Building a DIY Music Server

[ASRMichael]

@Nenon do you think Taiko case for dual CPU mobo will be available this year? 

[Nenon]

14 minutes ago, ASRMichael said:

@Nenon do you think Taiko case for dual CPU mobo will be available this year? 

 

I do.

When I visited Taiko last month they were expecting a delivery of a new state of the art CNC machine (guessing the stuff companies like MSB use). If I was to guess the DIY chassis will be one of the first projects they will do on the new machine once it is set up. It makes sense - they can play with their new toy on our chassis as we don't even know exactly how it will look like 😂.

[ASRMichael]

2 minutes ago, Nenon said:

 

I do.

When I visited Taiko last month they were expecting a delivery of a new state of the art CNC machine (guessing the stuff companies like MSB uses). If I was to guess the DIY chassis will be one of the first projects they will do on the new machine once it is set up. It makes sense - they can play with their new toy on our chassis as we don't even know exactly how it will look like 😂.

Thanks last question (for now) when you plan to do that write-up on windows Configuration for dual cpu? Sorry If I asked this before…

[Nenon]

Just now, ASRMichael said:

Thanks last question (for now) when you plan to do that write-up on windows Configuration for dual cpu? Sorry If I asked this before…

 

The original plan was to build a new server when I get the Taiko chassis and write the guide as I set it up from scratch. It has been a while since I installed mine, and I made a lot of changes over time, many of which I don't remember. But with two servers I can compare them side by side and make sure I cover everything. That is still the plan... it's just that with the global shortage of materials and parts every step takes much longer than expected. 

[Dev]

what happened to the Taiko's Noratel transformer ? Did they receive their orders yet ?

[Nenon]

2 minutes ago, Dev said:

what happened to the Taiko's Noratel transformer ? Did they receive their orders yet ?

 

Noratel were supposed to be shipped over sea (takes about 6 weeks) last week. I could not wait any longer and ordered some transformers from Toroidy, and they are stuck in Fedex Germany for nearly two weeks :(. Not the best times for impatient audiophiles like us.

[Dev]

9 minutes ago, Nenon said:

Not the best times for impatient audiophiles like us.

 

No doubt 😃

[lwr]

24 minutes ago, Nenon said:

 

Noratel were supposed to be shipped over sea (takes about 6 weeks) last week. I could not wait any longer and ordered some transformers from Toroidy, and they are stuck in Fedex Germany for nearly two weeks :(. Not the best times for impatient audiophiles like us.

Is the Toroidy that you ordered expected to be sonically the equivalent of the Noratel?  If so, would you please provide the Toroidy ordering information?  (I have never ordered from them before.)

 

Thanks

[bobfa]

32 minutes ago, Nenon said:

stuck in Fedex Germany

I have stuff stuck in Germany too.... up until today!!!

[Soul Analogue]

5 hours ago, Nenon said:

I enjoy reading what people do or have done. A lot of good stuff. There are probably people reading all these comments about chokes who are getting confused. So, let me clarify some things. 

 

1. There are some claims that the 5A choke I picked does not provide enough current and restricts the dynamics on the dual Xeon server. 

That is not the case with my build. I utilize about half of the current capabilities of this choke. And this is by far the most dynamic server I have ever built. 

There are two things to keep in mind here. First, this is an unregulated LPS + GaN based ATX, which is extremely efficient. Note the 95W vs. 270W comment in the post I am quoting below. Second, Windows (LTSC in my case) the way I have configured it is more efficient than Euphony. As a result of this I am using about 95W from nearly 200W available with this choke. 

More details here:

Unless you do heavy DSD upsampling with 120W TDP CPUs, chances are this choke will work great. As a general rule, if you can passively cool down your CPU with a HDPlex/Streacom chassis, you probably won't have any issues with this choke/design.

 

2. There are some claims that the 5A choke I picked will be very noisy.

This is also not the case in my system. Not at all! There are people here who have followed the DIY recipe I provided and can comment about their experience. No noise in my power supply. Please do share if your experience is different.

 

3. Very important - you SHOULD NOT just replace the choke with another part unless you know exactly what you are doing. You can potentially kill your Taiko ATX if you simply replace the choke with a much lower DCR one and don't make any other changes.

