Go Back   Home > Forums > >

Chip Amps Amplifiers based on integrated circuits

Neurochrome LM3886DR Build
Neurochrome LM3886DR Build
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
Old 20th October 2020, 12:36 AM   #471
tomchr is offline tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
Dang! You squeezed a lot of sardines into that can! Nice work. I like how all boards are supported. Also neat trick with the spade connectors on the Molex footprint.

As for the supply noise on the LM3886DR channels: The 'naked' LM3886 does not have great PSRR, so that could be what you're experiencing. I would try adding an LC filter on the supply to the LM3886DRs. You'll want a filter that offers meaningful attenuation at the switching frequency of the SMPS while not degrading the output impedance of the SMPS within the audio band. So you'll probably end up with a cutoff frequency in the 1-10 kHz range as the best compromise.
Given that you don't have a lot of room for additional components, I suggest focusing on the negative rail first. That's where the LM3886 has the lowest PSRR.

The reason the MOD86 works better is that the composite architecture of the MOD86 provides error correction, which basically eliminates the supply noise.

Tom
__________________
Modulus-86, 186, 286, & 686: 40-240W (8Ω), <-120dB THD. HP-2 Headphone Amp: 660mW, <-130dB THD. Taming the LM3886.
Neurochrome : : Audio - www.neurochrome.com - Engineering : : Done : : Right

Last edited by tomchr; 20th October 2020 at 12:39 AM.
  Reply With Quote
Old 20th October 2020, 01:55 AM   #472
greggtt is offline greggtt
diyAudio Member
 
Join Date: Dec 2014
Tom - Thank you for reading through my post and your tip of looking at rail vs. airborne contamination.

-Does the chip's PSSR "typical" to "limit" range apply to rail noise, possibly causing one chip within spec to be audibly worse? I never realized PSSR applied to noise rejection. I have spare LM3886's I could change out if it were a randomly noisy one (before trying an LC filter). As noted, the left pair are nearly as quiet as the Mod86's and 99.9% satisfactory.

-Are you thinking it's SMPS-specific noise or extraneous noise coming through the supply? SMPS noise shouldn't have a "sputter" character to it, or could it?

Thanks again.
  Reply With Quote
Old 20th October 2020, 05:50 AM   #473
tomchr is offline tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
Is what I've circled in the attached image what you consider the left LM3886DR?

PSRR = power supply rejection ratio. It is measured by applying an AC voltage in series with the power supply rail and measuring the perturbation on the amp output. If you have any noise or other AC signal on the power supply rails, some of it will make it through to the output of the amp. I doubt you'll notice any difference between individual samples of LM3886.

In addition, the amp generates its own noise. Electrons move randomly and that creates noise. Some electrons get stuck in traps (usually impurities in the devices) and take a while to get dislodged. That causes the 1/f component you see in noise plots. That's a different topic than PSRR, though.

Do you have the option of trying with a different power supply? Are you running the LM3886DR at its default 26 dB gain?

Given that the left and right channels behave differently but are fed from the same supply, I'm thinking you're having issues with noise coupling rather than conducted noise. The left input connection is right by the SMPS. Are you able to turn the left LM3886DR upside down for a quick test?

Tom
Attached Images
File Type: jpg top.jpg (323.4 KB, 175 views)
__________________
Modulus-86, 186, 286, & 686: 40-240W (8Ω), <-120dB THD. HP-2 Headphone Amp: 660mW, <-130dB THD. Taming the LM3886.
Neurochrome : : Audio - www.neurochrome.com - Engineering : : Done : : Right
  Reply With Quote
Old 20th October 2020, 02:03 PM   #474
greggtt is offline greggtt
diyAudio Member
 
Join Date: Dec 2014
I realize my last post obscured that it’s the RIGHT pair that’s noisy.
-Yes, you’ve circled my left channel, which is the good side.
-Thanks for the PSRR specifics.
-I do have a linear supply (from before adding the extra channels) and can give it a try.
-Gain is set to 29dB for all modules.
-Yes, the LM3886DR modules can be detached and moved around a bit, but I’ve not tried that yet. But interestingly, the noisy pair on right has less SMPS proximity.

Thank you.
  Reply With Quote
Old 20th October 2020, 02:53 PM   #475
poseidonsvoice is online now poseidonsvoice  United States
diyAudio Member
 
poseidonsvoice's Avatar
 
Join Date: Jul 2004
Location: NC
Neurochrome LM3886DR Build
Gregg,

Super nice build!

