LM4780 problem - need Help - diyAudio
Go Back   Home > Forums > Amplifiers > Chip Amps

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
Reply
 
Thread Tools Search this Thread
Old 12th January 2009, 02:50 AM   #1
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
Default LM4780 problem - need Help

When building PCBs for my 4 channel amp based on parallel configuration of LM4780 chips and Carlosfm’ NI LM3886 amp version 4 schematic I have encountered a circuit problem related with mentioned schematic and placement of additional components needed to configure the amp to function in parallel configuration. These are 0.1 Ohm output resistors, connected between each other. So, when I build a PCBs and soldered needed components I measure resistance with my DMM between Output path and -IN path of PCB at chip pinout. This is where gain setting resistors are located (R3 in that schematic (Rf - according to National schematics). Than I found that resistance was 1,045kOhm, when needed value is 2KOhm. This is because of gain setting resistors are connected between each other via feedback resistors (R2 (Ri - according to National schematics) which are also connected between each other through Ground (because of without feedback capacitors). R2 (Ri - according to National schematics) resistors show 98Ohm, when needed value is 100Ohm.
In other words, around every gain setting resistor (R3(Rf) there is the circuit which is formed by feedback (R2(Ri) and output (Rout) resistors and change the gain setting resistance from 2KOhm to 1.045kOhm. The same is with feedback resistors (R2(Ri) - from 100Ohm to 98Ohm. When there are feedback capacitors are present (Ci) this is not happening.
Therefore I guess this configuration is unacceptable for chips. However, the situation would be different with feedback capacitors, but they must be very big ones what is unacceptable too.
What do you think guys? Am I wrong about the problem and all is normal and correct. If I'm not wrong than is this possible to build the parallel configuration based on LM4780 chip without feedback capacitors?
Many thanks for your help.
Attached Images
File Type: png cfm ni lm3886_v4.png (12.2 KB, 652 views)
  Reply With Quote
Old 12th January 2009, 06:06 PM   #2
diyAudio Member
 
Join Date: May 2008
Default Re: LM4780 problem - need Help

Quote:
Originally posted by AndriyOL
What do you think guys? Am I wrong about the problem and all is normal and correct.
All is normal and correct. You are getting the wrong values, because the way you apply voltage during your measurements is different from the way voltage is applied during operation.
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 12th January 2009, 07:22 PM   #3
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
Thanks for your answer.
My way of measurement is not wrong because what is important this is the impedance between output and -IN pins of IC I suppose. If I would place a feedback cap the impedance will be different and right.
  Reply With Quote
Old 12th January 2009, 09:17 PM   #4
diyAudio Member
 
Join Date: May 2008
Quote:
Originally posted by AndriyOL
My way of measurement is not wrong
No, it is not wrong. It is only meaningless.

When you measure a resistance, you put a kown resistor in series, put a voltage across both and measure the voltage drop across the unknown resistance. In your first post you pointed out that you are also measuring everything that is connected in parallel to the resistor you are measuring, thus do not get the correct value. You got that absolutely right.

When the amplifier operates however, the voltage source is not across each resistor, but across the output terminals. You could simulate that by putting a low voltage across the speaker terminals (amplifier off and speaker disconnected of course) and measure the voltage drop at the inverting input. You will see that it is proportional to the resistors, even without Ci.
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 12th January 2009, 11:33 PM   #5
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
No, there is a circuit and output resistors, gain setting resistors and feedback resistors are connected in series which form a circuit. It can be seen at parallel schematic of LM4780 datasheet when imaging that feedback caps are omitted and ground of feedback resistors are connected together. Applied voltage goes where the lowest resistance is. When feedback cap is present in the amp the lowest resistance is gain setting resistor (2K) because there is no circuit. When it omitted the lowest resistance is other side of the circuit or other part where gain setting resistor is located.
BTW, in parallel configuration of LM4780 without Ci caps inverting inputs are shorted through feedback resistors.
  Reply With Quote
Old 13th January 2009, 06:33 PM   #6
diyAudio Member
 
Join Date: May 2008
Quote:
Originally posted by AndriyOL
No, there is a circuit and output resistors, gain setting resistors and feedback resistors are connected in series which form a circuit.
Once your power supply is connected those will be different circuits in parallel. Ground is the return path for all of them.

Quote:
Originally posted by AndriyOL
It can be seen at parallel schematic of LM4780 datasheet when imaging that feedback caps are omitted and ground of feedback resistors are connected together.
Then the current flows from the output through the resistors into the ground and back to the power supply.

Quote:
Originally posted by AndriyOL
Applied voltage goes where the lowest resistance is.
The voltage is present everywhere. The current is divided according to where the lowest resistance is.

