LM3886 questions

[sandyhooker]

I am building a 3886 amp on BrianGT's boards and have 220uf caps but no 100uf units. I dont see any problem but may be overlooking some factor. Also have 22uf but no 47uf . Would like to use what I have . Need your advice!!!

Thank you.

[ratza]

Go for 220uF. Actually, try them all and keep the ones you liek.

[AndrewT]

which circuit are you using?
I can only see the 1500uF on the supply pins.
There is a 4u7F listed in the BOM,
Where are the 100uF and 47uF?

[sandyhooker]

Originally posted by AndrewT
which circuit are you using?
I can only see the 1500uF on the supply pins.
There is a 4u7F listed in the BOM,
Where are the 100uF and 47uF?

Go to www.chipamp.com and find lm3886 kit. The assembly manual illustrates the schematic. As far as I know it's his only offering for that chip!

[AndrewT]

22uF for the feedback cap is about ten times too small.
Try the 220uF for all (100uF and 47uF: Cs, Cm, Ci).
Add 4u7F between the input pot and the 1k0.
Add 330pF to 1nF in parallel to R2 as RF filter.

[sandyhooker]

Thank you for your comments. In the construction notes there is consideration for two versions ,one with feedback,one without. The stated preference is the latter ,no feedback. If feedback is used ,the cap called for is 47uf. According to your notes ,that also would be small. I must admit it seems small to me. So it isnt an issue if we go with the reccomended config. What do you think?? I have very (none) limited experience with this type amp.

JTP

[AndrewT]

Originally posted by sandyhooker
In the construction notes there is consideration for two versions ,one with feedback,one without.

I think you have misunderstood what it is telling you.
You can't operate the chipamp without feedback, the gain is too high and the frequency response does not cover the audio band.

[sandyhooker]

You are correct...I did misunderstood the notes. Only the capacitor is optional not the feedback loop. Apparently the version sans capacitor "sounds" better.So the question of the capacitor is moot.???

Still your comment on the value of that capacitor is mysterious as the stipulated value is 47uf.Care to comment further???

I certainly appreciate your help.. it is comforting to know good help is only a keyboard away.

[panson_hk]

R3 and Ci set a low-freq 3dB point of the Fig. 1.2 amp circuit. Using the values given on the figure, you have it at 5 Hz. Though 5 Hz seems low enough, the corresponding phase shift at 20 Hz is 14 degree. It is probably too much. If you change the Ci to 220 uF, you will experience about 3 degree phase shift at 20 Hz.

[AndrewT]

Hi,
more importantly the input filters set the bandwidth of the amplifier.
That then ensures minimum AC voltage across the NFB cap.

If NFB is 47uF, there will be added distortion due to AC voltage across the cap and there is a risk of LF instability.
There is also the reduced bandwidth as Panson has said.