Richard, how are your speakers coupled to the floor? Have you ever done or thought of doing measurements of how well your speakers reproduce square waves?
Did you? Well, I don't do those either, to any meaningful degree. Triamped speakers, amps at the speakers. So still not a problem.
And - unless I missed it skimming to catch up all the overnight posts, no one has as yet put any numbers on these possible cable problems. JN threw some estimates into the pot, but he is not a protagonist of it being problem.
But no measurements from those that hypothesise that this is an audible issue?? And it's not like those should be difficult?
yes. I have done that. I made an analog cross-over which had variable slope and freq. With that and having the mid-high compression drivers on top of bass cabinet... I moved the drivers back and forth and adjusted the slope etc until I got a square wave at the X-over freq. on the scope via mic/mic-pre. Sounded great BTW.
I know most here would try to measure each driver's phase and freq and Z and import to a software program where they could simulate the drivers and a cross-over. About a year later maybe have something that works. A sq wave gen and excellent mic and mic preamp into a scope works fine and gets you there like right now. Besides, such hard/software didnt exist back in the day of early HiFi.
The only issue today is finding a continuously variable slope cross-over. You need it... Not a few fixed slopes. A good DIY project.
The speakers were on carpet over concrete.
THx-Richard
Last edited:
"Making square waves?" thread has recently been resurrected, I'm inspired by this post Making Square Waves? (shame the images are gone) to try to measure my Jordan Eikona VTLs. I only have a Panasonic WM-61A capsule and old oscilloscope, do you think the mic would be good enough? Another post: Making Square Waves?
Last edited:
Yes, that mic is fine... flat and wide band. THD or S/N is not a primary concern. it is good enough. There are many inexpensive measurements mics now available. You need to be able to time align and have variable slope and freq and level control. The phase summing has to be spot on to get a square wave. Approximation slopes or general purpose slopes from text book formulas wont be good enough.
I was able to get quite a bit better looking sq wave from the speakers than shown in thread you showed.
THx-Richard
Last edited:
There's no crossover 🙂 There's another link above where he says position is critical and on axis is no good.
A speaker or system which can do a reasonable square wave is doing somethings right.
-Richard
Last edited:
The Jordans have a rising HF on axis response and many say they sound better off axis, my ears enjoy the boost however 😉
I can measure to about 2ppm with a careful set up at 2V full scale on the QA401.
Can you provide your test regimen?
So you cannot perform the measurements. That's ok, I can understand your fear of failure.
It's unfortunate for all of us that your fallback schtick is to denigrate others.
Jn
For a single zip cable 4 or 5 meters long, most people do not have access to equipment capable of measuring Ls/Rs, as the reactance is too low in the audio band for tenth of percent accuracy and repeatability. Without that equipment, even careful measurement regimens may not be capable of seeing differences. With good experience and equipment, it is possible to see the changes when stranded cables are virgin, as well as when they are baked to cure the coating on the outer surface of the insulation. Vacuum impregnation is by far the worst player, as the sole purpose is to fill all the interstitial voids between strands, making the cable less prone to share among strands.
Since we are discussing internal and externally generated eddy losses, I would recommend a slight deviation first to increase sensitivity, that of measurement of single conductor coils.
If you examine my gallery you will find measurements of Ls/Rs using solid copper inductors. In that case, the purpose was to show how proximity to a copper clad board generates eddy losses. Note the Rs really takes off when proximity happens.
Using three made inductors, one solid 12, one stranded 12 as a standard, another stranded 12 as the test unit..
1. Measure Ls/Rs initially out to 100 kHz. The higher frequency is usually a throwaway, but the data up to there shows where the test setup begins to fall apart.
2. Bake the DUT in 24 hour increments, test all 3 coils every time to verify accuracy and repeatability. I recommend using the solid coil to zero cal the setup.
As the DUT ages, you will see Rs change even though end terminal resistance isn't changing much.
After a long enough period, you can use the coil wires to twist into pairs for cable tests.
In this form, running a 20 to 20k sweep into 4 ohms looking for eddy loss distortion components can be performed.
As I stated earlier, the absolute highest level one can expect of a simple twisted pair can never exceed 10e-2 to 10e-4, and my expectation is 10e-4 to 10e-5.
