hi
i sampled some LM 3886 for make a amplifier. but i only have a 50V 500VA toroid transformer, i haven't a symetric 25+25 transformer
the question is if i make a bridged amp with two LM3886, i can use single power supply with a floating GND? (virtual GND created with 2 resistors)
thanks
greetings
i sampled some LM 3886 for make a amplifier. but i only have a 50V 500VA toroid transformer, i haven't a symetric 25+25 transformer
the question is if i make a bridged amp with two LM3886, i can use single power supply with a floating GND? (virtual GND created with 2 resistors)
thanks
greetings
Hello.
You can not simply create a virtual GND with two resistors, as the currents involved make that impractical.
But you can feed the chips with an unipolar supply. (see Figure 2, "Single supply application circuit" in the LM3886 datasheet). That's usually avoided due to the necessity of removing the large DC component at the output, but in this case you can leave the big output capacitor out, as the Vcc/2 voltage at the output will be cancelled when wiring the amplifier in bridge mode.
You can also use a coldamp switching power supply as well ;-)
Hope this helps
You can not simply create a virtual GND with two resistors, as the currents involved make that impractical.
But you can feed the chips with an unipolar supply. (see Figure 2, "Single supply application circuit" in the LM3886 datasheet). That's usually avoided due to the necessity of removing the large DC component at the output, but in this case you can leave the big output capacitor out, as the Vcc/2 voltage at the output will be cancelled when wiring the amplifier in bridge mode.
You can also use a coldamp switching power supply as well ;-)
Hope this helps
hi
ssanmor: thanks for the explanation 😀
i had seen the "Single supply application circuit", but i think that these modification is because the ouput is betwen output and GND.
i seen the "aplication note 1192, picture 5.2.1 Bridged Amplifier Schematic" and i think that the GND current in these mode is really small because all power conections is only between +- power suply, GND is only for input signal and some preamp part. the current is small, because that, i think that i can use two resistors (100ohm) for create GND. or i'm wrong?
greetings
ssanmor: thanks for the explanation 😀
i had seen the "Single supply application circuit", but i think that these modification is because the ouput is betwen output and GND.
i seen the "aplication note 1192, picture 5.2.1 Bridged Amplifier Schematic" and i think that the GND current in these mode is really small because all power conections is only between +- power suply, GND is only for input signal and some preamp part. the current is small, because that, i think that i can use two resistors (100ohm) for create GND. or i'm wrong?
greetings
Hi,
If your transfomer is a toroid , you may cut the windings in the middle and make two-secondaries trannie.
Regards,
Lukas.
If your transfomer is a toroid , you may cut the windings in the middle and make two-secondaries trannie.
Regards,
Lukas.
ssanmor said:Hello.
You can not simply create a virtual GND with two resistors, as the currents involved make that impractical.
The currents from what? The ground current should be relatively low in a bridged configuration.
I think it's do-able though I've not done it myself.
virtual ground
To create a virtual ground referenced from 50VDC+ and ground at the proper amperage would take some pretty heffty resistors,quick math around 40 watts or so per resistor to float a center and then it will suffer mid to plus sag, would offer poor performance.
To create a virtual ground referenced from 50VDC+ and ground at the proper amperage would take some pretty heffty resistors,quick math around 40 watts or so per resistor to float a center and then it will suffer mid to plus sag, would offer poor performance.
Yes taro,
You can do this easily... You must use 2 resistors and each resistor must be in parallel with the lage capacitors for V+ & V-.
The resistors should draw about 10 to 20% of the total load current in order to maintain good balance between the + and - rails. It is actually the capacitors that form the virtual ground, the resistors are just there to dimnish the effects of capacitor tolerance/value. You should, of course use 2 identical capacitors from the same maker, date etc...
You will need some extra power (VA) from your transformer to waste heat across the resistors.
🙂
You can do this easily... You must use 2 resistors and each resistor must be in parallel with the lage capacitors for V+ & V-.
The resistors should draw about 10 to 20% of the total load current in order to maintain good balance between the + and - rails. It is actually the capacitors that form the virtual ground, the resistors are just there to dimnish the effects of capacitor tolerance/value. You should, of course use 2 identical capacitors from the same maker, date etc...
You will need some extra power (VA) from your transformer to waste heat across the resistors.
🙂
Re: virtual ground
What did your calculations use as the necessary current flow? 40W seems way too high for a bridged amp.
tiltedhalo said:To create a virtual ground referenced from 50VDC+ and ground at the proper amperage would take some pretty heffty resistors,quick math around 40 watts or so per resistor to float a center and then it will suffer mid to plus sag, would offer poor performance.
What did your calculations use as the necessary current flow? 40W seems way too high for a bridged amp.
Say the rails are +/-35. In a bridges config this would put 70 Volts across a 7 ohm speaker load... 10 amps.
Be generous and allow for 10% of this, 1 amp, to flow through the bleeder/balancer resistors. OK 70 Ohms... 70 Watts... 2 resistors @ 35 Watts each.
Now, the bridge configuration offers the opportunity to greatly reduce the resistors because symmetric currents are drawn from the +/- rails. Maybe a 1% of total current through the bleeder would be OK... 680~700 Ohms... 3.5 (7~10) Watts.
I think I would want to use 0.1% gain setting resistors to ensure symmetric current draw.
Total capacitance from rail to rail should be about 11,000 uF for a 2 Hz break with 7 Ohms... 2 - 22,000/35V caps should do it.
🙂
Be generous and allow for 10% of this, 1 amp, to flow through the bleeder/balancer resistors. OK 70 Ohms... 70 Watts... 2 resistors @ 35 Watts each.
Now, the bridge configuration offers the opportunity to greatly reduce the resistors because symmetric currents are drawn from the +/- rails. Maybe a 1% of total current through the bleeder would be OK... 680~700 Ohms... 3.5 (7~10) Watts.
I think I would want to use 0.1% gain setting resistors to ensure symmetric current draw.
Total capacitance from rail to rail should be about 11,000 uF for a 2 Hz break with 7 Ohms... 2 - 22,000/35V caps should do it.
🙂
- Status
- Not open for further replies.
- Home
- Amplifiers
- Chip Amps
- LM3886 Parallel Use