another bizarre audio circuit

p...

th

sue.

ing

ur

You are the one who complains all the time. You may have personal preferences about the nature of the threads that get started here, and the responses that get posted, but they are only of interest to you.

You are welcome to demonstrate your preferences by choosing to get involved with particular threads and in your particular reactions to other responses, but your whining about the nature of those responses doesn't make the group a more attractive or rewarding environment.

In the meantime, I'll post a circuit when I've got a circuit worth posting. Posting a example - without comnponent values - of a circuit that has been used in millions, for decades, doesn't strike me as a profitable use of bandwidth, but that is a personal preference.

-- Bill Sloman, Nijmegen

It doesn't. For lowest noise figure at the input, you use a single silicon pnp transistor, which has a lower base resistance than npn. rb is effectively in series with the source. Follow that with an op amp typically. You can also parallel input transistors, but the advantage soon drops off.

I'll see if I can find the circuit for a 12 channel mixer input stage that I designed around 1974. The noise figure related to a 600 ohm source was around 3 or 4 dB, from what I remember, with thd < 0.05% . Noise figure is quite important for pa applications, where you also have a lot of power gain from input to speakers...

Regards,

Chris

...

ue.

ng

Waste of time. Unfortunately, it isn't going to change.

I largely confine myself to criticising the alleged facts that you post from time to time - you are an uncritical consumer of right-wing propaganda, with a depressing tendency to recycle it here. This isn't a personality defect - though the fact that you keep on doing it does reflect an unfortunate strain of insecure vanity - but rather reflects you failure to learn critical thinking during your tertiary education, presumably because you confined your attention to subjects that you understood to be immediately useful.

-- Bill Sloman, Nijmegen

What I did was spin a signal-level bipolar circuit into a bipolar-mosfet power amp of similar topology.

The resulting dynamics is very interesting. Well, not to you.

John

Criticism would have some content. You know, something having to do with the circuit. All you've done is whine.

Electronic design is all about control. Of signals.

But you probably meant some sort of personal control. How does posting a circuit, and opening it for discussion, suggest control? I thought discussing circuits is what s.e.d. is for.

You're just a crabby old git who won't discuss electronics.

John

Wild. Sort of a single-slope ADC and a PWM driver. I wonder what the sensitivity is like.

I bet you do a similar thing with a single tiny-logic schmitt gate. Vaguely a superregenerative idea, namely triggering along a slowly decaying exponential.

John

That's funny.

It brought up a memory of work today, we saw that weird food guy and some "chef" somewhere was frying up a whole pan full of rooster mop tops (the little frilly things on their heads).

I can't believe some of the stuff that guy eats.

amp...

un

with

with

pursue.

ign

s?

ucking

do

your

to

I haven't got a relevant application at the moment, so of course it isn't interesting to me.

The universe is full of potentially interesting things. Some of them are also interesting to other people. If you weren't quite so self- obsessed, you might not expect everybody else to be interested in what happens to have caught your fancy today.

-- Bill Sloman, Nijmegen

Pretty good stuff.

It will go way over Sloman's head.

Along with the hundred other things a boy can do with a 555.

So someone has used a 555 to make a less than impressive radio- receiver. Why would anybody be interested, if they hadn't fixated on the device early in their career and never moved on?

-- Bill Sloman, Nijmegen

It's the base resistance that's important? Not that I disagree. I've been testing these 'low noise' power supplies with cap multipliers on the output and the negative supply with the pnp always seems to have a bit less noise.

George H.

amp...

un

with

with

pursue.

ign

s?

ucking

do

your

to

Yeah, I didn't get the inductor part. Do I have to spice it? Or does it have to do with head phone dynamics.

Say, and what about using the postive rail of an opamp as an output? I never heard of that.

George H.

amp

ne amp...

fun

ot.

ng with

g with

o pursue.

ssign

ors?

clucking

ow

t.

e do

ut your

r to

OK,

e.

have I told you I'm a terrible speller? (positive)

Oh, sorry, oh guru.

You are right, that is what 90% of the rest of the world has done. Not moved on.

Or could it be that it is *you* that has the problem?

You are the one that is not impressive.

..

I don't see the MOSFET being all that much of a change. And as for the inductor pull-up, this just doesn't make sense for low wattage high impedance headphone loads. Your inductive reactance needs to be a good few integer multiples of the load impedance, making these things prohibitively large if not unobtainable for a headphone app- you would use far less iron/ ferrite by boost switching your supply to accommodate the output swing...guess that's why I've never seen the inductive pullup here.

On 3/3/2011 12:04 AM, John Larkin wrote:

Okay, I put some values to it. It looks like a nice circuit, I admit. Good gain, low distortion, reasonable input impedance. Mind you, I didn't try to optimize it. I did notice that the feedback took higher than expected resistance and I was a bit surprised that the emitter capacitor of the output stage made the response do a camel hump at the beginning if too high.

So, critique away. I might learn something.

Version 4 SHEET 1 880 680 WIRE 32 -496 -240 -496 WIRE 352 -496 32 -496 WIRE -240 -400 -240 -496 WIRE 32 -400 32 -496 WIRE 352 -400 352 -496 WIRE -240 -304 -240 -320 WIRE 352 -288 352 -320 WIRE 352 -288 160 -288 WIRE 528 -288 352 -288 WIRE 560 -288 528 -288 WIRE 352 -224 352 -288 WIRE 32 -176 32 -320 WIRE 288 -176 32 -176 WIRE 160 -96 160 -288 WIRE 160 32 160 -16 WIRE 32 80 32 -176 WIRE -320 128 -336 128 WIRE -272 128 -320 128 WIRE -112 128 -208 128 WIRE -32 128 -112 128 WIRE -336 208 -336 128 WIRE 32 208 32 176 WIRE 160 208 160 96 WIRE 160 208 32 208 WIRE -112 288 -112 128 WIRE 128 288 -112 288 WIRE 352 288 352 -128 WIRE 352 288 208 288 WIRE 448 288 352 288 WIRE -336 320 -336 288 WIRE 32 384 32 208 WIRE 352 384 352 288 WIRE 448 384 448 288 WIRE 32 480 32 464 WIRE 352 480 352 464 WIRE 448 480 448 448 FLAG -240 -304 0 FLAG 32 480 0 FLAG 352 480 0 FLAG 448 480 0 FLAG -336 320 0 FLAG -320 128 in FLAG 528 -288 out SYMBOL npn -32 80 R0 SYMATTR InstName Q1 SYMATTR Value 2N3904 SYMBOL npn 288 -224 R0 SYMATTR InstName Q2 SYMATTR Value 2N3904 SYMBOL cap -272 144 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 0 32 VBottom 0 SYMATTR InstName C1 SYMATTR Value 10µ SYMBOL res 112 304 R270 WINDOW 0 32 56 VTop 0 WINDOW 3 0 56 VBottom 0 SYMATTR InstName R1 SYMATTR Value 47k SYMBOL res 16 368 R0 SYMATTR InstName R2 SYMATTR Value 1k SYMBOL res 336 368 R0 SYMATTR InstName R3 SYMATTR Value 1.8k SYMBOL cap 432 384 R0 SYMATTR InstName C2 SYMATTR Value 47µ SYMBOL res 336 -416 R0 SYMATTR InstName R4 SYMATTR Value 3.3k SYMBOL res 16 -416 R0 SYMATTR InstName R5 SYMATTR Value 4.7k SYMBOL voltage -240 -416 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 9 SYMBOL res 144 -112 R0 SYMATTR InstName R6 SYMATTR Value 150k SYMBOL cap 144 32 R0 SYMATTR InstName C3 SYMATTR Value .1µ SYMBOL voltage -336 192 R0 WINDOW 123 24 132 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value SINE(0 5m 1000) SYMATTR Value2 AC 1m TEXT -370 504 Left 0 !.tran 0 510m 500m TEXT -1072 8 Left 0 !;ac dec 100000 10 100k

amp...

pursue.

Yeah, C3 gives the overall amp response a low frequency bump, and C1 and C2 each contribute a low frequency rolloff. They all have to be balanced to make it flat. Probably eliminating C3 is a good idea, if the DC biasing still works. When I used this as a tape head preamp, the LF boost was an asset, part of the tape head response equalization.

R2 could be a lot lower. The open-loop voltage gain of Q1 is just R5/R2, which is only 5, which is pretty low... even lower when it's loaded by Q2. Or, another way to look at it, R2 kills the transconductance of Q1, and adds noise.

If you do my power amp version, with a mosfet for Q2 and an inductor for R4, there's another LF rolloff and the loop stability situation is horrifying.

John

amp...

pursue.

clucking

Old trick. Here's a bipolar-swing version.

ftp://jjlarkin.lmi.net/Opamp_boost_2.JPG

A similar thing is sometimes done to boost the current of an LM317, by using its input current to drive the base of a PNP "helper" transistor.

I use a variant of this circuit as a current splitter in my NMR gradient amps.

I can't spell, or type, either.

John