Aleph-X LC power supply for 4 channels - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

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 5th November 2003, 02:35 PM   #1
diyAudio Member
 
Join Date: Aug 2002
Location: New York, NY
Send a message via AIM to yldouright Send a message via Yahoo to yldouright
Default Aleph-X LC power supply for 4 channels

Gnomus, FEThead, Nelson Pass or anyone else with definitive reply capability

Do you have an opinion on using one larger transformer to feed all the amp channels instead of separate parts. Which is likely to sound better and what are the arguments pro and con each approach? I think I'm finally going to do the Aleph-X with 4 channel output, 16-20V rails and 2A though each FET. I was going to use a 3KVA toroid to feed the whole thing. Capacitance will be 140mF per channel. I've been told that using a 2.2mH coil in series with each capacitor bank will reduce the ripple by a factor of 100 but if all four channels will use 32A at maximum power which are the right kind of coils to use and how do I configure them? If I decide to wind them myself, which AWG should be able to pass 8A without much current loss? Does the wire need to be laminated or can standard household wire with a thin insulation be used? I'm also a little confused about the topology. Will four coils (one for each tap and before each capacitor bank) be enough or will I need 8 to do four amps because each amp needs a positive and negative rail? Will the 2.2mH value need to be higher if the rails are closer to 20V? What value should my taps be if I want to end up with 16-20V on my rails using an LC configuration? I was thinking 24V taps on the transformer would be perfect but from what I've read, alot depends upon the losses from the coils used. Please help.
  Reply With Quote
Old 6th November 2003, 02:59 AM   #2
grataku is offline grataku  United States
diyAudio Member
 
Join Date: Dec 2000
Location: -
It's not the voltage that kills you, it's the amps

Nobody's willing to touch this thread I guess, afraid of getting burned by the 32A?
Seriously, I wonder what is "standard household wire". I have 8 gauge magnet wire but I would hardly consider that standard. That may not even be enough for 30 amps.
The whole idea is ill concieved and I strongly advise against it.
Unless you are a professional used to dealing with high current wiring and well aware of all the issues involved you may end up shooting in your foot.
I have a mere 10 amps going around my AlephX and that scares the crap out of me already.
You are talking copper bars, insulators, ~30 watts dissipated in the rectifer alone, monster inductance monster caps, forced cooling for the whole thing etc.
Not to mention that for stereo separation it's much better to have one dedicated PS per amp.
  Reply With Quote
Old 6th November 2003, 04:08 AM   #3
diyAudio Member
 
Join Date: Feb 2001
Location: Columbia, SC
Huh! You'll be sinbin-ed you for talking carelessly about so much voltage and current.
Me? I'm fearless. Electrons and me are friends.
For my part, I think you'll find that 140uF is woefully inadequate, no matter what configuration you use for the power supply. 140,000 would be more like it.
2.2mH inductors are actually rather small as such things go. (I tend to think in terms of tube stuff where people aren't shy about using entire Henries.) Unquestionably, they will help but the factor-of-100 aspect is frequency dependent. What you need is a great deal of attenuation at 60Hz and multiples thereof, in particular 120Hz, which is what will come off of the rectifier. Using an inductor input filter on the power supply will give you a lower rail voltage than a cap-input filter.
Think of the power supply as though it were a lowpass crossover. A cap from rail to ground, by itself, will give you a 6 dB/oct rolloff. The exact frequency where you hit the nominal -3dB point will depend on the value of the cap and the impedance of the amplifier on the other side. Ditto for a single inductor in series. The caps/inductors in a typical amp have to be large due to the fact that the impedance of the amplifier circuit is so low. The equivalent of something like a 4 ohm load in your case, plus or minus a bit. Remember, you're not designing a crossover to 'cross over' at 120Hz on a 4 ohm load--you want it to already be way down the slope by then. Suppose you were to design for -3dB at 60Hz. You'd only be down about 5dB or so at 120Hz. That's not going to work well at all. If you set -24dB as your goal, you're going to have to design for -3 dB at around 7Hz. Just for the record, -24 dB at 120Hz ain't gonna cut it, either. So you're going to need some really big capacitors.
***Brief digression***
If you want to drive yourself crazy late at night, consider the idea that most amps' impedance changes as the music changes...so the ability of the power supply to filter out ripple will change dynamically along with the music. The original Zen amplifier is an example of a amp that doesn't change, but those are few and far between.
***
The next thing to do is to add an inductor to a cap or a cap to an inductor. Now you've got a 12dB/oct filter. You can keep adding alternating caps and inductors and get 18, 24, 30dB/oct...as sharp a cutoff as you want.
You can use a resistor in place of an inductor (but not in place of a cap), if you're wlling to pay the price. It will use up power. At the kind of currents you're talking about, you're going to need small resistances (though larger would be better for filtering action) and high wattage resistors if you should choose to go that route. Inductors are definitely superior.
All things being equal, air core inductors are better than iron core. Once you get to a certain point, the iron (whether a ceramic composition or actual layers of metal) will fill up with magnetic flux. At which point, the inductor will cease to function properly and you'll get more ripple. The core saturates at higher currents. Again, given the currents you're talking about, an air core is going to be the ticket. Air doesn't saturate. The problem with air core inductors is that the confounded things get physically big and heavy, fast. Depending on your frame of mind, that might count as a plus.
All things considered, I'd suggest at least two transformers; one per pair of channels. Better still would be one per channel. If you still want to go with one big one, I'd suggest hooking up one channel at a time and watching for hot spots. Keep an eye on currents and voltages, too, to make sure that nothing gets too high or too low.
I seem to recall having run into this question in a different thread. Either I'm cross-eyed from lack of sleep and getting the threads confused, or the same thing's posted twice. Or something like that.
What I meant, not what I said...

