Steel or other ferrous metals in wiring and connectors

[howard416]

So many of you probably are aware of that incident with NC252MP amps where some of those units were unintentionally assembled with binding posts that contained iron or steel or whatever kind of ferrous metal in them, which caused unusually-high distortion. My question is, when does the steel actually make a difference?

Examples:

- Speakon connectors (the cable side, anyway) typically have screws terminals, I bet the screws themselves are steel. Though they press down on cuprous tabs, I believe. And I keep hearing that they deliver the best distortion performance, even over gold-plated bananas.
- Wago lever nuts. It seems the conductor bar is cuprous but I'm sure the spring that acts on the lever/clamp is steel.
- Basically anything that uses a screw for tightening. Except some banana plugs, I'm sure they're not steel because I've stripped so many too dang easily.

Is it that the steel can be in electrical contact but not actually have any significant current go through it? Too many places where steel shouldn't be, if this isn't the case.

 

[MaxwellsEq]

So many of you probably are aware of that incident with NC252MP amps where some of those units were unintentionally assembled with binding posts that contained iron or steel or whatever kind of ferrous metal in them, which caused unusually-high distortion. My question is, when does the steel actually make a difference?

Examples:

- Speakon connectors (the cable side, anyway) typically have screws terminals, I bet the screws themselves are steel. Though they press down on cuprous tabs, I believe. And I keep hearing that they deliver the best distortion performance, even over gold-plated bananas.
- Wago lever nuts. It seems the conductor bar is cuprous but I'm sure the spring that acts on the lever/clamp is steel.
- Basically anything that uses a screw for tightening. Except some banana plugs, I'm sure they're not steel because I've stripped so many too dang easily.

Is it that the steel can be in electrical contact but not actually have any significant current go through it? Too many places where steel shouldn't be, if this isn't the case.

It's worth looking at the numbers involved. The distortion should have been very, very low, yet the steel crimp connector made it higher. BUT, even with the steel connector, the distortion was still very low and inaudible! For many amplifiers with higher natural distortion, the steel crimp would not show up in the measurements. These figures are tiny compared to the distortion inherent in loudspeakers and turntable pickup cartridges.

 

[somebodyelse]

For those that aren't aware, see here for the difference in distortion attributed to something ferrous (tabs IIRC) on the original binding posts Buckeye were using around 2021. They changed this once the cause was discovered. The results linked were from tests on an older 6 channel NC502MP based model with the original posts and a newer 8 channel NC502MP based model with the new posts. We're talking a few dB difference at around -90dB so you need a good amp for it to show up at all. Similarly when measuring high performance amps minor distortions caused by the dummy load become significant, as described by our host.

I don't think we've seen a direct comparison between the different binding posts on the same amp board to rule out sample to sample variations, or an investigation into whether the ferrous material needs to be conducting the signal, in contact with it (like the clamping screw) or just nearby.

 

[somebodyelse]

From internal pictures of the Benchmark AHB2 it uses PCB mount speakONs not the ones with screws.

 

[quattro98]

I am not sure, but I think the binding post issue was that the steel was the conductor. It had a small effect at high power. I doubt that having steel screws or clamps next to the brass or copper conductor matters at all, except when seeing if a magnet sticks.

 

[howard416]

From internal pictures of the Benchmark AHB2 it uses PCB mount speakONs not the ones with screws.

Yeah, sorry, to be clear, the Speakon jacks are almost never an issue in this regard; they will be typically be PCB mount or with QD tabs. And I doubt the tabs are ferrous. But the cable-mount Speakon connectors (i.e. plugs) are the ones with screws nowadays. I think older generations of Speakon had the (featured, anyway) ability to solder the wires, but seems screws-only now.

