19-05-2011, 05:56 PM
That's as clear as it could be, Mark. My LM317 data says it's stable with no external capacitors but 'recommends' them.
I tend to work to pretty low standards and if it works that's all I'm bothered. I dare say if I had an oscilloscope and could SEE how rough things are I'd take more care. That said my standards are raising as I get older as it gets more about the pleasure of making something than just wanting the finished item.
Some things just don't seem worth worrying about. For instance, I've never gone in for fancy speaker cable. I understand that the cheap stuff works like a capacitor and thereby blocks the high frequencies but on the sort of equipment I use there's probably not much up there anyway and it could even make things sound worse. Apart from that I can't help but wonder how much the benefits would be cancelled by room accoustics and even if my hearing would be sharp enough to notice.
As for de-coupling on the LM386 I've looked at a couple of manufacturer's data and not found it mentioned. Most of their circuits show a 10Ω resistor and a 0.05μF capacitor between the output pin and the speaker de-coupler. Fitting these just seemed to cause hideous oscillation so I left them off as some of the 'project' circuits on the 'net didn't have them.
I certainly will do further experiments as for what low voltage capacitors cost there's no point NOT fitting them, unless space is a major issue. My breadboard is coming from Hong Kong so it'll be a while but this allows time for more research and planning.
I've had that amplifier on a lot again today and still can't get over how good it sounds so will be very interested to see if it can be even better.
The other thing I arguably should fit is a heat-sink on the LM317 but it doesn't feel at all warm. My usual method is to think that if it's too hot to grip comfortably it needs a heat-sink or a bigger one if it's already got one. I first discovered these chips about fifteen years ago and have never blown one yet, though don't really know how!
Thanks for the reply.
- Joe
I tend to work to pretty low standards and if it works that's all I'm bothered. I dare say if I had an oscilloscope and could SEE how rough things are I'd take more care. That said my standards are raising as I get older as it gets more about the pleasure of making something than just wanting the finished item.
Some things just don't seem worth worrying about. For instance, I've never gone in for fancy speaker cable. I understand that the cheap stuff works like a capacitor and thereby blocks the high frequencies but on the sort of equipment I use there's probably not much up there anyway and it could even make things sound worse. Apart from that I can't help but wonder how much the benefits would be cancelled by room accoustics and even if my hearing would be sharp enough to notice.
As for de-coupling on the LM386 I've looked at a couple of manufacturer's data and not found it mentioned. Most of their circuits show a 10Ω resistor and a 0.05μF capacitor between the output pin and the speaker de-coupler. Fitting these just seemed to cause hideous oscillation so I left them off as some of the 'project' circuits on the 'net didn't have them.
I certainly will do further experiments as for what low voltage capacitors cost there's no point NOT fitting them, unless space is a major issue. My breadboard is coming from Hong Kong so it'll be a while but this allows time for more research and planning.
I've had that amplifier on a lot again today and still can't get over how good it sounds so will be very interested to see if it can be even better.
The other thing I arguably should fit is a heat-sink on the LM317 but it doesn't feel at all warm. My usual method is to think that if it's too hot to grip comfortably it needs a heat-sink or a bigger one if it's already got one. I first discovered these chips about fifteen years ago and have never blown one yet, though don't really know how!
Thanks for the reply.
- Joe
