LM3886 Oscillation? - 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 31st May 2014, 05:28 PM   #1
diyAudio Member
 
Join Date: Feb 2014
Default LM3886 Oscillation?

Hi everyone,
I had some issues a couple of weeks ago with wiring a torroid for a dual mono Chipamp.com LM3886 kit. I've gotten it working now, but I'm having some issues with hum. With open inputs, its dead silent. When I plug in my tester (mobile phone) it starts humming quite loud. It calms down significantly after I hit play, but the hum can still be heard at low volumes. In addition, the chips get quite hot (uncomfortable to touch for more than 10 sec) inside a minute or so if I leave it plugged in to the phone without music playing (while it's humming at the greater volume). If I play music, they still get pretty warm, warmer than I would expect, at a volume that would be comfortable sitting at a desk (not aggressively loud). When I turn the switch to the off position, the caps still have some charge and it'll play for another 2 or 3 sec, with no hum at all.

With music playing, DC offset at terminals was: ~30mv on each channel
Paused the music but still connected (loud hum situation):3.4V left, 300mv right WOW
with open inputs: ~100mv each
shorted inputs to ground:<30mV each. One side was near zero, other was ~30 but I don't remember which was which.

Any ideas? Good thing I'm using junk test speakers.


Here are some pictures of the test bed I have made for it with an old receiver chassis and the amp and PS schematics. If anyone is confused about the wiring I can draw up a quick diagram. The ugly board is a soft starter I build off of Rod Elliots design. Excuse the **** poor soldering and layout, it's the first thing I ever soldered.
Attached Images
File Type: png amp schematic.png (57.4 KB, 324 views)
File Type: png PS Schematic.png (44.4 KB, 320 views)
File Type: jpg IMG_6032.jpg (365.2 KB, 328 views)
File Type: jpg IMG_6035.jpg (283.4 KB, 304 views)
File Type: jpg IMG_6036.jpg (242.4 KB, 294 views)
File Type: jpg IMG_6039.jpg (350.1 KB, 149 views)
  Reply With Quote
Old 31st May 2014, 06:26 PM   #2
DUG is offline DUG  Canada
diyAudio Member
 
Join Date: Jan 2003
Location: mississauga ontario canada
It looks like you have a switch in the N from the power entry.

A switched N still leaves live power on transformer primary...not good idea.

It should be: power---fuse---switch---transformers.(heat shrink over all connections including power entry module and switch)

Good that GND goes to chassis but try to find some green or green/yellow wire for GND.

As for your noise issue...try bundling the AC feed from transformers to rectifier board...minimizing loop areas.

Change the sig-in feed to the amps to some kind of co-axial cable.

__________________
Doug We are all learning...we can all help
"You can't stop the signal, Mal. Everything goes somewhere..."
  Reply With Quote
Old 31st May 2014, 07:10 PM   #3
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Shorten the Protective Earth (PE) wire and bolt it next to the mains cable entry hole.
This Safety Earth is permanent and should never be dismantled.
Is Green/yellow sleeving available in the USA?
But ONLY use it on the PE wire, no where else.
__________________
regards Andrew T.
  Reply With Quote
Old 31st May 2014, 07:18 PM   #4
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
You have the minimalist schematic.
All the National declared "optional" compnents have been omitted.

Add two input filters. A High Pass to block DC and filter sub audio. A low pass to attenuate Radio Frequency (RF).
Make the HF decoupling route as short as possible. The junction between them is the Power Ground, to which you terminate the MF decoupling.
Make the Zobel route as short as possible from Output Pin to Power Ground.
Add an output R||L to convert the Output Zobel to a Thiele Network. This goes into the cable route from chipamp PCB to Chassis mounted Spkr terminals. Twist this cable before and again after the R||L, right up to the terminals.

Connect Ci to R2 and that to input signal return. Make sure that signal return actually returns to the Signal Input socket (or to the base of the vol pt if you have it inside the Power Amplifier).

