Class A headphone amp idea

WTF? Why don't you doodle some bias numbers onto that drawing?

Is the bottom device also a DN2540? ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Yessir, both DN2540 depletion MOSFETs. The important resistors are the two in the bjt leads, they should be of equal value and selected so that in combination with the battery, supply, and zener you get ~ -0.5 volts Vgs quiescent for both FETs.

----Android NewsGroup Reader----

You're making it too complex. Try this...

If you want further embellishment, let me know.

As for the DN2540, if you're into playing with depletion-mode MOS, maybe use a DN1509 or an LND01 instead ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Did you breadboard it to see if it works? If you did, why not include the S/N and THD specs?

I agree.

Or forget about "Class A" completely and look at the opa2228 datasheet, Figure 17. I breadboarded that and was amazed at the low noise and harmonic distortion, even when connected to my studio monitor headphones as the load.

A high-quality headphone amp is really not an issue nowadays. It is easy!

I tried the Supertex LND150 for a small signal audio amplifier, and found it far too noisy -- much noisier than the jFETs I'm used to. The invisible "hint" was the lack of any spec for noise listed in the LND150 datasheet! I also tried one or two DN25xx models, but I don't remember exactly what happened with those.

Supertex seems mostly interested in the LED lighting market, and not hi- fi audio, so watch out for that.

Here we're dealing with an output stage. Noise will not be an issue.

Seems like a good business plan! I just applied for my 19th Patent... Billboard Video (LED) Drivers ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

Can U show us?

----Android NewsGroup Reader----

Not until it issues. Let's just say I solved _all_ of the fault detection issues that plague such applications.

Besides which, it's a "hairball", THOUSANDS of MOS transistors on a single chip, driving, and monitoring 192 LED's ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   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.

And every time I am distracted while driving by some garish animated video billboard I will say "Thank you Jim Thompson, for making all this possible..."

--




----Android NewsGroup Reader---- 
http://www.piaohong.tk/newsgroup

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

Digi-Key doesn't carry any of the FET devices you mention. Is there one they do have that would be suitable?

[snip]

Maybe, availability-wise, you'll have to stick with the DN2540... but it has way-more voltage capability than you need. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
                        #BringBackOurBalls

Anybody notice how the feedback circuit is not going to work ?

The amplifier inverts from input to output, with a small signal gain of the DN2540s transconductance against the load resistor. The output should sit at 0 volts at no signal because of the servo. Can you elaborate why the feedback circuit wont work? It's certainly possible I'm missing something.

--




----Android NewsGroup Reader---- 
http://www.piaohong.tk/newsgroup

There _are_ the right number of inversions. What is unknown is stability of the loop. It fails the Mad Man Muntz test... too many parts ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
                        #BringBackOurBalls

One bipolar and two MOSFETs is too many parts?!

--


----Android NewsGroup Reader---- 
http://www.piaohong.tk/newsgroup

With equal load resistors the firsr stage has less than unity gain; the only gain comes from the transconductance of the output stage. It seems it would be hard to make it oscillate, but I guess maybe it could. Unfortunately I'm mobile now and I don't have LTSpice available at the moment to test it.

--




----Android NewsGroup Reader---- 
http://www.piaohong.tk/newsgroup

To which schematic are we referring? Your OP had two OpAmps in the feedback loop. Is there another version that I missed? Post Message-ID. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
                        #BringBackOurBalls

They're just a DC servo to make sure the output sits at exactly 0 volts with no signal. I dont know that they're strictly required...

--




----Android NewsGroup Reader---- 
http://www.piaohong.tk/newsgroup

You've got two OP AMPS running the feedback to the input. First of all, why are you going for a low impedaance input. It's so 1940s. Yes, you wantt ot subtract the output from the input at a specifc rate which will give you t he gain you desire. Loading the input harder is not necessary.

Of course to fix that you need another stage, but that's not the main probl em. The one OP AMP has a capacitor from the inverting input to the output. When you do that, the DC gain is at its open loop level. There is however, no AC gain. In fact, most decent OP AMPs will not even pass an AC signal co nnected like that.

The second OP AMP invertsd just fine, but so does the circuit. That would b e positive feedback, which you do not want.

the way you hgot it, if you want to use the feedback that way, just elimina te the OP AMPs and run it straight. You will need to set the resistance (wi th a resistor) higher than the source impedance of whatever feeds it, and t hen make a resistive divider based on the gain you want, ust like an OP AMP . Therre is no reason for those OP AMPs to be in that circuit except to int roduce noise and distortion.

You got the imnput to the resistor, to the negartive I think ans a cap directly from the outpuyt to that.

That'll about give you an inverted DC voltage signl. With near infinity gain, or whatever it is for that chip.