For example, modeling some of the bigger Hammond chokes turned out they will require 470,000 uF of capacitance instead of the 66,000 uF in V2. In some cases, with some chokes, you may reach the need to use 700,000 uF. 

 

4. A higher current choke does not sound better in my experience. It sounds different. There are always tradeoffs. What I have provided for DIY is a very well balanced power supply design. It's a good tradeoff between staging/envelopment and dynamics/naturality. It looks like a very simple design that I put together during my lunch break, but it actually took many months of development, testing, and most importantly - listening. It employs a lot of things I've learned from Emile and Sean, some of which I can't really share. But it doesn't really matter how I came up with this design. It's the end product that matters... and that's shared for anyone who wants to build it. I've shared details here (the cost of parts keeps going up, so my spreadsheet is already out of date):

 

Combine this unregulated LPS (v2 or v3) with the Taiko ATX to power up your computer. It hardly gets any better than that.

Once you hear what this sounds like, it will change your perspective on fully regulated linear power supplies for high current draw applications. You still can't beat a good LPS for low current draw applications (i.e. 1A-2A-3A), but for high current CPUs this is my new gold standard. 

 

Lastly, I encourage you to build this unregulated LPS (both v2 and v3) and try it with the Taiko ATX (powering up the EPS and ATX) before going into arguments or criticising a particular part I chose. No part I picked was just a random choice. They all work together and create a synergy. Once you do that, by all means feel free to experiment and if you find something that works better, please share. I am done with the unregulated LPS development and stopped tinkering with it (for now). No upgraded versions are to be expected by me for a while. But I hope other people will pick this from where I left it and share their improvements. 

 

@Soul Analogue  I admire your work. Feel free to post details and pricing about your unregulated LPS with handmade chokes. It might be a good alternative for people who are not comfortable building a DIY power supply. I would personally love to test it myself. Maybe someone in the US will buy one and will be kind enough to send it to me to try it :). 

 

Nenon… we were talking about Choke input psu yesterday… it is a total different story from capacitor input psu

 

our rationale of LPSU designing are very different… so there is no point to further discuss here without making true listening comparsions

[Exocer]

1 hour ago, Soul Analogue said:

Nenon… we were talking about Choke input psu yesterday… it is a total different story from capacitor input psu

Not to be argumentative but I was confused by some of your feedback. You mentioned choke size being too small around the time several of us posted pics of our CLC Unregulated LPS's. Glad this is cleared up and that perhaps the topology of those pics you saw was misunderstood at the time of your feedback. 

 

@Nenon - I did not realize the Noratels shipped. FOMO is settling in....Is it worthwhile to "upgrade" to it if you already have the Toroidy Supreme v2? What does the 5% improvement afford you?

 

Cheers,

-Rob

[Soul Analogue]

For your peace of mind..

 

Here is a screenshot of simulation by psudesiger ii of my cap input psu design for taiko module with high amp choke. the curve is the voltage at stepped constant load at t1 2a, t2 10a at 2s (therefore the drop shown).. 

It does not overshoot or ring, or voltage above 40v at any instance that could possibly kill a taiko module

 

no excessive capacitance is used

[Soul Analogue]

31 minutes ago, Exocer said:

Not to be argumentative but I was confused by some of your feedback. You mentioned choke size being too small around the time several of us posted pics of our CLC Unregulated LPS's. Glad this is cleared up and that perhaps the topology of those pics you saw was misunderstood at the time of your feedback. 

 

Cheers,

-Rob

There was no misunderstanding .. i was truely referring to the small size of hammond choke of cap input supply …  

 

until yesterday that Mag talked about choke input supply here so i chimed in…

 

you have to understand what psu topology is being used and what Is being discussed here in order not to be confused =)

 

but in either psu topologies, i dont see, measure or simulate, any overshoot, ringing issue by using high current choke…

[Exocer]

15 minutes ago, Soul Analogue said:

you have to understand what psu topology is being used and what Is being discussed here in order not to be confused =)

 

😁 I understood what topology was being used... I was certainly not confused about that aspect. More confused about your suggestion of changing one component in a circuit with an entirely different design than your own. Hence why I waited for Nenon's follow up before purchasing a larger choke, and purchasing the same 5A choke again to expand to V3. No hard feelings though...

 

19 minutes ago, Soul Analogue said:

but in either psu topologies, i dont see, measure or simulate, any overshoot, ringing issue by using high current choke…

 

Interesting. Maybe an opportunity for us all to learn from each other 😎. That's what this place is about.

 

Cheers,

-Rob

[Soul Analogue]

24 minutes ago, Exocer said:

There is no “entirely” different design …

My suggestion of a change of higher amp choke is still valid for Nenon’s cap input supply…

my cap input design is still transformer > rectifier > c1 > choke > c2

Very similar right? Just plain more simple i would say..

i have built over a hundred of this CLC topology psu for audio purpose for myself and my friends and i never encountered the issues mentioned by Nenon.. 

If these high current vs low current choke discussions has to be continued….perhaps Nenon shall share his findings of overshoot, ringing, burn out taiko issue more clearly.. 

what transformer secondary voltage was used? Which rectifier bridge? which high current choke(s) was tried? What were the symptoms of overshoot, ringing? How they were measured? Or even showing simulation screenshots would be helpful for troubleshooting

[Exocer]

1 hour ago, Soul Analogue said:

My suggestion of a change of higher amp choke is still valid for Nenon’s cap input supply…

my cap input design is still transformer > rectifier > c1 > choke > c2

Very similar right? Just plain more simple i would say..

i have built over a hundred of this CLC topology psu for audio purpose for myself and my friends and i never encountered the issues mentioned by Nenon.. 

If these high current vs low current choke discussions has to be continued….perhaps Nenon shall share his findings of overshoot, ringing, burn out taiko issue more clearly.. 

what transformer secondary voltage was used? Which rectifier bridge? which high current choke(s) was tried? What were the symptoms of overshoot, ringing? How they were measured? Or even showing simulation screenshots would be helpful for troubleshooting

All really good questions! 

[Nenon]

3 minutes ago, Soul Analogue said:

There is no “entirely” different design …

My suggestion of a change of higher amp choke is still valid for Nenon’s cap input supply…

my cap input design is still transformer > rectifier > c1 > choke > c2

Very similar right? Just plain more simple i would say..

V2 is: transformer --> rectifier --> 22,000 uF --> Hammond 159ZJ --> 66,000 uF.

The Hammond 195K12 choke was recommended as a replacement at some point. 

transformer --> rectifier --> 22,000 uF --> Hammond 195K12 --> 66,000 uF will require more than 66,000 uF capacitance to damp overshooting. 

 

6 minutes ago, Soul Analogue said:

If these high current vs low current choke discussions has to be continued….perhaps Nenon shall share his findings of overshoot, ringing, burn out taiko issue more clearly.. 

what transformer secondary voltage was used? Which rectifier bridge? which high current choke(s) was tried? What were the symptoms of overshoot, ringing? How they were measured? Or even showing simulation screenshots would be helpful for troubleshooting

I managed to destroy one of the rails of the Taiko ATX within an hour of receiving it. That was the first prototype of the Taiko ATX. My 5 AWG power cord created a spark while plugging it. I did not use a softstart at the time, which was a mistake. The mosfet on the Taiko ATX at the time was a 60V one and it would instantly fry if it gets anything over 60V even for a fraction of second. It's true - I was a witness of that :). Taiko added protection on the production releases, so that was a good lesson learned. Also, using a softstart here is a must!

As far as details on killing the Taiko ATX with a low DCR choke and not enough capacitance - those were Emile's words - and I am guessing the problem is overshooting. I am super careful not to kill my Taiko ATX again, as these things are not exactly cheap. 

 

And this takes me to the point I was trying to convey earlier. Not many people here understand power supply designs. When people started discussing chokes here, I received all kinds of crazy messages with crazy ideas from people who did not know what they were doing. I am really glad they asked before they tried some of them! 

Those who really know what they are doing (and probably know more than me) don't need my advice. Those who don't should just stick to the provided design.

 

2 hours ago, Soul Analogue said:

Nenon… we were talking about Choke input psu yesterday… it is a total different story from capacitor input psu

our rationale of LPSU designing are very different… so there is no point to further discuss here without making true listening comparsions

@Soul Analogue this is not a competition. We are all here with one goal only - to achieve the best music reproduction. 

Feel free to post a different design people can try. That's what this forum is about. It will be helpful to list all the parts. 

If you are willing to build your custom design, by all means post about that too. The more options we have, the better. 

 

1 hour ago, Exocer said:

@Nenon - I did not realize the Noratels shipped. FOMO is settling in....Is it worthwhile to "upgrade" to it if you already have the Toroidy Supreme v2? What does the 5% improvement afford you?

When I get some I will send you one to try. It has been a while since I tested that. I picked the Noratel at the time but the difference was not big. 

 

3 hours ago, lwr said:

Is the Toroidy that you ordered expected to be sonically the equivalent of the Noratel?  If so, would you please provide the Toroidy ordering information?  (I have never ordered from them before.)

I liked the Noratel a little better. But given how long it took Noratel, I don't think I will consider ordering from them. Hopefully Emile bought enough :). 

If you want to go with Toroidy (amazing company to work with), you can use those specs:

Supreme 400VA
Primary: 115VAC (assuming US usage)
Secondary: 24VAC @ 16A

+ side wiring plate

[Soul Analogue]

57 minutes ago, Nenon said:

V2 is: transformer --> rectifier --> 22,000 uF --> Hammond 159ZJ --> 66,000 uF.

The Hammond 195K12 choke was recommended as a replacement at some point. 

transformer --> rectifier --> 22,000 uF --> Hammond 195K12 --> 66,000 uF will require more than 66,000 uF capacitance to damp overshooting. 

How did you realize the voltage overshooting caused by swapping in a high current choke? 

Measure? Modelling?

more details would be helpful

 

I do not use as much capacitance but i could not measure (by Fluke RMS DMM and Rigol Scope) and model (PSUDII) any overshooting of voltage during power up, idle and under load

 

Quote

 

I managed to destroy one of the rails of the Taiko ATX within an hour of receiving it. That was the first prototype of the Taiko ATX. My 5 AWG power cord created a spark while plugging it. I did not use a softstart at the time, which was a mistake. The mosfet on the Taiko ATX at the time was a 60V one and it would instantly fry if it gets anything over 60V even for a fraction of second. It's true - I was a witness of that :). Taiko added protection on the production releases, so that was a good lesson learned. Also, using a softstart here is a must!

So there were a number of other possibilities that caused the burn out of Taiko

 

Quote

As far as details on killing the Taiko ATX with a low DCR choke and not enough capacitance - those were Emile's words - and I am guessing the problem is overshooting. I am super careful not to kill my Taiko ATX again, as these things are not exactly cheap. 

 

Could you quote the whole paragraph of Emile's words?

 

 

Quote

 

@Soul Analogue this is not a competition. We are all here with one goal only - to achieve the best music reproduction. 

Feel free to post a different design people can try. That's what this forum is about. It will be helpful to list all the parts. 

If you are willing to build your custom design, by all means post about that too. The more options we have, the better. 

 

 

I am not here for a competition at all.... I, last week, just made a kind suggestion of using high current choke instead of a small one on your cap input topology, based on my own experience.

I have no major comment (except the choke) on your C values, which is pretty standard ... so I do not have an urge to post something different 

 

I chime in again today because you quoted my post (about Mag's choke input PSU idea) and then you talked about my comment on your cap input supply...

So I was impelled to response, clarifying my post (yesterday) being about a different PSU topologies.

[Exocer]

Moving onto the chassis. I have measured all of the parts for V3 using all 22,000uf Mundorf HC caps and it is a snug fit in this enclosure:

https://modushop.biz/site/index.php?route=product/product&product_id=128

 

This enclosure is similar in size to the HDPlex H5 V2 many of us are using.

 

Going to order it and test it in the coming week. Worst case scenario I will be forced to swap out two Mundorf 22,000uF caps for a single 47,000uF cap to save space.

 

Cheers

[Nenon]

35 minutes ago, Soul Analogue said:

How did you realize the voltage overshooting caused by swapping in a high current choke? 

