PCB design of parallel of two inverted LM3886 with buffer and dc-servo. - diyAudio
Go Back   Home > Forums > Amplifiers > Chip Amps

Chip Amps Amplifiers based on integrated circuits

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th July 2006, 12:40 PM   #1
Roushon is offline Roushon  India
diyAudio Member
 
Join Date: Sep 2004
Location: Mumbai (Bombay)
Default PCB design of parallel of two inverted LM3886 with buffer and dc-servo.

I already put the PCB design in the thread `Bridgle or Parallel" where I had put the circuit design sometime back. But as I need suggestions and corrections for the circuit and the pcb design, I thought I should start a new thread with a different title.

The links are:

Cirsuit. http://www.diyaudio.com/forums/attac...amp=1102598305

Power Supply.

http://www.diyaudio.com/forums/attac...amp=1103622043

A .png file of the PCB.

http://www.diyaudio.com/forums/attac...amp=1154059199

Top side of the board:

http://www.diyaudio.com/forums/attac...amp=1154059264

Bottom side of the board:

http://www.diyaudio.com/forums/attac...amp=1154059327

I will make the following changes in the circuit/pcb suggested by Gajanan.

1. move the parts away from the heat sink.
2. place several PTH in power line wherever it ocures.
3. take the outgnd from one of the filter capacitor.

I will also redesign the circuits around the regulators in the PS to meet the minimal load requirements.

If you see any more troubles or mistakes with the circuits or PCB please point out.

thanks
Roushon.
__________________
S. Roushon
  Reply With Quote
Old 31st July 2006, 04:02 AM   #2
Roushon is offline Roushon  India
diyAudio Member
 
Join Date: Sep 2004
Location: Mumbai (Bombay)
Default simplicity

I have seen in the thread `Three-resistor LM3886 chipamp' by Mick_F a very simple circuit around LM3886 where not even the pin 5 is connected to +V. I have now some questions if anybody please clarify.

1. Is it possible to make parallel of two such circuits without making further changes?
2. If this is true then I, now, wonder why so much cry was made in this forum about complicated circuits and even more complecated PCB design for parallel/bridge of two LM3886.

I am in this forum for sometime and learn from here quite a lot, which prompted me to go ahead with the above circuit. I have no experience about high power amplifier.

I hope someone will answer.

Regards
Roushon.
__________________
S. Roushon
  Reply With Quote
Old 31st July 2006, 09:57 AM   #3
sangram is offline sangram  India
diyAudio Member
 
Join Date: Sep 2002
Location: India
1. Yes, provided you hand match the resistors to a few ohms of each other, to ensure equal current sharing and add output resistors to each output - or use 0.1% resistors...

2. To each their own, some people want buffers, servos and nulling curcuits, regulated supplies, DC protection, etc. The idea is to go what you're comfortable with.

Remember Mick's circuit does not even have a zobel, which I consider a minimum for any amplifier myself (since I don't really know where it's going to be connected always).

I would use the datasheet circuit, and use P2P wiring. really the 3886 is too simple to wire up to even require a PCB, just 6 physical connections are required to the chip.

My 2 cents...
  Reply With Quote
Old 3rd August 2006, 09:09 AM   #4
Roushon is offline Roushon  India
diyAudio Member
 
Join Date: Sep 2004
Location: Mumbai (Bombay)
Quote:
Originally posted by sangram
1. Yes, provided you hand match the resistors to a few ohms of each other, to ensure equal current sharing and add output resistors to each output - or use 0.1% resistors...

2. To each their own, some people want buffers, servos and nulling curcuits, regulated supplies, DC protection, etc. The idea is to go what you're comfortable with.

Remember Mick's circuit does not even have a zobel, which I consider a minimum for any amplifier myself (since I don't really know where it's going to be connected always).

I would use the datasheet circuit, and use P2P wiring. really the 3886 is too simple to wire up to even require a PCB, just 6 physical connections are required to the chip.

My 2 cents...
Thanks Sangram. I have one question regarding Zobel. How does one calculate the value of the resistor and the cap in the Zobel at the output stage of an amplifier. I think one needs to know the speaker characteristic also. Or one can calculate a universal value!

I am still waiting for suggestion and criticism regarding the circuit and the pcb design from anybody. Any doubt will be helpful for me. I have all the parts to go ahead with the etching of the PCB and assemble them. Just to make sure that everything is OK before that.

Thanks and regards
Roushon.
__________________
S. Roushon
  Reply With Quote
Old 3rd August 2006, 12:43 PM   #5
sangram is offline sangram  India
diyAudio Member
 
Join Date: Sep 2002
Location: India
f=1/(2*pi*R*C) where f = turnover frequency, R = resistor, C = Capacitor, pi = 3.14

f is selected first, then the C, then you can compute R.

Typically f is an iterative process, it should be high enough to keep any effects outside the audible range, and low enough so the amplifier does not go into oscillation. A lot of people don't use a zobel. I always use one, and I find a combination of 4.7 ohms + 100nF sufficient (about 340KHz) for most amps. The one PD supplies with his kits is 2.7 + 100nF, which is about 600KHz.

The zobel you're confusin it with is the one applied to the woofer terminals to equalise the rising impedance at high frequencies offered by the inductance of the voice coil, that is also called a zobel, its purpose is diferent from the one at the amp output, and it cannot stabilise an oscillating amplifier.

For that zobel you need to know the DCR and inductance of the speaker voice coil. The purpose of putting a zobel there is to ensure the series inductor in the crossover is still looking into the right impedance at high frequencies... The two can be served by the same set only if it's a full range system or one that uses no inductor on the woofer, I guess.

Like I said on your schematic, I think it's way too complex. I would ditch the regulated supply and the buffers as well, and take a regular stereo amp based on two LM3886. Then ditch the PCB as well, and assemble everything on P2P rather than PCBs.
  Reply With Quote
Old 3rd August 2006, 07:09 PM   #6
Roushon is offline Roushon  India
diyAudio Member
 
Join Date: Sep 2004
Location: Mumbai (Bombay)
Default thanks

Thanks a lot for the clear explanation. I will also add a Zobel. For a parallel connection is it enough to add one Zobel at the common output? What should be the wattage and making of the resistor?

Logging out from the forum...
__________________
S. Roushon
  Reply With Quote
Old 4th August 2006, 04:21 AM   #7
sangram is offline sangram  India
diyAudio Member
 
Join Date: Sep 2002
Location: India
Since there is an output resistor, I guess each half needs its own zobel.

Metal film 1/2 watt should be enough, but some designs specify 1 watt or so. Carbon film cold also be used, I don't know if it affects sound quality.

I've tried both carbon and metal film (from the spare parts bin) and cold not tell if there was a difference sonically.
  Reply With Quote
Old 4th August 2006, 05:29 AM   #8
Hurtig is offline Hurtig  Denmark
diyAudio Member
 
Join Date: Feb 2005
Location: Denmark - Århus
There are some thing I really do not like about this schematic.
Why would you want to use 2 paralell input buffers??? I see the idea of using paralell LM3886, since you get more power. But you should consider feeding them from the same buffer.
Even though both buffers are identical, there will always be some difference due to production tolerances. And if the 2 LM3886's are not driven be exactely the same signal, the will work against each other.
Also you should consider using only 1 DC-servo for the system. Not one for each LM3886. Paralell circuits will always try to correct each other, and therefor work against each other.
Also consider using a low speed op-amp for DC-servo. Like LF411, OP177...
  Reply With Quote
Old 4th August 2006, 02:55 PM   #9
Roushon is offline Roushon  India
diyAudio Member
 
Join Date: Sep 2004
Location: Mumbai (Bombay)
Default changed plan

Due to strong suggestions from so many people, I have changed my mind to a bridge circuit of two lm3886. As I am new in this, I am now afraid to take risk in a parallel circuit. I also have to change my speaker to 8 Ohm.

I will put the new circuit soon.
__________________
S. Roushon
  Reply With Quote
Old 5th August 2006, 04:51 PM   #10
palesha is offline palesha  India
diyAudio Member
 
Join Date: Nov 2001
Location: Pune India
Hv u got the LM 3886. If ok then let us know ur source.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
LM3886, XO, Buffer and Servo jackinnj Chip Amps 3 30th October 2008 12:14 AM
Non inverted, 3 x parallel LM3886 schematics and layout. markiemrboo Chip Amps 0 6th December 2007 12:42 PM


New To Site? Need Help?

All times are GMT. The time now is 05:02 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2