I found the gain on my eBay modules was way too high and it would actually oscillate. I reset mine back to the reference design and it works great now.
See this thread for details of what I did:
Fixing [and preventing] a fried Zobel network
Last edited:
Now that you mention it, at full output at 1kHz I did notice the carrier frequency was quite visible. Probably lower again than 175kHz. I'll just run and check.
-------
Okay - At just on clipping the ripple voltage touches 65V peak and you can definitely see the frequency change as the voltage rises toward the peak. Minimum frequency at the top of the waveform is 74kHz...
-------
Okay - At just on clipping the ripple voltage touches 65V peak and you can definitely see the frequency change as the voltage rises toward the peak. Minimum frequency at the top of the waveform is 74kHz...
I did see that post the other day but couldn't find it again.
Thanks!
Edit -> I just realised that seeing the other end of that resistor goes to a virtual ground, the input impedance of the amplifier would depend greatly on the value of that resistor. Increasing its value would reduce the need for a larger value input capacitor. Probably the cause of poor tone burst response as per here -> Lousy 30Hz tone burst.
Last edited:
I did see that post the other day but couldn't find it again.
Thanks!
Edit -> I just realised that seeing the other end of that resistor goes to a virtual ground, the input impedance of the amplifier would depend greatly on the value of that resistor. Increasing its value would reduce the need for a larger value input capacitor. Probably the cause of poor tone burst response as per here -> Lousy 30Hz tone burst.
Yes, correct. You should see _much_ better LF performance with higher input impedance.
This time driving a 380V 3-phase 6 pole 1000rpm motor at 860Hz and 16,800 rpm.
Link to EEVblog posting
Link to EEVblog posting
To answer my own question above, the 10 ohm resistor was cracked in half.., the three 16V zeners were broken and the buck converter too.
I repaired the amplifier by biasing +-12V auxiliary output from my power supply to -Vcc on the amplifier.
What can be done about a hot output inductor? Is the trim-pot at the input side of any use? Can the switching frequency be measured with a multimeter and adjusted?
I repaired the amplifier by biasing +-12V auxiliary output from my power supply to -Vcc on the amplifier.
What can be done about a hot output inductor? Is the trim-pot at the input side of any use? Can the switching frequency be measured with a multimeter and adjusted?
I got 2 new boards
I got 2 new boards form bangod. I had 2 already an I replaced the R13 from100Ohm to 4.1k due the Zobel network.
But the new boards have the R13 and R14 0 ohms. Is this correct ?
I got 2 new boards form bangod. I had 2 already an I replaced the R13 from100Ohm to 4.1k due the Zobel network.
But the new boards have the R13 and R14 0 ohms. Is this correct ?
An externally hosted image should be here but it was not working when we last tested it.
I've been spending half a year struggling with getting rid of power supply ripple noise in my car. You have a couple of things to try. Do the easiest thing first: lift ground from the RCA inputs, i.e. disconnect the black. Try this at both ends of the line audio cable.
Others things to try is to isolate the power supply with a isolation transformer (expensive). Or put a ground loop isolator ($5 ea.) in each channel.
Others things to try is to isolate the power supply with a isolation transformer (expensive). Or put a ground loop isolator ($5 ea.) in each channel.
I chose dual 65 V, 10 A supplies for bridged pair of amps at 8 ohms. Dual 65 V 5 A supplies would work for a single amp at 4 ohms. Slightly higher voltage would be better for higher impedance loads.
Last edited:
Here is a schematic for the board along with a couple of mods. The gain issue due to R13 has been documented above. I also found the negative current limit kicked in way too soon, positive being fine, and so suggest a change to R11, see attached pdf.
The inductor and heatsink are not up to job of continuous full power output but when testing with music via a limiter that just went to clipping, the temperature of the heatsink was low enough that I could touch it for quite a few seconds and the inductor was cooler. I don't see that as a problem. No problem getting clean 380W into 4 ohms.
I used 8AT HRC fuses in the power supply lines and 5AT in the speaker as an average maximum power of limit 100W to the speaker seemed reasonable for music to me.
The inductor and heatsink are not up to job of continuous full power output but when testing with music via a limiter that just went to clipping, the temperature of the heatsink was low enough that I could touch it for quite a few seconds and the inductor was cooler. I don't see that as a problem. No problem getting clean 380W into 4 ohms.
I used 8AT HRC fuses in the power supply lines and 5AT in the speaker as an average maximum power of limit 100W to the speaker seemed reasonable for music to me.