Linn Forums

Current time: 2017-12-14, 19:54 Hello There, Guest! (LoginRegister)

Linn Forums / Linn Music Systems & Hi-fi Separates / Network Music Players & Music Streamers v / Musing in the eye of Hurricane PaulSurround

Post Reply 
Musing in the eye of Hurricane PaulSurround
2017-09-02, 12:43
Post: #11
RE: Musing in the eye of Hurricane Paul
(2017-09-02 11:45)Briain Wrote:  Packet drops would require quite a large signal to be coupled into the cat5, so I doubt that's what's happening. If packet re-sends were to create any audible impact, then the easiest way to test for that would be to heavily load the network segment up with traffic (thus causing a few collisions and re-sends) and see if it changes the sound of the DS streaming. As I say, I doubt that's what's happening and in any case, that would be masked by the DS's data buffer, so by [thought] process of elimination, it leaves only what's sometimes referred to as the 'pin 1 problem' (though in this case, via the Ethernet socket). Google will find many articles on same (including radio ham microphone cables) and when I just tried it, the first link it found was on the Rane site (see here: http://www.rane.com/note165.html) which I haven't read, but a quick scroll down the page indicated it could be quite a good one for describing what's going on. Of course, an Ethernet cable doesn't typically have a screen (which terminates to the board instead of the case) so it's not exactly as described on that page, but it is 4 unscreened pairs extending into the case and terminated on the Ethernet chip (so more like 4 pin one problems Tongue ) and though I assume that Linn (and the Ethernet chip designer) have taken measures to deal with that possibility, I can only assume that once inside, a tiny fraction of it is being perhaps coupled into areas beyond that interface (as unless I'm missing something very esoteric about it all, I really can't think of any other mechanism whereby patch cables - or the screening thereof - could make an audible difference). Of course, I would be happy to be proved wrong, as the answer would be even more fascinating than a variant on the 'pin 1 problem'! Smile

Waving a radio about is quite interesting. The thing I like about the bank of FRIWO units is that they're quiet close up, but once you get a few feet from them, they are pretty much undetectable (I was using a nice old Roberts Radio for the test). Smile

The Sky+ box was a real racket producing horror, but I'm not sure if they're all like that, or whether my one has lost part of its mains filter (that can happen with SMPS units, be they the ones inside products, or of the wall-wart variety). I know that one of my friends has a selection of Sky boxes (he keeps them for helping out when a customer's one fails) so I must borrow one of the same type and see if it's similarly noisy.

It would be handy to identify a tight-braid coax (that you can buy by the metre, on line) as the stuff I used was a bespoke one used for the IF of NEC microwave radio units (similar arrangement to the coaxial link between a sky box and the LNB) and it has twin screens with almost no gaps between the wire strands (though for this sort of stuff, I think almost any grade of screen would more than suffice).

To cover it, I used something like the RS stuff (http://uk.rs-online.com/web/p/cable-sleeves/0408237/) and a short length of clear heat-shrink at each end (to keep it in place and stop it from fraying).

When I looked at the Sky+ box, I discovered that the case is grounded through the HDMI connection to my ADSM, so I terminated the box end of screen with a lug and fitted it under one of the existing screws at the back (so effectively 'extending' the Sky+ box's chassis to also surround its mains cable). Once you remove all the plastic, the Sky+ box is actually quite nicely screened by an internal metal box (only the back of which is left visible).

Bri

PS Poking the short wave radio at the front of the DS display window is rather fun (remember that you have to rotate it as the radio's internal ferrite rod antenna is directional) as you then hear the (actually quite funky) sounds from the digital works within. It's actually quite a musical cacophony, but I guess you would expect nothing less from a KDS! Big Grin

Facinating stuff Briain.

