Questions about multiple lm3886 chip amp
Hey everyone I'm currently designing 3 way active speakers compromising of lm3886 power amps and lm358 (or similar) preamps.
Im wondering whether its worth going to the effort of using a separate +/- toroidal with rectifier/regulator and possibly delay circuit or if using zenner diodes and caps off of the +/- 35V is suitable?
Thanks.
Hey everyone I'm currently designing 3 way active speakers compromising of lm3886 power amps and lm358 (or similar) preamps.
Im wondering whether its worth going to the effort of using a separate +/- toroidal with rectifier/regulator and possibly delay circuit or if using zenner diodes and caps off of the +/- 35V is suitable?
Thanks.
Last edited:
A simple resistor/zener/cap shunt supply off the main rails is ideal for low power opamp circuitry. Why add another transformer when you don't need it.
Just don't use the LM358 😉 Use something like the NE5532 or LM4562 instead. Any delay circuit should really be in the form of a speaker relay to 100% eliminate any switch on/off noises.
Just don't use the LM358 😉 Use something like the NE5532 or LM4562 instead. Any delay circuit should really be in the form of a speaker relay to 100% eliminate any switch on/off noises.
A simple resistor/zener/cap shunt supply off the main rails is ideal
Which should be calculated basing on the current absorbed by the circuitry
Thanks guys saves time space and money too!
I seem to prefer the NE5532's over the lm358's I guess ill see which I prefer, they're being used as balanced line driver receiver, 50hz cut, crossover, output buffers as well as DC offset (servo/MFB) correction.
I seem to prefer the NE5532's over the lm358's I guess ill see which I prefer, they're being used as balanced line driver receiver, 50hz cut, crossover, output buffers as well as DC offset (servo/MFB) correction.
Which should be calculated basing on the current absorbed by the circuitry
how can this be calulated? i will be running ~15 maybe more opamps.
Post #20
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2199280
Post #37
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2200615
Any DC servo should be based around a FET input opamp such as the TL071/2
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2199280
Post #37
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2200615
Any DC servo should be based around a FET input opamp such as the TL071/2
Maybe it's better to use a dedicated PSU 🙂
Post #20
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2199280
Thanks i think i will use a whole dedicated board for the +/- 35V and +/-15V rails loosly based off elliot sound projects.
Now where to link the grounds?
Keep isolated in PSU and run individual ground for 15 and 35 v to amp board where a ground plane layer is being used?
Or the grounds could be linked at the psu and the amp board with 1 ground wire or 2 ground wires.
Post #37
http://www.diyaudio.com/forums/analog-line-level/167495-simple-stereo-gain-stage.html#post2200615
Any DC servo should be based around a FET input opamp such as the TL071/2
Yes its based around the LM358 or similar its no simple dc servo though 😀
Maybe I'll share at the end?
It includes balanced input -> line driver receiver -> 50hz notch filter -> 3 way crossover (300hz/3000hz) -> output buffer -> 5 3* lm3886 paralell amps (1 tweeter, 2 mids 2 mid-woofers) each with its own DC servo monitoring the output/backemf made by the speakers to correct it.
A few more questions.
Is 50k uf per rail enough for 15*lm3886 amps on a dedicated PSU board?
Should i put maybe a 2200uf per rail close to the chips as well?
Is a 35V toroidal too much (yes I know its more after rectification.)
Is 50k uf per rail enough for 15*lm3886 amps on a dedicated PSU board?
Should i put maybe a 2200uf per rail close to the chips as well?
Is a 35V toroidal too much (yes I know its more after rectification.)
If you mean a 35-0-35 toroid then its definitely to much. Look at the data sheet for the LM3886. 84 volts DC as a maximum under signal conditions. That means a 30 volts AC or less transformer and typically a 25 volt would be a more sensible choice.
50,000uF is OK, even for fifteen LM3886's. Good hf decoupling at the chip is just as, if not more important than trying to add more uF's. A 0.1uf film cap and 100uf electro could be more beneficial as long the DC resistance of the rails is generally low from the reservoir caps to the chip. Its all layout dependent stuff though. Fifteen chips means there is going to be considerable wiring of print between everything.
50,000uF is OK, even for fifteen LM3886's. Good hf decoupling at the chip is just as, if not more important than trying to add more uF's. A 0.1uf film cap and 100uf electro could be more beneficial as long the DC resistance of the rails is generally low from the reservoir caps to the chip. Its all layout dependent stuff though. Fifteen chips means there is going to be considerable wiring of print between everything.
Okay great here is the design for the PSU (note I'll change 35V to 25V)
https://imageshack.com/i/pakJ2memp
Here is the crossover design tell me if you see any issues!
http://imageshack.com/a/img903/2629/xq6z2O.png
https://imageshack.com/i/pakJ2memp
Here is the crossover design tell me if you see any issues!
http://imageshack.com/a/img903/2629/xq6z2O.png
Here is the Power amp design there is 5 of these per speaker.
http://imageshack.com/a/img911/6266/zX3HGX.png
Note: I removed part of the feedback circuit which im not willing to publicly share.
http://imageshack.com/a/img911/6266/zX3HGX.png
Note: I removed part of the feedback circuit which im not willing to publicly share.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Chip Amps
- Dedicated +/- 15V toroidal or zenner diodes?