:22:21
1973shovel
Audiophile

Posts: 10117
Location: Greenville SC
Joined: February 25, 2007
 I don't think I have a 7812 (or even a 7815), but I'll take a look.

Your suggestion got me thinking. I do have a 12 VDC @ 3A Acopian industrial supply which someone gave me. I'll power that up and take the same measurement. IF (big if) it meets its rated 1.5 mV RMS ripple spec, it'll be a better indicator that something's going on with the LM317 circuit.

If it doesn't, it could be the supply itself, or my DMM, but it will give me more data to consider.

Thanks

 

RE: LM317 Regulator, posted on January 16, 2023 at 08:10:47
tubular.well
Dealer

Posts: 223
Joined: March 4, 2013
 The added cap will INCREASE hum. Remove.
As shown it lack proper loading , minimum load is a few mA.

 

Page processed in 0.020 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]