US Audio MartUS Audio Mart
It is currently Wed Oct 17, 2018 12:40 am

All times are UTC - 8 hours




Post new topic Reply to topic  [ 248 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 17  Next
Author Message
PostPosted: Wed Jun 13, 2018 6:48 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Quote:
Speedskater
Junior Member
Speedskater
1,073 posts
Report post
Posted September 12, 2017
Many pros are now going to unshielded bonded-pair Cat7 for balanced interconnect cables.
:mrgreen:

Quote:
Speedskater
Junior Member
Speedskater
1,073 posts
Report post
Posted September 13, 2017
Cat6 or Cat7 for a 4 channel balanced interconnect cable is fine, but for a current carrying DC cable not so much.

While Cat3 or Cat5 can mach a nice (and almost free) speaker cable, you have to parallel several cables.
Oh really? So much for the 'expurts' advice! Give PoE DC a try! :lol: :lol: :lol:


Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2018 6:55 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Quote:
BigGuy
Sophomore Member
BigGuy
438 posts
Report post
Posted September 23, 2017
On 5/4/2017 at 5:26 PM,

Thank you, John. This has got to be one of the best DIY improvement posts and one I will be trying.

A couple of questions...

For John: Given the thread topic, the JSSG has great applicability to DC power cables. Does it also offer value for other cables, e.g., USB, interconnects, AC power cables, etc., and, if so, how would they rank?

Which would be better...tinned or copper braid?



Quote:
JohnSwenson
Junior Member
JohnSwenson
1,433 posts
Report post
Posted September 23, 2017
The JSSG is primarily effective at lower frequencies. As mentioned in my post there is enough parasitic capacitance in most shielded cables to provide shielding at high frequencies. Where traditional doesn't work well is low frequencies. SO audio cables are a great place for it. High frequency cables (such as USB etc) no so much. If done right it would be effective at shielding power line frequency interference for high speed cables (which MAY be useful), BUT unless the cable was specifically designed for it adding it to an existing cable will more than likely mess up the high frequency performance of the cable.

John S.


Yet that didn't stop anyone from JSSGing Curious USB cables on CA! Heretics!
LOL!

That low freq shielding has yet to be proven for JSSG - now Mu Metal braided shielding has been tested to be effective.

Quote:
BigGuy
Sophomore Member
BigGuy
439 posts
Report post
Posted October 11, 2017
On 10/11/2017 at 5:59 AM, Speedskater said:
Fair Radio is not an internet store. It's a dusty old building with stuff sitting on shelf's for decades.

The braid (in random lengths & sizes) is in a 55 gallon cardboard drum near the door.

Send them an e-mail or phone.

Ah! That would explain why could not find "barrel" on their list! :D

I found this supplier of copper braid...

<https://www.techflex.com/land_metal.asp>


Speaking of copper braid, got some clarification on use of Mu-metal. Mu-metal is best used for EMI, which is low frequency. Copper is the best for RFI, which is high frequency and likely the biggest issue with cables.
Nailed it but was ignored.. as the JSSG frenzy continued! You need both kinds of shields Mu-Metal and tinned copper!


Now here comes Cornan the LDover LT3045 maven - a year late. Nice 'discovery' mate :roll: :roll: :roll: You guys missed out on these LT3045 babies for too long - should have been reading here... :lol:
Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,658 posts
Report post
Posted November 12, 2017
On 11/12/2017 at 9:02 AM, charlesphoto said:

Do you think I could go with a single 1A LT3045 set to 9V vs a series? Trying to save a few $ here and there (still have the rendu 1.4 board upgrade at the top of the list) and would like to put in couple of these in different places (office FMC as well). Thanks, CP


Yes, the LT3045s works nicely in single as well. The additional LT3045 in series just reduce noise a little further. Just mind that the LT3045s needs drop-down voltage. If you keep the drop-down to 0.7v - 1v they will not even get warm IME.

BTW. I have plans to try the LT3045s in series with one board on each side of the DC cable instead of directly in series at the powered device. It will be interesting to see if that improves it further? I am just waiting for additional plugs and sockets to get that project going.


