output network - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

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 14th September 2008, 08:26 PM   #11
icko is offline icko  Sweden
diyAudio Member
 
Join Date: Sep 2008
Quote:
Originally posted by guitar_joe
heya,
where do you buy the TDA8920?

cheers
Changed my mind and bought them from ebay. 45$ for 10 pcs with shipping.
  Reply With Quote
Old 16th September 2008, 11:37 PM   #12
mag is offline mag  Switzerland
diyAudio Member
 
Join Date: Jan 2008
Location: Chiasso
Hi Icko,

for the output inductor use a T106-2 toroid core and wind it by yourself.
There are plenty of commercial inductors with around 20uH and around 10A current capacity. The problem is that normally they are not suited for class-D.
The high switching freq (normally 300-400kHz) of a class-D amplifier need a output inductor with low losses at that frequency. Commercial inductors normally have too high losses at 400kHz and thus the output coil will dissipate a lot of power.
I have tried to use a Bourns 2300HT inductor, rated 22uH and 19A current; it was so hot that it melted its solder points...

If you use a T-106-2 core the losses are very limited,
for 20uH you need 38turns (AL=13.5nH).
If you need 7A wind 3x0.8mm diameter wires around the toroid.

You can get these parts easily on www.toroids.info and you can order them directly on www.kitsandparts.com.

I have already ordered some stuff from there, the shipping cost is low and in around one week you get what you need.

ciao
  Reply With Quote
Old 17th September 2008, 08:50 AM   #13
icko is offline icko  Sweden
diyAudio Member
 
Join Date: Sep 2008
Thank You!

Quote:
If you need 7A wind 3x0.8mm diameter wires around the toroid.
38x3x0.8mm = alot of cu

Where did you get that equation from? I mean how diameter of the wire and number of wires relate to the current.

Another question, when using multiple wires as you describe, you twist them before (to get a rope-like wire) you wind the toroid? Or just wire them flat.
  Reply With Quote
Old 17th September 2008, 10:14 AM   #14
mag is offline mag  Switzerland
diyAudio Member
 
Join Date: Jan 2008
Location: Chiasso
You can calculate exacly the wire diameter based on the current but you have to take into account the skin effect too.

Now I don't have the equation available but you can have an idea of the current capacitance of the wire on this link.

http://www.powerstream.com/Wire_Size.htm

A 0.8mm wire (AWG20) can be used for a DC current up to 11A. Here you are running at 400kHz so due to skin effect you use only a small portion of the wire area.

You are running your amp at +/-25V on a 4ohm load, this means that the maximum current will be around 6.3A + the ripple current in the output inductance.

By experience 2-3 parallel wires of 0.8mm diameter are enough.
I have used 2*0.8mm wires on a +/-40V and 4ohm (130Wrms) amp and a 5*0.8mm on a +/-55V 4ohm BTL amp (900Wrms) without any probems.

To wind the toroids don't twist the wires toghether it is harder because the twisted wire becomes less flexible. Electrically is more or less the same thing but practically is easier to wind them flat.

On a T106 toroid you can not wind 3*0.8mm*38t only on one layer, there is no enough place. Once you covered all the toroid with wires, start another layer until you have wound all the necessary turns.

ciao
  Reply With Quote
Old 17th September 2008, 11:40 AM   #15
icko is offline icko  Sweden
diyAudio Member
 
Join Date: Sep 2008
Thank you.

One last question about the wires. Do they need coating? I've seen toroid inductors with and without coating on the wires, but haven't found what the difference is.

If coating (or isolation) is needed i plan to use regular wire-wrapping wires of AWG20.
  Reply With Quote
Old 17th September 2008, 12:11 PM   #16
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
icko, mag!

Ripple current is negligible compared to load current, and load current is quasi DC, so skin effect is negligible.

Practically 1*0.8 mm is perfectly enough. And this way distributed capacitance and stray inductance will be significantly less. Increasing of resistive loss is only 0.3 W (approx.) at full power, wich is insignificant.

I use 3*0.8 mm too... for 1 kW.
  Reply With Quote
Old 21st September 2008, 11:05 AM   #17
icko is offline icko  Sweden
diyAudio Member
 
Join Date: Sep 2008
Thank you Pafi.

Then i'll will buy the T-106-2 and wind it with 1 or 2 layers of AWG20 wire which i'll assume should be isolated.
  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
Input / Output impedance, Resistor network Vs. Dynamics -_nando-_ Everything Else 3 25th March 2007 12:19 AM
advise on 2/3 way network kdlsrinu Multi-Way 8 24th August 2005 08:53 AM
BHC T-Network caps Fiberbyte Parts 0 26th January 2005 11:48 AM
Aleph 5--CRC network billy Pass Labs 2 24th June 2004 03:57 PM


New To Site? Need Help?

All times are GMT. The time now is 05:32 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