Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

A simple but low noise LT1761/LT1964 PSU design
A simple but low noise LT1761/LT1964 PSU design
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 17th October 2013, 10:22 PM   #21
jean-paul is offline jean-paul  Netherlands
diyAudio Moderator Emeritus
 
jean-paul's Avatar
 
Join Date: Sep 2002
Location: Germany
Ha I see, not only height is an item. I thought of weight but the 6 VA part is even lighter (to my surprise). Still I would like to suggest to use more pads for other standard transformers like EI38 types. You won't be limited to the Block type and it will make the PCB more versatile. We did it like that, see post #1:

http://www.diyaudio.com/forums/group...group-buy.html

For coils take a look at these:

http://www.digikey.com/product-detai...61CT-ND/354255

130 mA max. but you get the idea.
__________________
It's only audio. Member of the non modular PCB design committee

Last edited by jean-paul; 17th October 2013 at 10:27 PM.
  Reply With Quote
Old 26th October 2013, 11:16 PM   #22
maxw is offline maxw  United Kingdom
diyAudio Member
 
maxw's Avatar
 
Join Date: Jun 2004
Location: Berlin
Thanks Jean-paul, after doing a bit of reading I've added inductors to the design.

I have now:
- Added inductors for CLC Filter
- Added holes for EI38 trafo
- Changed Regs to LT1761ES5-BYP and LT1964ES5-BYP
- Added BYP caps to regs
- Replaced trim pot with fixed resistors

New schematic and unfinished layout are attached. Should output +/- 13.42V DC.

L1/2 = Panasonic INDUCTOR, 1812 CASE, 47.0UH
C1-4 = Panasonic FC series 35V, 470UF
C5-8 = Kemet CAPACITOR, 1206, 220NF, 50V, X7R
C17/18 = Reg BYP caps, TDK, 1206, C0G, 100V, 10NF
C9-14 = Panasonic FP series CAPACITOR, F CASE, 330UF, 25V
C15/16 = Wima Polyester 1812, 0.1UF, 63V
R1A/B = Panasonic 240R, 0805, 0.1%, 0.125W
R2A/B = Panasonic 2K4, 0805, 0.1%, 0.125W
Attached Images
File Type: png schematic.png (70.1 KB, 599 views)
File Type: png layout.png (433.5 KB, 577 views)
  Reply With Quote
Old 27th October 2013, 02:03 AM   #23
Mark Johnson is offline Mark Johnson  United States
diyAudio Member
 
Mark Johnson's Avatar
 
Join Date: May 2011
Location: Silicon Valley
A simple but low noise LT1761/LT1964 PSU design
I kinda think you will be (unpleasantly) surprised when you simulate your revised design in the frequency domain . . . . SPICE ".AC" analysis.

I kinda think SPICE will tell you that you have got a seriously underdamped resonant circuit. In addition to an unsnubbed "but I don't know what to do so I will do nothing" transformer.
  Reply With Quote
Old 27th October 2013, 07:39 AM   #24
jean-paul is offline jean-paul  Netherlands
diyAudio Moderator Emeritus
 
jean-paul's Avatar
 
Join Date: Sep 2002
Location: Germany
Hi I think this PSU PCB looks very good and it is well decoupled. If you will be having PCBs produced I would like to try one if possible. I have some doubt about the distance between the 230 V connector PCB pads and the mounting stud... You could use FR5 solderable fuses to avoid that. I prefer 5 x 20 fuses too but: safety first. Maybe you can move fuse holder F1 just 1 mm to the upper side and it will be fine.

* I would use smaller values for C9-C14. IMO 10 F is better*

I kinda think that people that kinda think Spice will prove this circuit to underperform better kinda do the Spice stuff before commenting.
__________________
It's only audio. Member of the non modular PCB design committee

Last edited by jean-paul; 27th October 2013 at 08:09 AM.
  Reply With Quote
Old 27th October 2013, 08:13 AM   #25
jean-paul is offline jean-paul  Netherlands
diyAudio Moderator Emeritus
 
jean-paul's Avatar
 
Join Date: Sep 2002
Location: Germany
Two caps 330 F in parallel at the output is too much. Make that a single 10 to 100 F cap. You are also using a bypass cap so why use 2 x electrolytic caps in parallel at the output ? You need capacitance at the spot where 100/120 Hz has to be buffered. So at the input.

Idea : you could use 2 x 330 F in parallel after the rectifiers, then the coil followed by again a 330 F cap ( for instance C9 and C10 that are now connected to the output of the 78/79Lxx). Use a 10 F after the 78/79Lxx/before the LT regs.
__________________
It's only audio. Member of the non modular PCB design committee

Last edited by jean-paul; 27th October 2013 at 08:25 AM.
  Reply With Quote
Old 27th October 2013, 10:51 AM   #26
maxw is offline maxw  United Kingdom
diyAudio Member
 
maxw's Avatar
 
Join Date: Jun 2004
Location: Berlin
Quote:
Originally Posted by transistormarkj View Post
I kinda think you will be (unpleasantly) surprised when you simulate your revised design in the frequency domain . . . . SPICE ".AC" analysis.

I kinda think SPICE will tell you that you have got a seriously underdamped resonant circuit. In addition to an unsnubbed "but I don't know what to do so I will do nothing" transformer.
Why? Just because I'm not using a snubber? Did you test it?


Quote:
Originally Posted by jean-paul View Post
I have some doubt about the distance between the 230 V connector PCB pads and the mounting stud...
Perhaps. I know it will fit but perhaps the distance needs to greater. I might make the PCB wider to accommodate some more space.

Quote:
Originally Posted by jean-paul View Post
* I would use smaller values for C9-C14. IMO 10 F is better*
Why? Can you have too much??

Quote:
Originally Posted by jean-paul View Post
I kinda think that people that kinda think Spice will prove this circuit to underperform better kinda do the Spice stuff before commenting.
Yes, or at least something more informative than "I kinda think you will be (unpleasantly) surprised"

Quote:
Originally Posted by jean-paul View Post
Two caps 330 F in parallel at the output is too much. Make that a single 10 to 100 F cap. You are also using a bypass cap so why use 2 x electrolytic caps in parallel at the output ?
Why not though? I know it might be overkill but the caps come in lots of 10 or 20 etc so I will have some spare.

Quote:
Originally Posted by jean-paul View Post
Idea : you could use 2 x 330 F in parallel after the rectifiers, then the coil followed by again a 330 F cap ( for instance C9 and C10 that are now connected to the output of the 78/79Lxx). Use a 10 F after the 78/79Lxx/before the LT regs.
Could do. From a technical point of view, as opposed to cost and PCB space, can you have too much uF?
  Reply With Quote
Old 27th October 2013, 12:28 PM   #27
jean-paul is offline jean-paul  Netherlands
diyAudio Moderator Emeritus
 
jean-paul's Avatar
 
Join Date: Sep 2002
Location: Germany
Hi, too much F after conventional regs will make them oscillate as the ESR will become too low for the reg to function properly. Google for the exact technical explanation. Check datasheet of 78Lxx for max. capacitance after the reg. Recent LDO's are more tolerant but why would you use much F ? They don't need to filter 100 Hz at their outputs (rectified 50 Hz AC will be 100 Hz pulsating DC). The input caps buffer the ripple of the pulsating DC voltage. At that spot you want good filtering of ripple and other stuff you need to get rid of.

Rule of thumb (can be calculated) is about 2000 F per Ampere load current for 1 V ripple. Old info that was learnt many years ago. I am not the only one using the rule of thumb. It is way easier to keep in the head than the actual calculation

http://lenardaudio.com/education/12_amps_7.html

I would use around 1000 F in total for 100 mA load but 2200 F is standard value and can be used too with the drawback that those will be higher than 3 x 330 f. Ripple will be very low and total voltage will be way above the minimum required voltage for correct working of the 78/79Lxx. Much larger value won't bring benefits. The coils will filter RF/HF and the regs don't need to do much work except for being preregs for the LT's.
__________________
It's only audio. Member of the non modular PCB design committee

Last edited by jean-paul; 27th October 2013 at 12:50 PM.
  Reply With Quote
Old 27th October 2013, 02:03 PM   #28
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Not wanting to overcomplicate, but I've gotten very interested in methods of calculating PSU and decoupling cap values, lately (for the last year or so). What type of load will this PSU be powering? Do you have any specs for the load? Are there any dynamic requirements for the output voltage and current?

i.e. If it will be an active load, what is the minimum voltage between the power rail and the output (the active device's "vclip" voltage, e.g. a few volts, for a transistor output stage), and what is the desired maximum signal voltage swing at the output of the load, and what impedance is the active load driving? Those might enable calculating the maximum ripple voltage at the onset of clipping, or load device "dropout", which would be caused by the load current drawing down the output caps' voltage. Is there a "maximum tolerable ripple" spec at the load output? Actually, with the regulator(s) there, I would probably have to just simulate the circuit as a whole. But that would be interesting, too. (Worst case would be to simulate with constant DC current that's equal to peak of output current swing, with long-period square waves.)
__________________
The electrolytic capacitors ARE the signal path: http://www.fullnet.com/~tomg/zoom3a_33kuF.jpg

Last edited by gootee; 27th October 2013 at 02:08 PM.
  Reply With Quote
Old 27th October 2013, 02:08 PM   #29
maxw is offline maxw  United Kingdom
diyAudio Member
 
maxw's Avatar
 
Join Date: Jun 2004
Location: Berlin
I welcome the complication Gootee! But remember, I have no measuring equipment beyond a simple multimeter.

The LT regs only do 100mA max. I plan to use it to power OPAMP based circuits like a simple linestage, PGA2320, CS3318 or my R2R DAC attenuator.

What values would you recommend in my design?
  Reply With Quote
Old 28th October 2013, 03:44 AM   #30
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Sorry. I was helping a friend do plumbing and electrical repairs and upgrades all day and evening and have to get up early. But I will try to tackle some simulations tomorrow evening!

I don't expect any major differences from what jean-paul suggested. But it's more-interesting than the simplest types of regulated and unregulated supplies and I will like seeing how the variables in each stage affect the other stages and the output.
__________________
The electrolytic capacitors ARE the signal path: http://www.fullnet.com/~tomg/zoom3a_33kuF.jpg
  Reply With Quote

Reply


A simple but low noise LT1761/LT1964 PSU designHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Simple, low noise power supply Zoodle Power Supplies 5 18th November 2012 09:05 AM
Simple Low Noise Voltage Regulator? lxmyers Power Supplies 16 27th November 2011 08:24 PM
Low noise PSU for dacs on TNT guglielmope Digital Source 0 14th September 2008 08:28 PM
Low noise PSU for XO Clock Sunsun22 Power Supplies 7 16th April 2006 09:18 AM
Suggestion for the Low noise PSU for the osillator.... hatasa Digital Source 17 22nd November 2002 08:47 PM


New To Site? Need Help?

All times are GMT. The time now is 06:25 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki