Hey!
I've an issue with my LM1876, the gain is less or equal to one, I've been searching around for 2 days now and I just can't find the problem!
Mute and Standby pins are connected to ground and I'm only showing channel A in my schematic. Standby and Mute were left floating before and I had the same problem. I did change the values for feedback to set the gain higher but no matter what I do the output is low.
I did put in all the components shown in "Auxiliary Amplifier Application Circuit" in the datasheet except the 10R and 0.7 uf inductor and still had the same low gain. Am I missing something?
Supply is +-18 from a toroidal transformer (170VA) with CT and signal is coming from a phone if that matters.
I've an issue with my LM1876, the gain is less or equal to one, I've been searching around for 2 days now and I just can't find the problem!
Mute and Standby pins are connected to ground and I'm only showing channel A in my schematic. Standby and Mute were left floating before and I had the same problem. I did change the values for feedback to set the gain higher but no matter what I do the output is low.
I did put in all the components shown in "Auxiliary Amplifier Application Circuit" in the datasheet except the 10R and 0.7 uf inductor and still had the same low gain. Am I missing something?
Supply is +-18 from a toroidal transformer (170VA) with CT and signal is coming from a phone if that matters.
The value of C6 is much too low if this amplifier is to amplify bass. The 10 ohm resistors in the supply lines will reduce the maximum output level a lot.
According to figure 1 of https://www.ti.com/lit/ds/symlink/lm1876.pdf and your schematic, you have feedback from Out B to - In A. Mistake in the drawing or is it really connected like that?
Mistake in the drawing sorry, PIN3 is connected to pin7 just like figure 1. I'll try to connect channel B and see if I still have the same problem.
Channel B behavior is the same as channel a.
I can get the same lvl of output if I connect the speaker directly to my phones audio jack!
Channel B behavior is the same as channel a.
I can get the same lvl of output if I connect the speaker directly to my phones audio jack!
Last edited:
Have you measured the resistor values to make sure they are as marked or as you think they are marked?
It is really just a power opamp and it has to obey the rules regarding gain and the feedback loop values. Also make sure the feedback return (1k and the cap) really does have continuity to ground.
It is really just a power opamp and it has to obey the rules regarding gain and the feedback loop values. Also make sure the feedback return (1k and the cap) really does have continuity to ground.
To what value? You need something like a 22uF or 47uF in that location.
In addition, c1 is useless as chip is powered with dual supplies.
stdby and mute pin should be open.
stdby and mute pin should be open.
I changed it to 100uf bipolar. And took it out completely to see if it helps. And it didn't!
All resistors are measured before and after placing.
I think the problem might be the chip itself or bad connections because I'm using a breadboard atm.
I'll solder down the chip after work. I got some prototype boards.
Chip is bought from Mouser so it's not a fake chip.
100uF is fine. Removing it would give a gain of 1 provided the amp remained stable.
This has to obey rules, the gain is the 20k divided by the 1k which is 20 plus 1. So it should have a voltage amplification of '21'. If it is not doing that then you have to have a basic error somewhere.
Have you got a scope that you can look at the signals with?
This has to obey rules, the gain is the 20k divided by the 1k which is 20 plus 1. So it should have a voltage amplification of '21'. If it is not doing that then you have to have a basic error somewhere.
Have you got a scope that you can look at the signals with?
It doesn't get warmer than 30c.
Unfortunately I don't have a scop yet.
I'll change out all the caps and resistors and solder down the chip and components on a breadboard and will report the results here.
You can still test it without a scope if you are careful by applying a low voltage DC input. The input coupling cap and feedback cap must be linked out for this. You must NOT connect a speaker though.
Before you do that can you measure the DC voltage on the output pin and tell us what you get?
Before you do that can you measure the DC voltage on the output pin and tell us what you get?
I'll be home in 10 hrs. And will measure and post it here.
Thank you for your help.
If I recall correctly I was reading 3v DC at output before the cap.
Multimeter probe was grounded.
Thank you for your help.
If I recall correctly I was reading 3v DC at output before the cap.
Multimeter probe was grounded.
OK 🙂
And if you do want to test at DC then rig up a potential divider with a couple of high value resistors in series such as a 10k and a 100 ohm and place across either 18 volt rail. The 100 ohm goes at the ground end. That would give about 180mv DC at their junction which you can then apply to the input and see if it is amplified to around 3.8 volts (for a gain of 21). If you put it across the negative rail the readings will be negative.
And if you do want to test at DC then rig up a potential divider with a couple of high value resistors in series such as a 10k and a 100 ohm and place across either 18 volt rail. The 100 ohm goes at the ground end. That would give about 180mv DC at their junction which you can then apply to the input and see if it is amplified to around 3.8 volts (for a gain of 21). If you put it across the negative rail the readings will be negative.
That shows a problem. It should be close to zero.
There is always a possibility that such a result could be because the circuit is oscillating but that really needs a scope to check.
The circuit you posted is very basic and more of a data sheet test example, Normally we would need a Zobel network on the output and perhaps a very small cap (say 15pF) across the 20k to make it more stable.
Have you tried with 4 ohm speaker ?
There is not much to change with chip amp.
Easiest way to make well working amp is to make everything with separate components.
Chip may be ill or fake china chip.
There is not much to change with chip amp.
Easiest way to make well working amp is to make everything with separate components.
Chip may be ill or fake china chip.
Last edited:
The chip is most likely damaged by me or was faulty. I'm giving up and ordering some new chips now.
I feed 180mv to the input and all I got out was 100mv.
Thank you all for your time.
We can go a bit further than that...
Next you measure on the IC pins themselves. If you have a measurable difference between the two input pins and yet the output is only a volt or two then the chip is definitely more than highly suspect.
So for example if you measure 180mv on pin 8 then you should see the same on pin 7. So whatever is on pin 8 must also appear on pin 7. Any difference between the two pins should send the output hard to one or other of the rails,
Also (and I don't think this has been mentioned), the first rule of fault finding is to confirm the supplies are correct and so that means measuring on the pins themselves to make sure they are getting there.
Please redraw and post the actual circuit you built.
The only schematic shown is incomplete and wrong.
No need for a software drawing: black ink on paper is fine.
At the moment we are guessing too much.
The only schematic shown is incomplete and wrong.
No need for a software drawing: black ink on paper is fine.
At the moment we are guessing too much.