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

Chip-amps suited as power stage in a composite amplifier, LM1875/TDA2050 excluded.
Chip-amps suited as power stage in a composite amplifier, LM1875/TDA2050 excluded.
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 4th April 2020, 10:34 PM   #111
Turbowatch2 is offline Turbowatch2  Germany
diyAudio Member
 
Join Date: Aug 2013
Location: Northern germany
The through hole looks fine, I would prefer it, as I like old style components and have tonīs of them. If you have to buy parts, the SMD is probably better. You get 0.1% resistors in SMD, which is a very important. The better you match components, the lower you can make the in line resistor at the output. .1-.2 Ohm/ 3 Watt should do. Only with real LM3886, off course!
If anyone wants to build such an amp, there are some things to think off. For example the output should get a L-R which can be at the output posts. Read at shine7.com and "taming the LM3886" before you start a build.

PS SMD board

1pcs lm3886 x3 Parallel 150w HIFI Pure DC Mono Endstufe Bare Board | eBay
  Reply With Quote
Old 4th April 2020, 10:36 PM   #112
aparatusonitus is offline aparatusonitus  Croatia
diyAudio Member
 
aparatusonitus's Avatar
 
Join Date: Jan 2004
Location: Split, Croatia
Hi Turbowatch2,

I don't think this pcb is made properly ... in the picture I have drawn where the positive going signal from + Vcc connector goes -> via pcb plane/trace -> Amp(s) -> balancing resistor(s) -> output trace(s) -> + speaker connector -> load -> then back via -> ground connector -> ground plane/trace -> main capacitor(s) - and finally again at +Vcc plane/trace, so the process is repeated again as specified above. There is no local decopling near each amp, so the (large) loop is the same for all frequencies that the amp can handle, from pure DC to xy Mhz. The same thing happens for the negative going signal also just on the other side of the pcb.
__________________
Non é mai abbastanza...
  Reply With Quote
Old 4th April 2020, 11:10 PM   #113
aparatusonitus is offline aparatusonitus  Croatia
diyAudio Member
 
aparatusonitus's Avatar
 
Join Date: Jan 2004
Location: Split, Croatia
Hmm, i have some unknown problem uploading pictures...I'll try tomorrow...sorry.
__________________
Non é mai abbastanza...
  Reply With Quote
Old 5th April 2020, 05:56 AM   #114
aparatusonitus is offline aparatusonitus  Croatia
diyAudio Member
 
aparatusonitus's Avatar
 
Join Date: Jan 2004
Location: Split, Croatia
Quote:
Originally Posted by aparatusonitus View Post
Hi Turbowatch2,

I don't think this pcb is made properly ... in the picture I have drawn where the positive going signal from + Vcc connector goes -> via pcb plane/trace -> Amp(s) -> balancing resistor(s) -> output trace(s) -> + speaker connector -> load -> then back via -> ground connector -> ground plane/trace -> main capacitor(s) - and finally again at +Vcc plane/trace, so the process is repeated again as specified above. There is no local decopling near each amp, so the (large) loop is the same for all frequencies that the amp can handle, from pure DC to xy Mhz. The same thing happens for the negative going signal also just on the other side of the pcb.
Let's see if it works now...I had some problems syncing the Onedrive app.
Attached Images
File Type: jpg fpcb147.jpg (119.7 KB, 112 views)
__________________
Non é mai abbastanza...
  Reply With Quote
Old 5th April 2020, 07:57 AM   #115
Turbowatch2 is offline Turbowatch2  Germany
diyAudio Member
 
Join Date: Aug 2013
Location: Northern germany
@aparatusonitus

This is not my design and I do not see it as the best you can do.
Anyway, it is very close to the one "shine7" used and this works well single ended and bridged.
Then, there is often a basic misunderstanding between high frequency theory and the reality of a band limited amps, cruising around 10-50.000Hz. Which is noting where you get exited about high frequency electromagnetic stuff.
As we say in German: " Leave the church in the village".
The chip is getting tired at 100 kHz.
Maybe have a look into your cell phone, there you go with .8-5 MHz all day long.

If you tune your Lada, not all rules from Formula 1 may fit.

I do not understand the problem you see, any amp has to run in such a loop.
Maybe you can explain the solution you see for the problem?

Please do not forget, there are film capacitors that have to be soldered directly at the chips legs from the underside of the PCB. Making the way as short as can be. Then, as I mentioned, some extra copper over the traces will be beneficial.

