Replace resistor by low-inductance resistor - Is it audible?

[gfx_1]

This is what I have in my car:
Those are sand cast amirite? I'm not an engineer but I do know that those speakers are expected to run loud (given the "nature" of car audio) and those xovers are expected to be exposed to high in-car temps (parked outdoors in summer). Maybe sand cast does a better job in that environment?

They are for higher power-ratings, the little ones can dissipate 1/4 Watt. The rectangular ones are usually a couple of Ohms and a couple of Watts.

 

[locoliberty]

can we do capacitor changes(in crossovers) next?

 

[DDF]

I was curious about this at one time and took objective measurements of the frequency dependent inductance of various power resistor types and calculated their theoretical response loss at 20 kHz.

Here's the data attached, and a link to the discussion at the time.
https://www.diyaudio.com/forums/parts/234740-power-resistor-inductance-measurements.html

Definitely no cause for concern.

 

[617]

I sort of feel sorry for speaker designer stuck in this way of thinking. So many exciting and measurable things happening in speaker design and guys are trying to make better binding posts and doing resistor listening tests.

I suspect there was a time in the 90s when you could make a living with this level of knowledge but I'm not sure how much longer that gravy train will roll along.

 

[Milesian]

I believe, as with most audio gear, some will hear a difference, some will not. Then, in this perfect world, each will respect the other’s opinion without rancor, give a friendly wave and move on.

 

[BDWoody]

I believe, as with most audio gear, some will hear a difference, some will not. Then, in this perfect world, each will respect the other’s opinion without rancor, give a friendly wave and move on.

When someone's poorly founded, completely subjective, (and often completely impossible) 'opinion' has a chance to further mislead people who are coming to a science based forum looking for science based answers, an answer beyond a smile and a wave may be appropriate...given the venue...

Those hearing differences that aren't going to exist with listening controls might at least want the chance to prove to themselves that the industry is awash in garbage passing as a nice word salad.
If they want to continue to embrace ignorance, then it's up to them.

If people want to believe in 'Audio Umami' ( as you've called it), despite any and all real evidence to the contrary, at some point you do have to just smile and say:

Have a nice day (with a friendly wave)

 

[phoenixdogfan]

I realize most of you guys can't, but at age 67 I can easily tell a .02 db difference at 20khz. Great work, Amir. Now I know I should place an order to Mouser for the metal oxide resistors I need to replace in all my gear. Better level up on my surface mount soldering technique as well.

 

[wwenze]

Time to do a leaded vs unleaded solder test in that case.

Lead helps to capture neutrons better.

 

[phoenixdogfan]

Time to do a leaded vs unleaded solder test in that case.

Lead helps to capture neutrons better.

Imagine how much better it will be when I cool the solder joint by immersing it in liquid nitrogen.

 

[wje]

There's a reason Danny will never participate here or submit his modded speakers for NFS measurement. He's convinced that his CLIO system is "professional" and good enough. Furthermore, he not only purports that resistor swapping has an audible effect, he also touts his "Electra Cable Tube Connectors" as being audibly to superior to binding posts. It's hard to reconcile such stances with his otherwise seemingly very accomplished design skills, but there it is....

It's all about sales promotion. The same with PS Audio - oops... did I just say that? - No, seriously, following speaker builders and modders over the past few decades had me convinced that sand cast resistors were horrible. Thanks to Amir for these tests, reality is put back into the picture. In my case, now, the only reason that I might not use a sand-cast resistor on a crossover board, would be due to space constrictions as some being a bit "blocky" can take up a bit more room. But, the numbers don't lie. No need to spend money avoiding the sand-cast resistors on a build. Back to my modification of a pair of Elac B6.2 speakers in the past, I opted to replace 2 of the sand-cast resistors on the crossover since I was working on the board, but it would appear that the only gain or improvement in the project came from the use of a better poly capacitor.

 

[617]

It's all about sales promotion. The same with PS Audio - oops... did I just say that? - No, seriously, following speaker builders and modders over the past few decades had me convinced that sand cast resistors were horrible. Thanks to Amir for these tests, reality is put back into the picture. In my case, now, the only reason that I might not use a sand-cast resistor on a crossover board, would be due to space constrictions as some being a bit "blocky" can take up a bit more room. But, the numbers don't lie. No need to spend money avoiding the sand-cast resistors on a build. Back to my modification of a pair of Elac B6.2 speakers in the past, I opted to replace 2 of the sand-cast resistors on the crossover since I was working on the board, but it would appear that the only gain or improvement in the project came from the use of a better poly capacitor.

Again the low inductance is not significant but the power dissapation is. Make sure replacement resistors have the same or better power rating.

I suspect resistors can change value when heated enough, perhaps someone can fill me in on that.

 

[DualTriode]

Hello All,

I have done quite a bit of testing of speakers, resistors and speaker cables with AP’s APx555, APx1701 and the upgraded APx500 driver test software.

The largest thing that that shows in the tests is not so much impedance or frequency response but phase or time. (Yes time domain stuff.)

I believe that phase difference between mid-range driver and the tweeter is definitely audible.