Quote:
sandyk
Masters Level Member
sandyk
10,976 posts
Report post
Posted Sunday at 10:40 PM
On 6/10/2018 at 10:35 PM, Cornan said:
Trust me. You won’t! Just try it and see how you like it. I can almost guarantee that you’ll like them with the LT3045. In single or series.
Micael


Not everybody uses ONLY a HF detail challenged streaming service !!!

Your solution may work well with Spotify for you, but may NOT be suitable for those who don't rely just on streamed services.
Classic CA entertainment! :lol: :lol: :lol:
Cheers!


Last edited by Tubelover2 on Wed Jun 13, 2018 12:46 pm, edited 1 time in total.

Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2018 9:31 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
New PoE DC thread up!

viewtopic.php?f=35&t=1465


Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2018 11:39 am 
Offline

Joined: Fri Jan 27, 2017 3:14 pm
Posts: 207
Location: Surrey, UK
Tubelover2 wrote:
Hi M, Well I disagree in this case. In fact my experience with silver/teflon vs stranded copper DC cables when both are in the same star quad shielded configuration is the silver/teflon wins the day SQ wise.


Hi Rob,

Well, we're not disagreeing there. I too replaced my stranded copper with teflon/silver cables, some made by you. Just replaced them with PoE.

M


Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2018 12:27 pm 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Mr Underhill wrote:
Tubelover2 wrote:
Hi M, Well I disagree in this case. In fact my experience with silver/teflon vs stranded copper DC cables when both are in the same star quad shielded configuration is the silver/teflon wins the day SQ wise.


Hi Rob,

Well, we're not disagreeing there. I too replaced my stranded copper with teflon/silver cables, some made by you. Just replaced them with PoE.

M
Hi M. I use a combination - PoE DC into the LT3045 box - the star quad silver teflon out to the IR, etc...

From these posts from JS's DIY thread - it looks like the high inductance of the twisted pair small gauge CAT 7 & 8 may be the secret. Which goes against all that the experts, JS included say. It's also the function of the varied twist rates used on the pairs in the Ethernet cable - versus the star quad uniform twist.
Attachment:
twist-rate.png
twist-rate.png [ 4.87 KiB | Viewed 1414 times ]

Cheers!


Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 13, 2018 1:11 pm 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
This guy Cornan gives out so much bad information on the LT3045's it's ridiculous and keeps repeating this bad info in multiple threads. Lame-o thinks he's the tweeker king and is just a poser.
Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,663 posts
Report post
Posted December 30, 2017
On 12/30/2017 at 4:32 PM, JohnSwenson said:
You can't just parallel two voltage sources, the higher value one will supply most of the current, which kind of negates the reason for using two in the first place.

To do it right you make sure the voltages are very close and then you put a small resistance in series with each. The differences in voltages cause slightly different currents to flow through the "ballast" resistors. The closer the voltages the more equally matched the current sharing. (or alternatively the larger the resistors the better the sharing, BUT that negates the advantage of the extremely low output impedance of the LT3045).

The LT3045s are uniquely good at current sharing because of the way they work, which is VERY different than most regulators. Putting more than one together ON THE SAME BOARD, guarantees the output voltages will be VERY close, so only very small ballast resistors are needed. (literally a half inch of trace on the board) BUT that only works when they are on the same board. You don't get that advantage when paralleling two boards. With two boards it depends on the accuracy of the voltage set resistors and the accuracy of the current source in the chip. This can be a WAY bigger spread than what you get when you parallel the chips on the same board.

Thus for LT3045s you are much better off using a single board with fours chips rather than using two boards each with two chips.

John S.


Does this mean that even if the two LT3045 boards are completely matched (voltage and amperage) from the maker they still need resistors in series with each board?

I understand that a single board is optimal, but does that neccessarily make two separate boards a bad idea? Better without it completely if you use it in the DC cable path (close to the powered device)? The only thing you want to acheive using two boards is for devices requiring more than 1A.
Confusing parallel with series - and LDovr offers a dual parallel LT3045 board that does 1A - has been for months before his clueless posts. This guy is the joker of jokers. Acts like he's the LT3045 inventor. :roll:

Like this recent one on the PoE DC thread at CA:
Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,663 posts
Report post
Posted May 24
On 5/24/2018 at 12:00 PM, mansr said:
You can also put two (or more) in parallel to both reduce noise and increase current.

