My "Mini Bookshelf" LM386 Speaker!

[FenderBender11]

Hello everyone,

Perhaps you recall my previous thread on my quest to make a decent portable LM386 amp. Well, I finished up everything earlier today, and I am very happy with the results.

I'm giving this one away as a gift to a close friend. It'll be hard for me to part with it, after many hours of work, but it's going to a good cause.

Specifications:

Amplifier Chip: LM386N-3 (National Semiconductor)
Gain: Default (20)
Power: 1x 9V Battery
Driver: 3 x 5" 8ohm Panasonic
Cabinet: 4.5" L x 4.5" W x 6.5" H

Amp Component Selection:
Electrolytics: Panasonic FM
Films: Wima MKP4
Resistors: Dale CMF55
Pot: Bourns PTD901
Jack: Neutrik 1/4"
Switch: NKK DPST

Onto some pictures:










I used 24AWG solid wire, which explains the rigidity of the wires, and how it looks a bit messy. I would have used stranded if I had it, but I don't have any thin stranded.






This genuine Panasonic speaker is a factory buyout from Parts Express. It is listed as $1.50 as of 7/7/2011. Shipping is going to cost you around $6, but still relatively cheap in the end.



That's about it. Really happy with this build.

[visioninscience]

i hate the 386 so... much, but that is actually a nice build.

sometimes those projects are the best, simple, cheap and complete. a 9.6v rc car battery might help too if you happened to stumble on a cheap one. having a bridged version too would be pretty funny.

[FenderBender11]

Haha...Well LM386 is a pretty robust chip deep down. It's not hi-fi really, but you can make it sound decent if you try. 99% of the LM386 projects out there are quick and dirty. There are many other factors that make a nice amp.

[visioninscience]

yes, i agree, and for this application it works fine.

[!]

Nice to see the finished project. How much current does it draw off the 9V at normal listening volume?

[FenderBender11]

Well "normal listening volume" is sort of subjective. What do you think is "normal"?

I think about 1/2-2/3 volume on this is about "normal". 2/3 volume is about the max that I find to be "comfortable". From 2/3 volume to 100%, it will easily cover a conversation you are trying to have.

Here's my measurements. Using an Amprobe meter, so I THINK it should be accurate. The measurements varied A LOT, depending on the song I was playing, so I'll give sort of a range. I determined that playing a lot of techno seriously increases power consumption, and I'm not sure exactly sure why. "All I Ever Wanted" by Basshunter, on 3/4 volume, I got around 70mA draw. Meanwhile, playing "Little Wing" by Jimi Hendrix, I got around 40mA average....So again, it's really hard to say.

1/4 volume: 7 - 10mA
1/2 volume: 15 - 22mA
2/3 volume: 30-60mA
3/4 volume: 35-70mA
100% volume: 40 - 80mA

I guess "normal volume" could be anywhere between 15 and 60mA. Quite a range...

So, it's really hard for me to say. It's very inconsistent, but that's the best I can do. Battery was measuring at about 8V if that matters. Source is my computer via iTunes. Source is on maximum volume. Volume adjustment was done via potentiometer.

If you listen to mostly classic rock, and that type of music, it seems that you will get much better battery life than if you listen to some more "club-like" music present in today's music scene. Maybe I'm wrong, but there is a bit of truth to it, going by my results.

[visioninscience]

bass uses a lot more power(club music), will cause it to clip easily at this power range.

i agree that's a very subjective question, but

the 386-3 at full rated output (0.5 watt) running from a 9v source draws 78ma, assuming an efficiency of 60 percent.

offhand guess 50ma at normal output, of course it always varies, but we need a guesstimate. a duracell 9v alkaline battery is rated at 565mah, so in 11 hours of total use it's essentially dead (550ma taken).

the 50ma loading of the 9v battery is probably within it's normal intended range so official guess is,

11 to 12 hours with consumer 9v alkaline batteries.

[sreten]

Hi,

If you want it to sound far better for peanuts read on :

Google "baffle step compensation", understand it, work out the baffle step
frequency. Modify the power amps feedback loop to have an inverse function.
This will make a huge difference in the apparent richness and accuracy.

rgds, sreten.

[!]

It seems that the crazybots found this topic, lol.

Ignore them. The project is done and met the goal. Further time is just wasted.

[FenderBender11]

11 hours isnt too bad. The LM386 can run down to a very low voltage, so battery life will be good, however sound quality might get a little messy towards the end of it's life.

LM386N-3 has a 0.75W output at 9V with a 8ohm speaker...which is what I have...so battery life might be a bit less, (if at max volume).

A rechargeable 9V would probably be ideal, and though it probably wont last as long, it will save you money in the end.

Uh yeah no offense guys. It's an LM386. xD. I tried my best with the National schematic. I used above average quality components and did my best with layout. It sounds good to me! I already gave it away by the way.

Off topic: Speaking of 9Vs, I was in Lowe's today and they have a crapton of batteries at the front. Just being curious, I wanted to see where Energizer's 9Vs were made. And I'll be darned, they're actually made in the US! What a shock.