What cap type have you used fore Cs?
What have you fitted across the rectifier?
Have you fitted the 0.1uF shown on Carlos' PSU schematic? Remove it.

Twist every pair of Flow and Return cables. Mains cables, Power Supply cables, Signal cables, Speaker cables, transformer secondary cables, rectifier cables. If the power supply becomes a +ve & -ve and Zero volts, then this becomes a twisted triplet, but it looks like your rectified power is all done on the PCB. What happens between the PSU PCB and the Chipamp PCB?
The only exception to this twisted pair is the PE wire. It is a solitary wire.
__________________
regards Andrew T.

Last edited by AndrewT; 31st May 2014 at 07:28 PM.
  Reply With Quote
Old 31st May 2014, 07:19 PM   #5
DUG is offline DUG  Canada
diyAudio Member
 
Join Date: Jan 2003
Location: mississauga ontario canada
Quote:
Originally Posted by AndrewT View Post
Shorten the Protective Earth (PE) wire and bolt it next to the mains cable entry hole.
This Safety Earth is permanent and should never be dismantled.
Is Green/yellow sleeving available in the USA?
But ONLY use it on the PE wire, no where else.
Yes, I seem to remember that there is to be no other connections to the Protective Earth Ground connection point.

I don't know about G/Y sleeving because we use G/Y wire at work.

If anything there is green vinyl marker tape available...or green heatshrink.
__________________
Doug We are all learning...we can all help
"You can't stop the signal, Mal. Everything goes somewhere..."
  Reply With Quote
Old 31st May 2014, 07:35 PM   #6
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Dug,
you are shown as Canada. May I presume you have adopted the british convention of Grn/Yel for PE?

But, does that apply across the border in the USA. They don't like changing anything from what they adopted two centuries ago.
__________________
regards Andrew T.
  Reply With Quote
Old 31st May 2014, 07:49 PM   #7
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by AndrewT View Post
You have the minimalist schematic.
Most of the National declared "optional" components have been omitted.

Add two input filters. A High Pass to block DC and filter sub audio. A low pass to attenuate Radio Frequency (RF).
Make the HF decoupling route as short as possible. The junction between them is the Power Ground, to which you terminate the MF decoupling.
Make the Zobel route as short as possible from Output Pin to Power Ground.
Add an output R||L to convert the Output Zobel to a Thiele Network. This goes into the cable route from chipamp PCB to Chassis mounted Spkr terminals. Twist this cable before and again after the R||L, right up to the terminals.

Connect Ci to R2 and that to input signal return. Make sure that signal return actually returns to the Signal Input socket (or to the base of the vol pot if you have it inside the Power Amplifier). Add a reference wire from Signal Return to Main Audio Ground.

What cap type/s have you used for Cs?
What have you fitted across the rectifier?
Have you fitted the 0.1uF shown on Carlos' PSU schematic? Remove it.

Twist every pair of Flow and Return cables. Mains cables, Power Supply cables, Signal cables, Speaker cables, transformer secondary cables, rectifier cables. If the power supply becomes a +ve & -ve and Zero volts, then this becomes a twisted triplet, but it looks like your rectified power is all done on the PCB. What happens between the PSU PCB and the Chipamp PCB?

The only exception to this twisted pair is the PE wire. It is a solitary wire.
Must be past my bedtime ! too many typos. Now corrected.

BTW,
a minimalist can be built with a chipamp and just 6 external components.
A typical implementation usually uses from 10 to 14 external components.
I start with a layout that has 41 external components and remove only those that are duplicated else where.
__________________
regards Andrew T.

Last edited by AndrewT; 31st May 2014 at 07:52 PM.
  Reply With Quote
Old 31st May 2014, 09:12 PM   #8
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
The schematic you have is fine. Don't start adding components arbitrarily until you have figured out why you get hum in your circuit. The schematic as-is will work in 99.9 % of cases. Once you have it working, you can start optimizing. The "optional" components that people refer to were copied from an AC test circuit in the data sheet. They are not required for functionality, but could improve performance under certain conditions. Which conditions and improvements are described in the data sheet.