Both series and parallel reduces ripple to a minimum.
http://www.diyaudio.com/forums/vendor-s ... ost5424495

Series is simple. Parallel is not as simple but this post might help you to acheive it if you do not want to buy Stammheims HPULN or HC-HPULN.

Also Alexey have a parallel LT3045 in the pipeline. You just need to wait for it!
This was posted on May 24th of this year! Alexey (LDovr) has been selling dual parallel boards for nearly a year!
https://www.ebay.com/itm/LT3045-A-Ultra ... SwWLBZ9auH
Attachment:
s-l1600.jpg
s-l1600.jpg [ 120.46 KiB | Viewed 1406 times ]


The CA clueless leading the even more clueless - bad info all the way. The PSRR of the parallel LT3045's is only slightly better then a single
Not this silly statement - "Both series and parallel reduces ripple to a minimum." A single filters out 99.99% of the noise out to 1Mhz - adding a second takes that to 99.9999% but the LT3045's own 0.8uv of noise far exceed that benefit on all but the worst LPS's.

Here is he is called out and just adds so much confusion and bad information!
Quote:
look&listen
Freshman Member
look&listen
98 posts
Report post
Posted May 24
On 5/24/2018 at 12:49 PM, Cornan said:
Parallel is not as simple but this post might help you to acheive it if you do not want to buy Stammheims HPULN or HC-HPULN

Already parallel!
Alexeys = 2 LT3045s for 1A

HPULN = 6 parallel for 3A

HC-HPULN = 10 LT3045s parallel for 5A

Look on LT3045 data sheet, see SET pin on chip used as reference to parallel chips to work together.



But he doesn't quit the disinfo campaign:
Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,663 posts
Report post
Posted May 24
On 5/24/2018 at 1:08 PM, look&listen said:
Already parallel!

Alexeys = 2 LT3045s for 1A

HPULN = 6 parallel for 3A

HC-HPULN = 10 LT3045s parallel for 5A

Look on LT3045 data sheet, see SET pin on chip used as reference to parallel chips to work together.


Yes I know. I use them all!

Each LT3045 is 500mA and not parallel. Alexey is coming with a 1.5A version soon though.

To get the 500mA boards (or 1A) to match you’ll need to wire them according to my link above.
This guy is a piece of work - it makes me nauseaus to read his tripe and condescending attitude! "Yes I know. I use them all! " :roll: :roll: :roll:

Just simply amazing!

Then this JA 'expurt' snarkmaster comes in:
Quote:
mansr
Supreme Member
mansr
8,146 posts
Report post
Posted May 24
On 5/24/2018 at 1:08 PM, look&listen said:
Already parallel!

Alexeys = 2 LT3045s for 1A

HPULN = 6 parallel for 3A

HC-HPULN = 10 LT3045s parallel for 5A

Look on LT3045 data sheet, see SET pin on chip used as reference to parallel chips to work together.


Read more
And if none of those suit your needs, you can always make your own PCB.


God reading those CA threads to sift out a few decent nuggets of info is yeoman's work - and I need a few shots of Whiskey and a handful of Dramamine after an hour or so.


Top
 Profile  
Reply with quote  
 Post subject: Ferrites
PostPosted: Wed Jun 13, 2018 11:20 pm 
Offline

Joined: Fri Jan 27, 2017 3:14 pm
Posts: 207
Location: Surrey, UK
I tried some small ones on the PoE legs just before the powered device. I DIDN'T like the effect. Small, but I felt removed a tad of the relaxed nature of the sound I am getting ....removed.

I hadn't mentioned it before but I do use a couple of these smaller ferrites on the adapter between the IFI Mercury and the SingXer-F1, I will try adding a few more later.

I have bought 20 more larger ferrites which I will use:

1. On my Spdif, upping from 5 to 16; and
2. Add 5 to the USB cable connecting my server to the USB3.0 hub.

M


Top
 Profile  
Reply with quote  
 Post subject: Re: Ferrites
PostPosted: Thu Jun 14, 2018 5:01 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Mr Underhill wrote:
I tried some small ones on the PoE legs just before the powered device. I DIDN'T like the effect. Small, but I felt removed a tad of the relaxed nature of the sound I am getting ....removed.

