Choke for Pi Filter - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 January 2004, 03:18 AM   #1
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Default Choke for Pi Filter

In power supply there are some people are using Pi Filter (L-C) filter for ripple elimination. The L is usually placed on +rail and -rail, while the Capacitors are to ground.

I looked in some commercial amps, they use the common windings for this L. That is the +rail and -rail chokes are wound on the same core.

A question rises in my mind, wheter it is the A input-output configuration, or the B configuration is the right one.
The winding have the same direction for the 2 magnet wire. In A, both + and - input are in the same place, so I think the magnetic field will eliminate with each other.

But in B configuration, the input and output of + and - are crossing. So the magnetic field will add to each other.
Which is the right input-output configuration that is the right one for Pi filter after bridging diode?
Attached Images
File Type: jpg gam2.jpg (31.5 KB, 952 views)
  Reply With Quote
Old 17th January 2004, 03:24 AM   #2
diyAudio Member
 
JOE DIRT®'s Avatar
 
Join Date: Apr 2003
Location: Brantford, ON
figure "A" looks right
  Reply With Quote
Old 17th January 2004, 07:05 AM   #3
mikelm is offline mikelm  England
diyAudio Member
 
Join Date: May 2003
Location: England
Send a message via Yahoo to mikelm
you need a dual secondary transformer with the two windings in series for a dual supply. the centre point of the two windings defines the earth in between the two rails. You cannot rely on caps alone to define the earth between the rails - it will not work

also you need a cap in between the bridge and the choke.

that is: transformer, bridge, C, L, C

A classic Pi filter has a single choke in in a single rail supply usually drawn at the top hence the name.

good luck
  Reply With Quote
Old 17th January 2004, 08:09 AM   #4
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Yes, you are right. I forgot to draw the transformer, but it is certainly the CT (Center Tap) type, so ground is taken from this CT, not only from the junction of capacitors.
From the commercial unit I observed, it seems like it used B configuration, but I'm not sure, because it was hard to see it.
Which configuration is the right one?
  Reply With Quote
Old 17th January 2004, 08:35 AM   #5
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
Lumanauw,

B is the right one if you want to use the L's for smoothing out the current pulse of the rectifier and improve the power factor. This is a very desirable thing, as you can get more power from the same transformer and the amount of conducted and radiated interference is seriously decreased. Also the rectifiers are less stressed.
Because you want to store magnetic field-energy in the core you want the fields to add and not cancel. Otherwise no energy is stored and you just get an common mode rejection coil and you are not interested in a common mode disturbance in this case but in the differential signal (the output power).

Steven
  Reply With Quote
Old 17th January 2004, 10:04 AM   #6
jez is offline jez  United Kingdom
diyAudio Member
 
Join Date: Jan 2004
Location: n.e england
You need two seperate chokes for this to work properly,I think what you have seen is a common mode choke . Idealy a choke input supply (i.e. TX to rectifier then choke to smoothing caps) could be used to give almost constant current draw from the transformer but this is nomally only done on supplys for valve amps because of the size of choke that would be necessary for a solid state power amp (huge and two of them!).Musical fidelity (British fidelity in the states I believe) used this however in the A470 power amp but it took two people to lift one.
  Reply With Quote
Old 17th January 2004, 10:18 AM   #7
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Thanks Steven for the explenation.
But what is really the merit and drawback between using B configuration and separate chokes for + and - rails ?
  Reply With Quote
Old 17th January 2004, 06:03 PM   #8
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
I suppose these coils are not used as common mode choke, to suppress common mode interferences, but really as input chokes to stretch the active charging current pulse through the rectifier to charge the capacitors. This used to be done in the past mainly for tube amplifiers because tube rectifiers could not handle big current peaks and it helped in smoothing DC with the relatively small value electrolytic caps. These chokes were even used to stabilize the output voltage under changing load conditions. This was done with a so-called swinging choke. At small DC currents the inductance was high, causing some voltage drop, at higher currents the core came slowly into saturation and the inductance dropped, causing a smaller voltage drop. When Si-diodes, able to handle large current peaks, appeared, the input choke lost some of its value. Interestingly, the swinging choke re-appears in switched mode power supplies as stepped choke with high inductance at low currents an low inductance at high currents. This is done via a stepped airgap. By using this it is possible to increase the efficiency of switched mode power supplies at low currrent demands, like in the stand by mode of TVs and VCRs.

