Yamaha NS-777 Crossover

[Purplehazeffc]

Hello All..

I would like to upgrade/change the crossovers in my Yamaha NS-777 tower speakers.
With dual 8″ woofers, 5″ mid, 1′ tweeter. I find the crossover values could be improved. The current values are 1000hz – 4000hz which I find for 8″ woofer, 1000hz a bit to high. I played just through the 8″ woofers & way to much vocals are coming through. Personally I think they should be crossed over at the 500hz with the 5″ mid easily playing the 500hz – 4000hz

As I'm coming from car audio where in a 3 way front stage it's crossed over at 400hz & 3000hz. Also the majority of new 3 way speakers are the same.

As their factory crossover has just coils & caps, without any resistors etc. Is it viable just to do the same with a new designed crossover? I have included the diagram of the factory crossover & the specs of the speaker. Is changing the crossover a viable thing to try?? As it would be way cheaper than possible buying new speakers.

Thoughts?? And how best to design a new crossover??

Cheers

 

[MaxwellsEq]

I think you will be wasting your time and money. If you are unhappy with the sound, use your money to buy something that measures better or meets your needs better. You haven't said what sounds wrong with the current speaker. Can you run REW with a Umik microphone to identify issues?

The crossover is always a compromise permitting the maximum from the drivers AND the cabinet. There is no rule about what frequency is best to crossover at, and it's OK for the bass driver to carry vocals as long as the driver isn't becoming non-linear or beaming too much.

 

[alex-z]

Disconnect the drivers from the stock crossover, get yourself some measurements of the raw drivers in the cabinet. Also do impedance sweeps.

How to make quasi-anechoic speaker measurements/spinoramas with REW and VituixCAD

Quasi-anechoic measurements are a way to capture the 'real' frequency response of a speaker without the influence of a room. They are called 'quasi-anechoic' because they have the potential to get you really close to the results one might achieve with a non-reflective environment like an...

www.audiosciencereview.com

With that data you can design a far better crossover using software such as VituixCAD.

What you will likely notice is that the stock crossover was chosen more for power handling and low cost, rather than a consistent radiation pattern and neutral tonality. The tweeter may need to be padded down slightly to match the mid-range sensitivity, and perhaps the lower woofer should have an inductor added, making a 3.5 way design, for better vertical dispersion.

 

[AudiOhm]

I would first start by doing some crossover homework, learn what each component is for, etc.
The Crossover Design Cookbook

Once you understand the crossover and components you will be able to decide what you want from these speakers.

As mentioned above the manufacture has chosen the crossover design that best fits the drivers and cabinets.

Others with the same idea, oh this is you, looks like the same response here...

Ohms

 

[AwesomeSauce2015]

I think you will be wasting your time and money. If you are unhappy with the sound, use your money to buy something that measures better or meets your needs better. You haven't said what sounds wrong with the current speaker. Can you run REW with a Umik microphone to identify issues?

The crossover is always a compromise permitting the maximum from the drivers AND the cabinet. There is no rule about what frequency is best to crossover at, and it's OK for the bass driver to carry vocals as long as the driver isn't becoming non-linear or beaming too much.

I'm going to have to agree here.
There are no exact rules regarding where you can cross over certain components. Also, if you disconnected the other drivers and played audio just through one of the drivers, then you will impact the crossover/driver system's response due to the impedance changes.

It is entirely possible that yamaha designed the woofer to play up to those frequencies without significant distortion.

If you absolutely must tinker with these speakers, my recommendation would be to get something like the Dayton DATS, which will allow you to measure the driver parameters, and a UMIK to measure the responses. Once you have all that information, you could possibly implement a different crossover, or build a DSP implementation.

Of course, once you have done all that, you might as well have just gone and bought a better speaker in the first place.

Now, I can help you if you know what you don't like about the speaker as a whole. So my question is: What don't you like about the speaker as a whole system when you listen to it? (With all the drivers playing through the stock crossover.)

 

[Purplehazeffc]

Cheers for all the replies. Much appreciated.
I do have a Dayton UMM-6 microphone that I have been using for car stereo usage. So, I can do the speaker measurements, as I also (thou learning) use REW software.
The main reason that got me thinking was watching a review of the Focal Aria 936's Where the crossover values are 270hz & 3100hz.
They also have 2 x bass, 1 mid, 1 tweeter set up. I also understand that Yamaha designed those drivers & crossovers for a reason.
The crossover components have been used for a certain price point. So, I may just look at replacing those like for like with better components.

