12V in switchmode power supplies - Page 3 - 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 1st July 2002, 04:59 PM   #21
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
You will have too much copper loss by using 6 or 7 26 AWG wires. You should think about using 7 or more 18-16 AWG. You are interested in about 350W. At about 12V in, the primary wire needs to be able to handle more than 30A of current. If you wire is too small, the transformer will have an unacceptable temperature rise of more than 60-70 degrees C. You may want to look into using Litz wire.


BeanZ
__________________
BeanZ
  Reply With Quote
Old 3rd July 2002, 06:42 AM   #22
diyAudio Member
 
Join Date: Nov 2001
Location: Harlowton, MT, USA
OK, 7 18AWG wires ought to do. What's Litz wire?
__________________
- "With power amplifiers, negative feedback is a good thing, and positive feedback can induce destructive oscillation, but with people, positive feedback is a good thing, and negative feedback can induce destructive oscillation"
  Reply With Quote
Old 3rd July 2002, 05:28 PM   #23
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
Litz wire is short for Litzendraht and is german for (woven wire). It is mainly smaller wires woven into a larger one. The winding technique assures that each wire spends equal time in the center and outside of the wire. The minimizes the skin effect. Speaker wire is typically a type litz wire, observe its woven patterns.

BeanZ
__________________
BeanZ
  Reply With Quote
Old 3rd July 2002, 05:29 PM   #24
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
By using Litz wire rather than just a few larger wires in parallel the current density of the wire can me maximized for the use of a smaller core window area.

BeanZ
__________________
BeanZ
  Reply With Quote
Old 3rd July 2002, 08:01 PM   #25
diyAudio Member
 
Join Date: Nov 2001
Location: Harlowton, MT, USA
But it's imperative that thin enamel insulation or something like that (like magnet wire) is used in the transformer, does any Litz wire have that? All speaker wire are insulated in thick insulation like vinyl or silicone.
__________________
- "With power amplifiers, negative feedback is a good thing, and positive feedback can induce destructive oscillation, but with people, positive feedback is a good thing, and negative feedback can induce destructive oscillation"
  Reply With Quote
Old 4th July 2002, 12:34 AM   #26
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
The strands in Litz wire are usually polyamide-nylon insulated (enamel). They are usually stripped with the use of a solder pot but this is something that is not usually available to the hobbiest. It is possible with practice to tin the wires (an thus strip it with a large soldering iron. Another choice on the litz wire is: served or unserved. Serving is a nylon sleeve that is around the entire wire to make it easier to work with and prevent damage to the individual wires. For magnet wire sources try:

MWS Wire- www.mwswire.com

These guys can sell by the foot or by larger bulk quantities.


Unlike speaker wire, Litz does not have the plasitc coating. Imagine working with just the copper strands inside that clear plastic sleeve and that is what working with litz wire is like. You have to be careful not to fray or break any of the strands or performance is severely degradated. This is not the greatest choice for a mass production effort but it will give the best performance if you get the manufacturing process down. It is typical for car audio manufacturers to use about ten (10) 18 AWG solid copper strands twisted together in parallel for the primary windings. The guage that is choosen depends on the frequency. The frequency determines the skin depth. The current density of the wire decays exponentially from the outside to inside and the skin depth is defined as the depth in which the current is 1/e (~37%) times that at the surface of the wire. The equation is approximately:

S=2837 / (sqrt f)

where S is the skin depth in mils (thousandths of an inch) and (sqrt f) is the square root of the frequency in hertz.

The purpose of using smaller wire is that the radius of the smaller wires are smaller than the skin depth and so more of the wire is acually used to carry the current, therefore a higher current density is achieved and the equivalent resistance is reduced and the copper losses are minimized and leads to a lower temperature rise in the transformer. Most of you probably already knew this but I hope this helped anyway.

BeanZ
__________________
BeanZ
  Reply With Quote
Old 4th July 2002, 01:45 AM   #27
diyAudio Member
 
Join Date: Nov 2001
Location: Harlowton, MT, USA
From your equasion, I'd need 15x22AWG wires to keep the radius just under the skin depth, MWS has lots of cool wire, but nothing like that. The closest they have is to get 4x18AWG Litz wires (60x36AWG each), but be dealing with 240 hair thin conductors, even though, a solder pot is something I can do easily. It looks like I'll have to get 15 22AWG wires and twist them together myself, 22AWG is a lot easier to work with than 36.
__________________
- "With power amplifiers, negative feedback is a good thing, and positive feedback can induce destructive oscillation, but with people, positive feedback is a good thing, and negative feedback can induce destructive oscillation"
  Reply With Quote
Old 4th July 2002, 11:21 PM   #28
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
Almost.... but not quite. The first step should be to choose frequency and its associated skin depth. Then choose a wire guage where the skin effect is not too excessive. Then determine how many of these wires have to be used in parallel in order to achieve the same copper area as a solid wire. This is chosen based on your maximum current needs. The amount of current you are switching determines the amount of cross sectional area of wire. All you need to do now is add up a bunch of smaller conductors in parallel to achieve enough copper. The actual "bundle" of wires will occupy more window area of a toroid (the open area in the middle which limits the amount of turns and wire and ultimately the maximum power) than if a single wire of the same area were used. However, the efficiency would be junk and the transformer would get too hot and your power output would be not that effective.....all because of the skin effect. I hope this helps some more. Have you chosen a type of ferrite material yet? I recommend P-type or F-type ferrites for you application. What kind of numbers are you working with so far for you transformer core area? I assume that you began with Faraday's Law....hope it all goes well for you!

BeanZ
__________________
BeanZ
  Reply With Quote
Old 5th July 2002, 02:37 AM   #29
diyAudio Member
 
Join Date: Nov 2001
Location: Harlowton, MT, USA
I'm mostly still thinking of wire. I have already done just what you said for my last post, figured out the skin depth, about 0.0127", then multipied that by 2 and chose a wire AWG that has that diameter or less (22AWG). I then figured out how many I would need to handle 30A, 15 would be plenty, I probably wouldn't even need that many.

For ferrite material, anything that will work well will do, I'm not picky, p-type is fine. I have not worked with core area yet, I will use Faraday's law and work on that stuff next. Is there a way I can figure out just how warm the transformer will get? It will be in a wind tunnel, by the way.
__________________
- "With power amplifiers, negative feedback is a good thing, and positive feedback can induce destructive oscillation, but with people, positive feedback is a good thing, and negative feedback can induce destructive oscillation"
  Reply With Quote
Old 5th July 2002, 05:20 PM   #30
diyAudio Member
 
Join Date: May 2002
Location: Cincinnati, OH now Columbus,OH USA
Send a message via AIM to BeanZ
You should be dealing with the radius of the wire and compare it to the skin depth. The wire does not have to be completely smaller than the skin depth just near it. 18-22AWG solid wire should work fine. There are calculations for temperature rise but they are too complicated for putting on the posting board. They contain crazy powers and stuff like that. For an optimized transformer, the core loss should be balanced with the copper loss. This assures the lowest temperature rise and the maximum power and transformer efficiency. The core loss is proportional to the flux density, frequency, area of the core, and specific core material. Different ferrite materials have differing amounts of core loss usually rated as mW/ kG. You can also find core loss curves for approximating core loss for flux density. The equations for all of this can be found readily on the web. Have you thought about the output inductors yet? The output inductors should be a high permeability material. These are your powder cores. Molypermalloy Powder (MPP), KoolMu, etc. Keep in mind also that the core area of a transformer or inductor can be increased by stacking two or more toroids on top of each other and winding the wire around the stack. This is good if you are trying to design to a minimum footprint size. Keep in mind that if this is done, the core loss is at least doubled. The numbers can be massaged to get you where you want to be. I also want to recommend a monolithic high recovery diode for your output rectifiers. I like to use dual diodes in a TO-247 package such as the RURG series from Intersil. For a dual supply, you will have two secondary windings just like a regular 60 Hz transformer. Also, two filter inductors, two banks of filter caps, etc. Are you isolating your voltage mode control loop? Or are you just going to feed it straight back? I recommend isolation using an optocoupler. This is usually how it is done.

BeanZ
__________________
BeanZ
  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
Elliott Switchmode power supply PCB red Car Audio 86 11th July 2013 05:10 PM
Switchmode power supplies - 13.5V/40W - Ideal for T-AMP Fin Group Buys 9 4th June 2008 07:43 PM
switchmode power supply Tekko The Lounge 1 9th January 2005 08:10 PM
Class AB car amp, switchmode power supply, what rails?? SkinnyBoy Solid State 4 8th December 2003 03:26 PM
Removing switchmode power supply noise ih Digital Source 1 25th April 2002 09:18 PM


New To Site? Need Help?

All times are GMT. The time now is 04:18 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