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 gainclone DC servo
LM3886 gainclone DC servo
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 28th October 2017, 11:30 AM   #11
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
The middle schematic shows two stages:
stage 1 is a non-inverting gain of 2 stage, with DC block on input.
Any DC offset generated in the opamp will be multiplied by 2.
stage2 is an inverting with a gain of 1.
It has a DC block on the output.

When you take the output from stage 1 and from stage 2 the total gain becomes 4rimes the input voltage (+12dB)
The two outputs are not balanced.
You should have identical components on the outputs as seen from the loads.
__________________
regards Andrew T.
  Reply With Quote
Old 30th October 2017, 08:20 AM   #12
belyakove is offline belyakove  Netherlands
diyAudio Member
 
Join Date: Jan 2017
Location: Amsterdam
Quote:
Originally Posted by AndrewT View Post
I don't think those chipamp schematics are correct.
take away the DC servo loop.
What is left is a 22k feeding back into the -IN node.
The feedback lower leg impedance/resistor is missing?
As shown you have a gain of 1 Follower that will be unstable.
Quote:
Originally Posted by AndrewT View Post
The middle schematic shows two stages:
stage 1 is a non-inverting gain of 2 stage, with DC block on input.
Any DC offset generated in the opamp will be multiplied by 2.
stage2 is an inverting with a gain of 1.
It has a DC block on the output.

When you take the output from stage 1 and from stage 2 the total gain becomes 4rimes the input voltage (+12dB)
The two outputs are not balanced.
You should have identical components on the outputs as seen from the loads.
Indeed these schematics seem odd. I wonder if actual PCB is correct. Parts in balanced line driver can be altered to have equal gains (of 1 I guess) on both lines. But if feedback lower leg resistor is missing from LM3886, then it makes this PCB just garbage.

Will check it after PCBs will arrive.

Also I was going to get rid of DC block capacitors both in input and output of balanced line driver.
  Reply With Quote
Old 30th October 2017, 09:07 AM   #13
belyakove is offline belyakove  Netherlands
diyAudio Member
 
Join Date: Jan 2017
Location: Amsterdam
I’ve checked BOM list supplied together with schematics and it includes 6 1k resistors that cannot be seen on schematics. Seems that this is missing feedback resistors so I hope that actual PCB is ok.
I also have another question. Why DC servo opamps on schematic 1 and schematic 3 are in different polarity? These blocks seem identical even though one of them drives “+” signal and another one drives “-“ signal so why dc servo is in different polarity? I thought all blocks have to be identical since they are driver by the balanced line.

Sorry, I’m just a hobbyist so I might ask stupid questions.
  Reply With Quote
Old 30th October 2017, 09:14 AM   #14
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by belyakove View Post
..................Why DC servo opamps on schematic 1 and schematic 3 are in different polarity? These blocks seem identical even though one of them drives + signal and another one drives - signal so why dc servo is in different polarity? I thought all blocks have to be identical since they are driver by the balanced line..................
Both chipamps are non-inverting.
Both take a feedback from output to -IN.
Both take the DC servo feed to the -IN.
The polarity (non-inverting) of the DC servos should be the same.
Check the PCB when it arrives.
__________________
regards Andrew T.
  Reply With Quote
Old 30th October 2017, 09:16 AM   #15
belyakove is offline belyakove  Netherlands
diyAudio Member
 
Join Date: Jan 2017
Location: Amsterdam
That’s what I thought. Thank you very much!
  Reply With Quote
Old 5th November 2017, 10:03 AM   #16
belyakove is offline belyakove  Netherlands
diyAudio Member
 
Join Date: Jan 2017
Location: Amsterdam
PCBs arrived yesterday finally so I went down them and have drawn the actual real schematics. It is in attachment. It appeared to be way better than the schematics supplied earlier.

So it does have additional feedback resistor on LM3886s so it's not unity gain, that's good.
Also all 6 blocks are the same including servo circuits so that's also good.
Also outputs of both lines of balanced line driver are the same so that's also good.

However I still have several different questions so I'll try to ask them all together.

1. Output of DC servo circuit has 100k resistor but AN-1192 for LM3886 says that this resistor should be 10 times larger than feedback resistor which is 22k so this means that it should be 220k. Is there a reason why it could be made smaller or it's better to stick to datasheet recommendations?

2. The same documentation (AN-1192) uses LF412 op amps for dc servo circuits and it seems that they fit for this purpose really well because they have very good dc performance. So I guess it would be good to use those op amps and values of surrounding elements according to TI's docs?

3. For balanced line driver does it make sense to use "audio" op amp like OPA2134? If I'll use JFET OPA2134 which has low dc offset can the balanced line driver output 2uF caps be omitted? Or maybe it's worth to use bipolar OPA1602 which has lower noise but keep those caps?

Minor:
4. I saw in some schematics that OS-CON capacitors are used for the power supply decoupling of op amps? Does it make sense to use those or it will be useless and it's better to stick to cheaper Panasonic FMs?

That's a lot of questions but I really appreciate is you guys can help me out with those. Thank you.
Attached Images
File Type: jpg IMG_6744.jpg (576.4 KB, 249 views)
File Type: jpg IMG_6743.jpg (919.0 KB, 241 views)
File Type: jpg IMG_6746.jpg (485.4 KB, 213 views)
File Type: jpg IMG_6747.jpg (625.5 KB, 205 views)
  Reply With Quote
Old 5th November 2017, 07:05 PM   #17
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
LM3886 gainclone DC servo
1/ Making it smaller increases the range over which the servo can operate. A good test is to measure the DC voltage on the output of the servo opamp. If it is getting near to the supply voltage then the servo is running out of range. Your basic circuit will have poor DC offset and relies on the servo to correct it... but that is the designers choice.