 

[AwesomeSauce2015]

Realistically, the quality of components in the crossover probably doesn't influence the quality unless they are absolutely horrible.
Yes, you probably could design a better crossover for a large cost, which may improve the sound quality.

Really, what I would do in your case, is get a DSP unit, and multiple channels of decent amplification. Then, you can rip out all the passive crossover components (except maybe a capacitor to protect the tweeter), and do a custom DSP crossover. This solution has the added benefit of you being able to easily experiment and change parameters with no real time or material cost, as well as you being able to implement EQ for each driver.

 

[gene_stl]

Why on earth, would you imagine, that you, as an non engineering trained hobbyist, could design a better crossover network, than the experienced team at Yamaha????

Just because you read somewhere the capacitors and coils could be inaudibly improved on, in an article probably written by an idiot. There are a lot of those and they are in the majority. Out there in internet land.

The way you word the question screams inexperience.

 

[AwesomeSauce2015]

There are real examples backed up by facts about how some crossover networks are compromised. For example, in the KEF Q100 speakers, they only use 1st order slopes, which by themselves aren't horrible, but they do fail to take care of a few response issues which do compromise the listening experience. When you replace the stock crossover with a better (read: more expensive) design, then you do improve the quality of the speakers both measurably and subjectively, since the KEF coaxial driver array is a very well designed driver and has lots of potential.

I already mentioned above that replacing the crossover components like for like will probably not improve sound quality, unless something is broken.
Yes, there are a lot of reviews out there saying that using fancy (expensive) resistors (or whatever other components) will somehow magically improve sound quality. These reviews are usually flawed since the reviewer does not perform proper A/B testing or proper measurements, and the reviewer will usually have either expectation bias, or they will have been sponsored by the manufacturer of the expensive parts.

With that out of the way, my opinion remains:
The crossover on the speaker looks decent, but if OP were to design a DSP crossover with EQ, they could definitely improve the sound quality of the speaker, even without special engineering training... Especially given the abundance of resources out there. It would not be a trivial project, since OP would need to understand the basics of sound reproduction in rooms to be able to properly design something, but it would be a nice DIY project to facilitate easy tinkering and learning.
Some areas I would look into:
EQ on the woofers, and using 2 different crossover points on them, as well as high-pass filtering them for lower distortion.
EQ on the tweeter and midrange, separately, with separate amplification for each.
If you do those properly, then the end result will definitely be an improvement over the stock system.

 

[MaxwellsEq]

Frequency Response is not the sole reason why crossover frequencies and slopes are selected.

• By changing values, the OP will almost certainly change the dispersion characteristics which will impact the fidelity of the reproduction, even if the on-access FR is "improved".
• There may changes to resonances within the cabinet and in its walls, again impacting fidelity
• There may be driver anomalies such as resonances which affect distortion, even if the FR is "improved"

 

[fubarnow]

Send them to Danny at GR Research and let him design a new crossover.

 

[valerianf]

You need to buy DATS and measure the impedance curve from each speaker individually.
Then you will import the values in Virtuixcad and simulate the existing crossover.
From there, it is up to your creativity to improve what was made for an industrial production.
You have a huge advantage: no time and money constraints.
But the learning curve could be long and you may have to change a driver if it's performance is the weak point.
REW+UMIK will allow you to validate your changes.
Welcome to this speaker tuning hobby.

 

[alex-z]

You need to buy DATS and measure the impedance curve from each speaker individually.
Then you will import the values in Virtuixcad and simulate the existing crossover.
From there, it is up to your creativity to improve what was made for an industrial production.
You have a huge advantage: no time and money constraints.
But the learning curve could be long and you may have to change a driver if it's performance is the weak point.
REW+UMIK will allow you to validate your changes.
Welcome to this speaker tuning hobby.

DATS is not required, you can use any PC soundcard + high precision resistors. REW supports impedance and T/S parameter reading.

Measure Thiele Small parameters using free software - Audio Judgement

Learn how to measure thiele small parameters, or just a speaker impedance measurement using your sound card. Software used: Room EQ Wizard.

audiojudgement.com

 

[Purplehazeffc]

There are real examples backed up by facts about how some crossover networks are compromised. For example, in the KEF Q100 speakers, they only use 1st order slopes, which by themselves aren't horrible, but they do fail to take care of a few response issues which do compromise the listening experience. When you replace the stock crossover with a better (read: more expensive) design, then you do improve the quality of the speakers both measurably and subjectively, since the KEF coaxial driver array is a very well designed driver and has lots of potential.

I already mentioned above that replacing the crossover components like for like will probably not improve sound quality, unless something is broken.
Yes, there are a lot of reviews out there saying that using fancy (expensive) resistors (or whatever other components) will somehow magically improve sound quality. These reviews are usually flawed since the reviewer does not perform proper A/B testing or proper measurements, and the reviewer will usually have either expectation bias, or they will have been sponsored by the manufacturer of the expensive parts.

With that out of the way, my opinion remains:
The crossover on the speaker looks decent, but if OP were to design a DSP crossover with EQ, they could definitely improve the sound quality of the speaker, even without special engineering training... Especially given the abundance of resources out there. It would not be a trivial project, since OP would need to understand the basics of sound reproduction in rooms to be able to properly design something, but it would be a nice DIY project to facilitate easy tinkering and learning.
Some areas I would look into:
EQ on the woofers, and using 2 different crossover points on them, as well as high-pass filtering them for lower distortion.
EQ on the tweeter and midrange, separately, with separate amplification for each.
If you do those properly, then the end result will definitely be an improvement over the stock system.

Running a DSP makes the most sense, not in a $$ point of view. But yes you can change values quite easily for best results,
As I already run a Helix DSP running a active 3 way in my car. With a 6'5" woofer, 3'5" mid & 1" tweeter with 450hz & 3000hz at 24db slopes.
I know a little bit about tuning a DSP. Running those crossover points in my car and looking into what my Yamaha's were. I was actually quite surprised of the
1000hz point for twin 8' drivers. So really it was just a thought & to see what others with more knowledge thought as well.

Send them to Danny at GR Research and let him design a new crossover.

That would be a bit difficult. As I'm in Australia. Shipping costs would be outrageous lol. I don't agree with everything that Danny says or believes.
But I do think that better crossover components can make a difference. Specifically, when you see the cheap Electrolytic capacitors that are used in the NS777's

So will be keeping the as is for now. They are a pretty good speaker for a 20 year old design.

 

[fubarnow]

Running a DSP makes the most sense, not in a $$ point of view. But yes you can change values quite easily for best results,
As I already run a Helix DSP running a active 3 way in my car. With a 6'5" woofer, 3'5" mid & 1" tweeter with 450hz & 3000hz at 24db slopes.
I know a little bit about tuning a DSP. Running those crossover points in my car and looking into what my Yamaha's were. I was actually quite surprised of the
1000hz point for twin 8' drivers. So really it was just a thought & to see what others with more knowledge thought as well.


That would be a bit difficult. As I'm in Australia. Shipping costs would be outrageous lol. I don't agree with everything that Danny says or believes.
But I do think that better crossover components can make a difference. Specifically, when you see the cheap Electrolytic capacitors that are used in the NS777's

So will be keeping the as is for now. They are a pretty good speaker for a 20 year old design.

So us all a pic of that crossover

 

[AudiOhm]

Yamaha NS-777 Crossover

Ohms

 

[Purplehazeffc]

Yamaha NS-777 Crossover
View attachment 268657

Ohms

Yep, that's the one.

Cheers

 

[voodooless]

The 1 kHz crossover likely was chosen to not need to use baffle step compensation on the midrange. That yields higher efficiency, thus preventing like a dozen more crossover components.

This is clearly a budget speaker. Creating a new crossover could easily cost you 1/3 of the price. You’ll never get it right the first time, so will need multiple component values.

One thing about the current crossover stands out: they did a decent placement of the coils! You don’t see the done correctly very much. That’s a lot of electrolytics though!

 

[fpitas]

Can you design a better crossover? Maybe. Will it take you years before you know what you're doing? Probably.

Don't believe those online calculators that use perfect 8 ohm terminating impedance. I have never seen a driver like that.

 

[Hillas]

Hi there, do you maybe want to sell your original NS777 crossover?