Power Supply Design - Off My Rocker? - Page 4 - 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 31st December 2004, 02:45 PM   #31
MikeB is offline MikeB  Germany
diyAudio Member
 
MikeB's Avatar
 
Join Date: Apr 2004
Location: GŁtersloh
Hi eva !

Phew, that's a lot of knowhow in one single post...

What i don't understand, i remember that a transformer outputs
the reverseintegral (just don't know the english word) of the input,
means sin -> cos. But this would be infinite for squarewave ?
A triangle should output a square. Am i wrong ?

Mike
  Reply With Quote
Old 31st December 2004, 03:23 PM   #32
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
if you are going to use Bipolar transistors you have to seriously consider using a "current mode" controller chip instead of an LM555 multivibrator -- unless the transformer is perfect etc., etc. the transformer could go into saturation and pop the transistors (semiconductors as fuses!). A current mode controller costs only a few cents more than a multivibrator, and controls the device on a pulse by pulse basis.

But firstly, there so many application notes on this type of design from Texas Instruments, National Semiconductor, On-Semi that before you put soldering iron to PCB trace it would serve you well to peruse the manufacture data-sheets and apnonte.

Lastly, off line switchers -- there is a good reason that there is a steel case around them and it isn't just EMI considerations. In addition to what EVA said, I would add that you put the device in a cage (used ATX supply) in case some capacitor gets unhappy and decides to blow its lid!
  Reply With Quote
Old 31st December 2004, 03:28 PM   #33
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
A ideal transformer outputs exactly the inputsignal scaled by the turns ratio, the current is also inversely scaled

If whe have an ideal transformer with two windings, one of 20 turns and one of 10 turns then :

The instantaneous voltage across the 20 turn winding is allways twice than in the 10 turn one

The instantaneous current through the 20 turn winding is allways half than in the 10 turn one

Real transformes also suffer from :

- Leakage inductance : Adding an inductor in series with each winding of the ideal transformer models this quite well

- Magnetizing inductance : It's modelled quite well adding a inductor with a certain saturation current in paralell to one of the windings of an ideal transformer. The voltage applied to this inductance must be flipped periodically to avoid saturation. In gapped transformers this inductance has a high saturation current while in ungapped transformers it saturates when a few mA are reached, altough this takes some time since its value is usually quite high


Infinite current with ideal square waves only happens in ideal capacitors

Voltage squarewaves produce triangle current waves in inductors
  Reply With Quote
Old 31st December 2004, 04:00 PM   #34
MikeB is offline MikeB  Germany
diyAudio Member
 
MikeB's Avatar
 
Join Date: Apr 2004
Location: GŁtersloh
Quote:
Originally posted by Eva

Voltage squarewaves produce triangle current waves in inductors
Ah, that was my missing point !
This means, after rectifying the output, it's "pure" DC ?
(I should send some bomb to my physic-teacher, i should have known
that ALL information from him is somehow wrong...)

Mike
  Reply With Quote
Old 31st December 2004, 04:09 PM   #35
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
No, I'ts a square wave filtered by LC filters made with non-ideal components

After a good LCLC filter the square wave may appear as a 1mv p-p triangle wave on top of the DC component...

Anyway, filtering ripple is very easy but preventing diodes and transistors from ringing at 15, 25 or 40Mhz is not so easy
  Reply With Quote
Old 1st January 2005, 12:08 AM   #36
diyAudio Member
 
Join Date: Aug 2004
Location: IL
Man, I'm think I got enough stuff here to write a book!

So, I am able to do this, however it will take some low-level experimenting first with some design modifications.

Well, I'm definitely not going to use this as my power supply in my first design, but definitely later on I'll give it a shot (in a low voltage/low current design).

So, it is possible to use this design though say with 30kHz switching? The reason I chose 200Hz was I didn't know if I could get good power output switching really fast.

Thanks to all!

Reece
  Reply With Quote
Old 1st January 2005, 01:07 AM   #37
Stocker is offline Stocker  United States
diyAudio Member
 
Join Date: Dec 2003
Location: Austin
The computer next to you gets its its power by switching really fast, why can't your amp?
__________________
Jesus loves you.
  Reply With Quote
Old 1st January 2005, 01:17 AM   #38
diyAudio Member
 
Join Date: Aug 2004
Location: IL
Because the people who designed the power supply for the computer sitting next to me knew exactly what they were doing. ;-)

It's definitely something I plan to do, and I'll probably even make my current amp so that I can pull out the current power supply and put a new one in. But for the time being I'll stick with the typical.

Reece
  Reply With Quote
Old 1st January 2005, 08:41 AM   #39
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: USA
> What i don't understand, i remember that a transformer outputs the reverseintegral (just don't know the english word)...

Differential, differentiation.

> of the input, means sin -> cos. But this would be infinite for squarewave?

That happens when you study the current in the transformer. That is commonly done in engineering courses: don't shoot the professor. In many real-world cases though, we apply a constant-voltage AC source and ignore the current.

Transformers are not that complicated. They also give no ripple reduction. As Eva and I have said, Reece can gain some economic advantage by doing ripple filtering at higher voltage, if he can use power at that voltage or transform it cheaper than simply adding more low-volt caps.

I do doubt his original plan will hold smoke, but getting the smoke to stay inside is a "simple" engineering chore (probably involving more/bigger parts).

But anything along these lines is not going to reduce ripple except the slight economic advantage from doing filtering at high voltage.

A full SMPS can adjust on-periods to regulate the output, but a 200Hz switching frequency isn't going to straighten-out 120Hz ripple. 35KHz will, but commodity transformers will just choke, and commodity transistors may not be fast enough, or need very stern drivers. It is a design specialty. And while some excellent switch-mode audio exists, keeping a 35KHz bee-buzz boxed up and out of your audio paths is an added complication.
  Reply With Quote
Old 1st January 2005, 09:31 AM   #40
diyAudio Member
 
Join Date: Oct 2004
For a DIYer, swtiching supplies aren't really practical. It would be nice if a company came out with supplies just for specific popular DIY amps like gainclones and Leach, etc.

I really don't think it's that important to regulate the power to the output stage. In the multi-voltage amplifier designs that I have seen they switch rail voltage midcycle. They keep the voltage gain and driver stages running at a fixed voltage, but the voltage to the output stage changes. There are just several comparators, and when the voltage crosses a threshold, the output stage is switched from one rail to the next. No alteration is made to the signal to the output stage is made.

Now these amps aren't the end all of amplifier design, but they will hold there own.

Now the moral of all this is, if an amp can switch from a 25 to 50 75 to 100 volt rails, back down again, and repeat for the opposite polarity, and still sound good, then surely an amp can handle a little ripple to the output stages.
  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
Power Supply Design Defbrett Tubes / Valves 28 10th October 2009 12:54 AM
I would like to design a power supply Kramerguy Power Supplies 9 14th January 2009 06:50 PM
48 VDC/10 A power supply looking for an Amp design Analog Addict Solid State 16 16th January 2008 11:39 AM
need help with a power supply design lochness Power Supplies 1 31st October 2007 06:12 PM
Power supply design LeakSt20 Parts 0 16th August 2005 04:06 PM


New To Site? Need Help?

All times are GMT. The time now is 11:12 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