2/ A TL072 or TL082 is perfect for a servo, as is the LF412. They are all jelly bean opamps but ideally suited to this use. DC precision of any of the opamps is better than the servo demands.

3/ Yes, you should use good 'audio' opamps. The output caps could be omitted for the FET opamps but you must still AC couple the input.

4/ Wouldn't like to say on that. I suspect standard caps will be just fine and they are what most designers use. Its more important to get the basics right than worry over boutique or expensive parts.
  Reply With Quote
Old 6th November 2017, 03:51 AM   #18
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
LF412 (LF147, LF347) is a very old part number.

TL072 was specifically designed (after a couple years of process improvement) to meet very similar specs at much lower price. Lot of audio gear that jumped on the LF parts when new quietly switched to the TL0 parts. Today I have half a suspicion that a "LF412" bought today is really off the TL0 process, tested to LF specs (not much different) and marked-up for price.
  Reply With Quote
Old 6th November 2017, 09:16 AM   #19
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by belyakove View Post
.......................... I still have several different questions so I'll try to ask them all together.

1. Output of DC servo circuit has 100k resistor but AN-1192 for LM3886 says that this resistor should be 10 times larger than feedback resistor which is 22k so this means that it should be 220k. Is there a reason why it could be made smaller or it's better to stick to datasheet recommendations?
The two feedback resistors are 1k & 22k for a gain of 23times (+27.2dB)
The ratio of DC servo resistor (100k) to the lower leg resistor (1k) is 100times. This ensures the output of the DC servo is reduced to 1/100th part to be input into the -IN node. This also limits the amount of correction that can be applied. eg. for a servo fed from 12Vdc supplies the max output from the servo would be ~+-11Vdc. Divide by 100 and the maximum correction at the -IN node is ~+-110mVdc.
Quote:

2. The same documentation (AN-1192) uses LF412 op amps for dc servo circuits and it seems that they fit for this purpose really well because they have very good dc performance. So I guess it would be good to use those op amps and values of surrounding elements according to TI's docs?
The very old 411 & 412 had excellent DC specifications for their time.
They make excellent DC servo opamps. There are many modern opamps that match that DC performance and some exceed it by a big margin.
All of these modern versions are more expensive. If you need ultra low DC offset, then choose a better opamp. If the 411/412 gives good enough performance, then I don't see any point in spending extra money.
Quote:

3. For balanced line driver does it make sense to use "audio" op amp like OPA2134? If I'll use JFET OPA2134 which has low dc offset can the balanced line driver output 2uF caps be omitted? Or maybe it's worth to use bipolar OPA1602 which has lower noise but keep those caps?
It is my view that a DC block exists some where in the link from Source to Receiver.
You can choose to put the DC block in the Receiver or in the Source/s .
If you put in both, then at least install a bypass option in one, or other. This option allows you to choose the better capacitor as your DC blocker & Audio passer.
Remember that for balanced impedance, that all of these output components need to be matched. I think the 560r is too high, I would be looking at installing 100r or less. Maybe as low as 10r, if I know the opamp can perform properly with a value that low.
Quote:

Minor:
4. I saw in some schematics that OS-CON capacitors are used for the power supply decoupling of op amps? Does it make sense to use those or it will be useless and it's better to stick to cheaper Panasonic FMs?...........
Oscons are reputed to be good for digital signal circuits, where switching harmonics are into the GHz region. Does audio get much above MHz? These ultra high frequencies are handled by the ultra low inductance X7R or similar, not by the electrolytics.
__________________
regards Andrew T.

Last edited by AndrewT; 6th November 2017 at 09:29 AM.
  Reply With Quote
Old 6th November 2017, 09:38 AM   #20
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by Mooly View Post
...............
2/ A TL072 or TL082 is perfect for a servo, as is the LF412. They are all jelly bean opamps but ideally suited to this use. DC precision of any of the opamps is better than the servo demands....................
Quote:
Originally Posted by PRR View Post
LF412 (LF147, LF347) is a very old part number.

TL072 was specifically designed (after a couple years of process improvement) to meet very similar specs at much lower price. Lot of audio gear that jumped on the LF parts when new quietly switched to the TL0 parts. Today I have half a suspicion that a "LF412" bought today is really off the TL0 process, tested to LF specs (not much different) and marked-up for price.
The published specifications say otherwise.

I would expect the LF411 to typically have half the offset (or better) of a TL071
The TL071 does not give a maximum for deltaT offset drift, typ~18uV/C

Here's the data I have on file.
. . . . . Typ . . . . Max . . . Typ . . . Max
LF411 0.3mV . . 2mV . . 10uV/C . 20uV/C
TL071 . 3mV . . . 8mV . . 18uV/C . . ?
__________________
regards Andrew T.
  Reply With Quote

Reply


LM3886 gainclone DC servoHide 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 Servo Design Redshift187 Chip Amps 67 3rd September 2017 11:13 AM
LM3886, XO, Buffer and Servo jackinnj Chip Amps 3 30th October 2008 12:14 AM
Gainclone LM3886 DC servo peranders Chip Amps 10 24th March 2004 04:59 PM


New To Site? Need Help?

All times are GMT. The time now is 03:39 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki