I finally settled on this one:
Questions:
Thanks,
Michael
s.html#computer
I imagine most any small NPN transistor can be used in place of the
4401. If you substitute 1 ohm resistors for the 2.2 ohm, the DC bias current (through the TIP 31,32) will be higher since it will require more current to get the same voltage and negative feedback to stabilize the current. The diodes are used to produce a voltage just enough to barely turn on the TIP 31,32. As they turn on, the voltage rises between the two emitters due to the 4.4 ohm resistance. At some point the current will be stable since the rising voltage produces negative feedback and the transistors cannot further increase conduction. If you short out the 2.2 resistors, the current may runaway due to temperature rise. So, you might try the 1 ohm resistors, but keep watch on the bias current as temp increases. It may work on cold days and not so good on hot days.The output peak voltage might be around 5 volts, so using an 8 ohm speaker the resistance is 8+2.2 or 10.2 ohms, with peak current of
5/10 or 500mA. Average current would be (500mA * 0.7) / 2 =3D 175mA. Power would be 175mA^2 * 2.2 =3D 67 milliwatts. 1/4 watt resistors should be ok.-Bill
mps.html#computer
Wow, thanks for the explanation and power dissipation calcs! Much appreciated.
Now I gotta go shopping...
Michael
mps.html#computer
Say... if I use a 4 ohm speaker, are these calcs correct?
peak current I =3D V/R =3D (5V) / (4 + 2.2 ohm) =3D 830 mA avg current =3D 830 mA * 0.707 / 2 =3D 293 mA P =3D I^2 R =3D 0.19W, 1/4W resistors still ok ?
What's the reason for dividing by 2 in the second equation?
Thanks again!
Michael
s.html#computer
Undoubtedly yes. The National Semiconductor 2N2222 and 2N4401 are both the exact same silicon chip (process 19), just in different packages.
I never knew this. For some reason, I had always thought the '4401 was somehow more "robust" than the '2222, maybe a difference in Icmax, or power dissipation, albeit the dissipation could be accounted for by the package.
Now I'll go and look up both data sheets and em-bare-ass myself. ;-)
Ah! The 1N3904 is the wimpy one. :-)
Thanks! Rich
oamps.html#computer
of
aProbably ok at normal volume level. The figures are worst case and assume the amp is driven at max output with a continuous tone. With a
4 ohm speaker at 293mA and 6 volts, the heat from the TIP 31,32 will be .293 * 6 =3D 1.76 watts, so you may need small heat sinks on the transistors if you crank the thing up for too long. But should be ok at normal low volume. You divide by 2 since only one transistor and resistor are conducting at the same time. They take turns on the positive and negative half cycles, so that each is on 50% of the time.-Bill
No it will not be OK. The resistors provide thermal stability for the output transistors. I just roughed it out, but with 1 ohm resistors, if you pump out any decent sound you have a good chance of the whole thing going "splat". Use the 2.2 ohm resistors AND a small heatsink for the TIP31/TIP32
I've got a
:On Sat, 09 Oct 2010 10:31:54 -0700, whit3rd wrote: :> On Oct 7, 12:00 pm, Michael wrote: : :>> I finally settled on this :>> one:
:>> Questions: :>> 1. Would a 2N2222 work instead of the 2N4401? I've got a bunch of :>> quad-2s in my bin. :> :> Undoubtedly yes. The National Semiconductor 2N2222 and 2N4401 are both :> the exact same silicon chip (process 19), just in different packages. : :I never knew this. For some reason, I had always thought the '4401 was :somehow more "robust" than the '2222, maybe a difference in Icmax, or :power dissipation, albeit the dissipation could be accounted for by :the package. : :Now I'll go and look up both data sheets and em-bare-ass myself. ;-) : :Ah! The 1N3904 is the wimpy one. :-) : :Thanks! :Rich
... 2N3904 perhaps...
Gack! I normally proofread - guess I was in a hurry or something. :-)
Thanks, Rich
Have something to add? Share your thoughts — no account required.
Ask the community — no account required