Super Regulator, collecting the facts - Page 24 - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 29th October 2003, 09:44 PM   #231
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Default Straight from the source.

PRR,

Actually I looked the ratio trend it does appear start to increase rejection below a couple of mA. Mr. Curl is right (absolutely no surprise) about the about the noise starting to rise below about 1 mA. Even at third of a milliamp a green LED still appears to be far below the noise of most of the common inexpensive voltage references that one might use for voltage above a couple diode drops. Of course if one desires a precise voltage value that is stable with temperature there is no substitute for a good voltage reference.

http://www.analog.com/UploadedFiles/...tion_V-Ref.pdf


Mr. Curl

I really don't see that any of this is really that misleading. I believe with your input on noise it would be probably desirable to shoot for a target of about a milliamp for the LED when biasing with a jfet current source. Right within the lower end of the Idss spread for the J202. Parts with Idss at the upper end of the spread can be adjusted down to 1 mA with a source resistor. Am I correct in understanding that this degeneration brings the output impedance up to the level of the
devices with Idss equal to 1 mA? I thank you for your as always, extremely valuable input. Maybe we managed to shine a little light on LED impedance and noise and also the characteristics of jfet current sources. A pretty good catch for one day of fishing in my opinion. For some data on jfet current sources:

J501 zip attached.

Thanks,

Fred
  Reply With Quote
Old 29th October 2003, 09:45 PM   #232
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Red face Almost

The missing attachment.
Attached Files
File Type: zip j501.zip (24.5 KB, 134 views)
  Reply With Quote
Old 29th October 2003, 10:32 PM   #233
diyAudio Member
 
john curl's Avatar
 
Join Date: Jul 2003
Location: berkeley ca
Fred, have you determined the 'breakpoint' where the effective capacitance of the fet starts to LOWER the effective impedance? I don't know where it is, but that must be taken into consideration. Why compromise the overall noise of the current source for a slightly improved 60Hz rejection, only for it to be compromised above about 1KHz or wherever it is?
For the record, GREEN LEDS are OK, but RED LED's would give you more 'headroom' if you needed it. This was my point. I often use green LED's myself, but then I usually have plenty of extra voltage at the input.
  Reply With Quote
Old 29th October 2003, 10:55 PM   #234
diyAudio Senior Member
 
fdegrove's Avatar
 
Join Date: Aug 2002
Location: Belgium
Hi,

Quote:
For the record, GREEN LEDS are OK, but RED LED's would give you more 'headroom' if you needed it. This was my point. I often use green LED's myself, but then I usually have plenty of extra voltage at the input.
Personally, I stay away from LEDs and use JFETs instead...

Just my preference based on sonic results, that's all.

Cheers,
__________________
Frank
  Reply With Quote
Old 29th October 2003, 10:56 PM   #235
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: USA
> the 'breakpoint' where the effective capacitance of the fet starts to LOWER the effective impedance?

From the J500 CC-FET specs Fred posted (Thanks), taking 1mA as a likely trial value, the capacitance runs about 3pF and the dynamic resitance about 1Meg. Breakpoint is 53KHz, if my slip-stick isn't sticking in the heat (check me on that). Likely an octave or two lower on a real circuit board, if layout and cleanliness are not top-notch, so it could droop at the top of the audio band.

If we thought we needed more supply rejection, and do not care about practical details like existing circuit boards: use a 2K resistor and a 12V Zener to get a semi-constant reference 12V below the input voltage, then use the snazzy CC-FET and LED, 1N914-stack, or a 2-pin voltage reference diode fed from the 12V Zener. It's a pile of parts, but in DIY work the parts are less than the true value of the labor which is hopefully less than the value of the music enjoyment (don't lose sight of the goal when in the swamps of design).
  Reply With Quote
Old 30th October 2003, 12:30 AM   #236
dimitri is offline dimitri  United States
diyAudio Member
 
dimitri's Avatar
 
Join Date: Oct 2002
Location: retired
Send a message via ICQ to dimitri
Default Thank you Fred!

here is the gem

quote-----------------
Electrolytic capacitors usually imply DC leakage errors, but the bootstrap connection of C1 causes its applied bias voltage to be only the relatively small drop across R2.
Attached Images
File Type: gif filter.gif (4.7 KB, 931 views)
  Reply With Quote
Old 30th October 2003, 02:07 AM   #237
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Default The easy way

"From the J500 CC-FET specs Fred posted (Thanks), taking 1mA as a likely trial value, the capacitance runs about 3pF and the dynamic resitance about 1Meg. Breakpoint is 53KHz, if my slip-stick isn't sticking in the heat (check me on that). Likely an octave or two lower on a real circuit board, if layout and cleanliness are not top-notch, so it could droop at the top of the audio band. "

Why not just put a 0.01uF (or maybe larger but still with good RF characteristics) cap across the LED. You can kill some of the high frequency LED noise as well and still have a circuit that comes up quickly. If a cascode seems a little costly, a 5K to 6 K resistor could be used as degeneration with the gate referenced to the filtered voltage reference and the other end of the resistor to ground. It probably is required to bootstrap the voltage at the gate until the reference comes up after the regulator output comes up and biases the LM329. It could be a pretty simple and inexpensive circuit to do this though.
Attached Images
File Type: jpg image27.jpg (35.3 KB, 982 views)
  Reply With Quote
Old 30th October 2003, 02:09 AM   #238
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Default Re: Thank you Fred!

Quote:
Originally posted by dimitri
here is the gem

quote-----------------
Electrolytic capacitors usually imply DC leakage errors, but the bootstrap connection of C1 causes its applied bias voltage to be only the relatively small drop across R2.
That's Walt's gem. I just linked it
  Reply With Quote
Old 30th October 2003, 02:20 AM   #239
jwb is offline jwb  United States
diyAudio Member
 
jwb's Avatar
 
Join Date: Mar 2002
Location: San Francisco, USA
Send a message via AIM to jwb
Truly this could soon be considered absurd. Fred's last schematic is a wonderful thing, but the thoughtful designer must consider that this design cycle could tesselate infinitely. For example, the LM329 is biased with a resistor in Fred's last, but surely noise would be reduced if it was biased with a JFET current source! And on and on and on.

Also I'm having a hard time seeing how the LM329 could be considered less costly than the cascode. Last I ordered some, they were more than the $1.60 that caused so much ruckus upthread. I'm inclined to think the cascode would perform better, but that opinion isn't beyond single-digit stuff on the back of this envelope. For that matter I'm modelling this with BJTs instead of JFETs. Call me a heathen.
  Reply With Quote
Old 30th October 2003, 02:48 AM   #240
andy_c is offline andy_c  United States
Banned
 
Join Date: Apr 2003
Quote:
Originally posted by jwb

(...)Also I'm having a hard time seeing how the LM329 could be considered less costly than the cascode. Last I ordered some, they were more than the $1.60 that caused so much ruckus upthread.(...)
If I understand Fred's schematic correctly, the LM329 is doing double duty here. It's the reference for the current source, and also the main reference for the regulator itself. IOW, it was already there in its previous role, so there's no increase in cost from just giving it more jobs to do. After completion of startup, D2 is off, so the 329 is biased from the regulated output as it was before. But it gets a shot in the arm from D2 at startup.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 07:15 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2