I hadn't mentioned it before but I do use a couple of these smaller ferrites on the adapter between the IFI Mercury and the SingXer-F1, I will try adding a few more later.

I have bought 20 more larger ferrites which I will use:

1. On my Spdif, upping from 5 to 16; and
2. Add 5 to the USB cable connecting my server to the USB3.0 hub.

M
Nice M! I'm doing quite a few experiments with various ferrites on the PoE injector input and output sides - some with very good results. The key is getting the right freq suppression - not all ferrites are the same of course.

I use the iPur2 between the Mercury and F-1 - so it acts like as the adapter and has other SQ benefits. Would not live without it!

Cheers!


Top
 Profile  
Reply with quote  
PostPosted: Thu Jun 14, 2018 12:51 pm 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Best (well most ridiculus) Cornan CA posts ever:
Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,663 posts
Report post
Posted April 18
On 4/18/2018 at 10:32 AM, BigGuy said:
Thanks for the sanity check, John. Read some of the USB spec til my eyes glazed over. As confirmation, heard from one USB cable fabricator that he is no longer producing them due to the complexity. Obviously it is more than just connecting 4 wires to the USB connectors! Begs the question as to how many available cables actually meet the 90ohms, etc. of the spec!?


Don´t let sanity let you down or spoil the fun! ;) My ATL starquad USB cable beats the USPCB in my setup between Aries Mini and ISO Regen. If I would have listen to pure sanity I would have missed a good chunk of the fun. The experimenting bit is 50% or more of the fun for me though. Win or loose.


Quote:
Cornan
Experimental audiophile & curious musical explorer
Cornan
4,663 posts
Report post
Posted April 19
On 4/19/2018 at 7:52 AM

Fair enough, but in any case which ever USB cable you will DIY you will have difficulties to biuld them according to specs unless you have equipment to measure it. IME the impedance match is not the most crucial thing for SQ with USB.

USB cable meeting 90ohm spec not important! Huh? :roll: :roll: :roll:

Quote:
JohnSwenson
Junior Member
JohnSwenson
1,433 posts
Report post
Posted April 11
On 4/11/2018 at 10:00 AM, BigGuy said:
Questions re DIY USB...
Planning on doing starquad for the 5+/GND with JSSG.
Does it make sense to use starquad for the D+/D- or is twisted pair with JSSG more than adequate?
Planning on bundling signal and power separately.
Has the issue of cable lengths <1/2m or >1m been resolved to any extent? I have read that 1m is really not a good choice.



DIY USB cables are probably the least likely to be done right (ie if you don't know what you are doing it may not work at all). You can't hurt anything if it is not done right, it just doesn't work right. The probability of getting the basics right is pretty low, so it is a useless enterprise to make your own with expensive wire etc and find out it doesn't sound very good, you have no idea if it is the materials or the basic properties of the cable such impedance etc.

If you really want to DIY USB cables I recommend you have a TDR and know how to use it to get impedance right.


You are certainly free to play around with DIY USB cables, you MAY get something that sounds really good, but making sweeping general statements about materials, geometry etc would seem to not be a good thing to do.

John S.
Again JS recommends against star quad for USB - and the need for impedance matching...didn't stop Cornan from making one and not testing - and now it's better then JS's (Uptone's) own - $35 USPCB (USB connector)! Posting this on Uptone's own sponsor thread! And the push back there? Zero, Zilch, Nada...even from Alex. :shock:

You just can't make this up - his whole 'experimental' thread is 200 pages of this non-sense!

Fun? Maybe for him...but has little to do with serious audio improvements.

And it just goes on and on there...the reason I mention it...his newest speaker wire DC cable of course bests any PoE DC he hasn't heard and he posts every page with 9 pictures of his speaker wire creation on the PoE DC thread there...and on it goes at old CA!

The 'ISO Regen performance Improvement Cheap!' thread has gone completely off the rails...and why I felt PoE DC needed a 'real' thread with accurate, consolidated information about it. And it's off to a great start! Hundreds of views in the first few hours...

Cheers!


Top
 Profile  
Reply with quote  
PostPosted: Fri Jun 15, 2018 12:55 pm 
Offline

Joined: Sat May 06, 2017 1:21 pm
Posts: 31
Location: PLYMOUTH, MN, US
Lmitche seems frustrated at the direction the thread is taking. Imagine that. Maybe he should just 'go back where he came from'? Not the first...problematic... comment there I've received recently. I had Chris close my account. Only forum I've ever done so. Don't regret it. That well is poisoned. Fun to watch though.


Top
 Profile  
Reply with quote  
PostPosted: Fri Jun 15, 2018 5:02 pm 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
wushuliu wrote:
Lmitche seems frustrated at the direction the thread is taking. Imagine that. Maybe he should just 'go back where he came from'? Not the first...problematic... comment there I've received recently. I had Chris close my account. Only forum I've ever done so. Don't regret it. That well is poisoned. Fun to watch though.
Yeah that was really unfair to be so arrogant to you...I did my best to call him out. That didn't play well there. And I'm so glad to be out of there. That thread has turned into the Jabbr show - now pages about him - with no mention of PoE DC. He's to big a star there to lower himself to actually hear it...and some there have called him out. Ironically Larry being one! And then Alex to come to Jabberish's rescue and defense...I saw this train coming weeks ago when I posted on my old thread:

Quote:
Tubelover2
Post subject: Re: Ultra USB Chain USB/Ethernet Hybrid Startech/ICRON/LanRPostPosted: 14 May 2018 01:10
Online
User avatar

Joined: 16 Aug 2013 16:23
Posts: 1921
Location: Seattle, WA, US
lmitche wrote:
We should probably be careful about current limits with the 26 awg Ibra cable and other cat 7 solutions. The POE splitter puts current on 2 26 awg conductors on each side, for an effective awg of 23. That yields a max current of 0.729 amps according to the following table.

https://www.powerstream.com/Wire_Size.htm

That may explain why my Netgear switch is flakey when powered with the Ibra cable and POE splitters. However there is more then enough current for an ISO Regen.


Wrong - the .729 is for a 1000ft run

Run the calc - 1M 6VDC 1A = .158VDC drop. Very close to the 0.1 VDC I measure. 1A no problem - which is what I power my SSD with 1A 5VDC output LT3045. Those with out LT3045's are missing a key element of the SQ equation. Since the a 0.15 VDC voltage drop is negligible for a LT3045 being fed 8VDC or even 7VDC input.

I sense the no nothings hijacking this thread. Go post your BS on CA. Oh wait you're to cowardice to try that!
Fearing the SoTM flame throwers.

I dare you - JM you too - post what you posted here there. Would love to see SandyK's and Jabbr's reactions...do it, do it, do it.

Na Chickens!

Meanwhile stop posting claptrap here thanks!
Well they did and look at the last pages - just a joke!

And I see Alex is finally calling out Cornan for all his bogus information...now on R-cores:
https://www.computeraudiophile.com/foru ... ak-thread/

Quote:
Superdad
Music lover; Owner, UpTone Audio
Superdad
5,870 posts
Report post
Posted Wednesday at 12:28 AM
On 6/12/2018 at 1:29 PM, Cornan said:
Works great out of the box but will get hot and create a lot of EMI that you will need to deal with by dressing nearby sensitive components and cables in copper or shielding.


Toroidal transformers do not generate that much heat...



Sorry but much of what you wrote about R-core versus toroidal transformers is incorrect.

For example, temperature rise is much greater inside a torrid because the core is not exposed at any point so there is no surface area for it to dissipate.

Leakage flux from an R-core is extremely small due to balanced windings cancelling out stray flux.

Plus R-core efficiency is higher than a toroidal trans.

And those are just the superficial advantages...9_9


It's just another instance for fail and bad information Cornan spews on an hourly basis...his latest creation a pair of speaker wires for DC...says it beats all the PoE DC cables he hasn't heard. Exclamation marks are not a substitute for actual knowledge and trail.

Just the way CA rolls!

