My First Design & Build Project

Hi,

I have recently started my first design & build project where I am aiming to produce a portable multi-way loudspeaker. Considering I have a background in mechanical engineering, I plan to use simulation software (ANSYS) to verify and optimise my enclosure design before fabrication using additive manufacturing methods.

At this very moment, I have a few conceptual enclosure designs that all use following Dayton Audio drivers:

  • ND105*4 4" Aluminum Cone Midbass Driver 4 Ohm
  • NHP25Ti-4 1" Titanium Dome High Power Neodymium Tweeter
    4 Ohm (x2)

I have also considered using the following passive radiator in one of the designs:

  • DMA80-PR 3" DMA Series Passive Radiator (x2)

Would anyone be able to help me identify an appropriate crossover circuit that I can buy for this driver arrangement? (Any other advice regarding my project would be greatly appreciated, this is my first project 🙂)
 
Hello,
I understand that you want to built low price speakers and you're looking for an assembled crossover.
DaytonAudio 2 ways crossover
Considering the data of the speakers you have chosen the above could make the job but will remain the difference in SPL (94 dB vs 86.8 dB) that this kind of crossover doesn't handle... I'm afraid the results will not be really staisfying (tiring listening).
Good luck
 
Off the shelf crossovers assume an ideal world - flat impedance and frequency response. This is never the case in reality (look at the manufacturer frequency and impedance curves). Ideally - you would construct the box, put the drivers in and measure yourself - to then simulate/design your own crossover.

A "next best" option is to use manufacturer response data (Frequency and impedance) - apply the effects of your enclosure and baffle shape to then simulate / design a crossover.

I wouldn't recommend buying a crossover. The frequency response will be all over the place and worse, impedance could dip below the level your amplifier can tolerate.
 
Hello,
DaytonAudio 2 ways crossover
Considering the data of the speakers you have chosen the above could make the job but will remain the difference in SPL (94 dB vs 86.8 dB) that this kind of crossover doesn't handle... I'm afraid the results will not be really staisfying (tiring listening).
Good luck

Hi Lionel,

Thanks for your help, unfortunately the hyperlink you have provided for the crossover is not working.

Also, I have not yet bought any drivers for this project, are you suggesting that I pick drivers that have similar if not the same SPL?
 
A "next best" option is to use manufacturer response data (Frequency and impedance) - apply the effects of your enclosure and baffle shape to then simulate / design a crossover.

I wouldn't recommend buying a crossover. The frequency response will be all over the place and worse, impedance could dip below the level your amplifier can tolerate.

Hi Dave,

Thanks for your advice! My project is fairly time-restricted and as as a result I don't think I have enough time at the moment to design a crossover circuit. I will be mainly looking at investigating how the enclosure geometry affects the frequency response.

Have you got any further recommendations or advice for choosing an off the shelf crossover?

Cheers 🙂
 
Hi Dave,

Thanks for your advice! My project is fairly time-restricted and as as a result I don't think I have enough time at the moment to design a crossover circuit. I will be mainly looking at investigating how the enclosure geometry affects the frequency response.

Have you got any further recommendations or advice for choosing an off the shelf crossover?

Cheers 🙂

The enclosure geometry is far less important than the enclosure volume and a decent crossover designed for the driver combination in mind.

It doesn't take much to simulate the enclosure volume. WinISD is very easy to use.

PS: Are you designing a 2 way with single woofer and tweeter? (MT) or 2 woofers (MTM) or something else?

The woofer is only 86dB sensitive. After bafflestep (unless you are using these in a bookshelf / hard up against a wall), the overall system sensitivity will be ~ 83-84dB. Your tweeter is 94dB sensitive - so you are going to need to add an lpad or series resistor to the tweeter in your off the shelf crossover... or your tweeter will dominate. by 10dB (i.e. twice as loud as the woofer).

I don't have any off the shelf / premade crossovers I'd recommend. If you insist on going that way and you are building an MT - you will want a 4 ohm crossover targeting 4th order slopes at 3KHz or greater.

If you are going MTM - then you will need to series your woofers (for 8 ohms) and get an 8 ohm (woofer) and 4 ohm (tweeter) crossover combination

I could simulate the crossover for you doing a quick and dirty usingthe Dayton supplied manufacturer files. If you choose an off the shelf crossover - I can show you (within a small margin of error) how it will measure and likely sound.
 
