Building a DIY Music Server

[basillus]

Hi, thanks for this informative and terrific thread. Maybe I am wrong, but if I look at your picture at the atx 20-pin connector and you are wiring pin 8,9 and 10, then these have to be at the atx socket nearest the external regulator boards and not in the other end at the atx socket. I hope it is understandable, what I mean 😉

[basillus]

elan120 wrote:

OP did say the wiring connections are as follows.

 

   On 1/11/2020 at 8:08 AM,  Nenon said: 

I connected it this way:

3.3V --> pin 1

5.0V --> pin 4, pin 8, and pin 9

12V --> pin 10

Ground --> pin 3

Yes I know, but according this:

and this there is a mismatch where the wires are placed...

 

[basillus]

1 hour ago, Nenon said:

I share mostly success stories, but there is a lot going on behind the scenes with my experiments. Here is a behind the scenes failure story... Tried to replace the clock on one of my Buffalo switches, and the switch board is fried.

 

 

The problem is I don't know why. I had someone else remove the clock, so that might be part of the problem. Also USPS was quite brutal with this package as it can be seen by the bended port. 

Well, one less precious Buffalo in this world. That specific buffalo breed is nearly extinct!

I will sacrifice another one, and hopefully it would work out next time. 

You have removed the 2 capacitors between the crystal and gnd? 

Heat shouldn`t have destroyed it...can’t you measure if there are bad connections/pcb lanes?

[basillus]

Is there at the www. some there have tested and compared the pink faun clock with other  clocks? 

 

An cheaper and maybe nearly as good or maybe better alternative I don’t know... is  Andrea Mori’s GB at diyaudio. He has a 25mhz AT-Cut crystal there is suitable for the old oscillator versions TWTMC-C and TWTMC-P and also for the new Pierce TWTMC-PXO PCB board. 

GB link: https://www.diyaudio.com/forums/group-buys/291925-tempered-master-clock-buy.html 

 

 

[basillus]

18 hours ago, Exocer said:

Very informative post! Thank you.

 

Taking a moment to process it all...

 

As for frequencies, we are interested in 25mhz LVCMOS output in this context. It would be great if it were at least possible to equal the (supposed, and unmeasured) -125dbc/hz of my reference clock.

 

As for programmable synthesizers, I would not know where to begin with this approach but it does sound intriguing to research further.

Andrea Mori at diyaudio have developed them the last 15 years, but still it looks like, that I am the only one, there is interested in his 25MHz or 6.25MHz crystal GB with his new clockboards and I don’t understand why and the price isn`t near PinkFaun, but maybe the crystal and clockboards are... Sorry, it is frustrating.     ;0)

[basillus]

1 hour ago, Exocer said:

 

It would be great if you (or Andrea) could provide some datasheets with specs. Is there a URL? Phase noise specs at 10MHZ appears to be one of the defacto specs which would interest us, also, frequency stability.

 

Perhaps that will allow some of the industry pros (and those of us learning) to weigh in on performance and increase interest level :).

It`s a long thread, but here are a couple of links, but there are many other “test” in his threads. 

Andrea does it in his sparetime and is a hobbyist.

https://www.diyaudio.com/forums/digital-line-level/261651-tempered-master-clock-building-low-phase-noise-jitter-crystal-oscillator-post6360033.html

 

https://www.diyaudio.com/forums/digital-line-level/261651-tempered-master-clock-building-low-phase-noise-jitter-crystal-oscillator-post6368700.html

 

https://www.diyaudio.com/forums/digital-line-level/261651-tempered-master-clock-building-low-phase-noise-jitter-crystal-oscillator-post6345475.html

[basillus]

 

 

1 hour ago, Exocer said:

The specs do look very good per my untrained eye. @ around 10hz they are all at least -120dbc/hz at 10Hz. Any info on the form factor?

 

Edit: I see the photos of the form factor. This could be an interesting alternative. Will do some further digging on the thread for pricing info etc.

The final prices, BOM etc. are coming soon. The Group buy had running a while, the problem is that no one have yet ordered a 25MHz or 6.25 crystal and the minimum order is 10 or 15. 

The latest prices were:

https://www.diyaudio.com/forums/group-buys/291925-tempered-master-clock-buy-post6255266.html

[basillus]

7 hours ago, realDHT said:

If you want for example Pierce AOI oscillator with 6.25 MHz crystal then why don't you add yourself to the group buy interest list, I dont see your nickname in the list? If you do that it would increase the chance that others follow. Its is not committing to buy anything at this point, just an interest list. I see 2 people (including me) that sofar added themselves to the list for 6.00 MHz and Pierce AOI (giving 24 MHz output). I will need the 24 MHz for usb communication.  

Thanks. You are right and I will do as you suggest. The reason I had not already done so is that I thought that it was more than a interest list and there were some sort of binding. 