But as you say it's a lot of fun to watch from a distance. And why I started the dedicated PoE DC thread- just to important to get drowned out in the noise there.

Cheers!


Top
 Profile  
Reply with quote  
 Post subject: Ferrites and Neatness
PostPosted: Tue Jun 19, 2018 12:35 am 
Offline

Joined: Fri Jan 27, 2017 3:14 pm
Posts: 207
Location: Surrey, UK
I have now filled my spdif with twenty ferrites, and it has worked VERY well for me, as well as the PoE DC power change.

On the Naim forum Simon has banged on about ferrites for an age and as a result I added some onto my power cords, with little or no effect. For some reason I didn’t add any onto my spdif cable. The spdif being digital means that there are no negative effects to adding ferrites, as there can be adding them to analogue cables.

I have been battling edge for over a year and have my usual suspect files for testing. Over the year a fair few of them dropped off the list as different tweaks ‘solved’ those files' issues; there are multiple overlapping issues that lead to this ‘sound’.

One of the best tweaks is Rob's PoE using certain CAT7 cables. This leads to a relaxation in what you hear that is profound, but it still left certain recidivist files, including my bete noire, ‘Thunder Child’ from ‘War of the Worlds’. The ferrite tweak has mainly sorted these files, or at the least substantially reduced the issue.

The theories as to this effect boiled down to two, one of which is BluDave specific, the other is RF pollution.

On Rob's new PoE thread he mentioned other peoples lack of housekeeping wrt there system components, which made me reflect on mine. Basically mine was set up on shelves, including:

Top Shelf
ultraRendu / Iso Regen / SingXer F1 + assorted cables and LT3045 ......PLUS LPSU for my music system switch;

Middle Shelf
Pre-Amp / music system switch; and

Bottom shelf

Border Patrol SE DAC / R-Core DC30W used to power the SingXer

Other equipment is on two other racks.

....soooooo, reorganisation:


Top Shelf
ultraRendu (uR) / Iso Regen (IR) / SingXer F1 + assorted cables and LT3045;

Middle Shelf
Pre-Amp / music system switch; and

Bottom shelf

LPSU for my music system switch / R-Core DC30W used to power the SingXer / ZeroZone R-Core LPSU for uR & IR.

I also added ferrites to the DC Cables from the LPSUs.

Additionally I added five ferrites to the USB cable connecting my Prestige 2 server to the StarTech.

The Effect:
It is still bedding in, I heard subtle changes through the evening last night. There is no doubt that there is instrumentation that I am now clearly hearing that just wasn't heard by me before. Bit, teh character of my system seems to have slightly shifted. Tonal density (OK, I am reaching for a way to explain in words what I am hearing!) appears to have reduced slightly, but the sound has opened out. It is easier to hear into the sound-field and follow what effects and instruments are doing. On the whole this is positive, perhaps there is some rock music that may suffer slightly, more listening required.

Next
I need to add more ferrites to the back end of my system and hear the effect.

I also want to redo the comparison of files streamed from Qobuz vs SSD (Prestige 2) vs StarTech.

M


Top
 Profile  
Reply with quote  
PostPosted: Tue Jun 19, 2018 6:39 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Hi M,
Which ferrites composite are you using? 61's? Have you tried different ones like 75's?

My set-up is very similar to yours - except my LPS for the IR, F-1 is on a separate shelf from my DAC. The other LPS's are on the floor away from my Startech, SSD, USB hub, Music Server. The long silver/teflon leads that come stock with the DC-30W's allow for placement many feet away.
My DAC sits on it's own shelf. The IR, LT3045's, F-1 on a dedicated top shelf the furthest from the LPS's.

My Music Server is also many feet from any sensitive digital gear - aside form the SSD which sits in it's own Faraday cage.

