Maida Regulators Redux

[drj759]

Eli,

Don't know if the following is critical to your application:

The LR8 datasheet specifies a minimum of 1uf on the output to maintain stability.

Jackinnj suggests bypassing the adjust pin here: http://www.diyaudio.com/forums/showthread.php?postid=1112318#post1112318

Dave

 

[Eli Duttman]

Eli,

Don't know if the following is critical to your application:

The LR8 datasheet specifies a minimum of 1uf on the output to maintain stability.

Jackinnj suggests bypassing the adjust pin here: http://www.diyaudio.com/forums/showthread.php?postid=1112318#post1112318

Dave

A 1.5 muF./250 WVDC metallized polyester part will be at the O/P. Stability is a non-issue.

What's a good value for the adj. pin bypass? Is 10 nF. enough?

 

[drj759]

Eli,

Jackinnj recommends 10uF per this post:

http://www.diyaudio.com/forums/showthread.php?postid=1156661#post1156661

That seems to be consistent with the LM317 datasheet on page 8 and Figure 3, page 9 here:

http://www.national.com/ds/LM/LM117.pdf

Adj pin bypass not required other than improving ripple rejection so it might be worth a try without it first.

Dave

 

[ctaudio]

Here's a very nice discussion of the LR8 and LM317:

http://groups.google.co.vi/group/sc...read/thread/30ce7fc10b66d1db/1d95888d774b5792

 

[drj759]

Eli,

I dusted off the LR8's that I haven't had time to play with and I experimented a bit with the adj pin bypass. I definitely would recommend using a 10uF across the R2 (the lower voltage set resistor) to bypass the adj pin.

I saw about 1.5 mVp-p hash on the LR8 output without the 10uF adj pin bypass. The hash dropped to less than 0.5 mVp-p with the 10uF cap.

Did the experiments at low voltage, 16V in, 5V out. A 1uF poly was used on the output of the LR8. The LR8 output was connected to a JFET configured as 3.5ma CCS load.

Dave

 

[jackinnj]

If you're going to hang another power device off it anyway, might as well go the Maida route and benefit from the 317's excellent performance.

We won't know until someone compares the LR8N3 flying solo with an LM317 a la Maida for line rejection, noise and impedance...

 

[SY]

What I meant by that was LM317 with a HV standoff transistor versus an LR8 with a current boost transistor.

Now who could do that measurement? Hmmm, whom could it be...?

 

[Eli Duttman]

Eli,

I dusted off the LR8's that I haven't had time to play with and I experimented a bit with the adj pin bypass. I definitely would recommend using a 10uF across the R2 (the lower voltage set resistor) to bypass the adj pin.

I saw about 1.5 mVp-p hash on the LR8 output without the 10uF adj pin bypass. The hash dropped to less than 0.5 mVp-p with the 10uF cap.

Did the experiments at low voltage, 16V in, 5V out. A 1uF poly was used on the output of the LR8. The LR8 output was connected to a JFET configured as 3.5ma CCS load.

Dave

I was making up parts order lists this A.M. for the owner to act on. 2X 10 muF./250 WVDC 'lytics to serve as adj. pin bypasses are on a list.

Being anal, I've called for 2 protective diodes on each regulator. No prisoners are being taken. Since this is a low current situation, saying the PSU will roar (like a lion) is inaccurate. However, this PSU design will howl, like a scorpion mouse.