Measure? Modelling?

more details would be helpful

 

I do not use as much capacitance but i could not measure (by Fluke RMS DMM and Rigol Scope) and model (PSUDII) any overshooting of voltage during power up, idle and under load

 

So there were a number of other possibilities that caused the burn out of Taiko

 

 

Could you quote the whole paragraph of Emile's words?

 

 

 

I am not here for a competition at all.... I, last week, just made a kind suggestion of using high current choke instead of a small one on your cap input topology, based on my own experience.

 

I chime in again today because you quoted my post (about Mag's choke input PSU idea) and then you talked about my comment on your cap input supply...

So I was impelled to response, clarifying my post (yesterday) being about a different PSU topologies.

 

Moved the conversation to PM. We'll share if any good stuff comes out of our discussion. 

 

For those building the v2 or v3 of the unregulated LPS - please don't change the choke unless you know exactly what you are doing 😂. 

[Dev]

One thing I can say is that I could not power my i9-10900k using a Hammond 5A choke in CLC configuration using two Mundrof 22,000uF, unless I ran the CPU at base clock and turned off hyperthreading, cutting down the power requirement significantly. Unlike the core i series which are designed for extreme ups and downs to find a balance between high performance, power requirement and heat, the Xeons are designed for server applications which typically consumes a constant power irrespective of load (think in terms of Class A amp). I don't have a dual Xeon system, so can't tell but its power requirements might be less stringent and very different from i9-10900k. I would think the choke/cap values would much depend on the boot power requirement of the system - if it can safety cross the check mark, runtime power is usually less for simple bit-perfect playback but if you up-sample using Hqp, that's entirely a different ballgame.

[Darryl R]

6 hours ago, Nenon said:

 

I do.

When I visited Taiko last month they were expecting a delivery of a new state of the art CNC machine (guessing the stuff companies like MSB use). If I was to guess the DIY chassis will be one of the first projects they will do on the new machine once it is set up. It makes sense - they can play with their new toy on our chassis as we don't even know exactly how it will look like 😂.

 

Wow, that's going to be a big project for them.  For the last ten years I worked at a company that made aircraft parts with CNC machines, and we had a staff of people to program them, and another staff to plan the parts for programming (coding dimensions and tolerances off of the drawings).

[Dev]

20 minutes ago, Nenon said:

Moved the conversation to PM. We'll share if any good stuff comes out of our discussion. 

 

I think it will be a useful discussion for many of us to learn :-)

[Nenon]

20 minutes ago, Dev said:

One thing I can say is that I could not power my i9-10900k using a Hammond 5A choke in CLC configuration using two Mundrof 22,000uF, unless I ran the CPU at base clock and turned off hyperthreading, cutting down the power requirement significantly. Unlike the core i series which are designed for extreme ups and downs to find a balance between high performance, power requirement and heat, the Xeons are designed for server applications which typically consumes a constant power irrespective of load (think in terms of Class A amp). I don't have a dual Xeon system, so can't tell but its power requirements might be less stringent and very different from i9-10900k. I would think the choke/cap values would much depend on the boot power requirement of the system - if it can safety cross the check mark, runtime power is usually less for simple bit-perfect playback but if you up-sample using Hqp, that's entirely a different ballgame.

What voltage? 

Did you do just 22,000 uF --> choke --> 22,000 uF ? I would not use anything less than 66,000 uF (after the choke) with this choke. 

 

You get about 200W at 40V. But as a general guideline:

10 hours ago, Nenon said:

Unless you do heavy DSD upsampling with 120W TDP CPUs, chances are this choke will work great. As a general rule, if you can passively cool down your CPU with a HDPlex/Streacom chassis, you probably won't have any issues with this choke/design.

The i9-10900k is 125W TDP ? 

 

[Dev]

8 minutes ago, Nenon said:

What voltage? 

Did you do just 22,000 uF --> choke --> 22,000 uF ? I would not use anything less than 66,000 uF with this choke. 

 

28v. yes, the config is correct. I already had two 22,000 uF from the SJ high current rail and that is what I was re-using to test it out. I still have to do a proper high quality unregulated lps.

 

8 minutes ago, Nenon said:

You get about 200W at 40V. But as a general guideline:

The i9-10900k is 125W TDP ? 

 

 

yes but boot can consume over 200watts.