Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by OAudio

  1. Active rectifiers can provide real up tick in SQ, but to be more confident of them doing this, they need to be designed into a supply from the outset. When swopped into existing circuits, particually higher current LPSs which were designed with diode bridges originally, the impact on performance and sound quality is variable. For higher current capacity supplies at least, keep in mind that dropping in an active rectifier can generate a really very significant jump in peak ripple current in the transformer, rectifier and filter components. The 'on' resistance (Ron) of a
  2. Hi R1200CL, These are great coolers. I use the Supermicro versions in 2u size. Unfortunatly they are not efficient enough to the cool 3647 cpus passively. They are designed for server cases with fans and ducts to make sure the coolers see some forced air moving through them. Whilst working on the server for the moment I am still using the Supermicro versions of these coolers with electrically isolated fans. It's far easier to strip and reassemble the server with the Supermicro 2U coolers than with the new CNC coolers and heat pipe system installed. OA
  3. Hi Chris, No commercial offerings yet but the situation could change moving into next year so I will pm you for guidance.
  4. DC ATX supplies can put in good performances but PC's are current driven devices and at the current levels required DC ATX and other switched / buck supplies will only take you so far. Linear is perhaps the best route but it's a significant task developing them for ATX supplies as an application. CLC is attractive but can occupy a lot of space and needs to be carefully designed but can be done. We have gone for multiple linears with a list of special features in the supplies. Performance even at high(ish) currents is good and it shows in SQ achived. If you c
  5. It's a great review serise, looking forwards to part 6 !
  6. I have servers in the late stages of development, there are single and dual socket c621 varients. SQ is comparable with the Taiko. Of the single and dual socket setups the dual socket is definatly the most felxible. As ray-dude says, it's satifying to hear SQ benefits when data paths and software processes and priorities are well setup.
  7. @ray-dude great points. The sage archtecture diagram or equivalent for other dual CPU motherboard are very important inputs to working up affinity strategies. Another hint @kyoya78would be not to forget that that LTSB's kernel threads, many SW device drivers and other OS owned services cannot have their CPU affinities set by user performance tuning tools. These OS processes are scheduled by the OS across available cores on both CPU sockets, which can be challenging and needs consideration. Attaching your critical replay hardware (eg OS disk, USB ports) and setting core
  8. +1 here for the 665p Intel drives. Their SQ is usage dependant, which isn't really a suprise, but they are capable supporting very good sq when used for media storage.
  9. Try 1 dimm per channel and then 2 just to check out any sound quality change. Often the extra dimm on a channel dosent add anything or makes sound a little worse, but that might not be the case with your board 😉
  10. Hi MarcelNL, The most important thing when deciding how much memory to install is how many DIMMs are needed to have one DIMM per CPU memory controller / channel. For instance on a c621 motherboard there are usually 6 memory controllers / channels so 6 DIMMs are needed. For an X99 / X299 there are usually four controllers / channels so 4 DIMMs. Etc. Best to look at your motherboard manual to work out how to populate the physical slots so as to use all available controllers / channels (physical DIMM sockets on motherboards may not nessasarily be mapped one to one to a cpu
  11. The time has come to deal with cooling of the c621 server I am working on. For the usual SQ and ambient noise reasons the server must to be entirely passively cooled. The cooling solution must deal with the heat dissipation requirements of a high power Xeon SP series CPU and it has to integrate into my server's quite dense packaging of power electronics in the area of the CPU. Of course there are no off the shelf passive cooling solutions for LGA 3647 socket CPUs 😦. Some time ago I posted details of the CPU cooling solution I had in mind.
  12. I figured you might have the burden voltage ruled but thought to mention it just in case 🙄. Decent current at a volt is tough with a linear solution, great that you have it solved.
