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 3rd August 2012, 01:26 AM   #351
fas42 is online now fas42  Australia
diyAudio Member
 
fas42's Avatar
 
Join Date: Jun 2012
Location: NSW, Australia
Blog Entries: 9
Quote:
Originally Posted by mightydub View Post
I'm late to the party and haven't made it through the entire thread, however here is my contribution from almost 3 years ago to simulation on this topic:

AC to DC Power Supply - diyAudio
Thanks very much for pointing that out, mightydub, I've just started playing with what you put up. As a simple start, just adding ESR to the smoothing cap's ...

So, now be afraid, be very afraid ...!!!

What I did was use a jazz track WAV file to drive the current in the load resistor, in other words, a real world situation!

This is rough, very rough, but gives a starting point to show what happens in real components ...

LTspice IV - [basicPowerSupply03-1.gif

Woah!! Scary stuff, what does this show? The top pane is the +V voltage rail!: green is when ESR of the 35,000uF smoothing caps is 30mR, a quite excellent figure for a typical electrolytic, blue is 3mR, red is 0.3mR. Bottom pane is the driving audio track, the voltage across the load resistor; a high volume but one that the amp should deal with ...

To "prove" this is not a fake, then go to the end of the track, audio drive drops to zero -- we then get the classic, textbook ripple across the smoothing cap, voltage building up again to unloaded level ...

LTspice IV - [basicPowerSupply03-2.gif

Now, zoom in to a small part, and really see how the load current is badly upsetting the voltage rails:

LTspice IV - [basicPowerSupply03-3.gif

This is the real world, and we haven't even looked at any of the other parastics in the game, yet ...

Frank
  Reply With Quote
Old 3rd August 2012, 01:31 AM   #352
diyAudio Member
 
Join Date: Oct 2011
Nice work. The voltage doesnt even drop by 2 volts
  Reply With Quote
Old 3rd August 2012, 02:16 AM   #353
fas42 is online now fas42  Australia
diyAudio Member
 
fas42's Avatar
 
Join Date: Jun 2012
Location: NSW, Australia
Blog Entries: 9
Quote:
Originally Posted by OnAudio View Post
Nice work. The voltage doesnt even drop by 2 volts
Whoops ... I did say this was rough ... !!

I forgot to bypass the soft start, the initial condition's voltage was set to 69V to get a reasonably straight waveform; to get a better idea of what the unloaded rail would look like:

LTspice IV - [basicPowerSupply03-4.gif

Frank
  Reply With Quote
Old 3rd August 2012, 01:12 PM   #354
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
Frank, do you concur that the power rails are modulated by the output current and that this can be observed in the time domein. It will be heard as well, but it may not be distiguisable from the signal in the frequency domein.

Frank, it would be interesting swapping out some capacitors to establish a cap size per amp of current required that would provide the most sensible choice.

Secondly, a single large cap could have higher ESL than several smaller caps in parallel and would prove what Tom and the guys were promoting by splitting caps into smaller equivalent value and startegically placing them on the board.
__________________
Kindest regards
Nico

Last edited by Nico Ras; 3rd August 2012 at 01:14 PM.
  Reply With Quote
Old 3rd August 2012, 01:28 PM   #355
fas42 is online now fas42  Australia
diyAudio Member
 
fas42's Avatar
 
Join Date: Jun 2012
Location: NSW, Australia
Blog Entries: 9
Quote:
Originally Posted by Nico Ras View Post
Frank, do you concur that the power rails are modulated by the output current and that this can be observed in the time domein. It will be heard as well, but it may not be distiguisable from the signal in the frequency domein.

Frank, it would be interesting swapping out some capacitors to establish a cap size per amp of current required that would provide the most sensible choice.

Secondly, a single large cap could have higher ESL than several smaller caps in parallel and would prove what Tom and the guys were promoting by splitting caps into smaller equivalent value and startegically placing them on the board.
Hi, Nico.

Indeed the rails are modulated, and it is strongly dependent in the first instance by the ESR, the lower the effective value the better. The impact of varying ESL is not anywhere as great, I'll put up examples tomorrow.

This is all very definitely pointing to the value of using multiple small caps in parallel, something I did years by a far more seat of the pants approach, and for a very definite subjective improvement.

What I've just been doing is swapping out the pure class B current sources, and inserting the output stage of Bob Cordell's THD20 amp. Might as well use the real thing, and see what happens!! And it's looking very interesting ...

Getting very late, so I'll post this variation tomorrow, when my head's a bit straighter ...

Frank
  Reply With Quote
Old 4th August 2012, 10:05 AM   #356
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
I am "thinking" that there are less subjective issues between amplifiers and these audible issues may be measurebale as well as explained.
__________________
Kindest regards
Nico
  Reply With Quote
Old 4th August 2012, 11:36 AM   #357
fas42 is online now fas42  Australia
diyAudio Member
 
fas42's Avatar
 
Join Date: Jun 2012
Location: NSW, Australia
Blog Entries: 9
Okaaay, got something a bit closer to a real class AB amp working, driving the power supply, and this is combining mightydub's power supply sim, and the back end of Bob Cordell's posted LTspice model, from another thread here:

basicPowerSupply04-BC.gif

Note that this sub-section of the amp, which is mirror reversed compared to how normally shown, to match the original class B source, is merely a current amplifier, no voltage gain here. The VAS, so to speak, is device E1, an ideal voltage gain stage, shown here with a gain of 75. Again, the source input is a jazz track WAV. And there's no feedback, so there would be a fair bit of distortion if you looked closely, but that's not the point at the moment, we're looking at what's happening to the power supply.

First off, to show this is true class AB, we'll drop the gain to 7.5, where the output devices are sharing the current drive to the load much of the time:

basicPowerSupply04-BC2.gif

This is not taxing the supplies severely, but there is still quite a bit of noise on the voltage rail; power supply as before, with ESR of the smoothing caps at 30mR, a quite realistic figure:

basicPowerSupply04-BC3.gif

Now, we'll drive the output stage hard, push it heavily into class B territory, by upping the gain to 75:

basicPowerSupply04-BC4.gif

Note the output transistors are really switching off hard now, but that's not the real story, here's what's happening to the voltage rail because of the high current drain:

basicPowerSupply04-BC5.gif

Next will be to start playing around with the parasitics, and the smoothing circuit area, to see what can be done to help matters ...

Frank
  Reply With Quote
Old 4th August 2012, 02:18 PM   #358
diyAudio Member
 
Join Date: Mar 2008
here's an output impedance plot for the LVDC bus on a +24V 100W smps. Both layouts are on a 4-layer 2Oz 1mm thick FR4 PCB. the Rev 1 (Red & Blue) layout had a small bank of X7R caps at the rectifiers, but the layout was not as good. The rev 3 layout (black) was re-arranged such that the load current was shared more evenly, and the X7R caps were moved to the load end.

Both measurements were taken at the load, with the rectifiers removed, scale in dB-ohms. although the sharp impedance trough in rev 1 might look like a good thing, in practice it means there are some complex resonances involved which are poorly damped - never a good thing. rev 3 reduces the impedance swing by 10dBOhms and its fairly well damped.

thats only 9*560uF 35V RubyCon ZLH caps = 5mF, and |Z| = 10mOhm @ 5kHz. Layout is important, but its clear that 50mF should drop this down to below 500Hz. LF performance is the parallel combination of the DC bus impedance and the transformer/rectifier output impedance, but I havent got any convenient plots to share.

this DC bus was designed to provide a specific lifetime at a given ripple current, with a nasty height constraint, for the output of a PFC flyback converter. Measured with an HP3577A + HP35676 50R signal divider. full one-port calibration at the DUT (I have a set of DIY "calibration standards" with vertical SMA sockets, and I attach an SMA socket to the DUT so the interconnect is pretty much excluded)
Attached Images
File Type: jpg +24V_raw DC Bus impedance #2.jpg (278.3 KB, 192 views)

Last edited by Terry Given; 4th August 2012 at 02:26 PM.
  Reply With Quote
Old 4th August 2012, 05:43 PM   #359
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Blog Entries: 1
Quote:
Originally Posted by fas42 View Post
Okaaay, got something a bit closer to a real class AB amp working, driving the power supply, and this is combining mightydub's power supply sim, and the back end of Bob Cordell's posted LTspice model, from another thread here:

<snipped>

Note that this sub-section of the amp, which is mirror reversed compared to how normally shown, to match the original class B source, is merely a current amplifier, no voltage gain here. The VAS, so to speak, is device E1, an ideal voltage gain stage, shown here with a gain of 75. Again, the source input is a jazz track WAV. And there's no feedback, so there would be a fair bit of distortion if you looked closely, but that's not the point at the moment, we're looking at what's happening to the power supply.

First off, to show this is true class AB, we'll drop the gain to 7.5, where the output devices are sharing the current drive to the load much of the time:

<snipped>

This is not taxing the supplies severely, but there is still quite a bit of noise on the voltage rail; power supply as before, with ESR of the smoothing caps at 30mR, a quite realistic figure:

<snipped>

Now, we'll drive the output stage hard, push it heavily into class B territory, by upping the gain to 75:

<snipped>

Note the output transistors are really switching off hard now, but that's not the real story, here's what's happening to the voltage rail because of the high current drain:

<snipped>

Next will be to start playing around with the parasitics, and the smoothing circuit area, to see what can be done to help matters ...

Frank
How about posting the .asc file(s)?

What diodes are used in the bridge? Maybe right-click on them and put some "real" power diodes in?

Need to model at least the main power-rail conductors, between the smoothing caps and the load, as series inductance and resistance (25 nH and 1 mOhm per inch).

Lower priority: Maybe try two of my transformer model, or similar, since the one being used doesn't have leakage inductance. (Might need a snubber, then, however.)
  Reply With Quote
Old 5th August 2012, 12:07 AM   #360
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
If yo ustart using large smoothing capacitors you then need to think about power on surge blowing fuses. People then add anti surge circuits which is fine but I have lost count of the number of threads where teh anti surge circuit had gone wrong.
I prefer to use lesser smoothing and get rid of the anti surge circuit. A good design will reject ripple on the power supply.

For 100watts I would use 2 off 4700uf capacitors.
__________________
http://www.murtonpikesystems.co.uk PCBCAD50 pcb design software.
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Valve power supply - How to size transformer? SanderW Power Supplies 25 4th January 2013 04:12 PM
How do you calculate choke size in a power supply? Original Burnedfingers Tubes / Valves 25 5th January 2012 12:23 AM
power supply bypass cap size BigE Power Supplies 11 5th July 2011 02:59 PM
Power Supply Case Size diymixer Power Supplies 1 10th October 2010 05:47 AM
What size power supply should I get for repair work? spooney Car Audio 3 6th December 2007 11:50 PM


New To Site? Need Help?

All times are GMT. The time now is 12:57 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