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

Single supply LM3886 layout
Single supply LM3886 layout
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 15th December 2017, 08:08 PM   #51
00940 is offline 00940  Belgium
diyAudio Member
 
00940's Avatar
 
Join Date: Sep 2003
Location: Liège
I've had a look. There is indeed a small blip at startup as Q4 starts conducting. In sims the voltage across the resistor simulating the relay goes up to 3.5V before being pulled again to 0. On a 48V relay that should be ok. I'm using the schematic of the esp33 as is in a power supply and it works without problems.

The voltage developed across the relay is function of the speed at which the supply rise but depends mostly of the value of R8.
__________________
All that is excessive is insignificant. Talleyrand
  Reply With Quote
Old 15th December 2017, 11:43 PM   #52
tomchr is online now tomchr  Canada
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Calgary
No problem then.

Tom
__________________
Modulus-686: 240W/8Ω @ <0.00025% THD. Modulus-86: 40W/8Ω @ 0.000054% THD. DG300B & Other Tube Circuits.
Neurochrome : : Audio - www.neurochrome.com - Engineering : : Done : : Right
  Reply With Quote
Old 2nd February 2018, 08:29 PM   #53
00940 is offline 00940  Belgium
diyAudio Member
 
00940's Avatar
 
Join Date: Sep 2003
Location: Liège
After being first in vacation, then busy and finally sick, I found the time to revisit this thing.

I managed to slightly simplify the dc offset detection circuit. The cost is a slightly more complicated ac-loss detector. It's all fairly basic.

For the dc-offset circuit: R1-C1 are the input filter. R2 allows to tailor the sensitivity of the offset detection. R5-C2 determine the startup delay. Once the voltage at the base of q3 reaches 25.5V or so, q3 turns on and turns on q2, allowing the 48Vdc relay to operate.

For the ac-loss circuit: as long as q5 operates, it keeps discharging c3, keeping voltage at the base of q4 low. When ac disappears, q5 turns off and the voltage shoots up, turning on q4 which shorts out the base of q3. The zener is needed as there's always a small dc voltage at the top of c3.

Besides the schematic, I attach a quick sim result to show it operates as it should. Blue is the input of the ac-loss detector, green is the input of the dc-offset detector. Red is the current through the relay. There is a 6.5sec startup delay btw.
Attached Images
File Type: jpg dc protect.jpg (43.3 KB, 112 views)
File Type: png dc protect2.PNG (13.1 KB, 111 views)
__________________
All that is excessive is insignificant. Talleyrand
  Reply With Quote
Old 3rd February 2018, 07:51 AM   #54
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
well done 940.
That implementation is very similar to D.Self's, that I have been using for many years.

It does not suffer chatter due to the delays.
It has delayed ON at start up and after reset following a DC detection.
It has instant OFF on loss of mains and/or DC detect.

The one problem I have found is that at some very low frequency and with high signal level the DC detect gets confused and sends a string of pulses to the the trigger transistor. In practice on a working amplifier I don't find this to trigger spurious triggering. But it is easily demonstrated using a sinewave generator feeding the DC detect.
__________________
regards Andrew T.
  Reply With Quote
Old 3rd February 2018, 11:22 AM   #55
00940 is offline 00940  Belgium
diyAudio Member
 
00940's Avatar
 
Join Date: Sep 2003
Location: Liège
Not all is quite perfect yet though. A point of problem is the load resistor charging up the output capacitor and how it mixes up with the dc offset protection. It has to be relatively low (<300R) otherwise the delay required gets very long. So what are the options ?

1. using a simple spst relay (or a spdt wired to gnd the speaker as often suggested), the load resistor has to be connected to the output all the time. Which means it has to deal with signal (up to 20vac sinewaves) and fault conditions (48vdc at absolute worst case if both the lm3886 and the output cap die).

2. using a spdt relay - to either connect the output to the load resistor (nc position) or to the speaker (no position) - means that the load resistor doesn't see signal conditions but still see fault conditions.

In any case, that means a 300R 10W resistor... Or I've to get more sophisticated and I'm not sure it's worth it.
__________________
All that is excessive is insignificant. Talleyrand
  Reply With Quote
Old 6th February 2018, 11:18 AM   #56
00940 is offline 00940  Belgium
diyAudio Member
 
00940's Avatar
 
Join Date: Sep 2003
Location: Liège
In the end, for the load resistor, I decided to go for 3 130r/2w resistors in serie, always connected. If things go completely wrong and 48v end up at the output, they should barely survive without setting anything on fire.

It also came to me that I could use the mute signal for the lm3886 mute pin, rather than relying on a rc filter.

Here is the current shape of the pcb. It is 9*5.5cm big.
Attached Images
File Type: png lmpcb.png (77.7 KB, 64 views)
File Type: png lmsch.png (13.2 KB, 63 views)
__________________
All that is excessive is insignificant. Talleyrand
  Reply With Quote
Old 6th February 2018, 11:21 AM   #57
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
The ratio of C7:C10 is not optimised.
__________________
regards Andrew T.
  Reply With Quote
Old 6th February 2018, 11:41 AM   #58
anti is offline anti  Slovakia
diyAudio Member
 
Join Date: Jun 2008
Wouldn't you have to connect the R2 from the Q1 emitter?
  Reply With Quote
Old 6th February 2018, 12:37 PM   #59
00940 is offline 00940  Belgium
diyAudio Member
 
00940's Avatar
 
Join Date: Sep 2003
Location: Liège
Andrew: concerned about very low frequency distortion ? You can drop the input cap to .22µ

anti: either way works and doesn't bring major advantages. That's simply copied over from the datasheet.
__________________
All that is excessive is insignificant. Talleyrand
  Reply With Quote

Reply


Single supply LM3886 layoutHide 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
LM3886 Layout - Single Supply north99 Chip Amps 6 7th January 2016 02:12 AM
LM3886 PA150 single sided layout request ramiro77 Chip Amps 4 20th February 2014 12:18 PM
My tiny single sided LM3886 PCB layout. :) Russ White Chip Amps 74 14th September 2008 06:53 PM
pa100 pcb single sided layout lm3886 dgaxeman Chip Amps 6 24th May 2006 01:45 AM


New To Site? Need Help?

All times are GMT. The time now is 05:20 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 70.00% / PHP 45.87%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki