An open source layout for LM3886?

The problem is their 3 boards minimum policy. Not really practical for stereo builds :/

I don't see that as a problem. Build two. Use the third as a coaster. Or build two, screw one up, build the third.

I often use OSH Park for prototype builds. Even if I only need one board, it's still lower cost than the overseas outfit I used to use. Granted, I didn't choose the absolute cheapest outfit as I did want good or at least decent quality as well.

Tom
 
Well to be fair, OSH Park are fairly competitive if you really want ENIG. Which imo isn't really required for such simple boards (things would be different with small smd).

Then, there is shipping to factor in, or rather the costs associated with customs. When I last ordered pcb for the lm3886, I was able to fit 10 amp pcbs, 10 dc protect and 5 PSU in two orders, both under the 22€ limit for tax free import. So the total cost really was 44€ for 5 stereo sets.

With OSH park, 3 psu boards start at 40€ or so, 6 DC protect are 20€ and 6 amps are 40€. So 100€ total. Add 20€ of flat rate custom presentation fee and 20€ of VAT and we are at 140€ for 3 stereo sets.
 
Someone pointed out to me that the revised BOM had C6 missing. Not sure what happened there, so apologies for that, but I've updated it now, please use this link
https://enjon.uk/wp-content/uploads/2020/09/Mouser-lm3886-updated-project-bom.pdf
Any real difference between X7S and X7R for C15, C16? I don't mind going for the cheaper option (X7S is almost a third the cost) but the R has slightly tighter C tolerance and potentially better decoupling performance? Or is it practically the same?

EDIT: C4 is listed twice, but with different part numbers. C5 and the second C4 are the same, but the first C4 is different from the other two.
 
Last edited:
Actually X7R has a lower temperature coefficient than X7S. Here's a good introduction: Capacitor Fundamentals: Part 8 – Dielectric Classifications.

In addition to the temperature coefficient, you also have to consider the voltage coefficient of the capacitor dielectric. X7R caps typically lose over 80 % of their rated capacitance (yes, -80%, i.e. they deliver only 20% of their rated capacitance) at their full rated voltage. I don't know how bad the voltage coefficient is for X7S, but you could answer that by looking at the manufacturer's data sheets (and/or simulation parameters).

Tom
 
Actually X7R has a lower temperature coefficient than X7S. Here's a good introduction: Capacitor Fundamentals: Part 8 – Dielectric Classifications.

In addition to the temperature coefficient, you also have to consider the voltage coefficient of the capacitor dielectric. X7R caps typically lose over 80 % of their rated capacitance (yes, -80%, i.e. they deliver only 20% of their rated capacitance) at their full rated voltage. I don't know how bad the voltage coefficient is for X7S, but you could answer that by looking at the manufacturer's data sheets (and/or simulation parameters).

Tom
Where are you getting that? R is +/- 15%, S is +/- 22%, and what I think you're referring to is V at +22%/-82%.

EDIT: But also I said that R had a slightly tighter tolerance than S... I don't think that's wrong.
 
Last edited:
Tolerance and temperature coefficient are two different things. Take a C0G, for example. You can have a 1 nF, ±5% with 0% temperature coefficient. The absolute value of the cap varies by ±5% (production tolerances) but it does not vary with temperature.

±15 % (R) is a lower temperature coefficient than ±22 % (S).

Then add the voltage coefficient.

Tom
 
Got it, thanks. At the end of the day it sounds like X7S isn't going to be much different from X7R, so I'll save the extra dollars.

Still curious about the BOM conflict on C4, though. For now, I'm going to go out on a limb and say that's an orphan line that should be deleted (comparing to post #129).
 
Last edited:
Hi all !,
Sorry for the disturbance. I seem to have a problem figuring the power amp schematic. I keep on getting dc on the output of the amplifier 19 V. i have eliminated R8 and the dc voltage disappears but the amplifier stay mute. With R 8 i am getting amplification but also DC output. Power supplay is a small transformer ( for testing porpoises only) that delivers after rectification an filtering 2x 24 V. Many thanks
 
dumb questions

Hi, some beginner doubts:
1. what is the board input impedance (seen by a preamplifier) with the suggested values?
2. My tube preamplifier has an output capacitor (1 uF followed by a 1M resistor to ground), the input cap on the amplifier board is still necessary?
3. if I use an input cap, how the series of the two (output and input) behave in terms of corner frequency? What value do you suggest?

Tks in advance for any replies
 
1. about 20K
2. if it is the only preamp or source that will be connected to it, not really. But I wouldn't omit it as you might need it in the future.
3. with a 1uf cap in serie with a 2.2uf cap, the -3dB point will be at 11hz and the -1dB point will be at about 20hz. I wouldn't worry too much about it and keep those default values.