I'll do some more experiments when I get home. Big Grin

Exakt Surround 5.0:
Katalyst Akubariks, AEDSM, Akurate Exaktbox 6/4200 for 225 Centre, Majik Exaktbox I for 212 Surrounds, OPPO 105 on StillPoints.
Power: Environmental Potentials 2460, Shunyata Triton, Zitron Cobras, Alpha Digitals
Room Acoustics: SPACE, SubDude HT's, Acoustic Panels
Find all posts by this user
Quote this message in a reply
2017-09-02, 13:00 (This post was last modified: 2017-09-02 15:12 by Briain.)
Post: #12
RE: Musing in the eye of Hurricane Paul
The first paragraphs on the Wikipedia page about the Ethernet 'transmission process' CSMA/CD (Carrier-sense multiple access with collision detection) gives a summary of how it works:

Quote:Carrier-sense multiple access with collision detection (CSMA/CD) is a media access control method used most notably in early Ethernet technology for local area networking. It uses a carrier-sensing scheme in which a transmitting station detects collisions by sensing transmissions from other stations while transmitting a frame. When this collision condition is detected, the station stops transmitting that frame, transmits a jam signal, and then waits for a random time interval before trying to resend the frame.

CSMA/CD is a modification of pure carrier-sense multiple access (CSMA). CSMA/CD is used to improve CSMA performance by terminating transmission as soon as a collision is detected, thus shortening the time required before a retry can be attempted.
(source: https://en.wikipedia.org/wiki/Carrier-sense_multiple_access_with_collision_detection)

So on a busy network segment (as in one carrying lots of actual traffic, not just lots of idle machines sitting on it) packet collisions and re-sending are expected activities. Interestingly, 802.11 (AKA WiFi) does not work in the same way, as with there being far less bandwidth available, such a re-sending scheme would - on a busy access point - end up being rather wasteful, so they instead use a collision avoidance system to make a more efficient use of the available capacity, but that is yet another story (and rather a jolly long - though very interesting - one, at that)! Smile

To do the test properly (as in see if excess packet re-sending makes any difference) you'd want to use an olde worlde Ethernet hub (as opposed to a switch) to ensure you had the worst possible scenario, then stream a few videos on a few PCs (with everything connected to the hub) or if you had a few NAS units, a full NAS to empty NAS backup would be a great way to generate lots and lots of traffic. Then see if that makes any impact on the sound of the third NAS streaming to the DS, though I'd doubt that it will due to the DS's buffer (unless you generated so much traffic that it prevented the buffer filling fast enough, then you'd get music drop-outs). Of course, all that activity would also mean that the network card in the DS was being rattled quite heavily, so that could also influence the outcome of the test. Wink

Anyhow, as per the previous post, I'd argue that if anything is going on, then likely it's down to high frequency noise getting in and impacting on something beyond the Ethernet interface (rather than packet drops and re-sends) so when time permits, that's what I want to have a bash at measuring (and experimenting with), mostly just for the fun of doing it.

Bri

KDS/1 (music) + ADSM (AV) -> KK/1 -> 350A + miniDSP time & phase aligned 345 rear sub
KDS Renew -> Homebrew fixed attenuator -> 2250/D -> 212 and Sizmik front sub (bedroom)
MDSI -> 104C (awaiting installation in my kitchen)
MDSI -> Shahinian Arc (installed at my mum's house)
Find all posts by this user
Quote this message in a reply
2017-09-03, 11:56
Post: #13
RE: Musing in the eye of Hurricane Paul
(2017-09-02 13:00)Briain Wrote:  The first paragraphs on the Wikipedia page about the Ethernet 'transmission process' CSMA/CD (Carrier-sense multiple access with collision detection) gives a summary of how it works:

Quote:Carrier-sense multiple access with collision detection (CSMA/CD) is a media access control method used most notably in early Ethernet technology for local area networking. It uses a carrier-sensing scheme in which a transmitting station detects collisions by sensing transmissions from other stations while transmitting a frame. When this collision condition is detected, the station stops transmitting that frame, transmits a jam signal, and then waits for a random time interval before trying to resend the frame.

