Fancy Ethernet Cable-Experiment with the last 1 metre
Posted by: Kiwi cat on 02 June 2018
I have spent a considerable amount of money over the last 6 years, starting with a Uniti and Unitiserve with Rega Jura Speakers , ending up with my current 272/555DR/250DR/ Ovator400 system. It is a constant source of joy and is my end point for the foreseeable future.
This has not stopped me tweaking, the latest experiment being with a fancy Ethernet cable 1m long , the Chord Signiture Super ARay. This is my experience, dear readers.
In my system, the Unitiserve and NAS sit in the back room of the house. This is where the fibreoptic cable comes in to the house from the street. They are connected to a modem with several Chord C-Steam cables. To get to my 272, which is in the front room, the modem then connects via another Chord C-Steam cable to a wall socket and the signal travels about 20m in wall and ceiling cavities to the front room via $2 a meter Cat5 cable.
The Ethernet socket behind my 272 has 2 Ethernet sockets, this makes comparison relatively easy between different Ethernet cables. Good old Chord C -Steam runs from this socket to the 272, and it really sounds wonderful. I have had the opputunity to live with the Chord Signiture Super Aray for 2 weeks running to the 272 from this wall socket. This $1300 1m cable costs about 5x more than the previous 35 meters leading up to it.
I was a bit cynical as to whether this 1 meter cable would be any better than the C-Stream which is Chords considerably cheaper entry level Ethernet offering. I lived for a week with the Signiture Aray then switched over back the the C-Stream. I had hoped that there would not be much of a difference, after all bits are bits aren’t they? Unfortunately there was a considerable difference. I would estimate as being about 50% as good as adding the 555DR to the 272. Specifically the music flowed more easily, as if it had been “ deconstipated”. The music was more open, detailed and relaxed.
I find this hard to rationalise It must be the improved shielding of the cabling from the electromagnetic radiation from the transformers of the naim boxes. This has been discussed ad nauseum on this forum before but I was a non believer. But even at the end of 35m of lesser cable the exotic 1 meter actually does what it says on the tin.
Sadly I will be returning the Signiture Super Aray due to discretionary funds being low after recent carbon mountain bike purchase, but eventually I will go for the overpriced fancy Chord cable.
Posted on: 06 June 2018 by Obsydian
FRENCHROOSTER Morning - forgot to say i had looked into the Uptone LPS, but understand i would need a power supply (possibly ifi) to run that anyway - so i disregarded it. Will look into the Wyred option as if it powers two TP Links thats a cleaner setup for me.
uptone js2 can power 2 items, 2X5v or 2X12v or 5v and 12v. uptone lps can power only one item. you don’t need other ps to run the lps. 2 lps are less expensive than the js2. i hope it can help you.
FrenchRooster morning (again), i am thinking of either one Uptone LPS1.2 (£425) to power one of my TP links (not sure if the first or second makes more sense), or did see the Teddy Pardo Dual5 that offer 2x 5v (£350), so could power both TP Links, but it does not appear to be a linear p/s.
Posted on: 06 June 2018 by Finkfan
I think all TP power supplies are linear psu
Posted on: 06 June 2018 by Obsydian
The media converters come with SMPS, i swapped mine for an IFI SMPS.
Posted on: 06 June 2018 by Mike-B
I think all TP power supplies are linear psu
Correct in as much they are not SMPS. But they are not linear in the true sense of the word; LM317 type regulators are linear which is where the linear word comes from. The TP 'S-T-Reg' design is not linear just as Naim's DR's are not, all consist of numbers of discreet (individual) components.
Posted on: 06 June 2018 by hungryhalibut
I think all TP power supplies are linear psu
As I understand it, the supply for the UnitiServe is an SMPS design.
Posted on: 06 June 2018 by Huge
I think all TP power supplies are linear psu
As I understand it, the supply for the UnitiServe is an SMPS design.
I believe it is linear, but this is getting too close to the AUP, so I'm not going to comment further as to why I believe that.
Posted on: 06 June 2018 by fatcat
Linear refers to the fact that the regulator outputs a constant voltage. Non linear regulators do not output a constant voltage. The voltage varies with load.
Posted on: 06 June 2018 by Simon-in-Suffolk
Are you sure, a voltage regulator provides constant voltage irrespective of load .. if there is no regulator the voltage will vary with load.... never heard of a ‘linear’ regulator before.
I think the referred comment refers to transformer power supplies, which some refer to as a ‘linear power supply’, as opposed to using a switched power supply.
Simon
Posted on: 06 June 2018 by Huge
I think all TP power supplies are linear psu
Correct in as much they are not SMPS. But they are not linear in the true sense of the word; LM317 type regulators are linear which is where the linear word comes from. The TP 'S-T-Reg' design is not linear just as Naim's DR's are not, all consist of numbers of discreet (individual) components.
Mike, they are linear (as opposed to switching), but they're not integrated circuit regulators.
They use a transformer to drop the voltage, rectify and smooth that, then use discrete components to form a linear regulator. The word linear refers to the transfer function (i.e. the mode of operation) of the components from which device is made and how they are used, not the regulator's physical construction (i.e. whether or not the linear components are physically located in an IC). The Naim DR is also an application of linear componentry.
Posted on: 06 June 2018 by Simon-in-Suffolk
The regulators actually are based around comparators and using negative feedback to adjust the gain of an amplifier, so as to maintain the output level constant within designed limits. The Naim DR regulator was about providing a responsive low noise feedback amplifier circuit using a low noise comparator and reference (simplistically)
Posted on: 06 June 2018 by Huge
It's not possible to make a power supply that has a completely constant voltage output. Even the best regulators have a degree of load dependency (and also input voltage, thermal and temporal dependencies). Their output voltage is always a multidimensional map!
In the best regulated PSUs, the variance of the output voltage can be contained within +/-0.1% when functioning within their designed operating envelope (but outside this envelope strange things can happen!).
Posted on: 06 June 2018 by fatcat
But some regulators are more linear than others. The reason the Naim DR and other types of super regulator offer a significant improvement in the SQ, is they do output a constant voltage, irrespective of fluctuating load. The 78xx's and 3x7's don’t.
Posted on: 06 June 2018 by Mike-B
Mike, they are linear (as opposed to switching), but they're not integrated circuit regulators.
.......... ???? ........... thats just the point I made, ".... consist of numbers of discreet (individual) components"
The company concerned make the point that they are not linear as in LM317 type(s), it has to be said this is only a play on the linear word, & is possibly were this confusion comes from.
Whatever, this is all thread drift & nothing much to do with "the last 1 metre" of ethernet.
Posted on: 06 June 2018 by Huge
The SQ improvement is mainly due to transient response and the noise density function. The load regulation figure is relatively less important.
Posted on: 06 June 2018 by Huge
Mike, they are linear (as opposed to switching), but they're not integrated circuit regulators.
.......... ???? ........... thats just the point I made, ".... consist of numbers of discreet (individual) components"
The company concerned make the point that they are not linear as in LM317 type(s), it has to be said this is only a play on the linear word, & is possibly were this confusion comes from.
Whatever, this is all thread drift & nothing much to do with "the last 1 metre" of ethernet.
Then the "company concerned" are using the term 'linear' in an incorrect manner.
Posted on: 06 June 2018 by French Rooster
FRENCHROOSTER Morning - forgot to say i had looked into the Uptone LPS, but understand i would need a power supply (possibly ifi) to run that anyway - so i disregarded it. Will look into the Wyred option as if it powers two TP Links thats a cleaner setup for me.
uptone js2 can power 2 items, 2X5v or 2X12v or 5v and 12v. uptone lps can power only one item. you don’t need other ps to run the lps. 2 lps are less expensive than the js2. i hope it can help you.
FrenchRooster morning (again), i am thinking of either one Uptone LPS1.2 (£425) to power one of my TP links (not sure if the first or second makes more sense), or did see the Teddy Pardo Dual5 that offer 2x 5v (£350), so could power both TP Links, but it does not appear to be a linear p/s.
i had the tpardo one before, with one imput, and it is a linear ps. you have also mcru linear ps with 2 imputs( british product). around same price.
Posted on: 06 June 2018 by Obsydian
FRENCHROOSTER Morning - forgot to say i had looked into the Uptone LPS, but understand i would need a power supply (possibly ifi) to run that anyway - so i disregarded it. Will look into the Wyred option as if it powers two TP Links thats a cleaner setup for me.
uptone js2 can power 2 items, 2X5v or 2X12v or 5v and 12v. uptone lps can power only one item. you don’t need other ps to run the lps. 2 lps are less expensive than the js2. i hope it can help you.
FrenchRooster morning (again), i am thinking of either one Uptone LPS1.2 (£425) to power one of my TP links (not sure if the first or second makes more sense), or did see the Teddy Pardo Dual5 that offer 2x 5v (£350), so could power both TP Links, but it does not appear to be a linear p/s.
i had the tpardo one before, with one imput, and it is a linear ps. you have also mcru linear ps with 2 imputs( british product). around same price.
Interesting i did see the MCRU but not a dual output version will investigate.
Assume the TP was not your liking.
Posted on: 06 June 2018 by DaveBk
My DIY solution seems much simpler.... it's been powering my FMC and Switch for 3 years. £50 in parts and a few hours work. PCB design and etching probably took the longest.
Posted on: 06 June 2018 by French Rooster
FRENCHROOSTER Morning - forgot to say i had looked into the Uptone LPS, but understand i would need a power supply (possibly ifi) to run that anyway - so i disregarded it. Will look into the Wyred option as if it powers two TP Links thats a cleaner setup for me.
uptone js2 can power 2 items, 2X5v or 2X12v or 5v and 12v. uptone lps can power only one item. you don’t need other ps to run the lps. 2 lps are less expensive than the js2. i hope it can help you.
FrenchRooster morning (again), i am thinking of either one Uptone LPS1.2 (£425) to power one of my TP links (not sure if the first or second makes more sense), or did see the Teddy Pardo Dual5 that offer 2x 5v (£350), so could power both TP Links, but it does not appear to be a linear p/s.
i had the tpardo one before, with one imput, and it is a linear ps. you have also mcru linear ps with 2 imputs( british product). around same price.
Interesting i did see the MCRU but not a dual output version will investigate.
Assume the TP was not your liking.
i just seen the dual output mcru on their site. But don’t know if it will power 2X 5v. You can ask them via mail.
The israeli ps was not so good to power my unitserve. But for a switch, it should be perfect.
Posted on: 06 June 2018 by Simon-in-Suffolk
Dave, interesting... I presume the chunky metal cases devices at the top are diodes... in which case why didn’t you use a heatsinked bridge rectifier as opposed to a two diode full wave rectifier which means you could do away with using a centre tapped transformer and therefor use a smaller transformer for the given power output.
Posted on: 06 June 2018 by DaveBk
Dave, interesting... I presume the chunky metal cases devices at the top are diodes... in which case why didn’t you use a heatsinked bridge rectifier as opposed to a two diode full wave rectifier?
They’re power resistors. I could not source a transformer with close enough secondary voltages to the optimum for the 2 regs, so worked out the average current draw and drop a volt or two on series resistors ahead of the regulators. The regs now run slightly warm rather than red hot as in my first version of this. There are 2 bridge rectifiers on the PCB, hidden behind the smoothing caps.
Posted on: 06 June 2018 by Simon-in-Suffolk
Ok... even more interesting.... 
Despite that I assume the regulators don’t go pop if there is nothing connected to the powersupply?
Posted on: 06 June 2018 by DaveBk
No popping. The power dissipation in the regs is proportional to the difference between input and output voltage, assuming the same current draw. No current, negligible thermal dissipation. The regs are 78xx series, so need about .8v headroom I recall.
Posted on: 06 June 2018 by fatcat
That's a nice neat job.
You've managed to get more components than there are in a flatcap, in a box, a tenth the size.
Posted on: 06 June 2018 by Simon-in-Suffolk
No popping. The power dissipation in the regs is proportional to the difference between input and output voltage, assuming the same current draw. No current, negligible thermal dissipation. The regs are 78xx series, so need about .8v headroom I recall.
Good stuff
