Hello guys.
I was looking at the LM3886 internal schematic and it came to my attention that the mute resistor interferes with the current inside the chip!
So, based in experience and knowledge, witch value do you guys think will suits best ?
In the schematic they use 1k resistor. (page 7 of the LM3886 schematic)
Regards
I was looking at the LM3886 internal schematic and it came to my attention that the mute resistor interferes with the current inside the chip!
So, based in experience and knowledge, witch value do you guys think will suits best ?
In the schematic they use 1k resistor. (page 7 of the LM3886 schematic)
Regards
Referring to the equivalent schematic on page 6 of National pdf documentation, I'm not sure if the 1k you're talking about is external or internal. Internal might be okay, if 1K were enough to turn the mute off. But if it were external, may be 1k is a safe value to ensure the muting function is off.
In case of carlos schematic, 10K is used. Higher R will draw less current but might not turn the mute off completely. I think, because the resistor draws current from the input bias (as is your concern), we will want to draw as small as possible current, but big enough (0.5mA from mute pin) to turn the mute off.
Now, who knows how this (drawing too much current than necessary) will affect the sound? I don't think it has so much effect on the sound so I don't think I need to calculate or experiment with other values. May be you want to calculate the exact current a 1K or a 10K draws from mute pin?
In case of carlos schematic, 10K is used. Higher R will draw less current but might not turn the mute off completely. I think, because the resistor draws current from the input bias (as is your concern), we will want to draw as small as possible current, but big enough (0.5mA from mute pin) to turn the mute off.
Now, who knows how this (drawing too much current than necessary) will affect the sound? I don't think it has so much effect on the sound so I don't think I need to calculate or experiment with other values. May be you want to calculate the exact current a 1K or a 10K draws from mute pin?
So.... can't see anything to be worried about.XELB said:
You'll have to check the datasheet. It's about current and is dependend of the the used supply voltage.XELB said:
10 k is OK but I have used a but more 33k becasue I wanted longer delay time with 100 uF.
As you can see from the simplified schema it might be an good idea to have this pin filtered. In other words: The softstart serves two purposes, Carlos.
This cap is maybe a bigger improvement than the snubber 
I can't see any reason NOT to have it. A real smooth startup is not wrong. Check my Gainclone at startup och powerdown.The pictures shows the supply voltage and the output signal. Can it be better?
Startup
http://web.comhem.se/~u33113170/gainclone/start.jpg
Power down
http://web.comhem.se/~u33113170/gainclone/shutdown.jpg
carlosfm said:
Attached, Page 7 of the LM3886 Schematic
As you can see, depending on the resistor value, the current inside the chip change! Or am I miss understanding the schematic ?
PS: I have search the forum and I could find any investigation about this particulary caracteristic.
Attachments
You miss only a bit . Remember that National shows only a simplified schematic and this is made only for understanding. I'm pretty sure that the bias voltage for this stage is in some way stabilized. One way to find out is to measure the current the IC consumes when you are changing this resistor. Since National doesn't mention this resistor especially much I wouldn't worry.
Read the datasheet carefully and go with thier recommendations. 10 kohms seems to be a value you can choose without thinking. I also recommend a cap, 100-470 uF.
. Remember that National shows only a simplified schematic and this is made only for understanding. I'm pretty sure that the bias voltage for this stage is in some way stabilized. One way to find out is to measure the current the IC consumes when you are changing this resistor. Since National doesn't mention this resistor especially much I wouldn't worry. Read the datasheet carefully and go with thier recommendations. 10 kohms seems to be a value you can choose without thinking. I also recommend a cap, 100-470 uF.
peranders said:As you can see from the simplified schema it might be an good idea to have this pin filtered. In other words: The softstart serves two purposes, Carlos.
P-A, I'm using a cap too, on my main amp, and I know it filters the PSU noise after the resistor.
I didn't include the cap on the schematic because it was not practical, and it doesn't seem important to me (it doesn't make difference, I tried), and here's why:
1. With or without the cap, the amp is absolutely silent and power on and off (with my PSUs).
2. My PSUs are high capacitance CRC and regulated (much less noise than yours).
3. I don't need a big delay at turn-on, and as I said, there's no noise at all.
peranders said:This cap is maybe a bigger improvement than the snubber
Maybe?
You mean you are not sure?
Maybe on your amp, not mine.
My implementation is based on the Carlosfm schematic.
I am using 10k but, I am willing to try different values and see how the LM3886 behaves.
I don’t know if Carlos have tried different values so, I will give it a try, measure it with AP and listen.
Something tells me that this resistor could influence the LM3886 performance
I am using 10k but, I am willing to try different values and see how the LM3886 behaves.
I don’t know if Carlos have tried different values so, I will give it a try, measure it with AP and listen.
Something tells me that this resistor could influence the LM3886 performance
because in this version there's a schematic in page 7, like I posted 
Well, as a matter of fact there is a simplified schematics on page 7 at October 2003 datasheet. Carlos, you probably have the one dated April 1995 which has graphs on page 7 and simpl. sch. at page 5.
But I think it is irrelevant, you both have the same simpl. sch. with 1k.
Another thing. Aren't we trying to improve something and check other meaningfull values??
But I think it is irrelevant, you both have the same simpl. sch. with 1k.
Another thing. Aren't we trying to improve something and check other meaningfull values??
peranders said:
I just don't like 'Gainclone' because that means a copy of the Gaincard. Which is a copy of the datasheet.
It is not what I'm doing, and your amp is not too.
Or is it?
Anyway, back to my point: the LM3886 alone is silent at power on, even without the mute cap.
You have an op-amp gainstage and a servo on your amp, so you did well in using it.
I think the Gainclone now has turned into an amp with LM3875 or LM3886.
The servo of my amp creates a minimal turn-on thump becasue it starts to work before the amp is working due to the built-in undervoltage protection of the LM3886. I have this mute capacitor as an option. Skip it if you don't like it.
The servo of my amp creates a minimal turn-on thump becasue it starts to work before the amp is working due to the built-in undervoltage protection of the LM3886. I have this mute capacitor as an option. Skip it if you don't like it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.