Best,
Anand.
__________________
Science is not a democracy - Earl Geddes
DIYAudio is a journey not a destination!
  Reply With Quote
Old 20th October 2020, 10:11 PM   #476
tomchr is offline tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
Quote:
Originally Posted by greggtt View Post
I realize my last post obscured that it’s the RIGHT pair that’s noisy.
Ah. I'm sure I did read that in your longer post from the other day, but promptly swapped left and right. I'm rather left/right confused. My bad.

Quote:
Originally Posted by greggtt View Post
-Yes, you’ve circled my left channel, which is the good side.
I see. I'll direct the rest of my comments towards the non-left side, then.

Quote:
Originally Posted by greggtt View Post
But interestingly, the noisy pair on right has less SMPS proximity.
Yeah. There went that theory. However, it does give rise to another theory: Could it be that UcD couples into the LM3886DR? Do you have a way to turn the UcD off while leaving the rest on? Preferably by disconnecting its power supply.

Tom
__________________
Modulus-86, 186, 286, & 686: 40-240W (8Ω), <-120dB THD. HP-2 Headphone Amp: 660mW, <-130dB THD. Taming the LM3886.
Neurochrome : : Audio - www.neurochrome.com - Engineering : : Done : : Right
  Reply With Quote
Old 21st October 2020, 02:35 AM   #477
greggtt is offline greggtt
diyAudio Member
 
Join Date: Dec 2014
Yup, that was the first thing I tried, but it had no effect. I will try substituting the linear PS to try to narrow it down. I can also try to isolate the problem to a single board.
  Reply With Quote
Old 21st October 2020, 02:50 AM   #478
greggtt is offline greggtt
diyAudio Member
 
Join Date: Dec 2014
Quote:
Originally Posted by poseidonsvoice View Post
Gregg,

Super nice build!

Best,
Anand.
Thanks for the kind words. It was one of the few fun things facilitated by Covid-19.
  Reply With Quote
Old 21st October 2020, 05:28 AM   #479
tomchr is offline tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
Quote:
Originally Posted by greggtt View Post
Thanks for the kind words. It was one of the few fun things facilitated by Covid-19.
You and me both, actually. I've caught up on a bunch of back burner projects. If this continues I may have to write/update my business plan or something.

Are the offending LM3886DRs the bridged ones? If so, I would try to test them one at a time. Perhaps you had the same thought when you mentioned isolating the amps.

Tom
__________________
Modulus-86, 186, 286, & 686: 40-240W (8Ω), <-120dB THD. HP-2 Headphone Amp: 660mW, <-130dB THD. Taming the LM3886.
Neurochrome : : Audio - www.neurochrome.com - Engineering : : Done : : Right
  Reply With Quote
Old 21st October 2020, 05:51 PM   #480
greggtt is offline greggtt
diyAudio Member
 
Join Date: Dec 2014
Quote:
Originally Posted by tomchr View Post
You and me both, actually. I've caught up on a bunch of back burner projects. If this continues I may have to write/update my business plan or something.
Difficult times for sure, I hope the smiley face means you have decent options.

Quote:
Originally Posted by tomchr View Post
Are the offending LM3886DRs the bridged ones? If so, I would try to test them one at a time. Perhaps you had the same thought when you mentioned isolating the amps.
Yes, the (4) LM3886DR's are bridge pairs. None are used single-ended.

I'm comfortable troubleshooting, and understand isolating stages is likely what's needed.

Really my query was:
1- if the type of noise I hear sounded of SMPS origin,
2- if the Mean Well units* radiate high frequency EMI from a specific area of the module, where my layout increases the potential for proximity coupling,
3- if the physical separation of the stacked bridged-boards, or my input wiring scheme, makes it hyper-susceptible to EMI.

*I'm using the EPP-400-27 as opposed to the low-EMI rated RPS-400-27 because I got them dirt cheap.

Thank you for your time,
Gregg
  Reply With Quote

Reply


Neurochrome LM3886DR BuildHide this!Advertise here!
Thread Tools

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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Neurochrome LM3886 Done Right - 45 W, 8 Ω, 0.002 % THD+N; 80 W, 4 Ω, 0.004 % THD+N tomchr Vendor's Bazaar 32 31st August 2020 06:49 PM
Neurochrome Parallel-86 Group Buy tomchr Group Buys 55 23rd August 2019 10:14 AM
Neurochrome Differential preamp build IanLenco Analog Line Level 39 10th March 2019 07:08 PM
FS: 4 channels of Neurochrome Mod-86 amp kits tubesguy Swap Meet 4 7th April 2017 05:46 PM
FS: Full set of Neurochrome PCBs (Parallel-86, Power-86, THAT Rx) FJHookah Swap Meet 5 7th July 2016 09:15 PM


New To Site? Need Help?

All times are GMT. The time now is 10:17 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki