Vintage Yamaha NS-1000M or Linkwitz Speaker System

[FrantzM]

If you want to do it, just create crossover from scratch and keep the old one for historic purposes. But i doubt you will see/hear much difference between old/new (maybe at first yes, but then you will get used to it)

Thanks, I have become increasingly skeptical of these tweaks ... My money will be spent elsewhere.

 

[SpaceMonkey]

Thanks, I have become increasingly skeptical of these tweaks ... My money will be spent elsewhere.

It is not like it is a bad idea to do it. But you need to check whether things are broken in the first place. I saw reports on both ends. Some people check crossovers and nothing broken. Some are reporting day and night difference (but apparently parameters did drift).

 

[gene_stl]

For what some crossover tweaks cost, a person could triamp the system.

 

[FrantzM]

Hi

I am building the room. Works will begin this coming Monday. 6 x 4.8 x 2.8 meter. Concrete walls to be covered by drywall on 2 x 4 studs with insulation wool in between...

Speakers: NS-1000M
Pre/DAC/Streamer : miniDSP SHD Studio
DACs: Weson Khadas Tone Board for mains and Fioo DK-3 for subwoofers
Amp: Rotel 5-channel )Interim) I plan to get a of Purifi or Hypex-based amplifier next year
Subs: One SUB-1500 + One SUB-1200 + One SUB-100) from Parts express for a Total of $469
The decision about these specific subs, came after a spirited. " Interstellar", session and a re-visit of This Thread.

 

[dualazmak]

Hi again FranzM,

Nice to hear your decision on dedicated audio room with NS-1000M!

BTW, what would be your idea and method on "synchronization" of the two DACs? USB devices almost always run asynchronous these days (by independent ASIO drivers), which means that each DAC has it's own master clock and there is a definite risk that these DACs will drift out of sync over time, sooner or later.

In my case, I have two methods for the "sync" operation of my main NS-1000+T925A and the sub YST-SW1000 (L and R);

1. Use DAC8PRO's high quality headphone OUT (CH-1 +CH-2) with an TRS to RCA adopter for RCA unbalance input into subs. The headphone OUT of DAC8PRO in under the gain and volume control function of DAC8PRO, so in this case I can use DAC8PRO's volume as master volume of entire system.

or,

2. Use DAC8PRO's digital AES/EBU OUT (CH-1 + CH-2) into a second DAC (with AES/EBU IN, my case ONKYO DAC-1000S) and connect second DAC's RCA unbalance analog line OUT into subs. Digital AES/EBU signal always contains "sync information" into second DAC, and the second DAC is in sync with the first DAC by digital AES/EBU connection. In this case, however, the digital AES/EBU OUT (through OUT of CH-1 + CH-2) is not under the gain and master volume of DAC8PRO, this means I need master volume control upstream before DAC8PRO. I can use software crossover EKIO's digital input gains as master volume, or most upstream JRiver MC's (or Roon's) nice digital volume as master volume.

Please visit my post here and here for the details of these synchronization and master volume issues.

 

[FrantzM]

Hi

The DACs will be fed by the miniDSP Studio, they will receive the same clock, that of the miniDSP SHD Studio. I assume drift will be small, pertaining to the different processing time of the two DACs, such differences are likely in the microseconds ... nanoseconds even, I suppose those to be inconsequential in an Audio chain...

 

[dualazmak]

Hi

The DACs will be fed by the miniDSP Studio, they will receive the same clock, that of the miniDSP SHD Studio. I assume drift will be small, pertaining to the different processing time of the two DACs, such differences are likely in the microseconds ... nanoseconds even, I suppose those to be inconsequential in an Audio chain...

Hello,

Thank you for your response. I do hope your two DACs would work in sync as you expected.

I do not know in detail about Weson Khadas Tone Board (looks very nice with ESS ES9038Q2M chip!). Also I know littel about Fioo DK-3, you mean Fiio DK3??

The sync operation of the two DACs may be OK if your upstream miniDSP Studio and/or JRiver (or Roon) rather frequently reset the sync timing for the two DACs, e.g. at the start and end of each short music tracks. I am a little bit worried, however, about in case if you would play a rather long track, like a very long opera track of say 30 - 60 min long, theoretically your two DACs will drift out of sync over the time unless otherwise synced by any sync mechanism such as precision unique sync clock input to the two DACs or AES/EBU connection between the DACs.

As far as I looked at information on Weson Khadas Tone Board at amazon site, it has no sync clock signal input nor AES/EBU input....

In any way, you would better to check and confirm carefully the sync operation of the two DACs.

 

[FrantzM]

This requires further investigation...
Warning! Rambling ahead

I have taken taken a Volte-Face about the anal tendencies of us, the audiophiles. I thought for a few years that my hearing was above average only to realize that I cannot reliably tell the difference between 256 Kb/s mp3 to lossless. Granted if I know what to listen to my threshold improves to 320 kb/s but ... There is an online test, someone posted on our abilities to detect THD and I didn't fare as well as I thought , I would I believe I got to .1% but under strain and repeated exposure to the tests...
I am certain , I am not alone. I will not spend a lot of money as in the past , when $20K was what I called with some kind of pride for being finacially responsible, mu upper limits for electronics. I saw a $20,000 amp or $20K preamp or $$20K DAC or even a $20K CD Transport!!!!! as almost normal, beyond $20K for a single component I would start to moan but ...
ASR has seen to it. We all know now, that a carefully assembled Audio system doesn't have to cost and arm and a leg. If Science is used rather than pseudo-mysticism and quackery, one usually attains, a system that sound far better than several over $100K Audiophiles shrines and that costs a fraction of those. Superiority in these cases is often subjective too. This is liberating.

Rambling Off...

This out of the way:
I already have the Khadas (cost me $100) and the Fiio DK-3 is $20, shipped via Amazon Prime. I don't think you need anything better for subwoofers. The clock in SPDIF is derived from the signal, both DACs should receive the same clock... perfectly in synch , coming in, How far would they drift and how critical could this be in the low bass? Below 80 Hz? I don't know and am inviting those who know more to educate me.

 

[dualazmak]

I assume you would be quite OK in synchronization since you will connect to both DACs by SPDIF routing which does have clock signal like AES/EBU. Thank you for your detailed response!

 

[gene_stl]

I have a Rotel stereo power amp. In my recent move two of the speaker terminals got whacked off. While having the amp open to replace them I have marveled at what a thing of beauty the Rotel amp is. I have been considering picking up some of their home theater amps for multi channel and multi amping use. If you don't want that five channel and it is in the US let me know. (I recommend keeping it and maybe even using it instead of "upgrading")

 

[dualazmak]

Hello FrantzM and friends,

Just for your info and reference, I posted rather detailed information and measurements of attenuators of YAMAHA NS-1000 (and NS-1000M) at my post #248 in my multichannel thread which would be somewhat valuable for many NS-1000 and NS-1000M owners, I assume.

 

[SpaceMonkey]

Hello FrantzM and friends,

Just for your info and reference, I posted rather detailed information and measurements of attenuators of YAMAHA NS-1000 (and NS-1000M) at my post #248 in my multichannel thread which would be somewhat valuable for many NS-1000 and NS-1000M owners, I assume.

Just for extra info, in this video a guy is showing effect of l-pad (attennuator) in the middle of the video:

I guess with a bit of measurements one could simulate effects of crossovers pn drivers in ns-1000 rather well with xsim.

 

[dualazmak]

Hello SpaceMonkey,

Thanks a lot for your info on xsim and the YouTube link.

Yes, xsim is an interesting LCR network simulation (educational?) tool. I have xsim installed in my PC workstation, and tried it several times.

With xsim, to configure and simulate LCR network is rather easy and straightforward. I always feel much frustration, however, to fully characterize the virtual SP units (drivers) with xsim even I would have detailed spec sheet of the driver. Also it would be almost impossible to simulate the characteristics of specific enclosure of the speaker(s), and also impossible to simulate the room acoustics, of course. I do not deny the fundamental educational value of xsim on PC.

Even in case I would have really nice, almost flat Fq response over 20Hz - 20kHz actually measured by ECM8000 microphone, I could never simulate it with xsim in PC.

I agree with you that xsim can be used to see only the "tendency" in Fq response change, e.g. tendency for connect or disconnect one resister in parallel with a SP driver. It would be almost impossible, to determine the best R value using xsim. I believe we need to trust our ears and brain in fine tuning procedures, as you would fully agree.

 

[anmpr1]

I cannot reliably tell the difference between 256 Kb/s mp3 to lossless. Granted if I know what to listen to my threshold improves to 320 kb/s but ...

I've never tried 256, but I can't tell a difference between 320 and lossless. Just to give doubt all the benefit it deserves, when I ripped my CDs to PC I used FLAC. I guess I sleep better at night for it.

 

[FrantzM]

I've never tried 256, but I can't tell a difference between 320 and lossless. Just to give doubt all the benefit it deserves, when I ripped my CDs to PC I used FLAC. I guess I sleep better at night for it.

The operative term is "reliably", On some cuts I know very well, I am able to tell but not reliably.. a bit over 50%, last I checked, but I am not able to tell the difference on, say a new album, if well recorded and playing though a transparent system. It doesn't sound as , "lacking" in anything. Now that I listen a lot to Spotify, I don't find it lacking either.
I do know , now, that the hearing of several audiophiles, mine included, is overrated; of course, we usually have above average eyes to ears coordination

 

[Adam_M]

Hello SpaceMonkey,

Thanks a lot for your info on xsim and the YouTube link.

Yes, xsim is a interesting LCR network simulation (educational?) tool. I have xsim installed in my PC workstation, and tried it several times.

With xsim, to configure and simulate LCR network is rather easy and straightforward. I always feel much frustration, however, to fully characterize the virtual SP units (drivers) with xsim even I would have detailed spec sheet of the driver. Also it would be almost impossible to simulate the characteristics of specific enclosure of the speaker(s), and also impossible to simulate the room acoustics, of course. I do not deny the fundamental educational value of xsim on PC.

Even in case I would have really nice, almost flat Fq response over 20Hz - 20kHz actually measured by ECM8000 microphone, I could never simulate it with xsim in PC.

I agree with you that xsim can be used to see only the "tendency" in Fq response change, e.g. tendency for connect or disconnect one resister in parallel with a SP driver. It would be almost impossible, to determine the best R value using xsim. I believe we need to trust our ears and brain in fine tuning procedures, as you would fully agree.

XSIM works fine - I've used both that and Jeff Bagby's excel spreadsheets to design speakers. The key is to NOT rely on manufacturer graphs - rather measure each driver in the box. Given gating limitations, nearfield measurements of the woofer/port responses will need to be spliced/blended to the gated farfield, and then minimum phase information extracted.

Before the sim will be accurate, the z offset of the tweeter needs to be empirically determined by comparing a number of measurements so accurate phase simulation can be achieved, which is particularly important in the crossover region - as even when targeting (for example) 4th order LR slopes, the slopes are a bit asymmetric to shift driver phase a bit so the drivers are in-phase for as broad of a region as possible in the crossover region.

Without doing this, XSim isn't going to give you predictable results.