It's by far not as bad as you imagine. And his imagination tells a lot about how things are going in life for somebody...
The protection that tripped was in the power supply and involves a current sense transformer with 200 turns and inherent antenna properties. Ironically, this transformer is working at mains frequency, it's a linear circuit involving op-amps and a PIC. It's calling for a couple of ferrite beads and caps to avoid rectification of ambient RF in the input LTP of a bipolar op-amp.
Protection against wrong oscillation modes is very useful as the product ages. Imagine the result of an open feedback resistor or a cold joint somewhere. Bang! or immediate shutdown.
btw: No, it does not make funny noises unless the phone is placed crazy close to the modulator
btw2: That's without an enclosure, bare pcbs on a wood bench
The protection that tripped was in the power supply and involves a current sense transformer with 200 turns and inherent antenna properties. Ironically, this transformer is working at mains frequency, it's a linear circuit involving op-amps and a PIC. It's calling for a couple of ferrite beads and caps to avoid rectification of ambient RF in the input LTP of a bipolar op-amp.
Protection against wrong oscillation modes is very useful as the product ages. Imagine the result of an open feedback resistor or a cold joint somewhere. Bang! or immediate shutdown.
btw: No, it does not make funny noises unless the phone is placed crazy close to the modulator
btw2: That's without an enclosure, bare pcbs on a wood bench
Last edited:
Today i built a little circuit that provides a delayed turnon to let the caps charge before the amp goes active.
Heres a short vid of the cir in action:
YouTube - Delayed turnon circuit
Heres a short vid of the cir in action:
YouTube - Delayed turnon circuit
It have turned out im still below 200kHz, had moved a knob on the scope without knowing it, so that it looked like the frequency had increased to 300kHz.
And now i noticed there were too many resistors in the input part so when i removed those i actually did climb to 250kHz, and it sounds like it improved ther sonics as well.
And now i noticed there were too many resistors in the input part so when i removed those i actually did climb to 250kHz, and it sounds like it improved ther sonics as well.
Last edited:
AP2, i used the enable/disable in the current source like in the philips ucd schematic: http://www.nxp.com/documents/user_manual/UM10155.pdf
pad 3 and 4 in my schematic, tho i guess using the sd pin on the IR2010 would yield similar results, i chose what i chose because i dident feel like digging up the ground plane around the SD pin.
And here are two new pics of the updated startup delay:
Amp in standby, no oscillation:
Amp active:
The white led was supposed to be yellow, picked the wrong bag.
pad 3 and 4 in my schematic, tho i guess using the sd pin on the IR2010 would yield similar results, i chose what i chose because i dident feel like digging up the ground plane around the SD pin.
And here are two new pics of the updated startup delay:
Amp in standby, no oscillation:
Amp active:
The white led was supposed to be yellow, picked the wrong bag.
It's similar to this layout:
And its rock solid, the carrier resudial is so clean that one could think it was the output of a function generator, not even the slightest hint of switching noise.
Are there eagle or gerber files to DL for this board?
AP2, i used the enable/disable in the current source like in the philips ucd schematic: http://www.nxp.com/documents/user_manual/UM10155.pdf
pad 3 and 4 in my schematic, tho i guess using the sd pin on the IR2010 would yield similar results, i chose what i chose because i dident feel like digging up the ground plane around the SD pin.
And here are two new pics of the updated startup delay:
Amp in standby, no oscillation:
Amp active:
The white led was supposed to be yellow, picked the wrong bag.
Hi Tekko,
I did not completely read the posts but try these;
i) put an electrolitic cap parallel to R6 for delayed turn on, also disconnecting R2 from gnd can be used as disable function.
ii)Or put an electrolytic cap from vdd to SD pin of IR2112 and let the SD pin go to -24v with a resistor. For both i&ii you need to make some calculation for cap and/or resistor values.
iii)To adjust the offset use offset balance fuction already supplied on LM311 (pins 5,6).
Regards
FH
The audio source is my computer, i mix the channels into mono, then run the mono signal through two opamps, one inverting and the other non inverting.
After that the balanced signal travels thround an old ethernet cable to the amp, around 7 meters worth, at the amp its received by the buffer that i chose to have an input impedance of 18kohm.
There might be a slight improvement in audio quality.
After that the balanced signal travels thround an old ethernet cable to the amp, around 7 meters worth, at the amp its received by the buffer that i chose to have an input impedance of 18kohm.
There might be a slight improvement in audio quality.