Anyway. You're building this on a PCB that, presumably, the designer tested before sending to market. It should work. The boards look familiar. Gainclone? Many have built those successfully, so the boards or the circuits are not likely to blame.

I bet one of the input connections is bad. You say the circuit is quiet with the source disconnected. This means that R2 is probably connected. As is the rest of the amp. So my guess is that the fault lies to the left of R2 is the schematic. Check the connections on R1 and the volume pot (if fitted), input connectors, and the cable going to the source. Check both the signal and signal return.

With a battery powered source, you're operating at the best case scenario for ground loops. I.e. the case that is the least likely to result in a ground loop. The fact that the amp is quiet with the source unplugged leads me to believe that the amp is good, but the source connection is bad.

If cleaning up and reflowing the solder joints in the input path doesn't clear things up, try another source just in case it should be a source problem.

Re wire colors: The electrons don't care about the color of the insulation, obviously. But it is mighty handy for the builder if a certain standard is followed. For mains wiring in the US, green is earth/ground, black is hot (marked L for Live), white is neutral (N). In the EU, yellow/green is earth/ground, brown is hot, blue is neutral as I recall.
Earth/ground from the mains inlet MUST connect to the metal chassis. If the chassis is non-conductive, it must connect to all metal parts that can be accessed by the user. That's a safety issue.
In DC circuits, red is commonly used for the positive supply and black for the supply return (minus connection on the battery). I use blue for the negative supply.
If you only have one color of wire, coloring the ends with electrical tape or, preferably, heat shrink works well.

~Tom
__________________
Modulus-86: Composite amp achieving 0.00018 % THD. Damn Good 300B, Novar Spud, 21st Century Maida Reg., Filament Reg., etc.
Neurochrome : : Audio - http://www.neurochrome.com/audio - Engineering : : Done : : Right

Last edited by tomchr; 31st May 2014 at 09:20 PM.
  Reply With Quote
Old 31st May 2014, 11:19 PM   #9
diyAudio Member
 
Join Date: Feb 2014
Quote:
Originally Posted by DUG View Post
It looks like you have a switch in the N from the power entry.

A switched N still leaves live power on transformer primary...not good idea.

It should be: power---fuse---switch---transformers.(heat shrink over all connections including power entry module and switch)

Good that GND goes to chassis but try to find some green or green/yellow wire for GND.

As for your noise issue...try bundling the AC feed from transformers to rectifier board...minimizing loop areas.

Change the sig-in feed to the amps to some kind of co-axial cable.

Maybe I'm an idiot but with AC does it matter which line off the mains I switch? One is always available to bite you if you are grounded correct? You're saying move the switch to the other AC terminal?

Quote:
Originally Posted by AndrewT View Post
Shorten the Protective Earth (PE) wire and bolt it next to the mains cable entry hole.
This Safety Earth is permanent and should never be dismantled.
Is Green/yellow sleeving available in the USA?
But ONLY use it on the PE wire, no where else.
Should I use this as a star ground still? or star ground elsewhere and run wire between SE chassis connection and star?

Quote:
Originally Posted by tomchr View Post
The schematic you have is fine. Don't start adding components arbitrarily until you have figured out why you get hum in your circuit. The schematic as-is will work in 99.9 % of cases. Once you have it working, you can start optimizing. The "optional" components that people refer to were copied from an AC test circuit in the data sheet. They are not required for functionality, but could improve performance under certain conditions. Which conditions and improvements are described in the data sheet.

Anyway. You're building this on a PCB that, presumably, the designer tested before sending to market. It should work. The boards look familiar. Gainclone? Many have built those successfully, so the boards or the circuits are not likely to blame.

