diyAudio (
-   Group Buys (
-   -   PCB 3xLM3886 (PA150) 150W (

marlowson 27th June 2012 07:49 AM

PCB 3xLM3886 (PA150) 150W
Ladies and Gentlemen

Let me put all the informations in one and a right place (group buys).

Basing on alexw88's ( experience, I've created my own project of PA150 (3xLM3886). It has been modified a bit, the size has been reduced, and some other tweaks have been made. I'd like to share the project with you (below). PDF file with switchable layers, try it! I'm gonna order the PCB soon, if anyone wants to grab a couple of PA150 PCB, let me know. Of course, tweaks still can be done before the order.

PCB layout:

Power supply up to +35/-35 VDC. LM3886 datasheet.
Details: PA150 - 3x LM3886 PCB assembly guide
PA150 - 3x LM3886 PCB assembly guide

I has been well tested by alex88. Testing and measurement:
DIY BPA300 6x LM3886 300W audio Amplifier

It is still the same project, except the capacitors which were moved from the bottom to the top, C10 (10u) replaced with wima (better solution), and some downsizing.

Anyone who is interested, just give me your e-mail address and I will send you some details.

Of course don't hesitate to ask questions :)

ByronInPortland 30th June 2012 06:05 PM

Comment on parallel chipamps
Please take the following advice as constructive criticism. It's not at attempt to rip on your design:)

All this information on bridging and paralleling LM3886's can be found in National Semiconductor's Application note 1192, available here:

When paralleling LM3886's the load is divided up among the chips. Section 3.0 reads "Number of ICs in Parallel * Load Impedance = Load Impedance seen by each individual IC". It does not increase the gain of each chip. The increased power available from parallel chips comes only from the ability to drive lower impedance loads.

For example: suppose you are using 35V rails with a single chip, which will give you 50W into an 8R load. Hooking this up to a 4R load to get 100W will exceed the thermal dissipation abilities of the chip and it will shut down. Paralleling chips will split the thermal load between the two chips and allow you to get the desired 100W. Output into an 8R load will still be 50W. Your 3 LM3886's will probably give you 150W into a 2.7R (8/3) load. The important thing is that you're only getting the benefits of paralleling chips if you are using a low impedance load. You may get a little better sound because chips have lower noise & distortion when kept cool, but not more power for a given load.

Bridging LM3886's will give you extra power into a given load. However, the bridged/Paralleled LM3886's (BPA200) has additional complications, and servo circuits are recommended to minimize power-robbing and heat-producing output offset voltages (see section 7.2.2 of AN1192). I couldn't tell whether or not your project has such servo circuits because I couldn't open the PDF, but with 3 paralleled LM3886's, such circuits would probably be important.

pinnocchio 1st July 2012 04:55 AM

Hi ByronInLawrence,

I totally agree with you on the theory and partly on practice for that particular application. I have built this exact circuit about two years ago and while it is not supposed to give you more power at say 6ohms by paralleling more ships, it did make a huge difference in dynamics. The amp feels faster, more "alive".

Using two of those modules and bridging them (not bridging the chips as per the datasheet) but the amp modules, see this thread DIY BPA300 6x LM3886 300W audio Amplifier, it gave almost twice the power at any given load that the amp can support. So basically, these boards are paralleled LM3886x3 and the amps are then bridged using a balanced signal (transfo or op-amp). I'm sure you already know all this... :D

When I did this a couple of years ago, I was using the DRV134 "I believe it is the one" to drive two amps in bridge mode. But then I wanted to do the same as Shine 7 and bought the nice LL1545A transfos which to this date are still nicely wrapped in their original packaging. I wanted transfos to eliminate ground loop which sometimes occurred with the DRV134 but mostly for removing the coupling caps since with transfos, there is no DC issues.

If anyone is interested in those two transfos, just PM me...

As for the servo, well there is none... Only three pots, one for each ship. You have to adjust them before soldering the power resistors. It was a good 20 min+ of trimming each pots until they were all flat almost equal. But it was all worth it!

Good luck with the project, it is really a nice amplifier and sounds great!


ByronInPortland 2nd July 2012 02:05 PM


Originally Posted by pinnocchio (
it did make a huge difference in dynamics. The amp feels faster, more "alive".

That's good to hear because I'm building an amp with pairs of parallel LM3886's right now. I'm looking forward to the better sound!

FormosaWest 2nd September 2013 05:03 AM

Are there any pcbs left? Id like 4 to build two balanced channels

mprivitera1 30th September 2013 05:51 AM


where to buy The PA150 PCB?

Thanks for the reply.

All times are GMT. The time now is 12:31 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio