Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Chip Amps Amplifiers based on integrated circuits

LM3886 & LME49720 composite amp
LM3886 & LME49720 composite amp
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 2nd February 2015, 12:50 AM   #11
papasteack is offline papasteack  France
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
wanted to go fast so not really clean soldering and made order errors...but here is mine. Not yet tested. I got 9 more board, will keep 5 so 4 left if you want to try one...
From upside view, we do not see stacked smd decoupling cap neither crappy serial cap to get votage ratings bellow big solder point
Attached Images
File Type: jpeg download_20150202_014528.jpeg (47.3 KB, 1113 views)
File Type: jpeg download_20150202_014506.jpeg (30.1 KB, 1108 views)

Last edited by papasteack; 2nd February 2015 at 12:55 AM.
  Reply With Quote
Old 2nd February 2015, 01:46 AM   #12
abraxalito is offline abraxalito  United Kingdom
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Quote:
Originally Posted by IO390 View Post
That's a good point, these ultra low noise regs are not exactly necessary. The opamp designers have worked on high PSRR so that you don't have to worry so much about the supply...
@papasteack: Some kind of reg's needed so that the opamp's supply voltage specs aren't exceeded. The 3886 can handle a lot more volts than the 49720. These chips do only have good PSRR at lower frequencies though.

Quote:
I've not seen the Bato MM board until now. It seems there is a lot more about these composite amps around than I realised. I really like the idea of having a sense line to the speaker. I've joined diysmps so I can read a bit more about it.
If THD is the overriding concern then to achieve lowest THD in-situ (with a real speaker load) sense lines become mandatory. Otherwise you'll only get your headline 3-zeros-after-the-decimal number with a purely resistive load.
__________________
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant - Alan Greenspan
  Reply With Quote
Old 2nd February 2015, 08:15 AM   #13
papasteack is offline papasteack  France
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
Quote:
Originally Posted by abraxalito View Post
These chips do only have good PSRR at lower frequencies though.
So does lm338 kind reg. Don't know about shunt regs for the moment with my limited knowledge; it seems pretty interesting, but there's too much different implementation on the web, and don't know where to start trying to make my opinion.

Why not good reg (shunt?) only for the lme49720 (and not for the lm3886), or since it doesn't need high power, a simple filter for it ( crcrc for example + good decoupling of course) ?
  Reply With Quote
Old 2nd February 2015, 10:15 AM   #14
abraxalito is offline abraxalito  United Kingdom
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Various discrete shunt regs exist but for an opamp methinks they're overkill. TL431s I love because they're seriously cheap - the output impedance at lower freqs is around 0.2ohms. They only need less than 1mA quiescent current so well suited to loads in the single digit mA range. Line regulation might be an issue - if so it can be improved substantially by using an LM317L as a CCS to feed them.
__________________
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant - Alan Greenspan
  Reply With Quote
Old 2nd February 2015, 10:49 AM   #15
IO390 is offline IO390  United Kingdom
diyAudio Member
 
IO390's Avatar
 
Join Date: May 2014
Location: Near Brighton, Sussex, UK
Quote:
Originally Posted by abraxalito View Post
@papasteack: Some kind of reg's needed so that the opamp's supply voltage specs aren't exceeded. The 3886 can handle a lot more volts than the 49720. These chips do only have good PSRR at lower frequencies though.
I meant that the LM317/337 is good enough (shunt regs not needed). I have a box full of LM317/337 so it's not an issue for me.

Quote:
Originally Posted by abraxalito View Post
If THD is the overriding concern then to achieve lowest THD in-situ (with a real speaker load) sense lines become mandatory. Otherwise you'll only get your headline 3-zeros-after-the-decimal number with a purely resistive load.
Of course, low THD like this is trivial (as I said in the OP). Sense lines to the speaker are a very good idea though - this board could quite easily be modded for this.
__________________
io390.co.uk
  Reply With Quote
Old 2nd February 2015, 07:02 PM   #16
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
I have a hard time believing the circuit in the pdf is stable with the given component values at least by quick eyeball and counting on my thumbs look at gains, time constants
  Reply With Quote
Old 2nd February 2015, 07:11 PM   #17
IO390 is offline IO390  United Kingdom
diyAudio Member
 
IO390's Avatar
 
Join Date: May 2014
Location: Near Brighton, Sussex, UK
Quote:
Originally Posted by jcx View Post
I have a hard time believing the circuit in the pdf is stable with the given component values at least by quick eyeball and counting on my thumbs look at gains, time constants
I would say the same thing. But it is stable and works fine, or I wouldn't have published it.

I would have posted measurements but my function gen is currently being repaired.
__________________
io390.co.uk
  Reply With Quote
Old 2nd February 2015, 08:14 PM   #18
IO390 is offline IO390  United Kingdom
diyAudio Member
 
IO390's Avatar
 
Join Date: May 2014
Location: Near Brighton, Sussex, UK
As I said before, the values are probably not optimal as it was designed in a morning.

I drew up the basic circuit and gave the values an educated guess, then built it point-point and adjusted the compensation until I had a nice crisp square wave up to 20KHz.

In addition to what I said above, I will provide more measurements when my function gen is fixed. My HP 34401A has also developed a fault on the ohms range so I might be busy for a while...
__________________
io390.co.uk
  Reply With Quote
Old 2nd February 2015, 08:23 PM   #19
tomchr is offline tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
Quote:
Originally Posted by IO390 View Post
If you're wondering why I'm using the LME49720 as opposed to the LME49710, it's because I use the other half as the integrator for the DC servo. Yes it doesn't have the same level of DC accuracy as an OPA227 for example, but that is totally unnecessary for a power amp IMO.
You may want to redo the math on that one... I doubt you get much benefit of the DC servo in terms of output offset unless you use an op-amp with low input voltage offset and low input bias current.

Quote:
Originally Posted by IO390 View Post
Oh and this isn't a copy of anything. I thought I should mention in case someone thought I had ripped off the Modulus-86 or something, which I haven't.
I'm not worried about competition... While your circuit uses a composite topology, it's pretty far from the Modulus-86, actually.

Quote:
Originally Posted by IO390 View Post
As for performance, the distortion at 15W RMS, 1KHz into 8 ohms was less than my sound card could measure, which is 0.0003%.
Try measuring at 20 kHz. That'll tell you a lot about the performance of the layout as well as the amount of loop gain available at 20 kHz.

Also look at the THD at 20 Hz. That's where any THD from the DC servo will show up.

Quote:
Originally Posted by IO390 View Post
I'll get some more graphs and upload them. Of course, distortion levels this low are trivial.
I'm guessing you dropped the 'not'. As in "not trivial".

Quote:
Originally Posted by IO390 View Post
Anyway, any suggestions/comments/criticisms?
Neat circuit. I like the form factor. There's room for improvement in the supply routing, it seems. I also suggest adding the Zobel and Thiele networks so your circuit will remain stable with a slightly capacitive load.

Have you looked at the performance as the output voltage reaches the rail. That's where trouble brews a lot of the time. I suggest testing both with a square wave and with a sine wave. Slowly increase the amplitude of the input signal until the amp clips and go a bit beyond clipping. Watch the output on an oscilloscope. It's common with oscillation in that region of operation and I suggest finding a solution for it...

~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; 2nd February 2015 at 08:26 PM.
  Reply With Quote
Old 2nd February 2015, 08:36 PM   #20
IO390 is offline IO390  United Kingdom
diyAudio Member
 
IO390's Avatar
 
Join Date: May 2014
Location: Near Brighton, Sussex, UK
Quote:
Originally Posted by tomchr View Post
You may want to redo the math on that one... I doubt you get much benefit of the DC servo in terms of output offset unless you use an op-amp with low input voltage offset and low input bias current.



I'm not worried about competition... While your circuit uses a composite topology, it's pretty far from the Modulus-86, actually.



Try measuring at 20 kHz. That'll tell you a lot about the performance of the layout as well as the amount of loop gain available at 20 kHz.



I'm guessing you dropped the 'not'. As in "not trivial".



Neat circuit. I like the form factor. There's room for improvement in the supply routing, it seems. I also suggest adding the Zobel and Thiele networks so your circuit will remain stable with a slightly capacitive load.

Have you looked at the performance as the output voltage reaches the rail. That's where trouble brews a lot of the time. I suggest testing both with a square wave and with a sine wave. Slowly increase the amplitude of the input signal until the amp clips and go a bit beyond clipping. Watch the output on an oscilloscope. It's common with oscillation in that region of operation and I suggest finding a solution for it...

~Tom
I was hoping you'd turn up (no sarcasm)

The input biasing current of the LME49720 is 10nA and the offset voltage is 100uV, (typical datasheet values) so not ideal for a DC servo application. But IMO it's "good enough", and it turns what would otherwise be an offset of up to several hundred mV into an offset below 1mV. Also board space is a problem for two opamps, and cost if I wanted a decent FET opamp.

As for the distortion, I would measure it at 20KHz but my sound card oddly won't measure anything above about 10KHz, it just drops off after that. That's only the standard motherboard line in though, I really need to build something better for making measurements.

As for the distortion values, I do personally believe that the difference between 0.03 and 0.0003% THD is not audible. I do however believe that the side effects of the lower distortion is what makes the difference - from the quick measurements that I have done, the fast opamp keeps the LM3886 "in check" at higher volumes where it might otherwise distort a bit. But that's my opinion.

There is certainly room for improvement on the board, although the supply routing seems OK to me (?).

I'll give the boards some more rigorous testing when my function generator is working again, as I said before. I suspect that only so much can be done with this topology though, it's quite simple compared to most of the ones that I see.
__________________
io390.co.uk
  Reply With Quote

Reply


LM3886 &amp; LME49720 composite ampHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
An LM3875 composite amp dellama Chip Amps 41 8th November 2014 11:17 AM
Bridged LM3886 + LME49720 + remote FB iggy111 Chip Amps 32 2nd September 2014 12:14 PM
LME49720 Pre amp PCB/Kit Jsixis Analog Line Level 23 19th April 2014 11:44 AM
how about LME49720 driving LM3886? mondogenerator Chip Amps 3 16th April 2010 07:31 AM


New To Site? Need Help?

All times are GMT. The time now is 05:13 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki