Dayton Audio ND65-8 Actual vs Advertised

[joshk03]

The enclosure definitely helped. I had to use a towel instead of the proper acoustic insulation since I am still waiting on that delivery.

As a reminder, here's the open-air data from the opening post in the thread. It's not very valid, but it's interesting to see side-by-side with the new enclosure data.

And here is the data with the cement enclosure and a substitute insulation inside.

And the speakers are not broken-in yet so this should improve.

 

[KEW]

The Dayton measurements do seem a bit optimistic for a 2-1/2" full range driver!

 

[shadyJ]

I'm guessing Dayton's measurements are more realistic. The CBT24 used 2.5" drivers with a higher Fs yet the CBT24's low-frequency elbow doesn't occur until 120 Hz or so.

 

[TLS Guy]

Well done! There seems to be some sort of interference reflection on the inside of the cabinet causing that null centered in 375 Hz. Hopefully better damping material will help. That effect though is a problem with symmetrical enclosures as a dominant reflection can arrive at the cone at the same time and cause those sort of problems.

I would think that your low frequency roll off measurement is about right for a sealed enclosure that size, given the Fs of 77 Hz.

I have researched that driver some and there are reports of success with a ported alignment for that driver. However you won't get a lot of bass from a 2.5" driver. But it should work for a sub sat system.

 

[joshk03]

I'm guessing Dayton's measurements are more realistic. The CBT24 used 2.5" drivers with a higher Fs yet the CBT24's low-frequency elbow doesn't occur until 120 Hz or so.

I'll keep working toward it and see where it ends up

Well done! There seems to be some sort of interference reflection on the inside of the cabinet causing that null centered in 375 Hz. Hopefully better damping material will help. That effect though is a problem with symmetrical enclosures as a dominant reflection can arrive at the cone at the same time and cause those sort of problems.

I would think that your low frequency roll off measurement is about right for a sealed enclosure that size, given the Fs of 77 Hz.

I have researched that driver some and there are reports of success with a ported alignment for that driver. However you won't get a lot of bass from a 2.5" driver. But it should work for a sub sat system.

Thanks. I had planned to add ribs inside the orb/egg shape to scatter the waves. But you mentioned asymmetry, maybe I will create a spiral rib and some other odd geometry. Good idea.
After I have a second enclosure I will burn-in the set. Hopefully with the burn-in, better insulation, and the asymmetry these things are solid performers.

 

[Johnny2Bad]

I'll keep working toward it and see where it ends up


Thanks. I had planned to add ribs inside the orb/egg shape to scatter the waves. But you mentioned asymmetry, maybe I will create a spiral rib and some other odd geometry. Good idea.
After I have a second enclosure I will burn-in the set. Hopefully with the burn-in, better insulation, and the asymmetry these things are solid performers.

Did you use an application to punch in the Thiele-Small parameters to determine your cabinet volume or just go by some advice from Dayton / Parts Express? The reason I ask is because the Thiele-Small parameters will change with break-in, which could affect the desired cabinet dimensions, port sizes, etc.

Most serious speaker designers will use a specific frequency / SPL tone to break-in a driver, but it can be annoying, especially in domestic situations where others have to put up with your experiments. You should be fine with full range music for perhaps 12 hours, you could go a bit longer if you have the time but that should be enough.

But if your enclosure dimensions are already fixed, you could just go ahead with using the speakers instead, same effective result. Break-in is more useful if the cabinet dimensions are still to be determined, and a crossover is yet to be designed.

Adding internal damping will effectively enlarge the enclosure, from the driver's perspective, which can benefit LF response, although with very small drivers, maybe not so much. Good luck with your project.

 

[joshk03]

The product page has this info I am going by:

Optimum Cabinet Size (determined using BassBox 6 Pro High Fidelity suggestion)

• Sealed Volume 0.02 ft.³
• Sealed F3 111 Hz
• Vented Volume 0.05 ft.³
• Vented F3 55 Hz

 

[Winkleswizard]

A couple of quick checks on your efforts so far...

1. What are your measurements conditions? Indoor or outdoor? If indoor, what is relation to surfaces? Are you making gated measurements?

2. What are you doing to ensure a tight driver seal with the enclosure? Any leaking or vibrating is more likely to cause frequency response aberrations than things like break-in.

Measurement of speakers can be very tricky. A lot of factors can affect the results. I think you have already experienced some of this. REW’s help section is pretty good, but there is no substitute for experience. Hope this helps!

Ww

 

[joshk03]

I'm doing it near the corner of a large room with the speaker and mic at 45 degrees to the walls. This way both the front and back radiation is reflected completely away from the speaker and not into the mic. I will probably repeat the tests outdoors once I have a final design.
The speaker has a seal. I allow a tiny leak near the wire exit. If it were completely sealed it would inflate like a balloon when the atmospheric pressure changed and really sound bad.

 

[Winkleswizard]

Will try to help more later tonight (am without power at home right now) but the frequency response notch is more likely a room interaction artifact.

How far is your mic from the speaker?

We

 

[joshk03]

20 inches from speaker to mic. The notch is not in both graphs though, and I used the exact same location.

