I wanted to build my DIY replacement for the now obsolete Yamaha NS10M using some existing drivers from (P)eerless and (S)EAS, as a result I am calling these the PS10M. This is my first attempt on Paper, I do have the drivers in hand, so I hope to cut some wood in the next few weeks and get this project rolling.
Attachments
I can place them in a smaller sealed box and get a similar low to the original Yamahas. The original NS10M is in a much smaller cabinet. 15" x 8.5" x 7.75" and sealed, on the other hand my design is 11.5" x 16.25" x 8" and ported. The real question is.....Is it more important to be the same or better (more extended frequency response)?
Member
Joined 2003
As this speaker appears to be a replacement for your old speakers, and not an attempt to replicate the speaker, I don't think the naming convention is a good one. What you have here is a new speaker, it shares nothing in common with the NS10M.
Now if you want to do something along the lines of a "clone", the very least you will want to do is a frequency response of the NS10M, and attempt to replicate that tonal character with your new speaker.
The crossover needs some work. There are a lot of things to consider when designing a crossover, a lot. At the very least, the tweeter is a couple dB more sensitive than the woofer, before any BSC, and you have no attenuation pad. I would recommend some more research before embarking on your own design.
Now if you want to do something along the lines of a "clone", the very least you will want to do is a frequency response of the NS10M, and attempt to replicate that tonal character with your new speaker.
The crossover needs some work. There are a lot of things to consider when designing a crossover, a lot. At the very least, the tweeter is a couple dB more sensitive than the woofer, before any BSC, and you have no attenuation pad. I would recommend some more research before embarking on your own design.
That is a question that only you can answer.MIKET said:
Crossover Design
DcibeL,
After reading your reply I went to Jay_WJ homepage - Jay's speaker building pages and did some reading. I'll saw the Designing Crossovers with Software Only section and he suggests initial design of the crossover using Speaker Workshop or Passive Crossover Designer. I downloaded both. As for the enclosure size, I think the smaller enclosure will be a better choice for what I want to build. (an NS10M replacement for nearfield mixing)
DcibeL,
After reading your reply I went to Jay_WJ homepage - Jay's speaker building pages and did some reading. I'll saw the Designing Crossovers with Software Only section and he suggests initial design of the crossover using Speaker Workshop or Passive Crossover Designer. I downloaded both. As for the enclosure size, I think the smaller enclosure will be a better choice for what I want to build. (an NS10M replacement for nearfield mixing)
Member
Joined 2003
Jay's "crossover design with software only" page is a good one, read through it thoroughly. You may also want to look here:
FRD Consortium tools guide
FRD Consortium
The FRD tools are not very user friendly, so be sure to read the documentation thoroughly to make sure you are doing everything right.
FRD Consortium tools guide
FRD Consortium
The FRD tools are not very user friendly, so be sure to read the documentation thoroughly to make sure you are doing everything right.
Member
Joined 2003
Well I guess you're done, great success!
Seriously though, I hope you take my "reasearch" comment to heart. There is a lot more to crossover design that what you have presented above. I am assuming that the graph you made above is just traced response from the datasheet, and applied the crossover values you've chosen.
What you'll end up with is a speaker that "works", but far from as good as it could be. Some general steps to use when looking at speaker design:
-trace response of drivers from datasheet or 3rd party mesurement. Make note of how this frequency response was generated, IEC baffle, free air, infinite baffle, etc, as this will influence the graph.
-trace impedance
-model bass driver on measurement baffle, as well as desired cabinet. save response (Unibox).
-model baffle diffraction using BDS (FRD tools) of your divers on the desired baffle. Save response.
-Combine response using FRC. Use FRC to extract minimum phase for the acoustic response and the electrical response (impedance). Save response. This will be your simulated response of the driver in the cabinet you designed for it.
-import resulting response to the Passive Crossover designer and start crossover design.
-take into account resulting electrical impedance and phase, this will tell you how easy the "load" is on your amp.
-take into account individual driver acoustic phase and power response. Enter driver offsets for accurate simulation (read documentation, the numbers entered here are weird)
-model frequency response on and off axis, off axis is just as important to know where the nulls occur among other things
This is just a basic run down, off the top of my head. For more detail, follow Jay's page as well as the pages I linked to above. Hopefully from that you can see why many people choose existing "proven" designs as first projects, rather than trying to spin their own. It appears clear to me that most people prefer to DIY build a speaker that works very well ie "proven" design, then have something they can say "I designed that" but doesn't work quite as well.
Seriously though, I hope you take my "reasearch" comment to heart. There is a lot more to crossover design that what you have presented above. I am assuming that the graph you made above is just traced response from the datasheet, and applied the crossover values you've chosen.
What you'll end up with is a speaker that "works", but far from as good as it could be. Some general steps to use when looking at speaker design:
-trace response of drivers from datasheet or 3rd party mesurement. Make note of how this frequency response was generated, IEC baffle, free air, infinite baffle, etc, as this will influence the graph.
-trace impedance
-model bass driver on measurement baffle, as well as desired cabinet. save response (Unibox).
-model baffle diffraction using BDS (FRD tools) of your divers on the desired baffle. Save response.
-Combine response using FRC. Use FRC to extract minimum phase for the acoustic response and the electrical response (impedance). Save response. This will be your simulated response of the driver in the cabinet you designed for it.
-import resulting response to the Passive Crossover designer and start crossover design.
-take into account resulting electrical impedance and phase, this will tell you how easy the "load" is on your amp.
-take into account individual driver acoustic phase and power response. Enter driver offsets for accurate simulation (read documentation, the numbers entered here are weird)
-model frequency response on and off axis, off axis is just as important to know where the nulls occur among other things
This is just a basic run down, off the top of my head. For more detail, follow Jay's page as well as the pages I linked to above. Hopefully from that you can see why many people choose existing "proven" designs as first projects, rather than trying to spin their own. It appears clear to me that most people prefer to DIY build a speaker that works very well ie "proven" design, then have something they can say "I designed that" but doesn't work quite as well.
Crossover Tweaking
Still tweaking the crossover values with Passive Crossover Designer, here is the best result I've gotten with software so far.
FRD and ZMA files from Driver FRD & ZMA files
I used the 850439 FRD and ZMA files since it is very close to the 830875 in many respects.
Still tweaking the crossover values with Passive Crossover Designer, here is the best result I've gotten with software so far.
FRD and ZMA files from Driver FRD & ZMA files
I used the 850439 FRD and ZMA files since it is very close to the 830875 in many respects.
Attachments
Member
Joined 2003
Those are traces from the manufacturer's data, as it says on the page. That would be the first two points in my post above, you still need to do points 3,4,5 to get something useful for crossover simulation. A frequency response without phase data is useless for crossover design.
I'm afraid I can't help you any more. Good luck with your project.
I'm afraid I can't help you any more. Good luck with your project.
its not an NS10M
MIKET,
you may get a result you like - and I hope you do - but you will never achieve an NS10M sound with a ported enclosure.
Regardless of what shape a Frequency Response plot may look like in comparison to the plot of another loudspeaker, a ported cabinet never produces the same type of bass sound as a sealed cabinet.
Their specific sealed cabinet sound is part of why NS10(M)s are used by professional recording and mixing engineers,
and often semi-pro and home-enthusiast recorders do not understand this.
I have heard about 20 pairs of original NS10 and NS10M loudspeakers.
If you really want to make an NS10M sound-a-like I will post more about them,
but you will not get that exact sound with the drivers you have, even with a sealed enclosure, and regardless of crossover type.
Why do you want an NS10M sound ?
It is a thoroughly mediocre sound - that is why the pros use it as a Reference - "reference" to common home listeners' use, not tonally and transient accurate.
You can make a somewhat more accurate reproduction loudspeaker with those drivers than an NS10M,
and that woofer will work quite well in a sealed enclosure, from which a better Transient response can be got than from any ported enclosure.
MIKET,
you may get a result you like - and I hope you do - but you will never achieve an NS10M sound with a ported enclosure.
Regardless of what shape a Frequency Response plot may look like in comparison to the plot of another loudspeaker, a ported cabinet never produces the same type of bass sound as a sealed cabinet.
Their specific sealed cabinet sound is part of why NS10(M)s are used by professional recording and mixing engineers,
and often semi-pro and home-enthusiast recorders do not understand this.
I have heard about 20 pairs of original NS10 and NS10M loudspeakers.
If you really want to make an NS10M sound-a-like I will post more about them,
but you will not get that exact sound with the drivers you have, even with a sealed enclosure, and regardless of crossover type.
Why do you want an NS10M sound ?
It is a thoroughly mediocre sound - that is why the pros use it as a Reference - "reference" to common home listeners' use, not tonally and transient accurate.
You can make a somewhat more accurate reproduction loudspeaker with those drivers than an NS10M,
and that woofer will work quite well in a sealed enclosure, from which a better Transient response can be got than from any ported enclosure.
And it's a reference to what a common home listener would use around 1980. Emulate the process don't imitate the sound of the great mixers. The process is more what alan is referring to. Find a reference to what the common home listener would use NOW. Which imo is light years away from an NS10.
Intent of project
I realize the NS10M is the default Standard in home and smaller studios for a near field monitor. My hope with the PS10M is to come up with an alternative that I built myself. It will be used exclusively as a nearfield monitor in my home studio. It does not have to be a dead knock off, of the NS10M, it just needs to flat, acurate, low distortion and a good analytical tool for mixing and tracking.
I did tune the box to a F3/Fb ratio of 1.325 based on the work of Joe Rasmussen. the link is Elsinore Box Alignment
This seems to be a good compromise for my application.
I realize the NS10M is the default Standard in home and smaller studios for a near field monitor. My hope with the PS10M is to come up with an alternative that I built myself. It will be used exclusively as a nearfield monitor in my home studio. It does not have to be a dead knock off, of the NS10M, it just needs to flat, acurate, low distortion and a good analytical tool for mixing and tracking.
I did tune the box to a F3/Fb ratio of 1.325 based on the work of Joe Rasmussen. the link is Elsinore Box Alignment
This seems to be a good compromise for my application.
A lot of the NS10M's sound is in the paper cone woofer. The cone similar in construction to a fostex or lowther driver, with a single sheet of "rolled up" paper. As such, it has a rising midrange response and an extremely fast "waterfall" measurement. These are two characteristics the peerless driver lacks.
Your speaker will not have the detailed, "see-into-the-mix", revealing characteristic of the genuine article. The sound of the NS10 is not really in it's frequency response chart (which is actually a bit different than your modeled crossover, click the first link. they have a better FR chart.), but in the construction of its woofer.
Must-read: The Yamaha NS10 Story and http://www.soundonsound.com/pdfs/ns10m.pdf
Your speaker will not have the detailed, "see-into-the-mix", revealing characteristic of the genuine article. The sound of the NS10 is not really in it's frequency response chart (which is actually a bit different than your modeled crossover, click the first link. they have a better FR chart.), but in the construction of its woofer.
Must-read: The Yamaha NS10 Story and http://www.soundonsound.com/pdfs/ns10m.pdf
Last edited:
Woofer Construction
Looking at what Fostex has to offer, there isn't a good match for the NS-10 woofer and lowther drivers are way too rich for my blood. I'm going to stick with my original selection of drivers at least for this go around and get the most out of the drivers that I've selected, thanks for all the insight, it hasn't fallen on deaf ears
Looking at what Fostex has to offer, there isn't a good match for the NS-10 woofer and lowther drivers are way too rich for my blood. I'm going to stick with my original selection of drivers at least for this go around and get the most out of the drivers that I've selected, thanks for all the insight, it hasn't fallen on deaf ears
NS10M is not "flat, accurate, low distortion".
Better can be made with your drivers.
Cross over at 2kHz or very slightly above there for your woofer.
If you want NS10 type sound, then use the crossover circuit you have posted,
but if you want "accurate" and "best transient response" you will have to use a crossover circuit which permits both drivers to be connected in the same Polarity -
{same Phase, some refer to this as, but that is confusing because Phase has other applications in understanding crossover circuits}.
A 1st Order filter crossover is not practical with these drivers for your intended application, so consider 3rd Order-same Polarity, or 4th Order.
The electrical filters do not have to be the entire Order of the crossover slopes - part of the filter slopes can be the natural roll-off of the drivers.
The Sum of the drivers' rolloffs + electrical filters is made to equal 3rd Order or 4th Order.
Is your Simulator device able to calculate this ?
The Transient response from Joe R's work is undoubtably better than the previous commonly applied for ported enclosures,
however better transient response is obtained from a sealed enclosure - and that was a positive for the NS10.
You do not need a ported enclosure for your application.
With a suitable volume sealed enclosure for that woofer you will get plenty of useful bass response in Nearfield position.
One needs to consider bass extension with regard to the actual Pitches of the notes played for the recordings.
We are not primarily concerned with making ideal shaped roll-offs in accord with filter theory, because for the particular driver chosen a theoretically perfect filter roll-off may miss some of the important notes played by the musicians.
One has to consider Sound and Psychoacoustics.
That 6 1/2" woofer will not reproduce the 41Hz fundamental of the open E string of a bass guitar, but that is not necessary if one includes how the human brain interprets small amounts of harmonic distortion - and the bass guitar itself produces harmonic distortion, as will the woofer.
For that woofer, choose an enclosure size that causes the in-box Fo Impedance peak of the woofer to be at 60Hz, or as close to that,{towards slightly higher}, as is possible.
That will likely be approximately a 20 litre enclosure, {allowing for the space occupied by both the drivers in the enclosure}, perhaps 18 litre but unlikely smaller.
What does your simulator predict for Fo = 60Hz in a sealed enclosure ?
- and I do mean Fo and not F3.
Forget about F3 for now.
If you will only use a ported enclosure then tune it to 40Hz, and that is permissable with that driver.
Vb will be somewhere between 80% <---> 100% of Vas.
But, you will find with ported enclosures that notes with pitches near to the tuned frequency of the port will sound out of tune, especially when the speaker is being driven hard,
and you will hear delayed midrange resonant sound coming through the port.
The frequency area/s and bandwidth/s of these resonances will be determined by the length and diameter of the port, and by the geometrical position of its open end inside the enclosure.
These are different for every different loudspeaker, thus mixing for acceptable sound balance on one particular midrange-resonant through port loudspeaker will not necessarily give balanced spectrum replay via a different ported loudspeaker.
Home-listeners will choose ported loudspeakers with resonant colourations that please them, and differently for each buyer, but you will have to mix to allow for all cases.
Yes, the large Monitors in most Recording Studios usually have ported enclosures, but those are always tuned to 30Hz or lower -
{usually tuned in the 20Hz <--> 25Hz region}- and have much larger enclosures than almost all Home-use loudspeakers.
The midrange resonant energy can be controlled in a less audibly intrusive manner when one designs a large enclosure than can be easily done in a small enclosure.
The small Active pro-monitors with ported enclosures have electronic high-pass filters to assist reducing some anomolies caused by the ports.
The Semi-pro/Home-studio small Active monitors are mostly not as good as their manufacturers claim them to be.
Regardless of whether sealed or ported, the internal ratios of Height versus Width versus Depth versus geometrical position of the woofer in the baffle all determine the degree of midrange resonant behavior back-effect on the woofer cone, and thereby the sound emmitted from the cone.
Does your simulator include calculations for this ?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.