Note that there will be a max level point as the interstrand contact begins the degradation from really good contact, into the start of noticeable degradation, then to almost no contact heading into a virtual litz regime of operation. I say virtual litz, as true litz has all conductors spending some time at bundle center, some at bundle edge, all at random.
A twisted wire typically has layers, like 13 around 6 around 1, so can never perform as a true litz. Instead, the end to end resistance will suffer at frequency because the inner wires will not carry current uniformly.
Jn
Last edited:
John,
Please be so nice to tell us.
The conductance quantum Go = 2e^2/h that you mentioned, happens also to be the quantum conductivity of a Multiwalled Stranded carbon Nanotube (MWNT) at applied voltages below 50mV.
Put N of such MWST's together and you probably have a Bybee.
However this quantum conductivity depends on the applied voltage according to dG/dV and is approx 0,5Go/Volt.
So starting at Go with practically zero volt applied, conductivity changes with a rate of 0,5Go/Volt.
I would call this a Quantum VDR, not exactly a device that comes to my mind to place in a signal line.
But never too old to learn, I would like to know why this would and could have an positive effect on sound reproduction.
Or was it just the result of a trial and error experiment that resulted in a product ?
Hans
Remember, their test showed ballistic for sub micron distances. And at the current densities they discuss, they show how the dissipation has to be at the end of the tubes.
Just taking a powder made of nanotubes and constructing a bulk device does not provide the end connections as described in that paper..
Pre 2000 paper, whoa is that old and out of date..especially in the quantum and nano material arena.
Jn
Just taking a powder made of nanotubes and constructing a bulk device does not provide the end connections as described in that paper..
Pre 2000 paper, whoa is that old and out of date..especially in the quantum and nano material arena.
Jn
Last edited:
Well the reason I’m so adamant about relaying the subtle differences I hear is to help support the quest for ‘better’ sound reproduction. The whole thats ‘good enough for most’ is annoying to say the least. Hopefully the younger generation of EE’s will bring a better game than the second hand lions that reside here sitting on the porch with shotguns waiting for the ‘audio’ salesmen to come along.
Maybe it does need to be looked at on a smaller scale, you know the things you’ve been taught didn’t matter?
Maybe it does need to be looked at on a smaller scale, you know the things you’ve been taught didn’t matter?
I don't think Syn08's critical way of thinking denigrates many others.
On the contrary, it helps together with other contributions to get an overview of all physical processes taking place, as seen from many different aspects.
Comparing a stranded cable to a dirty connector, seems like apples and pears because all individual strands keep conducting and the connector is not.
That Ls and or Rs may change per individual strand because of oxidation can be agreed.
But take a stranded wire from a different manufacturer and you will find complete different figures for Ls/Rs versus frequency.
So far there is not even an absolutely consensus about an optimum for Ls/Rs behaviour as far as I know.
So if the absolute value of Ls/Rs or its dependency versus frequency are influencing the sound, the consequence would be that each cable would have a different sound signature.
If that being true, we would still have no clue for the cause of these possible acoustical differences and thus why oxidation worsens the sound reproduction.
Hans
Last edited:
There is a distinct difference between critical thinking and trashing others. He will trash others when he has no ability to discuss. Just ask so many here who become his targets. Honestly, I cannot understand how he is allowed to trash this thread on a consistent basis.
It is virtually impossible to measure distinct differences in manufacturers products off the shelf, as the ability to measure below 5% Ls/Rs accuracy in the audio band with a 5 meter cable is beyond practically everybody, reactance is just too low.
Your statement "as all individual strands keep conducting" is very telling. It is not strand conduction along the length that is the concern as in DC, but rather how the interstrand contact integrity affects the current distribution during dI/dt.
As to the Ls/Rs of the cable affecting the sound, that is what this discussion is about, yes?
I do not claim it makes an audible difference. I have detailed the real world mechanisms, how to test, what to look for, level of effect possible, as well as procedural errors that can spoil the test accuracy. The last sentence born of 25 years of actual measurements necessary to advance SOTA.
Jn
Last edited:
Bob is going to be sooooo disappointed, but really he should have seen it coming.
A few pages back:
If that’s “denigrating”, then indeed I’m too old for this sh1t.
- Status
- Not open for further replies.