People see all kind of problems with these chips that do not exist if you buy a real one, no China clone. I bet the Chinese fakers sell a lot more fake 3886 than TI real LM3886.
If you trust these chips, you ask for trouble.
The real one is quite tolerant to parts value, component quality and layout. This is most important for OEM producers and a cause while it is still around.

The Boucherot cell is external, usually at the speaker posts at the back of the case.

The board is much smaller than it looks, it is just 6 cm x 12 cm. You get two on a 25 cm / 10" heat sink. So parts are close together.

Last, in a bridge amp the current flow is from one amps + to the other ones inverted +. So ground is not directly involved.

Please, donīt get me wrong, share your ideas, that is why we communicate here!

Thank you!

PS my single chip, low cost China 3886 build has an output offset of .5 mV. With input closed, output open (same with 4 Ohm load). Without any servo or trimmer. I have seen other builds measuring 16 mV after trimming. Something to think of. As my projects go slow, I can not share more about these 3x boards. I will sure buy my chips at Mouser, DigiKey or RS and select them for groups of similar offset.

Last edited by Turbowatch2; 5th April 2020 at 08:08 AM.
  Reply With Quote
Old 5th April 2020, 09:23 AM   #116
aparatusonitus is offline aparatusonitus  Croatia
diyAudio Member
 
aparatusonitus's Avatar
 
Join Date: Jan 2004
Location: Split, Croatia
Quote:
Originally Posted by Turbowatch2 View Post
Maybe you can explain the solution you see for the problem?
The solution is not to use that pcb...you will never achieve datasheet performance using it. Due to the large, high ac current loop inside which there are very sensitive signal nodes, various spurious signals will be generated and superimposed to an input signal that is yet to be amplified, and depending on whether such signals nodes are inside or outside the negative feedback loop, the same will be corrected or not.
You will be better served if you make the p2p variant yourself, using knowledge from neurochrome sites, using low impedance, close coupled traces for all input and output high ac current signals, also taking into account power supply decoupling and how and where to connect signal gnd to amplifier power gnd...exactly as Tom Christiansen explained.
__________________
Non é mai abbastanza...
  Reply With Quote
Old 5th April 2020, 01:34 PM   #117
Turbowatch2 is offline Turbowatch2  Germany
diyAudio Member
 
Join Date: Aug 2013
Location: Northern germany
See, Iīm a technician, no religious fanatic with HIFI.
Ok, you have your knowledge from neurochrome. Tom C. is a gifted engineer and a great sales man. Iīm happy that such wonderful creations like his exist, even if they may not be for anyone.
For people with lower expectations, even a simple LM3886 like TI/ National have it in their datasheets will work, for others, like you maybe, not. Even if you only guess that they are worse, not know. Did I mention religion?
Poor people like me have only an oscilloscope capable 20 MHz, so what is even further up there is out of my view. But for others this may be the region important for ultimate sound quality.
I have been building kit amps for more than 45 years now. Small details can make huge sound differences and some huge construction details can badly disappoint in an objective A-B test. Which is the final criteria for me.
For example, I bet my house, car and two daughters, that you will not hear the difference of an .2mV output offset and 50mV.
That does not mean you may not feel better with .2mV!
  Reply With Quote
Old 5th April 2020, 02:00 PM   #118
chermann is offline chermann  Austria
diyAudio Member
 
Join Date: Jan 2017
Location: Vienna
Quote:
Originally Posted by Turbowatch2 View Post
The through hole looks fine, I would prefer it, as I like old style components and have tonīs of them. If you have to buy parts, the SMD is probably better. You get 0.1% resistors in SMD, which is a very important. The better you match components, the lower you can make the in line resistor at the output. .1-.2 Ohm/ 3 Watt should do. Only with real LM3886, off course!
If anyone wants to build such an amp, there are some things to think off. For example the output should get a L-R which can be at the output posts. Read at shine7.com and "taming the LM3886" before you start a build.

PS SMD board

1pcs lm3886 x3 Parallel 150w HIFI Pure DC Mono Endstufe Bare Board | eBay

Thanks Turbo.
I guess for non composite amp - an amp with "just" 3x parallel would be sufficient for me...
chris
  Reply With Quote
Old 5th April 2020, 04:43 PM   #119
aparatusonitus is offline aparatusonitus  Croatia
diyAudio Member
 
aparatusonitus's Avatar
 
Join Date: Jan 2004
Location: Split, Croatia
Quote:
Originally Posted by Turbowatch2 View Post
See, Iīm a technician, no religious fanatic with HIFI.
As such, I perceived you too...otherwise I wouldn't bother.

Quote:
Originally Posted by Turbowatch2 View Post
Ok, you have your knowledge from neurochrome.
Your assumption is wrong...I can say, without false modesty, that I have mostly learned the important bits and piaces from the real giant (it is a real big loss for all of us that he is no longer able/want to participate in the discussion here and elsewhere as much as I would like)...his forum nick is ilimzn, here and elsewhere. As luck would have it, he was also from Croatia, so an entire generation from Croatia had the opportunity to study online basic and advanced electronics for over a decade.

To the vast majority of us, he clearly explained and practically showed why /how the ac currents in the amplifiers/psu flow...neurochrome was able to implement this from a to z, so I mentioned him as a good reference.

That's all, no religion involved, no audiofoolery...purely technical.

Quote:
Originally Posted by Turbowatch2 View Post
For people with lower expectations, even a simple LM3886 like TI/National have it in their datasheets will work, for others, like you maybe, not.
Believe it or not, a correctly implemented schematic from NS datasheets works for me to.

Quote:
Originally Posted by Turbowatch2 View Post
For example, I bet my house, car and two daughters, that you will not hear the difference of an .2mV output offset and 50mV.
That does not mean you may not feel better with .2mV!
It's not clear to me why you even mention the DC offset output every once in a while...we're not talking precision opamps here...and to answer the question, I could not care less for a DC offset within reasonable limits for an output amplifier.
__________________
Non é mai abbastanza...
  Reply With Quote
Old 5th April 2020, 09:51 PM   #120
Turbowatch2 is offline Turbowatch2  Germany
diyAudio Member
 
Join Date: Aug 2013
Location: Northern germany
I think technically we are not so far apart as it might seem, it is just another view with weight put on different edges.
What you say is all very true, if you build a discreet amp from scratch, but with such a chip, the advantage is that 90% of the critical stuff done inside. On the outside there are just a few variable parts and such of high volume. If someone starts to tell me about nH inductance and shows me advantages in the MHz region of some combined parts, I scratch my head and ask if this is relevant for me and the LM3886 amp that does not even operate at more than 100kHz. Also, why do I have to design my a*s off, just to get distortion from .001 to .0005 %. All these improvements are real, but how do they turn out to the listening audience?
The answer is easy: For some it matters, for others not. Iīm more the "not" type of guy. Iīm happy that out there are talented engineers that push limits, but some may only matter as interesting data. I donīt believe in overestimating single parameters, but count on the combination of the whole package.
If you have read my previous posts carefully, you will know that I questioned this board from the first moment it was mentioned and advised not to use it as is. There are quite some things to be improved and modified. For you it may be the best way to design your own, new, better board, for others, modifying something worth 20$ is a better way. In my country even the PCB material costs more than the Chinese board. I have etched my first own design PCB when I was not even 14 years old. By the way, was a STK Hybrid amp that worked more than 10 years, until I disassembled it. So I know a bit about this stuff.
From what I see in this specific board is a nice, simple opportunity to build something interesting. If it turns out to sound bad, what I really donīt expect, even that does not matter, as it is "hobby". You do not loose anything, as you do it for recreation, not profit. Worst case, I will take the more expensive parts for another project and throw, maybe 40$, in the bin.

If you want to add something constructive to this thread, fine, if you only want to troll, no problem, go on, there is an ignore button.

Tomīs neurochrome amps are not the theme here, he has enough pages filled with his premium stuff. Maybe in half a year I will buy a few of his boards, but now I want to play...
  Reply With Quote

Reply


Chip-amps suited as power stage in a composite amplifier, LM1875/TDA2050 excluded.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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
LM1875 in parallel configuration and used in a composite amplifier. FauxFrench Chip Amps 550 17th March 2020 04:09 PM
TDA2050 vs. LM1875 for my gainclone lineup Chip Amps 11 1st May 2018 06:01 PM
Stereo LM1875 / TDA2050 PCB with Power Supply kuldeepsingh Chip Amps 35 27th May 2012 11:29 AM
Power output test LM1875, TDA2050, TDA2040 johnr66 Chip Amps 5 6th May 2011 06:28 PM
LM1875, TDA2050 on 2 Ohms johnr66 Chip Amps 6 18th February 2010 01:55 PM


New To Site? Need Help?

All times are GMT. The time now is 04:54 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki