PSU RC multistage filtering - Page 2 - 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 28th February 2011, 11:45 AM   #11
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Quote:
Something's not right here.

Circuits B and C have the same amount of capacitance (1760uF) after the rectifier but B has additional RC filtering action due to the 14.1 Ohms of resistance before the caps. The cutoff frequency of the RC filter in B is a low 6Hz.

You would expect, therefore, the spectrum at P14 to have much lower HF content than that at P24, but they're almost identical. This cannot be correct.
Gopher thank you for commenting on it.
I was sceptical myself with the similarity of P14 Spectrum to the P24 Spectrum before I post them.
I had to retake both measurements, but the new measurement spectrums were the same as the old ones.
Then I looked at the wave files (all wave files are 10sec duration) from which these two spectrums were generated.
As you can see, they look quite the same. But the P24 waveform has sharper leading edge and smoother trailing edge compared to the P14 waveform.
I don’t know if this should translate to a different spectrum in theory.

Regards
George
Attached Images
File Type: jpg Waveform P14, P24.JPG (35.3 KB, 406 views)
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 28th February 2011, 11:58 AM   #12
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Quote:
Hi George,

in the sims I've done having the bigger cap closest to the rectifier seemed to work best, could you reverse the order of the caps in A) and post the difference (if any)?

Tony.
Tony, here they are.
Thank you for the stimulus .
By the way, what you wrote about the regulator OUT-IN voltage difference is spot on, as I found out these days that I am doing these measurements.
So much, that my measurements after the regulator (P6,P16, P26, P36) are to be taken with little credibility.
Some more measurements will follow

Regards
George
Attached Images
File Type: jpg SCHEM A, D.JPG (61.4 KB, 408 views)
File Type: jpg P31, P32, P33, P34 Spectrum.JPG (162.9 KB, 389 views)
File Type: jpg P4, P6, P34, P36 Spectrum.JPG (157.5 KB, 358 views)
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 28th February 2011, 11:58 AM   #13
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
The transformer is described as 15V, 0.15A. This is 2.25VA, which is a small transformer. Maybe this is just one secondary. Anyway, small transformers tend to have poor regulation. Let us assume 10% (20% might be more realistic, though). This means an effective secondary resistance of around 10 ohms, so the difference between B and C will not be huge - 7.6dB if other things are equal. Both should drop at 6dB/octave, but from different corner frequencies.

The plots show a much smaller difference than this, with some harmonics increasing! Maybe the OP can give us the figures, but my estimate from the plots is
100Hz 2dB
200Hz 1dB
300Hz -9dB
400Hz 3dB
500Hz 0dB
600Hz -5dB
700Hz 1dB
800Hz -10dB
900Hz 4dB
1kHz -7dB
The two plots show a very different harmonic structure, so comparison is difficult. I guess electrolytic ESR is a factor. Also, at the higher frequencies, details of grounding. Were the caps grounded exactly as shown - i.e. to a ground bus with the rectifier at one end and the regulator at the other?

PS the waveforms in post 11 suggest that the total resistance for B is about twice that for C, as the charging period takes about twice as long.

Last edited by DF96; 28th February 2011 at 12:04 PM. Reason: charging period comment
  Reply With Quote
Old 28th February 2011, 12:25 PM   #14
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
This is the last (I promise) set of measurements.

Regards
George
Attached Images
File Type: jpg SCHEM E, F, G.JPG (80.4 KB, 378 views)
File Type: jpg Spectrum P41, P42, P43, P44.JPG (160.3 KB, 199 views)
File Type: jpg Spectrum P46, P56, P66.JPG (120.3 KB, 174 views)
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 28th February 2011, 12:46 PM   #15
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi,
your last set of results make perfect sense.
The multiple RC version delivers ~-13dB of 100Hz harmonic and ~-19dB of 200Hz.
The higher harmonics are down in the noise.

Then look back at the earlier postings and you will see this all falling into place.

I am surprised that the 7818 passes so much @ 50Hz. I think this indicates a layout problem. Charging circuit sharing routes with the regulator sense circuit??
__________________
regards Andrew T.

Last edited by AndrewT; 28th February 2011 at 12:49 PM.
  Reply With Quote
Old 28th February 2011, 05:04 PM   #16
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
If I would mind gross differencies (which was my initial intention), I would make the following sum-up.
I check for better spectrum (that is, no high frequencies and low amplitude at low frequencies) at the input of the voltage regulator.
Thus I compare the three RC multistage circuits (A), (D), (E). See first attachment.
It is circuit (E). Quite obvious, as it has the largest total capacitance, thus largest RC constant. But for low power circuits, the cost and size penalties due to large capacitors are not great (*PS).

Then, I compare circuit (E) over (F) and (G), all having the same total capacitance. See second attachment.
It is circuit (E) again.

Regards
George

(*PS) If cost and size are a factor, then I would choose circuit (A) over circuit (D), as I think that (A) will behave better under dynamic conditions as well changes of Rload because the largest capacitor is closest to the changing load, with no resistor in between.
Attached Images
File Type: jpg Spectrums -x4.JPG (141.2 KB, 238 views)
File Type: jpg Spectrums E, F, G.JPG (139.9 KB, 209 views)
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 1st March 2011, 02:01 AM   #17
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Thanks George! good to see that what I am seeing in the sims translates into similar results in real tests. The result with the four 1000uF caps shows that brute force does work

That is what I was doing with my Sim, Finding what amount of capacitance I needed to use to get effectively no ripple on the output of the reg... It is an excercise in overkill and goes somewhat against the "conventional" logic of using a small amount of capacitance pre regulator and let the reg do the job of smoothing the output.

My original sim I did something dumb, and was halving my capacitance so was using double the sizes I actually needed. but when I re-ran the sim I did find that dropping the initial 10,000uF down to 4700uF did result in more ripple on the output... however there may have been a subtle drop in pre-reg voltage which gave me this result. I'll have to re-visit the sims with reduced capacitance keeping an eye on the input voltage and see if the results are any different. The main idea was to find the point where diminishing returns were making the expense of additional capacitance un-worthwhile.. Unfortunately I already have bought the caps based on my initial "dumb" sim, so will go ahead and build with 10,000 - 3r3 - 4,700 - 3r3 - 4,700... As I said an excercise in overkill!

Another observation I had when doing the sims (after realising the in-out difference needed to be 5V at least) was that it was better to have slightly lower resistance between the caps and higher ripple before the reg, if it meant that the pre-reg voltage was 5V or more higher, once the magic 5V was hit, any additional voltage (pre reg) made no difference... Of course this gets tricky as voltage drop increases as load current increases, so you need to choose the resistor values based on the max current draw expected.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 1st March 2011, 02:58 AM   #18
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 96
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Quote:
Originally Posted by DF96 View Post
The transformer is described as 15V, 0.15A. This is 2.25VA, which is a small transformer. Maybe this is just one secondary. Anyway, small transformers tend to have poor regulation. Let us assume 10% (20% might be more realistic, though). This means an effective secondary resistance of around 10 ohms, so the difference between B and C will not be huge - 7.6dB if other things are equal.
Just mention here something I recently discovered about transformers. The resistance at DC is way lower than the AC resistance. With my LCR bridge measuring losses, I see a considerably higher value at 50Hz compared with what my multimeter shows me (the DC value). I think this is down to two factors - core losses and proximity effect.
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler
  Reply With Quote
Old 1st March 2011, 03:06 AM   #19
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
Do you do that measurement with the primary shorted (assuming you're measuring the secondary)?
__________________
You might be screaming "No, no, no" and all they hear is "Who wants cake?" Let me tell you something: They all do. They all want cake.- Wilford Brimley
  Reply With Quote
Old 1st March 2011, 03:12 AM   #20
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 96
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Ah thanks for jogging my memory SY No, first off I didn't. Then I later realised my error - it certainly does reduce the reading considerably. I'll do it again and get back to you.

<edit> Shorting the primary when measuring the secondary means the flux is much reduced, so the core losses go down considerably. What I don't know is how this affects the proximity effect.
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler

Last edited by abraxalito; 1st March 2011 at 03:15 AM.
  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
Canton ERGO RC-1 (frequency room compensation) for RC-L - Schematic wanted tiefbassuebertr Multi-Way 2 20th September 2010 11:20 AM
RE: RC-19 or RC-30 chungtat Tubes / Valves 6 6th August 2005 03:51 AM
'Local Feedback and Nonlinearity of Multistage Feedback Amplifiers' mikeks Solid State 62 20th July 2005 08:03 AM
filtering switched mode PSU for SI 5066 amp supply jives11 Parts 0 24th January 2005 09:19 PM


New To Site? Need Help?

All times are GMT. The time now is 02:28 PM.


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