John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Patents were always intended to be for bright ideas - that means bright technical ideas, not bright commercial ideas.

The funny thing about this particular patent is that it relies on a low shunt impedance for the filtering action, which in turn only works if you have a sufficiently high series impedance for the mains supply. This impedance is poorly defined, and the patent description doesn't seem to highlight this basic flaw. All good mains filters have series and shunt elements.
 
I'm glad that you are amused, PMA. It suits you better than your usual disposition. '-)
Today, a fellow engineer who has more degrees in both engineering and physics than most anybody, told me that he acquired a new $13K D-A with a successive approximation converter and a tube output. He is very happy with it. It is certainly out of my league, but I am intrigued.

John,

When you see your multi degree engineer mate - Any chance you
can find out from him what a successive approximation DA actually is?

SAR is an ADC architecture.

Which ADC Architecture Is Right for Your Application? | Analog Devices

cheers

Terry
 
DAC in my Devialet amplifier sounds fantastic. They use PCM1792. Which is a delta-sigma. It's just all in the execution, layout, mechanical grounding, parts with certain sound character, etc. Everything matters.

Looks like they also have a Cirrus WM8740 delta-sigma DAC inside as well.
 
Last edited:
That video was not meant for an audience of the type that would frequent this thread.

We can put it in the background while we demagnetize all our LP's. Some lose sight of how narrow and audience is that is represented here, there are stupidly priced hi-end audio products that are full of IC's and other horrifying things and they get rave reviews and customers. There are also people passionate about music who listen live when they can and defer to reproduction when they have to that don't hear .01uF caps bypassing 1000uF power supply caps. They don't bother with this kind of chat, I wonder why?
 
Last edited:
I am told that my ESS 9038 evaluation board costs a lot more than that, but on the other hand, I was also told by Kozned that all D-A's like the 9038 are junk and I also should go back to successive-approximation converters for best sound quality. By the way, WHY they sound better can't be easily measured, but I believe him because he listens as well as measures.

John, You really need to try Benchmark DAC-3. It uses some digital methods that are not used on the ES9038 evaluation board. They do make a difference, so I would agree that you are probably not hearing 9038 at its best. If you were to get someone who knows how to fix the digital stuff for you, then you might have some better audio to work with. As it stands, I would say you are probably working at a disadvantage.
 
Here is US Patent 5,260,862 issued to Richard Marsh in 1991

us pat 5260862 .png

View attachment US5260862.pdf

My most sited patent should have had prior art from the 1890's, I was surprised.

John, I think Scott's reply to my surprise hits the nail on the head. I'm more surprised by the awarding of said patent than anything, which speaks more to our patent system than Richard's contribution. Also why I wonder about its enforceability, given prior art.

As Richard wrote, though, patents are "complicated".


Patents like that remind me why my PhD supervisor told me not to take seriously any US patent. He said that they were all regarded as jokes in Europe.

LC traps like that were routinely used in TV circuits since the 1950s (and probably in radar circuits before that). How could a standard textbook circuit be patented? The patent examiner must have been astonishingly incompetent. Maybe he just searched patents for prior art?


Thanks, Mark. And, yeah, if the referee goes by Manny, that's just the guy you want doing all the patents for utility-level electrical systems. :D

Richard, that's what I was trying to say--the link I gave just showed the parasitic series inductance/resistance of the respective capacitors and the self inductance of the traces. Yours is a many-tap explicit RLC array. I'm admittedly surprised you were able to get this patent, truth be told, but 1991 means it's well past its last renewal.


I also have given up on patents for anything I do.
There was an interesting article in a trade magazine, suggesting that we all just do away with patents entirely...
There would be both benefits and hardships.

This kind of shows how you guys really think about the efforts and successes of others. Richard got a patent on something that he developed. You should congratulate him for his success, rather than berate the patent.
Of course, most patents are just annoyances that people inflict on each other, AND if you don't have big money to back it, others will just steal it with impunity. That is why I have never bothered to patent anything other than the one that Mark Levinson paid for. And he only wanted the patent to attempt to take control of the circuit from me by having me 'assign' the patent to him, which I did not. In his disappointment of not being able to take advantage of me, he kept the original paperwork just to annoy me.
Patents can be useful in deterring people from directly copying only if you are prepared to sue. I was not, so Ortofon and others copied my patented JC-1 design over my protest. That's life folks!


Patents were always intended to be for bright ideas - that means bright technical ideas, not bright commercial ideas.

The funny thing about this particular patent is that it relies on a low shunt impedance for the filtering action, which in turn only works if you have a sufficiently high series impedance for the mains supply. This impedance is poorly defined, and the patent description doesn't seem to highlight this basic flaw. All good mains filters have series and shunt elements.


I wish Sy was still postin. He seemed to have a deep knowledge of the US Patent System.

Cheers,
Jeff
 
Last edited:
John, You really need to try Benchmark DAC-3. It uses some digital methods that are not used on the ES9038 evaluation board. They do make a difference, so I would agree that you are probably not hearing 9038 at its best. If you were to get someone who knows how to fix the digital stuff for you, then you might have some better audio to work with. As it stands, I would say you are probably working at a disadvantage.

If you read the datasheets you'll find the ES9038 is basically 4x ES9028 DAC sections in parallel. Given that the ES9028 already has dynamic range far beyond need, I would probably use the 9028 over the 9038 in a stereo application. The output current of the ES9038 is quite high in stereo mode. The ES9038 was probably designed primarily to be used in an 8 channel system while still yielding ES9028-level stereo configuration specs.
 
@chris719 Yes. Actually, DAC-3 probably does use ES9028 now that I think more about it. Anyway, seems like most people tend to focus attention mainly on analog output stage stuff and disregard the importance of the digital processing that is a big part of DAC-3 sound quality.
 
Last edited:
Status
Not open for further replies.