I bought one of these EMI detectors - this I liked as the freq range is targeted to 30Hz to 300Hz - just the bandwidth local power related EMI would operate. Some have freq bandwiths too high or to wide. I want to measure the EMI from AC power cables (mine use very special ferrites and two layers of copper braid shielding and copper plug shields), DC cables, LPS's, DAC, etc...
Attachment:
s-l1600 (12).jpg
s-l1600 (12).jpg [ 45.39 KiB | Viewed 1043 times ]

An example from my office system. Note the copper sleeves need to make contact with the device chassis.
Attachment:
IMG_1217.JPG
IMG_1217.JPG [ 316.69 KiB | Viewed 1043 times ]


So I can test and measure then isolate any EMI LF around my digital gear.
If I have further issues I can add more Mu Metal to the bottom of my shelfs - it's cheap enough.

I also want to also test this JSSG theory to see if it really works. According to JS it block emitted as well as received LF EMI. I can find an unshielded power cord, run power through it and test with and without JSSG. The ferrites as well.

That should be interesting!

Cheers!


Top
 Profile  
Reply with quote  
PostPosted: Tue Jun 19, 2018 7:01 am 
Offline
User avatar

Joined: Fri Aug 16, 2013 8:23 am
Posts: 2234
Location: Seattle, WA, US
Looking at these folks systems I notice little attention to AC power isolation as well. All LPS's have AC leakage back to the wall, and the DC-30W's I use have a fair amount (I believe Alex mentioned the low cost diodes used as a factor). Most gear like DAC's have their own power supplies that cause leakage. But the biggest culprit are the large power amps! With massive capacitor banks - some in the 100,000uf range (my old Krell amps).

These dump large quantities of AC leakage back to the wall locally. So all digital gear should be protected by a significant AC moat to block this noise from having to dealt with my the gear itself. I use dedicated PB4X4Pros with 40dB of noise attenuation - one for My DAC, one for my music server, and one just for the DDC gear. These act to not only filter the incoming AC - but to isolate each of these banks from each other.

I have and will never have any SMPS in my audio system! They are also horrible noise generators. Those Meanwell's that energize the LPS-1/1,2 are pretty bad. As Elberoth's test showed. And they pulse when energizing the LPS-1 banks!
https://www.youtube.com/watch?v=3xVCZtj5qAA
And measured by AMir at ASR:
Attachment:
Regen SMPS noise.jpg
Regen SMPS noise.jpg [ 76.91 KiB | Viewed 1042 times ]


At the wall I use a Blue Circle Puck with another 60dB of wide freq noise suppression. A total of 100dB.

My Amps are direct to the wall, on the other side of the moat.

All this attention to noise reduction really pays of on the digital source transparency and very low level of floor noise.

Cheers!


Top
 Profile  
Reply with quote  
 Post subject: HiFi Reorganisation
PostPosted: Wed Jun 20, 2018 3:05 am 
Offline

Joined: Fri Jan 27, 2017 3:14 pm
Posts: 207
Location: Surrey, UK
Well, I went to bed last night in high dudgeon and disgust, my system quality had taken a dive.

Sometimes changes can lead to tonal changes that raise instruments into higher relief, whilst not representing a genuine improvement; and, this is what I was now hearing I decided.

I stripped off all the small ferrites I had placed on my PoE DC Loops. The tonal weight immediately returned.

I decided to play Billy Joel, 'River of Dreams' - as I went to see him on Saturday. This is NOT a well recorded album and can be sharp in my system; this had been largely addressed by PoE and ferrites - not now. I did notice that if I added ferrites to the DC cable from teh ZeroZone the issues diminished, but still not to the quality I had before.

In fact the main thing I noticed is that the feeling of relaxation that I had gained had lessened - DAMN.

So, what had changed?

1. LPSUs together on one shelf
.....could they be interacting?
2. Changed cable runs
3. DAC moved
.....My DAC had been sitting above my turntable. Now it was underneath and just above my BluRay player, AND far nearer a couple of power amps.
4. Ferrites added to DC on digital front and back end.

I removed all the ferrites from the DC cables
I swapped the LPSUs with the DAC, bringing the DAC further from the power amps and nearer the Digital front end.
Moving the LPSUs meant that the one powering my local switch neaded a longer DC lead = PoE using a generic CAT7 patch lead.

......and?

As soon as I started playing 'River of Dreams' I knew the relaxed quality had returned ......phew!

......now going to play with some ferrites on teh ZeroZone DC lead.

M


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 248 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 17  Next

All times are UTC - 8 hours


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group