Tantalum Resistors: An Audio Advantage?

[Wirth's Law]

Hello all,

Lately I've been doing some personal research on passive components using tantalum (and its close cousin: niobium). It's come to my attention that there are many people in the audio field who state that tantalum resistors (yes, resistors) offer a better quality of sound compared to other metal oxide or carbon film resistors.

Without going into great detail: I've seen the tantalum sound described as being anything from "less dry" to "happier" than other options. This seems like an overly-subjective way to put things. Although we've also seen from the "Yannie vs. Laurel" and "Brainstorm vs. Green Needle" viral videos earlier this year, human audio perception is extremely subjective anyway.

After searching through a lot of data sheets, journal articles, patents, and related books; I wasn't able to find a satisfying explanation for why tantalum resistors might be favored. From what I gathered, tantalum resistors are a thin-film type of resistor using tantalum nitride in place of the more-commonly-used nickel-chromium (nichrome) material. Since thin-film resistors are generally more precise and stable than other constructions; could the improvement in audio quality simply be due to having a high-precision part in the signal chain? Or is something else at work here? Thanks in advance.

 

[dknguyen]

Hello all,

Lately I've been doing some personal research on passive components using tantalum (and its close cousin: niobium). It's come to my attention that there are many people in the audio field who state that tantalum resistors (yes, resistors) offer a better quality of sound compared to other metal oxide or carbon film resistors.

Without going into great detail: I've seen the tantalum sound described as being anything from "less dry" to "happier" than other options. This seems like an overly-subjective way to put things. Although we've also seen from the "Yannie vs. Laurel" and "Brainstorm vs. Green Needle" viral videos earlier this year, human audio perception is extremely subjective anyway.

After searching through a lot of data sheets, journal articles, patents, and related books; I wasn't able to find a satisfying explanation for why tantalum resistors might be favored. From what I gathered, tantalum resistors are a thin-film type of resistor using tantalum nitride in place of the more-commonly-used nickel-chromium (nichrome) material. Since thin-film resistors are generally more precise and stable than other constructions; could the improvement in audio quality simply be due to having a high-precision part in the signal chain? Or is something else at work here? Thanks in advance.

I don't think precision would really improve sound quality unless the circuit relied on balanced components of some kind. But thin film resistors tend to have less noise. But when I say thin film, I actually mean vacuum deposited thin film resistor, and not a paste-deposited thick film resistor which still produces a thin film to human eyes.

Could just be placebo effect or confirmation bias as a result of tantalums being harder to obtain that even metal film resistors.

"Happier" is probably the worst descriptive term I've ever heard used to describe sound quality though.

 

[gophert]

Hello all,

Lately I've been doing some personal research on passive components using tantalum (and its close cousin: niobium). It's come to my attention that there are many people in the audio field who state that tantalum resistors (yes, resistors) offer a better quality of sound compared to other metal oxide or carbon film resistors.

Without going into great detail: I've seen the tantalum sound described as being anything from "less dry" to "happier" than other options. This seems like an overly-subjective way to put things. Although we've also seen from the "Yannie vs. Laurel" and "Brainstorm vs. Green Needle" viral videos earlier this year, human audio perception is extremely subjective anyway.

After searching through a lot of data sheets, journal articles, patents, and related books; I wasn't able to find a satisfying explanation for why tantalum resistors might be favored. From what I gathered, tantalum resistors are a thin-film type of resistor using tantalum nitride in place of the more-commonly-used nickel-chromium (nichrome) material. Since thin-film resistors are generally more precise and stable than other constructions; could the improvement in audio quality simply be due to having a high-precision part in the signal chain? Or is something else at work here? Thanks in advance.

Do we get to make fun of an audiofool in this thread. I'm really looking forward to it. PT Barnum would have had a field-day with the opportunity to make a buck off of audiofools. No need to haul around a three-ring circuis, just some oxygen-free copper, Tantalium resistors and big, fat Monster Cables.

 

[carbonzit]

Do we get to make fun of an audiofool in this thread. I'm really looking forward to it. PT Barnum would have had a field-day with the opportunity to make a buck off of audiofools. No need to haul around a three-ring circuis, just some oxygen-free copper, Tantalium resistors and big, fat Monster Cables.

Took the word right out of my mind; "audiophool" was what I thought immediately upon reading the original query here.

Y'know, it's true that resistors (like all components) do generate measurable noise, and it might be true that tantalum resistors produce less noise than, say, carbon or other materials; but the difference is going to be completely negligible and, to most folks, totally inaudible.

But there will always be a market for gold-plated connectors, oxygen-free wire and all the rest of that nonsense. Can't stop people from buying snake oil.

 

[audioguru]

I am an audio guy and have used carbon film and metal film resistors in all my audio circuits and they work perfectly.
Sound that is "less dry" to "happier" is absolutely hilarious! Is sound happier when it or you are wet?

 

[crutschow]

Or is something else at work here?

Something else.
It used to be called snake oil.

 

[dr pepper]

The only 'incorrect' kind of resistor I can think of in audio is a wirewound in a zobel network, the inductance might be an issue.

 

[alec_t]

Surely audiofools would settle for nothing less than platinum resistors? The more they cost the better the perceived sound quality.

 

[dknguyen]

Surely audiofools would settle for nothing less than platinum resistors? The more they cost the better the perceived sound quality.

But is not platinum known for it's resistance change over temperature? Hence it's use in RTDs. That doesn't sound very ideal to me.

 

[Nigel Goodwin]

The only 'incorrect' kind of resistor I can think of in audio is a wirewound in a zobel network, the inductance might be an issue.

As it's a low value resistor, than the inductance will be too small to make any noticeable difference.

But I agree with the general consensus, this entire thread is ludicrous, and even more so is the imaginary descriptions of what it supposedly improves

 

[ronsimpson]

this entire thread is ludicrous

If this thread was about what types of capacitors to use in audio ........... but resistors?

 

[JonSea]

Good job guys. Really great. The original poster asked a serious question, supposedly of people with knowledge in the field.

One reply attempting to provide some education, the rest of the posts mocking the OP. If your goal is to drive people away, you've done an awesome job. How far down is participation from previous years? Gee, I wonder why.

 

[Wirth's Law]

Well yeah, I was going to mention that. Mostly I'm looking for an unbiased opinion on these since it would be a cool/niche component to test out if the "legends" are true.

The only 'incorrect' kind of resistor I can think of in audio is a wirewound in a zobel network, the inductance might be an issue.

This was roughly my thinking as well. A parasitic inductance would be unwanted in many resistors. After doing some more digging, it looks like the tantalum nitride thin film variety is also featured as "non-magnetic," and this time, there were some reputable data sheets to support it. By non-magnetic, I'm guessing this is referring to ferromagnetism, meaning they would avoid materials like iron, nickel, cobalt, and some lanthanides. Even a non-wound resistor can have inductance, so this would really eliminate thin film resistors with nickel or nickel-chrome in their construction that would make things even worse. Unfortunately, no one seems to publish any detailed (proprietary) diagrams showing how much magnetic material is or isn't included in their resistors; so an actual physical measurement may be the only way to tell for sure.

I don't think precision would really improve sound quality unless the circuit relied on balanced components of some kind. But thin film resistors tend to have less noise. But when I say thin film, I actually mean vacuum deposited thin film resistor, and not a paste-deposited thick film resistor which still produces a thin film to human eyes.

Could just be placebo effect or confirmation bias as a result of tantalums being harder to obtain that even metal film resistors.

Right, so far what I've been finding seem to refer to sputtered or vacuum-deposited thin films around 100 to 1000 angstroms thick. But they mostly refer to the depositing process itself and don't describe the resistor performance in circuits.

 

[gophert]

Well yeah, I was going to mention that. Mostly I'm looking for an unbiased opinion on these since it would be a cool/niche component to test out if the "legends" are true.

Wirth's Law , I'm confused, JonSea seems to think we scared you off and you were somehow upset that we mistreated you. Yet you just jumped right in to the existing conversation you must have at least average and normal levels of self-confidence to not be sobbing in the corner. Jonsea said...

Good job guys. Really great. The original poster asked a serious question, supposedly of people with knowledge in the field.

One reply attempting to provide some education, the rest of the posts mocking the OP. If your goal is to drive people away, you've done an awesome job. How far down is participation from previous years? Gee, I wonder why.

 

[dr pepper]

This kind of subject on this forum might not get you far, audiophilia, over unity, and supernatural electronics dont seem to go down well.

 

[Wirth's Law]

Regarding wirewound resistors in Zobel networks--which are effectively there for impedance-matching--this got me looking more into non-ideal behaviors in the common resistor types. And I think I see how the whole "tantalum resistors for audio" thing got started.

It's well known that in the real world, resistors have non-ideal (parasitic) characterisitcs. Each one will have some detectable amount self-inductance, and even the flat, non-coiled chip resistors aren't exempt from this (per Ampere's Law).

Where I suspect tantalum factors in is: the permeability of any nearby material has a direct impact on its self-inductance; hence why inductors and transformers often feature ferrite. For thin film resistors, the two leading chemistries as of 2018 seem to be the popular nickel-chromium (NiCr) and the moisture-resistant tantalum-nitride (TaN). In nearly every source I've read, the NiCr type was using along with pure nickel in its end-caps. Not every source, but most. TaN end-cap materials were rarely specified, opening the possibility for non-ferromagnetic metals like gold or copper. As I said before, nickel is one of the only ferromagnetic metals. Its permeability is nearly as high as iron.

I decided to put this to the test using random 0805 resistors I had in stock. I took 6 rare-earth magnets and some basic-grade thick film resistors--brought them together, and sure enough, the resistors clacked right onto place on the magnet. I repeated this using 0603 precision Vishay TNPW series thin-films. They were attracted too, albeit with way less force. This makes me wonder how often people place potentially magnetized resistors near hall sensors or digital compasses.

For the most part, you only hear about tantalum capacitors. Tantalum-nitride and nickel-chromium resistors are both casually bought and sold as "thin film" devices, and consequentially, shockingly few engineers are aware tantalum resistors even exist. So basically, where I am now is trying to sort out: (1) exactly how much of an inductance difference there is, and (2) how likely will it might affect audio. The 'audiofoolery' assumptions reflect sloppy reasoning rather than science. And anyway, the idea of a totally non-ferromagnetic resistor is too good to pass up. Below is a simplfied cross-section of some chip-style resistors I found (re-drawn for copyright-avoiding purposes). The mystery now is what the exact inductance/impedance implications are for NiCr vs. TaN.

 

[KeepItSimpleStupid]

The problem may be with current noise. I found two relevant articles:

1. http://www.vishaypg.com/docs/49997/49997.pdf
2. http://preserve.lehigh.edu/cgi/view...le.com/&httpsredir=1&article=2754&context=etd

 

[dknguyen]

The problem may be with current noise. I found two relevant articles:

1. http://www.vishaypg.com/docs/49997/49997.pdf
2. http://preserve.lehigh.edu/cgi/view...le.com/&httpsredir=1&article=2754&context=etd

well, the first article just puts tantalums in the same category with metal foil thin film which the pro-tants specifically delineate between. The second article doesn't seem to have a control to compare against?

 

[KeepItSimpleStupid]

I pretty much agree. The thin metal films have a wide range of noise.

Procesing, apparently has a lot to do with it. I've done personally sputtering of metals and metal oxides particularly Mo, ZnO and ITO. RF sputtering for oxides and DC sputtering for metals. We electron beam evaporated metals (e.g. Gold, Nickel and Moly) but in that case, I just fixed the machine. I personally thermal evaporated magnesium and apparently better than anyone else could at work. They kept asking if I evaporated the wrong metal. We did carbon too using an arc.

 

[Wirth's Law]

Right, that pretty much squares with other Tantalum Nitride papers I had found. Even if the resistor is guaranteed to use TaN, the exact structure/phase of the TaN could be influenced by the specifics of the sputtering process itself; or by any subsequent heat treatment. One paper (link below) cited seven different phases of TaN. So there doesn't appear to be any exact look-up table for the current noise of TaN. Or if there is, it may be hiding.

https://iopscience.iop.org/article/10.1143/JJAP.10.248

To actually compare some physical resistors, I'm looking for a circuit that is good at measuring parasitic inductances. I have a function generator and oscilloscope, so I was thinking of using something like a bridge circuit with precision foil resistors as the reference. Any suggestions?