^ +1 on the above.

Enclosure geometry is pretty trivial. Between Hornresp (lower frequency stuff) and Edge (baffle diffraction etc), you can cover the full frequency range in terms of acoustic simulations.

Since you don't want to get into crossover design, I'd suggest using full-range drivers in the 3-4" range.

Chris
 
Here is some quick food for thought for your proposed drivers, but they do not consider driver separation or the resulting phase differences etc between drivers nor cabinet dimension caused diffraction, or flush mounting of drivers etc.

Treat as a basic starting block.
 

Attachments

  • 4 ohm daytons.PNG
    4 ohm daytons.PNG
    11.1 KB · Views: 172
  • 4ohm dayton stuff.PNG
    4ohm dayton stuff.PNG
    40.1 KB · Views: 176
The enclosure geometry is far less important than the enclosure volume and a decent crossover designed for the driver combination in mind.

Are you designing a 2 way with single woofer and tweeter? (MT) or 2 woofers (MTM) or something else?

I don't have any off the shelf / premade crossovers I'd recommend. If you insist on going that way and you are building an MT - you will want a 4 ohm crossover targeting 4th order slopes at 3KHz or greater.

If you are going MTM - then you will need to series your woofers (for 8 ohms) and get an 8 ohm (woofer) and 4 ohm (tweeter) crossover combination

I could simulate the crossover for you doing a quick and dirty usingthe Dayton supplied manufacturer files. If you choose an off the shelf crossover - I can show you (within a small margin of error) how it will measure and likely sound.

Initially I had a 2-way speaker in mind consisting of one woofer and two tweeters but now I'm thinking the 2nd tweeter is not very necessary considering the sensitivity difference. I haven't purchased any parts yet so I am fairly flexible and open to any alternative driver combinations that you think is more appropriate.

It must be said that size is one of the most important constraints for me as I plan on making a portable speaker that is of a similar volume to a lot of the Bluetooth speakers you see on the market nowadays.

It'd be great if you could suggest some better drivers that fit within the scope of this project? (max cone diameter of 4")
 
^ +1 on the above.

Enclosure geometry is pretty trivial. Between Hornresp (lower frequency stuff) and Edge (baffle diffraction etc), you can cover the full frequency range in terms of acoustic simulations.

Since you don't want to get into crossover design, I'd suggest using full-range drivers in the 3-4" range.

Chris

Cheers for sharing the info on simulation software.

From the research I've done thus far, it seems like full-range drivers are incapable of covering the entire audible frequency range hence why I am opting for a multi-way speaker.

I am hoping that the benefits of a multi-way arrangement outweighs the negatives of using an off the shelf crossover.
 
Here is some quick food for thought for your proposed drivers, but they do not consider driver separation or the resulting phase differences etc between drivers nor cabinet dimension caused diffraction, or flush mounting of drivers etc.

Treat as a basic starting block.

Thanks for this Raymond, much appreciated :up:

Which simulation software have you used for this?
 
I took the driver data from the Dayton website unzipped the files and imported frequency and impedance files into the Fine-X programme from Loudsoft.
See here:
FINE X-over™ | Intuitive X-over Design at your fingertips

Its a great tool, available for purchase.

For free software for personal use it is probably worth learning to use Kimmo's VituixCAD. Available From here:
Software
The drivers you have initially chosen may be in the speaker directory folder, which would be nice.

There is also Xsim which many also use and like from bwaslo:
http://libinst.com/Xsim/XSimSetup.exe

I could have used either of the above software packages to provide a simulated crossover as Dayton produce nice files to work with.

The dayton files provided for downlaod will be measurements taken against their reference baffle. Doing it this way does not take into account the front baffle dimensions of a design as these influence how the sound propogates from the baffle giving diffraction loss/Baffle step and other ripples associated with the cabinet shape and edges.

The two pieces of software mentioned above are free. Both designers designers are contributers to discussions here at diyaudio. I imagine they would appreciate some voluntary payment in appreciation of their work.

Bwaslo also has the OmniMic system for taking meaurements see below.

It should be noted to use data from your own design and verify the response characteristics you will definitely need a measurement microphone and suitable software which is another learning exercise. You could use the OmniMic V2 to do this or you would need to purchase a seperate mic usually phantom powered, a matching dual channel sound card with phatom power mic feed and other software such as REW or ARTA.

Good luck with your project.

Merry Christmas.