Quote:
Originally posted by AndriyOL
When feedback cap is present in the amp the lowest resistance is gain setting resistor (2K) because there is no circuit.
Yes, there is a circuit. Your multimeter works with DC, and a capacitor blocks DC. That is why your measurements are different with or without capacitor.

For the AC signal that will be present, when you play music through that amplifier the capacitors will only make a difference at low frequencies. At audio frequencies there will be none.

Quote:
Originally posted by AndriyOL
BTW, in parallel configuration of LM4780 without Ci caps inverting inputs are shorted through feedback resistors.
They are not shorted, because the ground is the return path for those signals. Don't get confused with the power supply. Imagine the split power supply as two single supplies. One part is from plus to ground and works for the positive half cycle, the other is from ground to minus and works for the negative half cycle. Then follow the signals again for each half cycle, i. e. as if only one of the two power supplies were present at each time. Or to make it more visual, cover V- and see, where the curent flows. Then uncover V- again, cover V+ and see, where the current flows.
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 13th January 2009, 09:15 PM   #7
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
Ok. You convinced me. Maybe I was wrong.
I'll try to power it and tell you results.

Many thanks for your help.
  Reply With Quote
Old 17th January 2009, 01:23 PM   #8
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
Hi, pacificblue.

The amp is not working. Chips are drawing much current (because toroid humming like never in normal use) and speakers reproduce only hum (like radio signal). Inputs disconnected. I can’t imagine where the problem is.
Could you help me?
  Reply With Quote
Old 17th January 2009, 07:03 PM   #9
diyAudio Member
 
Join Date: May 2008
Can you post the layout or good photos of everything?
Did you check the power supply without amplifier connected?
Is there hum with something connected to the input? And with an input signal?
Do you use a lightbulb tester? Or a limiting resistor in place of the fuses?
__________________
If you've always done it like that, then it's probably wrong. (Henry Ford)
  Reply With Quote
Old 17th January 2009, 11:46 PM   #10
diyAudio Member
 
Join Date: Jul 2008
Location: Lviv
Here are some photos. Sorry, poor quality I have camera only in cell phone. Layout is like in pdf file added but with other values of resistors which are according to carlosfm schematic. This is a rebuild version of amp. I already had it working under National parallel schematic. When I designed it according to National schematics without PCBs it worked pretty well, however with low frequency distortion on lower power (lower loud) levels in some cases (when there is low frequency buzz reproduction). Than I redesigned it in accordance with carlosfm LM3886 schematic with PCBs.
Till I try only two channels with carlosfm schematics and with one PSU. Have 4 channels, 2 PSUs.
PS rails of chips I make with cooper wire (1.3mm) – “- V” is on left “+V” is on right of the chip. Mute and Ground from the chip are also with wire. PCB is for “OutA” – ”-INA” – “+INA” – “-INB” – “+INB” – “OutB” traces only. Star ground also is with wires. Signal ground is on PCB which formed by Ri, Rin resistors, source ground, than direct wire without RG resistor to star ground. Before, when I placed RG resistor the distortion was becoming worse. So, I disconnected it.
All resistors are correct. I already checked and rechecked. All is fine and like it should be, I guess. I measured all and nothing is shorting.
PSU is working well with amp disconnected with 1 or 2mV difference between voltage rails. I have 300 or 400VA transformers custom wired by myself (only secondaries). I don’t exactly know how to measure VA ratings. I use 2A 250V fast blowing fuses. 1.6A were blowing up during start up from time to time. 2A value is Ok.
Maybe the problem is caused by 3.3uF MKP cap placed between voltage rails regardless to Ground. I never used it before. Or low input impedance (330Ohm) and amp is oscillating?
I don’t have a lightbulb tester or limiting resistor (I don’t know what they are). Fuses are Ok.
I didn’t try to connect to amp any input source yet as it’s not working. It worked well before without input source connected and any hiss or hum.
However, once I tried to use it without feedback caps – feedback resistors (Ri) directly to Signal ground, because one diyaudio member (dxvideo maybe) claimed that without Ci cap amp worked just fantastic.
As a result I had 0,5V on speakers’ outputs and distortion instead of low frequency reproduction, but without that hum what is now with no inputs connected.
My design of entire amp on photos is pretty fun.
Attached Images
File Type: jpg foto-0028.jpg (91.9 KB, 455 views)
  Reply With Quote

Reply


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
Problem with LM4780 Cyberzim Chip Amps 14 24th January 2010 06:40 PM
Problem with LM4780 Dielkius Chip Amps 8 13th January 2007 02:06 PM
LM4780 problem Lamicz Chip Amps 6 6th November 2005 02:50 AM
LM4780 problem Faber Chip Amps 8 16th August 2004 05:39 PM


New To Site? Need Help?

All times are GMT. The time now is 04:36 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2