[basillus]

Seeteeyou  You have read my mind 😉, because to combine the Schröder Method with the diy helix have been on my to do list for a long time based on a 2 bare mundorf silver-gold wire each in their own silk tube with 2 carbon roving 50k filaments sprayed with a diy graphene solution there enclosed the wire and to keep the graphene encapsulated a shrinkable ptfe tube or ptfe tape around the silk tube and then twist the 2 wire together. Maybe a JSSG screen around it, wooden beads and use 2 Jantzen ptfe wire(I live in EU) as neutral and/or gnd. I also thought of to use it in a computer as dc cable. 

[basillus]

1 hour ago, ASRMichael said:

Is there any risk of using the Schröder Method on RCA’s? As in potential damge to amp or DAC? Just want to make sure before I try it? 

 

In the beginning Douglas Schröder warned about to use his metode to class d amplifiers, but later I haven’t see that class d should be a problem with his metode. There could maybe be a problem if the preamplifier output and the amplifiers input impedans have a big mismatch. If not I don’t think there is a potential risk to use his metode...

[basillus]

Quote

19 minutes ago, Nsxturbo said:

Ok I’ll admit I’m kind of shooting in the dark here, but does any DIYer know what brand this capacitor is? There is no name on it just the logo.

 

 

It’s look like this: 

https://www.amazon.com/pcs-1000uF-Radial-Electrolytic-Capacitors/dp/B00I4QYRT2

[basillus]

9 hours ago, Soul Analogue said:

so the main LPS unit for the EPYC server has been completed

It comprises of regulated voltages:

1) 12vdc 8a

2) 5vdc 8a

3) 3.3vdc 8a

4) 5vdc 3a standby

 

as well as 4 unregulated voltages to power the regulators inside the server chassis for add-on cards

 

Around 5 - 6 sets of my discrete regulators will be installed inside the server chassis:

Hi, which PI-filter chokes are You using?

[basillus]

Exocer Said: 

I also have a Sean Jacobs DC3 12v LPSU in the works. After all of the recent talk of 10v I wish I would have had Sean tune the modules for 10v output ;).

 

 

Me: You just need a little screwdriver and a multimeter to read the dc output. 

[basillus]

1 hour ago, One and a half said:

 

Many others have considered a balanced power feed on the AC side as having benefits of a noticeable reduction in the purging of noise. For this topic, I had thought of ways to produce a balanced DC supply for computer use. DC has the advantage that it is 0 Hz, so in theory there's nothing to cancel. However. The DC in power supplies contains garbage from other pieces of equipment connected to the AC network, coupled capacitively through linear elements such as power transistors on heatsinks, mainly from the PC's PSU.

 

A Balanced 'DC' rail is actually DC+ a small amount of AC. It's the AC that can be effectively dealt with a balanced network, any asymmetrical voltages are 'gone', leaving pure(er) DC. This is especially important in ATX supplies where high currents are involved.  There's also load noise to consider, would also benefit from a balanced supply.

 

How this would be setup, is the centre connection is at 0V earth potential, wired back in a star scheme to the earth of the wall outlet or balanced AC supply mid point.

The 5V would be +/- 2.5V, 12V +/-6, the tricky part is 3.3V, since many 3 terminal regulators have a minimum of 1.25V, although with a bit of work and thought, can be overcome. 

 

Linear supplies would work well, and SMPS, but finding 6V or 2.5V off the shelf are not so easy to source, unless custom built like from Acopian. Like with a linear supply the fixed switching frequencies on a balanced supply are asymmetric from an SMPS, so the noise factor could reduce nicely as well.

 

I'm just putting this out there as a point of discussion as an improvement to remove noise. 

Interesting, thanks for your input. You don’t have a picture/schematic about the balanced dc-rail setup, because I am not 100% sure that I have understand it correctly. 

[basillus]

 

 

Great news and just the size I would have made it in myself. 

2cm holes above and under the heatsinks will probably be to expensive...

An option for a black or “silver” cabinet or just the front plate in the two colors.

 

[basillus]

@Nenon at the first picture with the Asus C621e motherboard,

it looks like that the CPU nearest the front,

when its cpucooler and heatpipes get attached,

that the pipes need to be bend,

if it has to be connected to the right big heatsink,

because the pipes has to get past the two aluminiums “screening” sides,

where the softstart, trafo etc. have to be.

[basillus]

31 minutes ago, Nenon said:

 

Marcin - please share more details as those become available. That will be a neat solution for those using the JCAT ATX power supply as it will be matching cases it seems like. 

 

The Taiko chassis is designed for the ASUS SAGE / dual CPU as there is no other solution on the market for this. BTW, it's available on the Taiko website now - https://taikoaudio.com/taiko-2020/product/diy-audio-server-chassis/

 

HDPLEX H5 does support Asus Sage C621E Dual CPU, when they are able to make the CPU cooler... First it was December last year, then May this year and now it is....?

Nenon do You know, if it is just the heatsink near the C621E MB there is used to both cpues at the Taiko Diy, like in the Taiko Extreme?