Thanks DT

 

[Bruce Morgen]

It's all about sales promotion. The same with PS Audio - oops... did I just say that? - No, seriously, following speaker builders and modders over the past few decades had me convinced that sand cast resistors were horrible. Thanks to Amir for these tests, reality is put back into the picture. In my case, now, the only reason that I might not use a sand-cast resistor on a crossover board, would be due to space constrictions as some being a bit "blocky" can take up a bit more room. But, the numbers don't lie. No need to spend money avoiding the sand-cast resistors on a build. Back to my modification of a pair of Elac B6.2 speakers in the past, I opted to replace 2 of the sand-cast resistors on the crossover since I was working on the board, but it would appear that the only gain or improvement in the project came from the use of a better poly capacitor.

That's what I told Danny in several U-Toob threads: Inductors (air vs. ferrite vs. iron cores) and caps (NP electrolytics vs. low ESR film caps) can make an audible difference, especially in the signal path -- not so much when wired as shunts -- but resistors and speaker wire connectors don't do squat. The biggest difference comes from a crossover circuit properly designed to take into account the characteristics of the drivers and the enclosure. Of course, getting rid of enclosure resonances -- by whatever means -- often results in the biggest improvement at all.

 

[Lao Lu]

You might also test two resistors of similar wattage to compare the effects of the temperature coefficient.

The greatest difference is the one you didn't measure - profitability.

I normally use the non-inductive resistors in tweeter circuits but more for a tighter tolerance rating than the slight difference in inductance. The encapsulation can affect the cost as some resistors are made for more demanding environments; however, this isn't an issue / advantage in a home audio speaker.

Hi Rick---Are expensive crossover components more reliable?

 

[SIY]

That's what I told Danny in several U-Toob threads: Inductors (air vs. ferrite vs. iron cores) and caps (NP electrolytics vs. low ESR film caps) can make an audible difference, especially in the signal path -- not so much when wired as shunts -- -

Since a shunt appears in the voltage divider just as much as a series, this assertion seems more than a bit odd.

 

[pma]

Pick your poison -
differences between wirewound resistors are huge.

I am showing 2 examples:
- resistor #1 is a 0R22/5W most used worst class cheapest white brick type,
- resistor #2 is a 0R22/7W much better but still affordable type.

Measured with square wave current injected into resistors. Comparative measurements made at 100mV/div, for #2 measurement at 20mV/div added because of much lower inductive peak. Make your choice, yourself, I made mine years ago. Such resistors are used as emitter resistors in output stages of power amplifiers, would anyone make a choice of #1? If you answer no, then be sure they are used most often. What do you think about high frequency stability of the amp, then? So again, this thread is an example of the typical ASR oversimplification.

 

[SIY]

The question for that particular application (different than the loudspeaker stuff we're all talking about) is, what's the effect on the overall amplifier performance? Any data on that for your amp?

 

[solderdude]

I see about 500ns period = 2MHz.
Not something to worry about with speakers but definitely in the feedback path of an amplifier (see post#2)

For speakers and audio frequencies (and even well above) I would not say differences between wirewound resistors are huge (as in audible).

 

[pma]

I see about 500ns period = 2MHz.
Not something to worry about with speakers but definitely in the feedback path of an amplifier.

Repetitive frequency is 800kHz (distance between 2 negative peaks), which is totally unimportant.

V = R*i + L*di/dt

i = 4.5mA
di/dt = 4.5mA/10ns (450 000A/s)

So you can calculate L and consider if it is OK for you. I am speaking about amplifiers, as I have already stated.

 

[Francis Vaughan]

...
Measured with square wave current injected into resistors. Comparative measurements made at 100mV/div, for #2 measurement at 20mV/div added because of much lower inductive peak. Make your choice, yourself, I made mine years ago. Such resistors are used as emitter resistors in output stages of power amplifiers, would anyone make a choice of #1? If you answer no, then be sure they are used most often. What do you think about high frequency stability of the amp, then? So again, this thread is an example of the typical ASR oversimplification.

Depends upon what you are trying to prove or say. Everything has inductance. A straight piece of wire has inductance. You are showing that one resistor has ten times the inductance of another. But so what? It isn not exactly a surprise. Is the lower inductance small enough not to cause problems? Is the larger inductance too small to cause problems? Where we came in, the question was inductance in resistors in crossovers. The answer for that is that there is no difference. Your results would support that.
Now you raise a totally difference use case. Emitter resistors for power amps. Does inductance matter there? Clearly eventually enough inductance will exceed the stability margin of the amplifier. So too much is obviously bad. But, does even the bad resistor you measure have enough inductance to cause stability problems? Maybe evne the better resistor will still cause stability issues. Depends on the amplifier. Certainly can depend on the choice of output device. But also on the feedback loop parameters. What about self oscillation of an output device? The acceptable value of inductance should be known, if the resistor exceeds it, don't use it. If it doesn't exceed it, use it. It isn't as if this is a new problem. You use the components with the parameters that are appropriate for the use. If you need low self inductance in a resistor, you specify it. If you don't, you don't. But you don't need to over-specify. That is just poor engineering.
Eventually it comes down to what the numbers say. Maybe tens times matters in some circumstances. But in others it matters not at all.