Grey
  Reply With Quote
Old 6th November 2003, 04:29 AM   #4
The one and only
 
Nelson Pass's Avatar
 
Join Date: Mar 2001
I understood that to be 140 milliFarad.

In any case, at 32 amps bias, you are dancing with the
devil

I suggest you start breaking it down into supply sections
devoted to each channel.

I have sampled a number of coils on the market, and the best can
get up to 6 amps or so. It's not just the wire resistance, but
also where the core material will saturate.
  Reply With Quote
Old 6th November 2003, 07:53 AM   #5
UrSv is offline UrSv  Sweden
diyAudio Member
 
UrSv's Avatar
 
Join Date: Jan 2002
Location: Sweden
I would also split it into several supplies. With 32 A continous you would have some pretty scary ripple, currents and spikes going through bridges and into caps. Secondly the quad mono could possibly yield better sound quality for other reasons.
__________________
UrSv
Those who say it can't be done should not stop those who are doing it.
  Reply With Quote
Old 6th November 2003, 11:42 AM   #6
diyAudio Member
 
Join Date: Aug 2002
Location: New York, NY
Send a message via AIM to yldouright Send a message via Yahoo to yldouright
First of all, thank you all for participating in this thread. Allow me to address each of your replies.

UrSv

I have a general dislike for using separate transformers primarily because they're difficult to match. I would be interested to know the qualifications of the quad mono power supply that would make it sound better than the LC supply I have outlined above. Most of the sonic benefit of dual mono should come from the separate capacitor banks as I understand it.

Nelson Pass

If I were using IRFP240 devices I would never use this much bias but seeing that the Aleph-X circuit boards allow me only four devices, I thought I could push 40W thorough each '150 fairly safely. My heatsinks are rated conservatively at 0.21C/W and so should be able to handle a worst case 160W dissipation. Your input on what is generally available on the market is disheartening. Unless the discussions here convince me otherwise, I will try and wind my own inductors with 10AWG household wire. The question that needs to be answered is how much current will the coils actually see? The 8A demand will only occur at max power. There are only 4 capacitor banks and each has a positive and negative rail so would it be possible to use 8 coils that only burden 4 amps each? Any suggestions on the topology to use for this scheme would be appreciated.

GRollins

I would have used 140F if that were what I meant but each bank has 4x35000F for 140mF and a total capacitance of 0.56F for all four channels. Your points about Z changing with the signal are important but I feel that those issues would be be better handles in the preamp. I opted for the LC topology because I like its simplicity and it will fit the locality and positioning of the parts better than any other C-X-?-X-C variation. You are absolutely right in your perception about seeing something like this elsewhere. I have been posting my problems on several threads because I'm eager to start and finish my "Tri-AX" (name is heretofore copyrighted along with 3-AX, THRAX, thrAX, 3-clone, TRIclone and tryClone). The "Tri-AX", "3-AX" or "thrAX" denotes the designed physical layout of the amp described above in the three enclosures. As is customary here, readers may duplicate the layout and design as it is disclosed but cannot use the aforementioned names and descriptions for commercial purposes without my express and written consent. Failure to do so will bring about legal actions against you for forfeiture of gains made with the names and descriptions above. I'm serious about this one guys. Intellectual property is hard to come by so don't test me.

grataku

Household wire would be any wire that can typically found in a house such as 10AWG THHN. This is what I would plan on using to wind my own coils if I had to. I don't think I stated that only one bridge rectifier will be used and it's implied that at least four bridges will be used. The three box scheme allows the heat problems associated with the rectifiers and the sonic detriments of the stray inductances from the coils to be handled rather elegantly. Simple but effective. Most good ideas usually are
  Reply With Quote
Old 6th November 2003, 12:07 PM   #7
grataku is offline grataku  United States
diyAudio Member
 
Join Date: Dec 2000
Location: -
Why are trafos difficult to match? Just buy 4 identical trafo from plitron or whatever.
The transformer is an integral part of the PS the secondary impedance affects the output impedance VERY MUCH.
You were looking for definitive anwers I think you got a few concurring opinions, even from the master himself that usually is very open to extreme electronic experimentation: it's a silly assed idea.

Anyway best of luck with your construction, show us some pictures when you are done.
  Reply With Quote
Old 6th November 2003, 12:29 PM   #8
diyAudio Member
 
BobEllis's Avatar
 
Join Date: Sep 2003
Location: Upstate NY
Quote:
Originally posted by yldouright
I'm eager to start and finish my "Tri-AX" (name is heretofore copyrighted along with 3-AX, THRAX, thrAX, 3-clone, TRIclone and tryClone). The "Tri-AX", "3-AX" or "thrAX" denotes the designed physical layout of the amp described above in the three enclosures. As is customary here, readers may duplicate the layout and design as it is disclosed but cannot use the aforementioned names and descriptions for commercial purposes without my express and written consent. Failure to do so will bring about legal actions against you for forfeiture of gains made with the names and descriptions above. I'm serious about this one guys. Intellectual property is hard to come by so don't test me.

I hope this part was meant as humor. Smilies would have helped.

If not, how about calling "your" creation AmpthrAX - since it might kill you to have our grand master take YOUR profits from trading on his patents, after you have spent a small fortune recovering them. The AX part of your names probably infringes on Mr. Pass' trademark, registered or not.

Commercialize what Mr. Pass has so generously shared with us for personal use? Don't go there.
  Reply With Quote
Old 6th November 2003, 01:03 PM   #9
diyAudio Member
 
Join Date: Aug 2002
Location: New York, NY
Send a message via AIM to yldouright Send a message via Yahoo to yldouright
grataku

Plitron makes nice transformers but that doesn't mean that they're perfect. There will be variations in matching 4 transformers that don't come into play working from one big transformer. Whether or not this is sonically detrimental is another matter and one of the reasons I posted this. What exactly do you mean by secondary impedence?

BobEllis

First of all, I am not producing a commercial product so curb your trepidations. Only the names and the topology were copywrighted. If anything, my copywrights help Mr. Pass by preventing others from using his supersymmetry patent, putting it in three boxes and making a competing and possibly superior product with a catchier name. I had thought of "ampthrAX" as quite a funny variation but it didn't sufficiently describe the three enclosure design and I suspect that its already in the public domain. I am not out to hurt Mr. Pass and if he wishes to discuss this with me privately, my email is public. I am quite serious in my purpose and intentions when it comes to the names and descriptions submitted above.
  Reply With Quote
Old 6th November 2003, 01:59 PM   #10
diyAudio Member
 
BobEllis's Avatar
 
Join Date: Sep 2003
Location: Upstate NY
Glad to hear you have no commercial intent. However your post shows a misunderstanding of intellectual property law. Here is a brief outline not intended as legal advice.

Generally, copyrights protect expressions of ideas. That's why you can make so many variations of the same movie theme, but you cannot use the same characters as someone else's movie. Claiming a copyright on a single word is pretty well meaningless. You can copyright the design of the word on a logo, though, since that is an expression of an idea.

What would be meaningful as a single word is a trademark. Trademarks are used to permit consumers to readily identify the source of a product. They apply within a field, or closely related fields. If I had a medical lab supply company that I wanted to call Pass Labs, I probably could, since there is little chance that customers would confuse my products with Pass Labs audio gear. A Pass Labs that made electronic test equipment would probably be a different result.

Mr. Pass can claim trademark rights in the expressions "Aleph" and "XA," among others, as they apply to audio equipment, whether they are registered marks or not (registration gives additional benefits). If he permits others to use them in identifying audio gear. these rights are diluted and his property may become worthless.

It is up to the trademark owner to protect his intellectual property. If you were to use 3-AX, etc. on commercial audio gear and NP didn't sue you to make you stop, he'd be reducing the value of his company. Would it be worth the legal fees to sue? That is for him to decide.

That's the catch with trademarks - You actually have to use them in commerce to have rights in the mark. Yes, you can file an intent to use application, but if you don't actually use it, you do not aquire rights in the name just because you thought it up and left it on the shelf.

Sorry, your copyright claim doesn't really help protect Pass Labs.

End of IP primer. See a lawyer to determine your intellectual property rights.
  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
aleph 1.7 power supply jarthel Pass Labs 8 29th June 2007 04:20 PM
5 channels class-T amp Power Supply niiico Class D 2 15th November 2006 01:20 PM
Aleph X power supply Mach_Y Pass Labs 42 29th March 2004 08:43 PM
Aleph 5 Power Supply martinschki Pass Labs 66 25th January 2003 06:34 AM
Power Supply for Aleph 4 Gabster Pass Labs 27 17th March 2002 12:40 AM


New To Site? Need Help?

All times are GMT. The time now is 08:44 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