 

Is the Diy Taiko chassis already gone?

I got this massage: 

• Sorry, we do not have enough "DIY Audio Server Chassis - Option 2: 2x LGA3647 square cooler + heat-pipe bending pliers + 9x heat-pipe (includes one spare heat-pipe) + heat-paste" in stock to fulfill your order (0 available). We apologize for any inconvenience caused.

[basillus]

37 minutes ago, Nenon said:

That was fast! 

There will be a new batch I am told (with around 2-3 months lead time). 

I happened to see your link 15 minutes after it was shown and a little later it was out of stock and yes, I do not think the next time you post a link on the new batch,  I will see it as fast as this time ...

Shit happens,  waited so long at this diy cabinet, well, then it must be a 2 cabinet solution with HDPLEX H5, when their LGA 3647 cpu coolers are made or maybe a large diy cabinet with 2 of these HDPLEX coolers...

[basillus]

On 7/8/2022 at 6:20 PM, JayM said:

 

Unfortunately the first batch of the Taiko Audio DIY Chassis got sold out way faster than expected.

 

Some people already contacted me, but to prevent this happening at the second batch as well please send me a private message if you are seriously interested in buying a DIY chassis.

 

If more people than expected contact me I still have the ability - right now - to acquire more raw materials, so that this won't happen again. I will need to know during this weekend though. Since the parts with the longest delivery time (about 1,5 month) will be ordered on Tuesday. As soon as these parts come in we can ship out the orders which means we can ship everything in 2 months from now (3 months maximum, in case of set-backs). So it won't be as long a waiting time as last time...:) 

I send You a PM saturday, but You still haven’t read it...

Does JayM have a e-mail adress?

[basillus]

On 10/25/2022 at 4:08 AM, Exocer said:

When I used Ghent Audio cables, i ended up removing the jssg360 shielding on my ATX cable as well and I thought it sounded better. From that day forward, I built my own ATX, EPS and DC input cables without JSSG360 shielding.

 

For my two EPS cables, I use relatively thin wire but I think it sounds great and I haven’t moved away from it:

https://www.partsconnexion.com/DUELUND-81388.html

 

Mundorf SG for ATX

Mundorf Angelique for DC Input

 

The SG is from before I discovered uninsulated wire (lol), but if I were to build my ATX cable again I would go with oversized teflon tube and uninsulated SG wire. Another member on the forum has introduced me to this method and it sounds wonderful. I have not tried silk so perhaps that is even better.

 

For anyone who’s on the fence with wire, try 15.5awg Angelique. I’ve even used it to build Helix XLRs (1.5mm for signal wires) and I’m astonished at how good these ICs sound. It is equally good for DC wiring but SG still takes the crown for DC (imho).

 

Lets get back to having fun and experimenting.

 

Cheers,

-Rob

The Mundorf SG wire for ATX You use, is it the 1mm or 1.5mm? 
I am thinking about to use  Mundorf Angelique 1.5mm or 2mm bare wire to the EPS cables and maybe ATX, but I haven`t find the bare wire in Europe yet…

[basillus]

1 hour ago, MarcelNL said:

I have not encountered any delay in cold start, neither with the Lundahl chokes or with these ones.

(keep in mind that the PSU supply has to deliver far less amps than the 12V to the CPU etc makes it hard for me too to consider chokes as rate limiting)

 

the DO improve SQ, and by quite a bit....

counter to keep track of interested folks: (please qoute this post adding your nick so I can keep track)

 

volunteer to manage things if it comes to a larger group buy:

----vacant----

 

serious interest in Hi-B cut chokes for LPSU

1 - @rico25

Basillus

[basillus]

1 hour ago, MarcelNL said:

for V3 you need two, I paid 160 euro a piece plus shipping (within my country), the weight is considerable so weight will kick international shipping up.

 

given the raw material frenzy cost may go up a bit, I dunno.

 

 

 

Thumbs up 😉 Do You remember how much 1 choke approx weight?
Shipping from Nederland to Denmark isn’t expensive.
A max. 10kg parcel cost approx 25€ at postnl.

If it had been the other way around from Denmark to Nederland it will cost approx 87€ 😳

[basillus]

@Gavin1977

 

is missing on the last list 😉

[basillus]

2 hours ago, MarcelNL said:

quick question, are the two EPS power connectors powering the CPU's on the MB linked or do I need two separate cables to feed both CPU's?

Yes, they are linked, also with the 6pin at the top middle. 
In the beginning I tested the sage with a normal atx pc psu and then one cable was enough, but The atx connetors 12v is also linked to the 2 EPS power connectors.

[basillus]

Yes, a multimeter is a good thing if you are in doubt. If you just want to test whether it works, an 8 pin cable is enough, but as you well know, the more cables the better... What do you find confusing that the 12v on the atx connector also has contact with the 12v on the EPSs?