Prev: PICAXE 08-M (12F683) Input buffering for ESD protection
Next: Big crowbar module for <4VDC?
From: Phil Allison on 17 Dec 2008 23:19

<morris.slutsky(a)gmail.com>
"Phil Allison" <
>
>
> > Schematic:
> >http://img213.imageshack.us/img213/2480/10wpowerstageym6.png
>
> > Resistors R1-R4 work very hard here. They set gain, linearity, input
> > impedance, output impedance, and bias. This thing idles along at
> > about 150 mA through Q3 and Q4, according to SPICE. A reasonably warm
> > class AB.
>
> > SPICE thinks that this is actually fairly linear.
>
> ** Shame how SPICE lets you operate devices well beyond their ratings and
> capabilities.
>
> The maximum collector voltage is being exceeded for the BOTH driver
> transistors as neither can tolerate 70 volts as your schem requires.
>
> Their puny current and power dissipation ratings will be exceeded in real
> operation too, since loudspeakers present quite a reactive load at some
> low
> audio frequencies.
>
> I strongly suggest you use the TIP31C /TIP32C pair in place.
>
> These are rated at 100 volts, 3 amps and 40 watts max & Hfe is also high
> at
> over 100 and fairly linear in the range up to 500mA.
>


70 Volts? I've got it using a 25 VAC transformer here, rail voltage
25-30V in operation.

** It can not be less than 35 volts DC.

Sorry, I misread the schem thinking it was a voltage doubler.


Wouldn't that transistor pair make a fine output transistor pair?

** For low power only.


Bit big for just driving the output pair.


** You are clueless about the requirents for driver transistors.

They are in fact just adequate for the job and widely used as output
drivers.


NTE carries complementary Darlington's. Might look into those,
actually. Keep that parts count down!


** Do whatever you like - pal.



...... Phil


From: Phil Allison on 18 Dec 2008 00:20

<morris.slutsky(a)gmail.com
"Phil Allison"

>> I strongly suggest you use the TIP31C /TIP32C pair in place.
>
> > These are rated at 100 volts, 3 amps and 40 watts max & Hfe is also high
> > at
> > over 100 and fairly linear in the range up to 500mA.
>
> 70 Volts? I've got it using a 25 VAC transformer here, rail voltage
> 25-30V in operation.
>
> ** It can not be less than 35 volts DC.
>
> Sorry, I misread the schem thinking it was a voltage doubler.
>
> Wouldn't that transistor pair make a fine output transistor pair?
>
> ** For low power only.
>
> Bit big for just driving the output pair.
>
> ** You are clueless about the requirents for driver transistors.
>
> They are in fact just adequate for the job and widely used as output
> drivers.
>
> NTE carries complementary Darlington's. Might look into those,
> actually. Keep that parts count down!
>
> ** Do whatever you like - pal.
>

> Come on, Phil. Take it easy. I am happy to take your advice. That's
> why I'm here.


** Then do take it and stop changing your mind.



> It's just - this is a 10 Watt amplifier, after all.


** Needs to be well designed, like any other.


> Yeah, it looks
> like +- 17 Volts on the rails, 35, like you said.


** Ok.

> So at balls-out
> hard clip square wave I could be putting 17/8=2.2 Amps into the load.


** What if the speaker wires, the connecting lead or jack connection shorts
out ??

How much current flows then - eh ??

Your circuit will self destruct in milliseconds.


Only it will, almost certainly, drop the rail voltage when heavily
loaded. Especially with a real, resistive transformer as opposed to
an imaginary SPICE one. If it's a really, really reactive load, worst
possible speaker at the worst possible frequency, double that. Say 4
Amps. And maybe I decide, for some insane reason, to use a 4 ohm
speaker, now it's 8 Amps. Worst-case beta for a power transistor is
what, 20?


** Nope.

At Ic = 8 amps, a typical 2N3055 has a beta of only 10.


> So I'm getting 400 mA, max, through the driver transistor.

With the phase shift of a reactive load, that current may be simultaneous
with a fair fraction of the supply voltage.

Goes way over the max dissipation figure for your puny devices.

Bang.


>Due to the configuration of this circuit, there's not going to be much
>Vce at that current,

** See above.


>Yeah, I see
> your point, I guess that is pushing dissipation limits, but . . . just
> seems like a big jump that's all.

** Wrong.

So . . would you be kind enough as to let me know how you pick driver
transistors? Am I reasoning this incorrectly?


** You need to figure out the worst case scenario for the drivers.

Typically, this will be with the output shorted and the drivers passing
whatever current is needed to make the output devices reach the limit -
this limit may be because of fusing or the use of additional VI limiting
circuitry.

Then you estimate the driver power dissipation, resulting temp rise and also
check the SOA data for the driver.

If you use puny drivers, they WILL fail first under overload conditions -
and take the outputs with them.


> Heck maybe I should just make a Darlington from 2 TO-3 power
> transistors. That'd be safe, kinda.


** Many commercial high powered amplifiers use TO3 pack devices as
rivers - sometimes another transistor the same type as the output devices.

Then there will be a TO220 or TO126 pack device driving the combination
forming an output "triple".

Very rugged.



..... Phil





From: morris.slutsky on 19 Dec 2008 22:10
On Dec 19, 5:52 pm, Jim Thompson <To-Email-Use-The-Envelope-I...(a)My-
Web-Site.com> wrote:
> On Fri, 19 Dec 2008 14:17:05 -0800, JosephKK <quiettechb...(a)yahoo.com>
> wrote:
>
> [snip]
>
>
>
> >I saw it.  I thought about it.  SPICE probably would say it works, and
> >it even may work.  It looks seriously Muntz'd.  
> >What's that you say?  There are many stories about "madman Muntz";
> >basically he was an extremist of circuit minimalization.  And twisted
> >every circuit to lowest parts count / minimum cost.
>
> From my father's stories it appears that Muntz used selected tubes.
> They were nearly impossible to repair... any tube change required an
> IF realignment.
>
>                                         ...Jim Thompson
> --
> | James E.Thompson, P.E.                           |    mens     |
> | Analog Innovations, Inc.                         |     et      |
> | Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
> | Phoenix, Arizona  85048    Skype: Contacts Only  |             |
> | Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
> | E-mail Icon athttp://www.analog-innovations.com|    1962     |
>
>  I love to cook with wine     Sometimes I even put it in the food

That's right about the whole Muntzing thing. I'm already considering
other parts to uglify the design. Maybe a couple clamping Zeners to
prevent Vbeo breakdown, perhaps even a small ballast resistor in the
collector circuit of the driver transistors to limit current.

And yeah, I don't know if I really want integrated Darlingtons. I'd
rather sort through a bin of driver transistors in the hopes of
finding 2 matching hFe combinations at idle current. Would be better
than no matching :)

So yes, totally Muntzed.



From: JosephKK on 20 Dec 2008 00:16
On 16 Dec 2008 10:05:19 GMT, Jasen Betts <jasen(a)xnet.co.nz> wrote:

>On 2008-12-16, BobW <nimby_GIMME_SOME_SPAM(a)roadrunner.com> wrote:
>
>> That's pretty funny. I would have liked to have seen their faces.
>>
>> The wireless stuff would have to be a fairly sophisticated technology like
>> spread spectrum to avoid that type of interference. I haven't done any
>> checking, but I would bet that it's already available.
>
>it also preferably needs to be realtime, if there's a way to do spread-spectrum
>in realtime then yeah that'd work.

That is actually quite easy if you know what you are doing. It can be
done FHSS, DSSS or combined modes in real time. Not really any worse
than 100 feet of wire. And that is with stable and constrained
latency.

First  |  Prev  | 
Pages: 1 2 3 4
Prev: PICAXE 08-M (12F683) Input buffering for ESD protection
Next: Big crowbar module for <4VDC?