CSMA/CD is a modification of pure carrier-sense multiple access (CSMA). CSMA/CD is used to improve CSMA performance by terminating transmission as soon as a collision is detected, thus shortening the time required before a retry can be attempted.
(source: https://en.wikipedia.org/wiki/Carrier-sense_multiple_access_with_collision_detection)

So on a busy network segment (as in one carrying lots of actual traffic, not just lots of idle machines sitting on it) packet collisions and re-sending are expected activities. Interestingly, 802.11 (AKA WiFi) does not work in the same way, as with there being far less bandwidth available, such a re-sending scheme would - on a busy access point - end up being rather wasteful, so they instead use a collision avoidance system to make a more efficient use of the available capacity, but that is yet another story (and rather a jolly long - though very interesting - one, at that)! Smile

To do the test properly (as in see if excess packet re-sending makes any difference) you'd want to use an olde worlde Ethernet hub (as opposed to a switch) to ensure you had the worst possible scenario, then stream a few videos on a few PCs (with everything connected to the hub) or if you had a few NAS units, a full NAS to empty NAS backup would be a great way to generate lots and lots of traffic. Then see if that makes any impact on the sound of the third NAS streaming to the DS, though I'd doubt that it will due to the DS's buffer (unless you generated so much traffic that it prevented the buffer filling fast enough, then you'd get music drop-outs). Of course, all that activity would also mean that the network card in the DS was being rattled quite heavily, so that could also influence the outcome of the test. Wink

Anyhow, as per the previous post, I'd argue that if anything is going on, then likely it's down to high frequency noise getting in and impacting on something beyond the Ethernet interface (rather than packet drops and re-sends) so when time permits, that's what I want to have a bash at measuring (and experimenting with), mostly just for the fun of doing it.

Bri

I suspect you are correct.

I was reading about issues cable installers were having with networks when they had run Ethernet cables bundled with mains cables, and that was the explanation they gave.

Electromagnetic interference is probably the real culprit.

Exakt Surround 5.0:
Katalyst Akubariks, AEDSM, Akurate Exaktbox 6/4200 for 225 Centre, Majik Exaktbox I for 212 Surrounds, OPPO 105 on StillPoints.
Power: Environmental Potentials 2460, Shunyata Triton, Zitron Cobras, Alpha Digitals
Room Acoustics: SPACE, SubDude HT's, Acoustic Panels
Find all posts by this user
Quote this message in a reply
2017-09-03, 12:37 (This post was last modified: 2017-09-03 12:40 by Briain.)
Post: #14
RE: Musing in the eye of Hurricane Paul
(2017-09-03 11:56)Paulssurround Wrote:  I suspect you are correct.

I wouldn't assume that; I very seldom am!

(2017-09-03 11:56)Paulssurround Wrote:  I was reading about issues cable installers were having with networks when they had run Ethernet cables bundled with mains cables, and that was the explanation they gave.

Electromagnetic interference is probably the real culprit.

Well the situation will get worse as you add more and more mains cables (particularly so in the age where many of them will be carrying lots of noise from SMPS devices) and as the length over which they are bundled with the cat5; both these factors will increase the level of noise coupled into the cat5. As far as I am aware, in the UK you are not permitted to run data cables in the same trunking as mains cables (for safety reasons, not noise reasons).

Bri

PS There was an interesting programme on the Burj Al Arab hotel (I think it might have been part of the 'Megastructures' series) where in one part of it, they showed a huge equipment room dedicated to power factor correction and from memory, the interviewee stated that without PFC, the noise from SMPS and light dimmers (etc) would cause the cables (e.g. bunched cables running between floors) to overheat and likely catch fire. It was quite a while back, but I'm sure that it occupied most of a floor! Hoping to find a photograph of the room, I did just rattle a search engine, and though I didn't, I did find what might be an interesting document about it (link below) so I'll read that later this evening.

https://library.e.abb.com/public/ed3d8f7f20a380c4c1256ddd00346c66/46-51%20M644.pdf

KDS/1 (music) + ADSM (AV) -> KK/1 -> 350A + miniDSP time & phase aligned 345 rear sub
KDS Renew -> Homebrew fixed attenuator -> 2250/D -> 212 and Sizmik front sub (bedroom)
MDSI -> 104C (awaiting installation in my kitchen)
MDSI -> Shahinian Arc (installed at my mum's house)
Find all posts by this user
Quote this message in a reply
2017-09-03, 12:56
Post: #15
RE: Musing in the eye of Hurricane Paul
(2017-09-03 12:37)Briain Wrote:  
(2017-09-03 11:56)Paulssurround Wrote:  I suspect you are correct.