I bet one of the input connections is bad. You say the circuit is quiet with the source disconnected. This means that R2 is probably connected. As is the rest of the amp. So my guess is that the fault lies to the left of R2 is the schematic. Check the connections on R1 and the volume pot (if fitted), input connectors, and the cable going to the source. Check both the signal and signal return.

With a battery powered source, you're operating at the best case scenario for ground loops. I.e. the case that is the least likely to result in a ground loop. The fact that the amp is quiet with the source unplugged leads me to believe that the amp is good, but the source connection is bad.

If cleaning up and reflowing the solder joints in the input path doesn't clear things up, try another source just in case it should be a source problem.

Re wire colors: The electrons don't care about the color of the insulation, obviously. But it is mighty handy for the builder if a certain standard is followed. For mains wiring in the US, green is earth/ground, black is hot (marked L for Live), white is neutral (N). In the EU, yellow/green is earth/ground, brown is hot, blue is neutral as I recall.
Earth/ground from the mains inlet MUST connect to the metal chassis. If the chassis is non-conductive, it must connect to all metal parts that can be accessed by the user. That's a safety issue.
In DC circuits, red is commonly used for the positive supply and black for the supply return (minus connection on the battery). I use blue for the negative supply.
If you only have one color of wire, coloring the ends with electrical tape or, preferably, heat shrink works well.

~Tom

These are BrianGT boards available as a kit from chipamp.com. I would prefer not to add components arbitrarily as others have absolutely used these PCBs without issue with the "minimalist" design so I must have an issue specific to my soldering and layout. I will try a few of these in about an hour and get back to you guys. Thanks for the advice.

Regarding wire color, i'm fairly colorblind and electrons seem to share this ailment. Since I'm moving the ground anyway, I'll use green this time to satisfy those with better eyes among us
  Reply With Quote
Old 31st May 2014, 11:31 PM   #10
diyAudio Member
 
Join Date: Feb 2014
Quote:
Originally Posted by AndrewT View Post
You have the minimalist schematic.
All the National declared "optional" compnents have been omitted.

Add two input filters. A High Pass to block DC and filter sub audio. A low pass to attenuate Radio Frequency (RF).
Make the HF decoupling route as short as possible. The junction between them is the Power Ground, to which you terminate the MF decoupling.
Make the Zobel route as short as possible from Output Pin to Power Ground.
Add an output R||L to convert the Output Zobel to a Thiele Network. This goes into the cable route from chipamp PCB to Chassis mounted Spkr terminals. Twist this cable before and again after the R||L, right up to the terminals.

Connect Ci to R2 and that to input signal return. Make sure that signal return actually returns to the Signal Input socket (or to the base of the vol pt if you have it inside the Power Amplifier).

What cap type have you used fore Cs?
What have you fitted across the rectifier?
Have you fitted the 0.1uF shown on Carlos' PSU schematic? Remove it.

Twist every pair of Flow and Return cables. Mains cables, Power Supply cables, Signal cables, Speaker cables, transformer secondary cables, rectifier cables. If the power supply becomes a +ve & -ve and Zero volts, then this becomes a twisted triplet, but it looks like your rectified power is all done on the PCB. What happens between the PSU PCB and the Chipamp PCB?
The only exception to this twisted pair is the PE wire. It is a solitary wire.
Ci was omitted as it was optional. I completely forgot about that decision until just now. Could that be the issue?

There are three 0.1uF caps used in the PSU schematic, which one? and why?
  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
oscillation merkur Tubes / Valves 28 10th May 2006 01:19 AM
Oscillation raveenvijendren Solid State 15 18th December 2005 11:16 AM
Oscillation udip Solid State 5 31st January 2005 06:43 PM
Are two separate lm3886 = 2*lm3886 in parallel or bridge mode rs1026 Chip Amps 11 21st October 2004 11:24 AM
What is oscillation? jnxw2 Solid State 10 19th May 2003 12:31 AM


New To Site? Need Help?

All times are GMT. The time now is 08:31 AM.


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