Nelson Pass said:
Well, don't keep us in suspense -- how did they sound?
I know a lot of DIY'ers have been reporting good results by replacing one op-amp or another with these op-amps. But the real question is whether they sounded better than your discrete circuits. In other words, is your next preamp going to use these parts because you can't build a better sounding circuit than National can?
stoolpigeon said:
Yes. All of the control microprocessors are IC's. All of the FPGA's are IC's. All of the DAC's are IC's. All of the FET switches are IC's. All of our video amplifiers have been made using (zero feedback) IC's. All of the PLL's are IC's. All of the logic gates are IC's.
As far as the analog audio circuitry goes, we use IC's in our two least expensive products. These are not op-amps, but rather a strange adaptation of an Analog Devices IC. By using two of them in a balanced configuration in a way that they were never intended to be used, we are able to use them with zero feedback. Their internal circuitry is fairly close to what we were using at that time.
We used them to save money. All of our other products are fully discrete.
We also compared them to the real thing (our discrete circuit of that time that was very similar, but made with discrete parts, metal film resistors, the ability to adjust all circuit values and bias currents for best sonic performance, et cetera). And guess what? The IC's didn't sound as good as our discrete circuit. Not really much of a surprise...
Oh, and by the way, even these products use our own discrete, zero-feedback power supply regulators.
Power supply regulators are critically important to getting good sound. For example, if you ever read a review of a "tube" preamp that has an "edge" or "glare" or "solid-state sound", you can pretty much guarantee that it is because the power supply regulators are based on high feedback IC regulators.
Re: Coal was used to construct microphone cartridges into the early days
Actually, it was carbon granules, not coal. And they were used well into the '80s in this country. All of the Western Electric (Bell Systems) telephones had carbon microphones. And by the way, those are still the best telephones ever made.
destroyer X said:
Actually, it was carbon granules, not coal. And they were used well into the '80s in this country. All of the Western Electric (Bell Systems) telephones had carbon microphones. And by the way, those are still the best telephones ever made.
Charles Hansen said:
Then why not make your own? Get the best of both words. You can do your own design with discrete components and get the advantage of very short circuit paths and a very small geometry.
Check these guys out. They say they specialize in short to medium length production runs and that initial start up costs are minimal.
In addition to circuit size and weight reductions over standard PC board packaging, a custom thick-film hybrid circuit simplifies assembly. Improvements in circuit performances are achieved by shortened circuit paths and closer spacing that yield reduced noise pickup, enhanced thermal coupling, and improved stability.
The hybrid circuit is a real and practical method to combine, in a single package, digital, analog, and even optical functions.
Flexibility for design changes and modifications are other real considerations in making your choice. Hybrids lend themselves to laser active-trimming of resistors to achieve precise circuit performance. With fewer interconnects and solder connections, improved operational reliability is inherent to the hybrid circuit.
Just think. You could be the first in this industry (that I'm aware of) to do this and not only could you pooh-pooh those who are using monolithic integrated circuits, you could also pooh-pooh those who are using traditional discrete circuits.
Hybrid Circuits, Inc.
se
Steve Eddy said:
Interesting idea. I have no idea what the costs involved are.
But there is at least one deal-breaker for me here. Check out their web page:
http://www.hybridcircuits.com/capable.shtml
Please note "Thick Film Print -- Resistors". Ever listened to those? Not very good...
Charles Hansen said:
The answer is "marketing". And if you can do your marketing for free via a letter to the editor of Stereophile, so much the better!
Sure, the engineers reading it might have to go running for their barf bag. But engineers are notoriously cheap anyway. They aren't the target audience.
Charles Hansen said:
Give 'em a call next week and talk to someone there. I'm sure they should be able to give you some kind of ballpark estimate.
That's only if you want them to print your resistors onto the substrate. That's not what you're limited to. You can also use surface mount devices.
se
andy_c said:
Andy, you only live 20 minutes from our factory. I have invited you several times to come up and listen to what we do. Please feel free to bring whatever op-amp based products you'd care to, and we will be glad to do some listening comparisons and see what you think.
And if you think Atkinson's article was spot-on, you can live happily ever after with the Crown IC-150 preamp. Measures darn near perfect...
SY said:
As far as I know, all of the thick-film SMT resistors are made with ruthenium oxide using a silk-screen like process.
When we first had the need to use SMT resistors, I got about a half-dozen major brands of thick-film resistors and about a half-dozen major brands of thin-film (ie, metal film) resistors. Of the latter, some were nichrome and some were tantalum nitride. (I was trying to cover all the bases.)
While there were some minor differences in the thin-film resistors, all of the thick-film resistors I tried sounded pretty much the same. Nothing offensive -- no glare, no edge, no grit. But at the same time, nothing to write home about -- no resolution, no inner detail, no harmonic complexity.
I figure about the only place where it would make sense to use them (besides for non-audio purposes) is if you were going to make $300 integrated amp that would be used with a really cheap, nasty sounding CD (or DVD) player. Then those resistors would tend to mask the sins of the source. But that's not the kind of gear that I design.
The way I listened to them was to make a series-shunt attenuator that could be placed between a CD player and a power amp (in place of a preamp). This allowed me to listen to the resistors in both series and shunt applications in one fell swoop.
Since then I have found a much better sounding SMT resistor. It is a bulk-foil type from Vishay. Of course it is horrendously expensive and only available in a limited range of values.
For some reason, I don't think that is what National is using in their new op-amps....
Steve Eddy said:
If the picture you posted is representative of a hybrid circuit, I guess I'm not seeing the advantage of using one compared to just using SMT parts on a good PCB substrate.
I seem to recall that hybrid circuits are usually built on an aluminum oxide substrate. I suppose that this would have better thermal conductivity than a PCB. But I would have to wonder about how good it sounded. In my experience, low Dk (dielectric constant) is important for a good sounding PCB material. But aluminum oxide is pretty poor in that regard.
Thick film SMTs, yes, but hybrids can use other technologies. There are simple resistive inks (similar to those used in potentiometer manufacture), thermoplastics, and fusible glasses with varying metallurgies (gold, platinum, palladium...). It might be worth exploring the options if you think that the presence of even a good-sounding chip won't turn off potential customers.
I doubt that you are missing anything Charles. These people think that we don't know or hardly ever use IC's. I took my first course in IC design in 1967. I have 41 year old linear IC's in my lab along with thousands of other IC's of all kinds and types. But then, how come I don't like to use linear IC's for everything that I do? Maybe, because when I try them, I find that discrete sounds better. What a concept!
- Status
- Not open for further replies.