I wouldn't assume that; I very seldom am!

(2017-09-03 11:56)Paulssurround Wrote:  I was reading about issues cable installers were having with networks when they had run Ethernet cables bundled with mains cables, and that was the explanation they gave.

Electromagnetic interference is probably the real culprit.

Well the situation will get worse as you add more and more mains cables (particularly so in the age where many of them will be carrying lots of noise from SMPS devices) and as the length over which they are bundled with the cat5; both these factors will increase the level of noise coupled into the cat5. As far as I am aware, in the UK you are not permitted to run data cables in the same trunking as mains cables (for safety reasons, not noise reasons).

Bri

PS There was an interesting programme on the Burj Al Arab hotel (I think it might have been part of the 'Megastructures' series) where in one part of it, they showed a huge equipment room dedicated to power factor correction and from memory, the interviewee stated that without PFC, the noise from SMPS and light dimmers (etc) would cause the cables (e.g. bunched cables running between floors) to overheat and likely catch fire. It was quite a while back, but I'm sure that it occupied most of a floor! Hoping to find a photograph of the room, I did just rattle a search engine, and though I didn't, I did find what might be an interesting document about it (link below) so I'll read that later this evening.

https://library.e.abb.com/public/ed3d8f7f20a380c4c1256ddd00346c66/46-51%20M644.pdf

Damn dimmer switches!!!

Perhaps you could invent a new type of dimmer switch that doesn't cause issues, make your fortune, and upgrade to Exakt Katalyst with 2 massive subs controlled by an Exakt subbox? Tongue

Perhaps a digital dimmer switch, powered by your newly invented zero noise SMPS power device? Rolleyes

Exakt Surround 5.0:
Katalyst Akubariks, AEDSM, Akurate Exaktbox 6/4200 for 225 Centre, Majik Exaktbox I for 212 Surrounds, OPPO 105 on StillPoints.
Power: Environmental Potentials 2460, Shunyata Triton, Zitron Cobras, Alpha Digitals
Room Acoustics: SPACE, SubDude HT's, Acoustic Panels
Find all posts by this user
Quote this message in a reply
2017-09-03, 13:29 (This post was last modified: 2017-09-03 13:32 by Briain.)
Post: #16
RE: Musing in the eye of Hurricane Paul
Sitting right next to me is a beautiful old variac transformer; that would dim without any creating any additional noise. This one came from a 1950's microwave radio system (one that occupied a suite of racks) but if you Google 'variac', you'll see some pictures of what they are, but basically, it's like a very big mains power volume control (but a variable transformer as opposed to a variable resistor). I'm about to use this to feed an old item of test equipment (a late 1970's HP spectrum analyser) at 1/3 mains Voltage for 15 minutes, then 2/3 mains Voltage for 15 minutes, then if all looks okay, I'll whack in the full mains Voltage (to help avoid the elderly capacitors from having a crisis). Wink

The one I have would not look at all out of place in an old, black and white Frankenstein film.

KDS/1 (music) + ADSM (AV) -> KK/1 -> 350A + miniDSP time & phase aligned 345 rear sub
KDS Renew -> Homebrew fixed attenuator -> 2250/D -> 212 and Sizmik front sub (bedroom)
MDSI -> 104C (awaiting installation in my kitchen)
MDSI -> Shahinian Arc (installed at my mum's house)
Find all posts by this user
Quote this message in a reply
2017-09-03, 14:54
Post: #17
RE: Musing in the eye of Hurricane Paul
(2017-09-03 13:29)Briain Wrote:  Sitting right next to me is a beautiful old variac transformer; that would dim without any creating any additional noise. This one came from a 1950's microwave radio system (one that occupied a suite of racks) but if you Google 'variac', you'll see some pictures of what they are, but basically, it's like a very big mains power volume control (but a variable transformer as opposed to a variable resistor). I'm about to use this to feed an old item of test equipment (a late 1970's HP spectrum analyser) at 1/3 mains Voltage for 15 minutes, then 2/3 mains Voltage for 15 minutes, then if all looks okay, I'll whack in the full mains Voltage (to help avoid the elderly capacitors from having a crisis). Wink

The one I have would not look at all out of place in an old, black and white Frankenstein film.

