Hi Bimo,
Nice pcb layout. Stereo on single pcb. great work. Any particular reason why you have used 100uF capacitors throughout? Original apex design uses 10uF and 100uF capacitors. what is the effect on amplifier if we use 100uF
reg
prasi
You mean 100uF on feedback?
It determine the -3dB of feedback frequency:
f-3dB = 1/(2 * pi * 100uF * 1K) = 1,59Hz.
I choose lower f-3dB to get lower THD and phase shift at lower frequency.
May be only subtle change or just placebo
You should try your self.
I agree, getting the F-1dB to around 4Hz seems to be good for very extended bass and is lower than I can hear any further improvement.
Hi Bimo,
I was looking at the eagle file of your PCB (post no. 233 ) and came across two tracks on the bottom on layer named "bCream" connecting 15k resistor and Speaker out pad. what is the purpose of this this track or is it just an insulation layer and also is it absolutely necessary to keep this track. I know its been a long time since you created this PCB and you may not be remembering. If you can have a look and answer my question, that would be most helpful.
regards
prasi.
I was looking at the eagle file of your PCB (post no. 233 ) and came across two tracks on the bottom on layer named "bCream" connecting 15k resistor and Speaker out pad. what is the purpose of this this track or is it just an insulation layer and also is it absolutely necessary to keep this track. I know its been a long time since you created this PCB and you may not be remembering. If you can have a look and answer my question, that would be most helpful.
regards
prasi.
Good afternoon! Need help!!!
Have a question about the LM3886. Never encountered anyone with such a phenomenon: The amplifier works quite efficiently, sound happy. But at times the volume slightly increases, after a while returns to the previous level. I suspected electrolytic capacitors, changed them, but deffekt not left. Perhaps LM3886 defective? What else could be wrong?
Have a question about the LM3886. Never encountered anyone with such a phenomenon: The amplifier works quite efficiently, sound happy. But at times the volume slightly increases, after a while returns to the previous level. I suspected electrolytic capacitors, changed them, but deffekt not left. Perhaps LM3886 defective? What else could be wrong?
Thank you Andrew! But no. Rewire all possible contacts, cold solder is left exact. Transformer like this how to 200W 24-0-24, in the process of determination of the cause of what is happening changed electrolyte capacitors 3 times. Elna - Black Gate - Film. Set and large denominations (C4-100uf instead of 47uf, and vice versa) the way the sound film won even in BG. The amplifier is not heated. I'm used to the tube amps, and for me this new experience))) It sounds quite exactly an hour, but at some point for 1-2 seconds the volume increases (weakly, but you can hear it) so I think that something with caps. But now I do not know what to think. With all caps is the same.
Hi Mark! I'll try to take a picture later amplifier (but there topology like sardines in a tin) Schematics most simple - a XY kit. The picture below. The values of the resistors and capacitors are set in accordance with the diagram. Preamplifier is Audio Research SP-10. Due to the this preamplifier (4 X ECC88; 26dB (± .25 dB 5Hz to 30kHz), I had to put a divider resistors at the input of the amplifier (50K / 1K5)
Horn speakers Altec "Voice of the Theatre of the A-5" in which compression and paper drivers -16 ohms.
Yes! I must say that the sound that appears for 1-2 seconds seems not only louder, but also better quality. More open. That's interesting. Mark, if you do not complicate please write more about feedback loop. I need to understand what's wrong. And I'm not a great expert in this. What to look for?
[/URL]
[/URL]
[/URL]
Horn speakers Altec "Voice of the Theatre of the A-5" in which compression and paper drivers -16 ohms.
Yes! I must say that the sound that appears for 1-2 seconds seems not only louder, but also better quality. More open. That's interesting. Mark, if you do not complicate please write more about feedback loop. I need to understand what's wrong. And I'm not a great expert in this. What to look for?
Last edited:
Hi vsinyagin,
Take care, this schematic is wrong, I said that cause it's mine :
http://www.diyaudio.com/forums/chip-amps/247535-lm3886-kit-am.html
C6 must be in // with R5.
I propose to change C1 and C2 (470 to 1000 uF, with 0,1 uF in //) for better bass and midbass.
C4 seems too small, maybe 100 to 220 uF..
Phil.
Take care, this schematic is wrong, I said that cause it's mine :
http://www.diyaudio.com/forums/chip-amps/247535-lm3886-kit-am.html
C6 must be in // with R5.
I propose to change C1 and C2 (470 to 1000 uF, with 0,1 uF in //) for better bass and midbass.
C4 seems too small, maybe 100 to 220 uF..
Phil.
Hi Phil! Thank You! I guessed about the C6 and P5))) I tried to put 470uf on the position of the C-1, 2. Bass becomes larger, but midrange scores. In the end I settled on a film capacitor as the C-1, 2, and 3 (100uf) and C-4 paper in oil at 50uf. Before that was Elna 50uf 63v, Elna 100uf 50v, then Black Gate 47uf 50v and paper in oil won by the sound. As in the film instead of electrolytes 1 and 2.
Try to let go of the high-end aspirations. The more you understand how an amplifier works the less you understand why these expensive parts are used.
The feedback loop is formed by R3 and R4 (and C4 for DC).
The gain is 1+R3/R4 = 1+22k/1k = 23
This means that the signal at the output is 23 times greater than at the input. If either R3 or R4 change then the gain will change and so also the output.
One other cause could be the thermal protection that is build into the LM3886, but I have never heard it switch on and off in the way you are describing.
What I would do is use the parts that were intended for this PCB, try upgrades step by step and understand the upgrades.
You could test the amp without the pre-amp. The pre has a gain of 20 where a gain of 1 should be enough.
The feedback loop is formed by R3 and R4 (and C4 for DC).
The gain is 1+R3/R4 = 1+22k/1k = 23
This means that the signal at the output is 23 times greater than at the input. If either R3 or R4 change then the gain will change and so also the output.
One other cause could be the thermal protection that is build into the LM3886, but I have never heard it switch on and off in the way you are describing.
What I would do is use the parts that were intended for this PCB, try upgrades step by step and understand the upgrades.
You could test the amp without the pre-amp. The pre has a gain of 20 where a gain of 1 should be enough.
Thank you Mark! Well, it is clear about the details. In fact hi-end circuit technique relates not, but rather to materials science.Materials and their properties. It's just a toy.
I had considerations regarding the thermal regime, but 3886 seems to be securely fastened to the cooler with thermal paste. And there is no heating. Relatively unobtrusive thermal fluctuations can not say ...
Making changes I test performance without preamp on the test speakers. But unfortunately they are very quiet and I did not hear the presence / absence of differences have become the cause of my posts.
Initially, the scheme were electrolytes and resistors recommended by the kit manufacturer denominations. The only change is the C4 instead of 22uf now set 47uf. As a result, changes were made types of capacitors. But ratings have remained the same, that are recommended.
Mark, whether I understand that by using a combination of resistors R3 and R4 can be made amplifier volume is half smaller without the help of a divider at the input? The fact is for me amp too loud, and my acoustics quite normally gets a tube amp, where max is 2.5W per channel. If this amp will output 25W, I think this will be enough. But whether it is necessary in this case to calculate the face value of C4? If yes, under what formula?
I had considerations regarding the thermal regime, but 3886 seems to be securely fastened to the cooler with thermal paste. And there is no heating. Relatively unobtrusive thermal fluctuations can not say ...
Making changes I test performance without preamp on the test speakers. But unfortunately they are very quiet and I did not hear the presence / absence of differences have become the cause of my posts.
Initially, the scheme were electrolytes and resistors recommended by the kit manufacturer denominations. The only change is the C4 instead of 22uf now set 47uf. As a result, changes were made types of capacitors. But ratings have remained the same, that are recommended.
Mark, whether I understand that by using a combination of resistors R3 and R4 can be made amplifier volume is half smaller without the help of a divider at the input? The fact is for me amp too loud, and my acoustics quite normally gets a tube amp, where max is 2.5W per channel. If this amp will output 25W, I think this will be enough. But whether it is necessary in this case to calculate the face value of C4? If yes, under what formula?
Last edited: