Unusual PWM audio

Status
Not open for further replies.
H

hari poter

New Member
Hi everyone..
I'l tried to modify 350W computer power supply (under 20 eur).
Input is 220V~, gretz and capacitors makes DC.. Push-pull mosfets driving ferrite core transformer, about 40KHz frequency with pwm.
Base of this circuit is ka7500, similar TL494 chip. Active cooling with 8cm fan.
Output capacitors are removed, output shottky diodes too.
Output impedance is very low, and driving homemade neodium magnet+alu folly speaker. Thru low-pass filter, of course.
Vref DC input of tl494 is modulated with audio. Chinch and trimer pot.
...BUT...
why output power is not strong enough? Before changing this power supply given 25Amps on +5V, and now is ~20W audio maximum! Why?
Best regard.
 
Rather a strange idea, but it sounds like the output voltage is far too low to get more power, plus the percentage of PWM change possible will limit power as well.

It's not intended for this use, so isn't likely to work well.
 
hari poter said:
why output power is not strong enough? Before changing this power supply given 25Amps on +5V, and now is ~20W audio maximum! Why?
Best regard.
Click to expand...
What's the impedance of the speaker?

Assuming 8Ω.

What's the peak-peak voltage?

Assuming 5V.

5²/8 = 3.125W

The RMS power will be 3.125/√2 = 2.2W

I hope you used a DC blocking capacitor too.
 
Last edited:
Output is NOT DC, output is secondary coil of ferrite transformer. 40KHz modulated.
No need serial capacitor.
Transformer is torus, and no problem to add few wind of wire. Output impedance is adjustable by adding few inches of wire.
Only low-pass filter is connected between torus and speaker. To stop 40KHz, i want only envelope on speaker. No DC, just audio signal.
 
Nigel Goodwin said:
Rather a strange idea, but it sounds like the output voltage is far too low to get more power,
Click to expand...
Output voltage is not matter. Output power (as DC power supply) is 5V*25A. Adjusting output impedance you can receive all of this power to driving speaker.
Nigel Goodwin said:
plus the percentage of PWM change possible will limit power as well.
Click to expand...
i'm not underatanding this.. pwm percentage (signal/pause ratio) is depending of audio input signal. you mean deviation of signal?
10x
 
hari poter said:
Output voltage is not matter. Output power (as DC power supply) is 5V*25A. Adjusting output impedance you can receive all of this power to driving speaker.
Click to expand...

Yes, but you need massive cables to feed it (to prevent high loss) and a special custom made speaker. What impedance is your speaker?.

i'm not underatanding this.. pwm percentage (signal/pause ratio) is depending of audio input signal. you mean deviation of signal?
10x
Click to expand...

Yes, mark/space ratio change in response to the audio signal - a PWM amplifier will go to almost 100% either way, a PSU chip probably only has a limited span?.
 
40khz is a very low PWM frequency to make your own class-D amplifier.
Commercial class-D amplifiers and ICs use a frequency that is 10 times to 40 times higher. Then the output filter can work properly.
 
Nigel Goodwin said:
Yes, but you need massive cables to feed it (to prevent high loss) and a special custom made speaker. What impedance is your speaker?.
Click to expand...
Nevermind, my speaker is homemade.
Adding few winds of wire to secondary may increase output voltage, and output impedance. You can adjust output voltage/impedance by changing number of secondary winds.
Nigel Goodwin said:
Yes, mark/space ratio change in response to the audio signal - a PWM amplifier will go to almost 100% either way, a PSU chip probably only has a limited span?.
Click to expand...
This is probably problem. PSU chip is no limit (over the 1.5MHz) but other components can becomes critical. 10x for idea.

@audioguru: 40KHz seems like enough for "dirty" 300Hz-4KHz middle range, just for stadions, and public meeting.
I am not expecting HI-FI from 10eur amplifier.
 
I haven't listened to a low-quality "dirty" AM radio station for about 50 years.
 
hari poter said:
Nevermind, my speaker is homemade.
Adding few winds of wire to secondary may increase output voltage, and output impedance. You can adjust output voltage/impedance by changing number of secondary winds.
Click to expand...

So presumably you have no idea what impedance it is?, so how can you expect to get any meaningful results from it?. Bear in mind, lower impedance means more losses in the cables, and particularly in the low pass filter feeding the speaker.
 
Yeah.. a lot of problems.
Speaker is thin alu- foly between two neodium magnet. Foly is from opened electrolytyc capacitors - thiny and clean. This is tweeter, of course. Feeding with kilohertz, low impedance, xtremly clean sound... but eficiency is depending of mechanical construction and, at the start of development, is not enough sensitivity.
Maybe add a impedance transformer?
And yeah, tl494 and switching transistors must working much faster than 40khz to improve sound quality.
 
Status
Not open for further replies.

Similar threads

T
Could use checking of my schematic for Teensy Audio Shield pcb
Replies
17
Views
4K
TonyAme
T
T
Need help checking over schematic for Arduino Nano Audio Sequencer PCB
Replies
6
Views
4K
liamjerr
L
G
  • Question
Impedence mismatch problem in audio project
Replies
12
Views
3K
audioguru
R
  • Question
Charging a Lifepo4 with a power supply and control current with a solid state relay controller by pwm. Good idee?
Replies
4
Views
2K
Rory1984
R
D
  • Question
Bluetooth integrated FM transmitter project
Replies
19
Views
4K
audioguru
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…