Why do I get the feeling that Edinburgh is about to experience a catastrophic power failure? Tongue

Exakt Surround 5.0:
Katalyst Akubariks, AEDSM, Akurate Exaktbox 6/4200 for 225 Centre, Majik Exaktbox I for 212 Surrounds, OPPO 105 on StillPoints.
Power: Environmental Potentials 2460, Shunyata Triton, Zitron Cobras, Alpha Digitals
Room Acoustics: SPACE, SubDude HT's, Acoustic Panels
Find all posts by this user
Quote this message in a reply
2017-09-04, 09:48
Post: #18
RE: Musing in the eye of Hurricane Paul
Wiltshireman kindly gave me a length of electrical mesh shielding, which I have now wrapped two of my ethernet cables in. The switch->DSM, and the DSM->Exaktbox. I have no idea if it has made a difference. But seeing as both runs run close to mains cables, no harm Smile

Akurate EDSM | Majik ExaktBox-I [on Custom Design iRAPs] | Majik 140, REL T5
Rega RP1, [QNAP HS-251+, Jitterbug, MinimServer, Asset], Kazoo [Linx1010]

AV: Marantz NR1506 | B&W 620(C), M1(S), REL T5 (LFE) | Fronts powered by Linn
Net: NetGear DGN2200 | GS108 | AQ Forest x3

Hello to Jason Isaacs
Visit this user's website Find all posts by this user
Quote this message in a reply
2017-09-04, 14:04
Post: #19
RE: Musing in the eye of Hurricane PaulSurround
(2017-09-04 09:48)timster Wrote:  Wiltshireman kindly gave me a length of electrical mesh shielding, which I have now wrapped two of my ethernet cables in. The switch->DSM, and the DSM->Exaktbox. I have no idea if it has made a difference. But seeing as both runs run close to mains cables, no harm Smile

As a general rule of thumb,

Plan A: I seperate the power cords as far away from other cables as possible.

Plan B: Wrap sections of the cables that are close to power cords with layers of aluminum foil.

Plan C: I wrap Ethernet cables their entire length in aluminum foil. Big Grin

Exakt Surround 5.0:
Katalyst Akubariks, AEDSM, Akurate Exaktbox 6/4200 for 225 Centre, Majik Exaktbox I for 212 Surrounds, OPPO 105 on StillPoints.
Power: Environmental Potentials 2460, Shunyata Triton, Zitron Cobras, Alpha Digitals
Room Acoustics: SPACE, SubDude HT's, Acoustic Panels
Find all posts by this user
Quote this message in a reply
2017-09-04, 15:19
Post: #20
RE: Musing in the eye of Hurricane PaulSurround
(2017-09-04 14:04)Paulssurround Wrote:  
(2017-09-04 09:48)timster Wrote:  Wiltshireman kindly gave me a length of electrical mesh shielding, which I have now wrapped two of my ethernet cables in. The switch->DSM, and the DSM->Exaktbox. I have no idea if it has made a difference. But seeing as both runs run close to mains cables, no harm Smile

As a general rule of thumb,

Plan A: I seperate the power cords as far away from other cables as possible.

Plan B: Wrap sections of the cables that are close to power cords with layers of aluminum foil.

Plan C: I wrap Ethernet cables their entire length in aluminum foil. Big Grin

You might as well go the full mile and also adopt plan D

Plan D: Wear a foil hat.

I don't think it necessary to wear a foil hat in order to keep radio signals from damaging my brain (it was shot to bits many years ago), but when I am seen wearing one whilst standing on my balcony and waving a dangerous looking antenna about, it certainly does seriously perturb my neighbours, so I reckons it's well worth wearing one for that reason, alone! Big Grin Big Grin

KDS/1 (music) + ADSM (AV) -> KK/1 -> 350A + miniDSP time & phase aligned 345 rear sub
KDS Renew -> Homebrew fixed attenuator -> 2250/D -> 212 and Sizmik front sub (bedroom)
MDSI -> 104C (awaiting installation in my kitchen)
MDSI -> Shahinian Arc (installed at my mum's house)
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)