Re: B40 LM386 and µB TDA2822 and that 470µF Cap

Tim Gorman
 

If you insert a mono plug into a TRS jack the ring terminal of the
jack is almost always shorted to ground (i.e. the sleeve). If you wire
the output of the 2822 to both the tip and ring terminals and then
insert a mono plug there simply isn't any doubt that you are probably
going to short the output of the 2822 to ground. Bad things will
happen.

There is no reason to wire the output of the 2822 to both the tip and
the ring terminals. If you have a stereo headset then wire both headset
leads to tip of the plug and only wire up the tip lead on the socket.
Then if someone inadvertently plugs in a mono plug you won't short the
output of the 2822 to ground.

I agree that the large charging current going into the 470uf cap could
stress the 2822, perhaps only causing a failure after a number of
on/off cycles. That's something that would be hard to diagnose!

I have modified my schematic to show using a 47uf cap being fed with a
1 amp fuse from the 2822. If I ever get a chance to actually finish up
my ubitx it will be interesting to see if the 1 amp fuse is sufficient.

tim ab0wr

On Thu, 01 Mar 2018 15:39:50 -0800
"Jerry Gaffke via Groups.Io" <jgaffke=yahoo.com@groups.io> wrote:

And subsequent replies have suggested this may not be sufficient.

Some seem to fry when plugging something into the headphone jack even
if there is no short to ground involved. Or when powering up the rig.
Our best guess now is that there are significant currents involved
when that 470uF cap suddenly has to charge up to Vcc/2 And that some
clone TDA2822's may be weaker than mainline manufacturers.

On Thu, Mar 1, 2018 at 03:31 pm, Christopher Miller wrote:


One of the first replies suggested not wiring the ring, and using a
mono to stereo adapter cable if stereo headphones are required. 
 

Join BITX20@groups.io to automatically receive all group messages.