For some decades this choke approach was hardly used because of high cost and the availability of cheap high current diodes and high value electrolytics. About 5 years ago the input choke gained interest again, also for low voltage equipment, because of the European and Japanese mains harmonic requirements (IEC 61000-3-2). This standard asks for low harmonic distortion of the current wave form as it is drawn from the mains and sets limits to the values of all harmonics up to the 39th. The simple bridge rectifier with capacitor does not fulfil these limits and manufacturers have started to use input coils again to to be compliant. This is in fact a passive power factor correction circuit. Just like switched mode power supplies use active power factor correction circuits to be compliant. A few years ago the introduction of the standard was postponed, however, for most commercial equipment, because of objections from the manufacturers/market due to increased costs.

In the past most of the (tube) equipment used and still use single power supplies that need only one choke, while most modern solid state amplifiers use double power supplies. Double power supplies need two chokes but, I think, there is no reason why these chokes could not be on the same core, if you make sure that the magnetic fields are added and not cancelled. Of course the core needs to be bigger for two windings to give them both the same inductance as each would have had when wound on a separate core. In Lumanauw's circuit double phase rectification is used in both supply lines so the fields allways add up (if the proper winding direction is used) and the core should not saturate because of that. In single phase rectification it is possible to drive one coil during half a mains period and the other coil during the other half, in which case the maximum density would not increase, but with the disadvantage of more ripple.
So, I think, it makes no real difference whether you use two smaller cores for each supply or one bigger for both supplies at the same time. In commercial equipment the designer often prefers to use one component instead of two, because that is cheaper.

Steven
  Reply With Quote
Old 17th January 2004, 07:02 PM   #9
mikelm is offline mikelm  England
diyAudio Member
 
Join Date: May 2003
Location: England
Send a message via Yahoo to mikelm
thanks Steven for that explaination, I had not seen chokes in this position before and I will definetly do some research on it.

I have found that 1uF - 4.7uF cap directly across the secondary has a hugely beneficial effect on the sound quality of every PSU that I have tried it on.

Although in theory it is less elegant that a carefully designed snubber cct it seems to do 90% of the job by drastically reducing the frequency of the resonance in the coils after the diodes switch off.

I will now do some spice research into how to these two elements can be combined to best effects.

have you ever tried caps directly across secondaries ?
  Reply With Quote
Old 17th January 2004, 08:26 PM   #10
mikelm is offline mikelm  England
diyAudio Member
 
Join Date: May 2003
Location: England
Send a message via Yahoo to mikelm
I just played around with these ideas in spice.

This is what I found

One 4.7uF cap across secondary still gives ringing but much much lower frequency.

1uF in paralell with 10uF + 47 ohms ( see picture below ) reduces frequency and nearly kills the ringing. The general rule of thumb is that the single cap has to 1/10 of the value of the cap in series with the resistor for effective damping. Damping can be optimised to reduce ringing to about 1 cycle or less by adjusting resistor value.

Adding in a 100mH choke certainly can extend the pulse duration to about twice the time (8ms to about 16ms ) but be prepaired for big loss of voltage at the o/p. In the cct below the o/p voltage was almost halved. Choke values that did not reduce the voltage much did not significantly increase the pulse time.

Using the choke without the caps does nothing to address the high frequency ringing when the diodes turn off so there still may be a rough quality to equiptment using such a supply.

In the high voltage valve equiptment the choke would have to be a higher value and therefore would naturally have a higher resistance and this would have a damping affect on the ringing.

I'm a big fan of simple RLC type passive designs so next time I am doing a suitable supply I will try this choke idea.

cheers

mike
Attached Images
File Type: jpg capture.jpg (17.2 KB, 812 views)
  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
Choke Input Filter. G Tubes / Valves 8 11th February 2008 08:57 PM
Ps Filter Cap or Choke? Dr.Frankenstein Power Supplies 3 11th March 2006 07:51 AM
Filter Choke BC Parts 2 22nd September 2004 09:36 PM
SOZ Choke input filter? gnomus Pass Labs 48 14th May 2004 06:41 PM
humbucking filter choke - how different? zobsky Tubes / Valves 4 19th October 2003 09:23 PM


New To Site? Need Help?

All times are GMT. The time now is 10:16 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