Simple bridged amplifier design question

Someone makes a full-bridge rail-to-rail chip. Unfortunately it appears I didn't save the info :-(

I was looking at building my own before I ran across a ready-made unit that would drive two sets of headphones... wife and I watching movies on my laptop while flying ;-)

...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 at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

Best to post a schematic. You can do this but the diff pair will have to have lots of quiescent current if you want to yank 4ohms around. I'd go all out and do a class D :-)

Or maybe a 5V to 100V bridge converter, then a class D, then get a Honda low-rider with pitch black windows and put it all in there, pop in a J.J.Cale CD with "Cocaine" on there, cruise down the road, wait for the cops to notice ... ;-)

When I was a kid I did something like this with a massive tube amp. No car, but the cops did show up because someone from half a mile away complained about the tchk ... tchk ... *BOOM*

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

I figured someone did -- hence the disclaimer about it just being educational. Some decent IC version probably has much lower distortion as well. The fun is in the design though, not just soldering down someone else's canned solution. :-)

I tried several sets of headphones for flying and found that (1) if you don't mind the "in the ear" rubber-plug style, they're just as effective at jet engine noise cancellation as the active noise cancellation headphones (that require a battery) and (2) there's a surprising range of sensitivities in headphones, more than 10dB (for a given signal power).

I ended up with some pretty cheap ones (~$25) that I like, although I certainly don't begrudge the folks who want to spend the big bucks on Bose noise-cancelling headphones... my boss has a set of those and loves 'em.

---Joel

want

"textbook"

a

East Campus MIT dorms... on "Field Day" we had amplifier competitions. A friend and I built a plywood horn enclosure that fit into a dorm window when the sashes were removed... driven by a 100W tube amplifier.

Only after I turned my system off could anyone else be heard (after the Cambridge cops came from Central Square ;-)

...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 at http://www.analog-innovations.com |    1962     |
             
Obama clearly blames insurance companies for his mother\'s death
from cancer.  One then has to wonder if this whole health bill
isn\'t a personal vendetta against private insurance companies?

OK, will do.

The diff pair just drives emitter followers, so it (the diff pair) probably doesn't need more than 5-10mA.

That sounds suspiciously like digital to me... gaak!

Just kidding, class D is fine, I just want to do a little "historical" development here.

I was pulled over by some cops while riding my bicycle home some years back becase they claimed my bike's light (which was a high-intensity discharge bulb... but store-bought item, not something I designed!) was "as blue as our police lights." (...blue lights being illegal and all...) Yeah, right. I think they just weren't accustomed to seeing such a bright light on a bicycle.

---Joel

It takes 0.5A RMS to deliver 1W into 4 Ohms... 0.707A peak... tricky to do.

...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 at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

I have a pair of battery-powered Sony earphones that work great.

...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 at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Yeah, 8 ohms is a bit more attractive (500mA peak).

I was thinking those ubiquitous 2 1/4" speakers from transistor radios could handle upwards of a watt, but most of them seem rated at only about 1/4W or

1/2W. Perhaps I will just do an single-ended amp first.

---Joel

Musing. Wonder how many 'HC04's in parallel, with a class-D modulator, would it take ?:-)

...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 at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

I believe that this is how some of the audiophile "fully balanced" preamp designs work - they're differential all the way through from the phono-cartridge input, to the balanced outputs.

Seems as if it ought to work fine in your application, with a differential-pair single-sided-to-differential converter up front, and emitter-follower stages to drive the speaker.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Not many. Or a single LM311, from 30V, with an output transformer.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

One question is, since you have two independent outputs, what defines common mode voltage? The answer is to use common mode feedback, something like this.

This circuit is just the diffamp (balanced in/out), add emitter followers and feedback for power output purposes.

Tim

-- Deep Friar: a very philosophical monk. Website:

to use

the

textbook"

dd an

s me of a

=A0 =A0 =A0 =A0 =A0 =A0 =A0 AE6EO

r

For a fully differential output, you need to establish a common mode voltage. This is generally done with common mode feedback.

This is actually a little more difficult than it sounds if you expect the output stage to swing rail to rail. That is, a simple diff pair input with resistive loads will produce is differential signal that always needs to be above the common mode of the input voltage. Otherwise the input diff pair will not stay biased.

A better approach is to use something similar to a folded cascode, but obviously not exactly since the typical folded cascode has a single output. Hopefully you can follow this description:

Assume a NPN differential pair. Each leg of the differential pair feeds a node consisting of a current source from the top rail and a cascode device that feeds the difference of the current "downward" for lack of a better description.

Now if the loads on the output are resistive and the currents are well controlled, you can set the common mode voltage without feedback. (E=3DIR) If you use current source loads, then you need to sense the common mode of the output and force it to a reasonable value using common mode feedback.

You can see why a lot of designs just use an inverter and two amps.

I want

nking of

er

this off

ut of

to use

the

textbook"

dd an

s me of a

Another proof that USA is a police state :))

On a sunny day (Wed, 12 Aug 2009 15:54:30 -0700) it happened "Joel Koltner" wrote in :

In the early transistor days we used push pull with 2 transistors and a transformer. And a transformer with a thrid transistor to drive the thing too. I dunno bout these transformers, but those must be obtainable still.

transformer.

That was a tube amplifier implemented with transistors :-), i.e. common emitter output stages, with a center tapped transformer primary serving as the collector load (battery voltage connected to the center tap).

The nice feature is that in class AB the total primary peak to peak voltage was nearly twice the supply voltage.

To drive those output transistors/tubes, instead of an input phase splitting transformer, a single transistor/tube could be used with equal emitter and collector resistors. One output stage driven from the emitter of the splitting stage, while the other is driven from the collector of that stage. Of course all signal paths need to be AC coupled and the phase splitting stage operating around Vcc/2 bias voltage.

Paul

In reading schematics, there is another approach to the inversion, and I think it works better. Take the output of one and feed it into the inverting input of the other. I have seen this improve distortion specs.

greg

want

"textbook"

a

Ahm, this was in Europe ...

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[...]

They pulled me over in Germany because my bicycle had a yellow headlight (French). After I showed them proof that my principal residence was in the Netherlands they let me off the hook.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.