NL2MPXX

This connector is fully compliant to the device standard IEC 62368-1, due to its component certification according to IEC 61984.2 pole chassis connector, black D-size flange, countersunk thru holes, 3/16" flat tabs. Does not intermate with the 4-pole cable connector.The speakON panel mount...

www.neutrik.com

 

[MaxwellsEq]

I doubt that having steel screws or clamps next to the brass or copper conductor matters at all, except when seeing if a magnet sticks

I agree. I think the ferromagnetic item needs to be in the signal / return path (so conducting). Screws, clamps, clips etc. will not have the same effect.

 

[somebodyelse]

I think older generations of Speakon had the (featured, anyway) ability to solder the wires, but seems screws-only now.

That was my recollection too, but couldn't see it in current docs. It's still there for the now discontinued NL4FC though:
https://www.neutrik.com/en/product/nl4fc

The industry-standard for loudspeaker connections offer extremely reliable and robust cable connectors with a reliable locking system. They feature solid contacts with screw-type terminals including a stranded wire protection which offer also solder termination.

RS still list the NL8FC as solder terminated too.

 

[Speedskater]

The original statement is suspect. Are there any replications of the test? An audio power amplifier's output impedance is a fraction of one Ohm, it would take a large amount of magnetic material to have an impact.

Many audio systems use cable TV RG6 coax for audio interconnects. Most cable TV RG6 has a Copper Clad Steel central conductor. Where are all the distortion measurements for these interconnect cables.

 

[howard416]

The original statement is suspect. Are there any replications of the test? An audio power amplifier's output impedance is a fraction of one Ohm, it would take a large amount of magnetic material to have an impact.

Many audio systems use cable TV RG6 coax for audio interconnects. Most cable TV RG6 has a Copper Clad Steel central conductor. Where are all the distortion measurements for these interconnect cables.

See somebodyelse's post above with "see here" link.

Also, and I didn't know this until looking it up just now, a lot of Canare's coax cable has stranded center conductors. I assume it is all-copper. And likely the iron distortion is proportional to the amount of current? Which is very small in interconnects, of course.

 

[kyuu]

See somebodyelse's post above with "see here" link.

Maybe I'm missing something but to my eye the difference in those distortion curves is incredibly minor, and that was with the steel binding posts which were obviously in the direct signal path. Not sure why you'd worry about steel screws just being in the vicinity somehow causing an issue.

 

[howard416]

Maybe I'm missing something but to my eye the difference in those distortion curves is incredibly minor, and that was with the steel binding posts which were obviously in the direct signal path. Not sure why you'd worry about steel screws just being in the vicinity somehow causing an issue.

Perhaps there is another chart or other info available that I don't readily have/know about? I seem to remember a pretty big hubbub about it, and it may be in other scenarios that the performance more significantly compromised.

Honestly I was just going off my recollection, regretfully I did not actually go through all the history before making this post

 

[Sokel]

Story goes that some seasoned engineers here got it right away by the type of distortion,apparently it's a known issue going way back in time.
That's all.

 

[howard416]

Ah, I was totally wrong. Not about the performance degradation but about which model.

Original, distortion rise: https://www.audiosciencereview.com/...ds/buckeye-3-channel-purifi-amp-review.40293/

Fixed: https://www.audiosciencereview.com/...-3-channel-purifi-amplifier-review-2nd.43834/

 

[Speedskater]

For the test to be the least bit meaningful, it has to be:
a] same day
b] same test bench & setup.
c] same unit with the only change being the type of metal.

 

[Speedskater]

Also, and I didn't know this until looking it up just now, a lot of Canare's coax cable has stranded center conductors. I assume it is all-copper. And likely the iron distortion is proportional to the amount of current? Which is very small in interconnects, of course.

It's the coax cable that comes off the cable company's installers truck that has the Copper Clad Steel solid central conductor.
Over the years, I have read about many people using it for audio cables.
* * * * * * * * * * * *
for analog RCA interconnects it doesn't matter if the central conductor is stranded or solid, thick or thin.

 

[howard416]

* * * * * * * * * * * *
for analog RCA interconnects it doesn't matter if the central conductor is stranded or solid, thick or thin.

Right. But if it's stranded (in the sizes we're talking about), chances are much lower that there's any steel in the conductor.