Need help with 3875 kit and DC - diyAudio
Go Back   Home > Forums > Amplifiers > Chip Amps
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Chip Amps Amplifiers based on integrated circuits

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
Thread Tools Search this Thread
Old 1st March 2009, 12:06 AM   #1
diyAudio Member
Join Date: Jan 2009
Question Need help with 3875 kit and DC

Hi to all,

I know this is a very common issue and it has been discaused many times, but i seached all the treads i could find and so far i can't find the "cure"

I have assembled today a LM3875 kit from Audiosector and i had a problem with the psu,tried another way and worked, but now i have a 93mV DC at the output.

The kit is not in a case, i was planning to try out different psu set-ups (original psu board, regulated with higher or lower voltage, pot or just 220R resistor ect).

I have just a simple +30 0 -30 DC supply (using 2 transformers, 2 bridges and 2 10000uF caps) and the amp pcb.

I used a 10uF cap insted of the jumper in R1, the multimeter rading was still 92mV, but when when i connected a speaker i heard nothing (the DC "hum" was gone, too little current to move the voice coil maybe?).

I thought the problem had passed,i connected a test source, and then nothing. i tried putting a cap in the output (read it in a post), but nothing besides a high-tone sound from the speaker.

I could try all night on my own but my knowledge is limited and all this has started to take it's toll on the PCB (soldering-unsoldering stuff).

Any help would be appreciated...
Attached Images
File Type: jpg setup.jpg (32.2 KB, 345 views)
  Reply With Quote
Old 1st March 2009, 02:56 AM   #2
diyAudio Member
jackinnj's Avatar
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
The DC output you see at the output of any opamp is the result of amplification of the input DC voltage (or some other DC component on the input) -- remember that current flows FROM the pins of these opamps on steroids so you have a DC potential at the opamp inputs -- ergo hoc -- you have to carefully balance the value of the resistors on the non-inverting and inverting inputs OR use a servo.

Peter's boards are very carefully designed and not to blame.
  Reply With Quote
Old 1st March 2009, 07:05 AM   #3
diyAudio Member
Join Date: Jan 2009
Hi jackinnj,

I don't blame Peter's boards, i know that they are good, i searched a lot to find out witch gainclone kit i will buy.I 'm sorry if i gave that impression on my post.

I 'm just trying to find out what the problem is, i don't know that much about amp design.Yesterday night i finished the second board and that worked fine!The only diffence was that i installed the 220R resistor instead of the jumper, but beside that the boards are the same.
  Reply With Quote
Old 1st March 2009, 08:17 AM   #4
Nuuk is offline Nuuk  United Kingdom
diyAudio Member
Nuuk's Avatar
Join Date: Feb 2003
Location: Somerset, SW England
You may just be unlucky with the chip itself. 93mV isn't that high (I've had 80mV here).
The truth need not be veiled, for it veils itself from the eyes of the ignorant.
  Reply With Quote
Old 1st March 2009, 08:59 AM   #5
diyAudio Member
Join Date: Jan 2009
Maybe that is the case Nuuk. I'm testing with speakers now, I can't "hear" the dc, is that 90mV a hazard to the speakers?

The amp sounds great, I'm sad that the speakers are not ready yet to do a proper test (now I'm using some old technics 6ohms).
  Reply With Quote
Old 1st March 2009, 09:19 AM   #6
diyAudio Member
Join Date: May 2008
The DC you get at the output is the DC offset at the input that is amplified with the gain. If you have the NFB combination 22k/680R than your gain is 33,35 times. 93 mV at the output means that you have 2,8 mV at the input. Well below the 10 mA that are stated as typical value in the datasheet.

To get rid of that offset you can use several methods. The simplest is to use Ci like in the datasheet, i. e. a capacitor in series with Ri (or R3) that should lead to a similar roll-off frequency as the combination of Rin (or R2) and Cin (your 10 F). Since you chose Cin very big, the corresponding size for Ci will be hard to handle. Try 82 F or 100 F.

Other methods are to employ a DC servo or to add a potentiometer and two resistors (AN-31 page 6).

DC does not give hum.

Remove that "capacitor in the output".

Check, if all solder joints are working. Measure, if the signal reaches R1 and pin 7 of the LM3875. Measure pin to pin, without touching the solder joints.
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 1st March 2009, 09:53 AM   #7
diyAudio Member
Join Date: Jan 2009
Hi pacificblue,

I have allready get rid of that cap on the output.i 'm testing both boards now, one with peter's psu and no other caps (only those 1500uF on the amp pcb) using a pot and jumper as R1, and the second with mine psu (post1) no pot and 220R R1.

I will try to do DC and AC coupling later, i 'm reading theads on how the input cap (high pass filter) is changing the bass on the output.I have lots of small (1uF - 47uF) to try out.Thanks for the help.

Also, what do you think about heatsink? the small one i use for testing now is not getting too hot, do you think that the one on the right will cope with both channels, or do i need two of those?
Attached Images
File Type: jpg heatsink.jpg (99.7 KB, 262 views)
  Reply With Quote
Old 1st March 2009, 09:53 AM   #8
diyAudio Member
Join Date: Jan 2009
ps. solders are fine, i tested them yesterday
  Reply With Quote
Old 1st March 2009, 11:33 AM   #9
diyAudio Member
Join Date: Jan 2009
Pic missing from above post
Attached Images
File Type: jpg heatsink_resize.jpg (91.7 KB, 251 views)
  Reply With Quote
Old 1st March 2009, 12:04 PM   #10
sangram is offline sangram  India
diyAudio Member
Join Date: Sep 2002
Location: India
Originally posted by Nuuk
You may just be unlucky with the chip itself. 93mV isn't that high (I've had 80mV here).
I've had 130mV on Peter's 4780 boards, the same board gives 20mV when connected to a source with no offset.

Since the board does not use the Ci cap, it will amplify the DC by its gain factor. This can be solved with

1. a DC blocking cap (but be prepared for about 30mV even with this),

2. a much lower Rin (which makes the amp harder to drive) or

3. a Ci cap (this results in the lowest practical offset, but the cap will have an impact on the sonic signature of the amp)
  Reply With Quote


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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Similar Threads
Thread Thread Starter Forum Replies Last Post
3875 PCB: Suggestion please? xneakers Chip Amps 14 31st October 2006 01:26 AM
3875 GC Hum Crash&Burn Chip Amps 5 7th September 2005 10:12 PM
another buzzing 3875 GC... cameo Chip Amps 10 30th August 2005 10:34 PM
LM 3875 finished... dhole Chip Amps 0 24th February 2005 07:32 PM
New Amp: Hex-Fet or LM 3875 Higo Solid State 8 2nd July 2003 09:36 PM

New To Site? Need Help?

All times are GMT. The time now is 06:30 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2