John Curl's Blowtorch preamplifier

[AndrewT]

I'd be more interested in comparing sound with discrete.

why don't we see more of these hi speed opamps being used for headphone drivers?

Was it CarlosFM that published a design based on Hi speed opamp?

 

[john curl]

Jan, I have been working professionally with IC op amps since 1966. I track what is going on. I have hundreds of data books, old and new, to compare, perhaps 1000's of individual data sheets on IC's. I know where the bear sits.

 

[syn08]

why don't we see more of these hi speed opamps being used for headphone drivers?

Because:

a) those performances are not required.
b) they are not easy to handle in consumer applications (e.g. PCB layout and quality is critical for keeping them stable). Paper PCBs made in China won't do.
c) there are specialized ICs for such applications, e.g. TPA6120A2. 1.4 ppm distortions, 120dB dynamic range, 120dB S/N ratio, 1300V/uS slew rate, 80mW in 600ohm load.

Do it with discretes.

 

[syn08]

I know where the bear sits.

You don't. Meantime, the bear has left the building.

 

[syn08]

I'd be more interested in comparing sound with discrete.

So go ahead, compare and let us know.

 

[john curl]

Pointless

 

[SY]

They do, please look again:

No, they don't. They are asserting that something that depends on an uncharacterized parameter (and the designer of that opamp has confirmed this) will work consistently from batch to batch. Without a LOT of supporting data, that's an extremely dangerous assertion. Read Pease for some horror stories about that.

 

[scott wurcer]

You know John, it is a bloody shame that you still feel the need for such messages. Just lash out, right, and the hell with interpersonal relations.
How would you feel if we asked you what you have invented in audio design that we didn't essentially know after your long tailed pair in 1964?

Jan Didden

John and Bernie Gordon might have one thing in common, Bernie swore for years that a REAL 12 bit converter is all you ever need for audio. For all I know he would still say it.

 

[scott wurcer]

How long has Bob Widlar been dead? Do you know how many years? What have you, Scott Wurcer, added to IC design that was not essentially known by Harris in the 1970's? Please provide evidence, and listening feedback. Please don't say that you invented low Rbb'. We known about that since 1966.

Some Harris op-amps had mediocre DC performance, noise, settling tails, marginal stablity, some lots drifted to the rails after 10min of operation. You won't find one instrument maker that longs for their return.

It was fun selling rebranded 883B parts to the Navy for $75, paid the bills. Will you pay $75 for an IC op-amp?

 

[syn08]

It was fun selling rebranded 883B parts to the Navy for $75, paid the bills.

Interesting AD7520 rebranded?

 

[scott wurcer]

Interesting AD7520 rebranded?

It is a funny story. We 100% guaranteed settling time on a fixture that was the last (or almost) thing Dick Burwen built for ADI. A real bench test, plug it in, trim a pot, and read off the scope. When it broke I asked Paul Brokaw to help me fix it (it was about my third month on the job) and he just laughed (expletive deleted).

 

[Rick Miller]

Hi Jan,

Glad to meet you too.

 

[Rick Miller]

http://blog.stereophile.com/ces2009/on_the_lamm/

John,

Here is the link to the Lamm review that you asked for.

 

[john curl]

The Lamm exhibit was the best sound at the show. Why?


Thanks Rick.

 

[reinhard]

Have a close look at the GM 70,

Reinhard

 

[vuki]

The Lamm exhibit was the best sound at the show. Why?

Could it be because Vladimir Lamm knows everything there is to know about relationship between sound and objective measurements? In few of his interviews he even said that he doesn't have to listen to his designs because he knows how they sound just looking at to measurements. Those GM70 monoblocks must have pretty good measurements...

 

[jan.didden]

Some Harris op-amps had mediocre DC performance, noise, settling tails, marginal stablity, some lots drifted to the rails after 10min of operation. You won't find one instrument maker that longs for their return.

It was fun selling rebranded 883B parts to the Navy for $75, paid the bills. Will you pay $75 for an IC op-amp?

Most of those early Harris were not even ICs. They were discretes in a box, hah!

Jan Didden

 

[jan.didden]

Could it be because Vladimir Lamm knows everything there is to know about relationship between sound and objective measurements? In few of his interviews he even said that he doesn't have to listen to his designs because he knows how they sound just looking at to measurements. Those GM70 monoblocks must have pretty good measurements...

Stan Curtis, the guy who put Rotel back on its feet many years ago, claimed the same. His secret, as much as he confided (IIRC; t was an interview in EW I think) was manipulation of the freq response and Zout, or the combination of it. His reviewers hated him, his amps didn't measure particularly linear but they couldn't stop listening to them. An enviable achievement.

Jan Didden

 

[jan.didden]

OPA847 - bipolar, 3.9GHz bandwidth, 950V/uS slew rate, 100uV offset, 0.85nV/rtHz noise, 18mA supply, probably 10mA output stage bias, distortions ppm in the MHz range.

THS4631 - JFET input, 325MHz bandwidth, 1000V/uS slew rate, unity gain stable, 7nV/rtHz noise, 12mA supply, certainly over 5mA output stage bias, distortions ppm in the MHz range.

Do it with discretes.

That OPA847 looks superb, too bad the (bipolar) input bias is so massive.

Jan Didden

 

[reinhard]

Could it be because Vladimir Lamm knows everything there is to know about relationship between sound and objective measurements? In few of his interviews he even said that he doesn't have to listen to his designs because he knows how they sound just looking at to measurements. Those GM70 monoblocks must have pretty good measurements...

Good promotion,
at this price point measurement and sound must be OK.