  13. Hi Soul Analogue, Appologies in advance if you are aware of DMM current range "burden voltages" and have already taken this into account. Is the 0.4v drop you are seeing is just the "burden voltage" for the 6ish Amp current and the meters 10amp measurement range on your Fluke ? There will be specs for the meter that explain how to calculate the voltage drop across the DMM when measuring current in a current measurement range. A 0.4v drop for a 6 or 7 amp current is in the right area for burden voltage on many meters. If it were the normal drop ac
  14. That is an interesting comment. Do you mean A DC-ATX using switched buck converters eg another HDPLEX type device. or A true fully linear DC ATX system that plugs into the motherboard and is fully controlled, managed and monitored as an ATX supply ?
  15. Alex hi, You may have misconception about how complex / expensive these server PSU measurements are to perform. Server LPSs are not ultra low noise units and the measurements we are discussing are simple, so they are straight forward to make and do not need sophisticated or expensive equipment. An oscilloscope and an electronic load for say £1-2k in total sourced from good manufacturers will do this. Its worth remembering as well that the same pieces of equipment are the tools required to develop, test and produce supplies, so they should already be on hand.
  16. The first 2 sets (copied below again) are my personally thoughts on what I would look for. The first case 1) for me would get the job of powering a DC-ATX or a single music server rail done, whilst the second case 2) would be for a more expensive class of device. Transient response, ripple and sag under load are quite basic parameters but I think they do have some correlation with SQ, and so are useful. The last set of figures 3) below are actual measurements taken from the 12v, 10 amp LPS module of the system I have built. 1) - to get the job done. "Start at < 20
  17. Defiantly agree, but has been addressed. For the project I am working on reliability and flexibility are two of the key requirements, so this challenge has had to be addressed. I think the result (described below) is possibly one of the most flexible server power system I have come across for high end audio servers. The system is certainly not a set of discrete power supplies, hot wiring ATX control wiring and crossing fingers for the "Kerr-thunk" as power is applied is too exciting / potentially expensive . All very important points that have
  18. Hi One and a half, Having said it would probably be helpful for people trying to choose a PSU for their systems if measured performances were given more, you are right, I really I ought to say where power measurements are at here 🙂. I have 5 LPS modules in the server I am working on that provide 5 linear ATX rails plus a few auxiliary rails for other stuff. Each module has 3 pcbs in the design which have been developed over a three+ year period. The modules are all specifically designed and optimized for their respective rail voltages (ie circuit differences - not a trim pot o
  19. I think it would be good to see measured performance specifications provided as well. As a general observation there are some large ticket LPSs out there for audio systems but few vendors provide measured specs for products. At the price levels we are operating at, it's not unreasonable to be looking for similar measured performance to lab class LPSs from companies like R&S Tektronics and Agilent. So ball park or better than: Sub 50 uSec 60% load transient recovery times. Sub 2mv rms ripple and noise. Sub 2mv rail sag from 10 to 100% of rated load.
  20. They look very well presented but there is an absence of measured performance figures, which seems to be the standard for audiophile supplies. The hifi wires, caps cryo parts etc etc, may be good tick boxes for some potential customers but measurements should be the starting point. Steady state noise figures at rated load would be nice as a start, but transient load response performance is proberbly the better paper indicator. Start at < 200usec for 5 amp transient rail voltage recovery with a critically damped response and <20mv over / undershoot. This is is typical for low to medium co
  21. The production slot has moved back a week because of a small delay in a prior jobs, machining starts a week on Monday, I will be dropping by the factory then to decide the finish and look over some other work. So all good thanks. I will post some updates as things progress.
  22. That's very nice close in performance for a ~50Mhz carrier. Is this with a sinusoidal or HCMOS output ?
  23. I spotted this too, but was charitable and assumed a typo, 100HZ should have been 100KHZ perhaps ? But even then is "strong" performance getting to -172dBc at 100kHz. Regardless of the clock details I think this is worth a look.
  24. @StreamFidelity great post this is an interesting route to look at. I had actually considered spinning up a PCB for a to UTP to fiber converter for isolation but hadn't realized how many are available and how cheap they are to buy . Personally I have been focused on improving timing and SI of network transmission by working inside the devices that are transmitting and receiving the stream (music server / audio server). Putting additional devices in to the network path such as switches and hubs introduces a multitude of additional variables that then need work. I guess y
  • Create New...