 

[joshk03]

So what is the standard distance for taking measurements? 1 meter? Is more than one distance used then combined?

 

[Winkleswizard]

So what is the standard distance for taking measurements? 1 meter? Is more than one distance used then combined?

Yes, 1m is pretty standard and yes you sometimes may combine different measurement techniques...

Hoping you did all your soundcard calibration and are using a calibrated mic with REW, then you should read this: https://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements

Not sure about your earlier statement, but if you are comparing a free air measure and one in an enclosure, you may see comparable higher frequency response, but will not on the lower frequency range. Any near room boundary (wall, ceiling or floor) will affect your measurements. You want to be as far away from the boundaries as you can be. Otherwise, you will see notches/bumps as one or more reflections come into play. One way to check this is to change the position of the speaker in the room and keep as much else the same. If the frequency response changes dramatically, you are dealing with room interference. Otherwise, you can start to contemplate driver/enclosure issues. Read the link above for how to gate measurements too.

As for leaving pin holes in a sealed design, this is may affect the frequency response, but is more likely to show up as distortion and/or some other noise. Usually drivers or enclosures are sufficiently leaky to deal with atmospheric pressure changes. Are you following design instructions that recommended this?

Finally, while not wanting to discourage some technical learning, if you are following a known, tested design for this speaker, would follow it. If it plays well enough, enjoy it and would not worry too much about measuring it.

Ww

 

[joshk03]

My mic is USB, so the soundcard is not involved with recording. I don't think you can calibrate the headphone jack... ?
I haven't used gated. That minidsp article says it ruins readings below ~230hz anyway. I don't need high precision yet.
I am designing the enclosure myself since I wanted a cement enclosure. My goal is great DIY speakers. The measuring helps me improve. I am on version2 and expect there will be a version3 before I am done.

 

[Winkleswizard]

If you have a line in, you can calibrate the output from the headphone jack with a loopback cable.

The point of using a gated measurement is that you are removing the effects of reflections. Yes, it does affect the low frequency measurement, but it is invalid due to room effects anyway. So, if you do not gate your measurement in a room, you either have to use near field measurements or do outdoors (either away from surfaces or using ground plane approach).

Since many of us do not have access to anechoic rooms, we have to make compromises. Usually these compromises mean that exact duplication of other's results is improbable. As I have indicated, this means understanding your room environment and that comparisons are relative to it. As an example, the Dayton FR graph for the says that the response under 450 Hz was done near field. This is because the lower frequency response is more affected by the room conditions and a near field measurement is how they chose to compensate for them.

Aside from the floor/ceiling artifacts, wall reflections can impact the accuracy of your measurements. So, if you measure in a corner, you will get more reflections and the resulting measurements will be less accurate.

One other factor for DIYers is measurement consistency. Unless you have a dedicated measuring equipment and space, you need to re-establish the measurements conditions consistently. This means establishing a baseline and checking against it every time you measure. This means documenting positions, distances, computer and amplifier settings to be able to produce comparable results between design sessions. Otherwise, you will not even be able to compare your own results meaningfully.

While you are wise to plan for design iterations, you will save time and materials (and stress) by improving your understanding of acoustics and being consistent.

Enjoy!

Ww

 

[joshk03]

No I don't have any line-in or mic port on my laptop.
But my Denon AVR will be here Wednesday, and after it arrives I will use it as the DAC (via HDMI) for my measurements. I trust that more than my laptop. And my UMIK mic is arriving today!
Also, I document everything about my measurements in the filename so I never forget.

Many homeowners have a anechoic room and don't even realize it. Their attic. My attic is huge, with blow-in insulation covering the whole 'floor' and studded wooden 'walls' that reflect sound toward the insulated floor. That beats the heck out of any room in the main living area. Clever, ey?

 

[Winkleswizard]

Your attic might be better than your other rooms, but despite what might seem better, it is unlikely to help where you need it most - low frequencies. It is not going be close to anechoic at bass frequencies.

Outside is likely the best you can do. If not doable, then try your attic. Even it were a major improvement, mine is too small and too hot to tolerate for very long!

Ww

 

[joshk03]

Still waiting on the mic, but I thought this subject would be interesting to some... You remember those incredible pictures Apple presents during their product launches? They say "Shot on iPhone7" with the fine print that "Additional equipment was used". Have you seen the additional equipment?

The Truth About ‘Shot on iPhone’-Style Ads

Now I don't have enough experience to suggest Dayton Audio does the same, but my point is that you can't know the lengths they went to so they could take advantage of your lack knowing.

 

[joshk03]

Yay. Today I got my UMIK microphone and my proper insulation. Sadly, my version2 cement enclosure is still too wet for assembly. But version1 got the full test!
I went outdoors and fired at the UMIK from 1 meter away. The noise floor in my backyard seems to be 55db according to the umik/rew SPL meter. That's louder than my attic, but hopefully acceptable.

Here are the results

As you can see, I am getting very close to satisfaction. Now, for sanity sake, here's the exact same data, but not stretched over the Dayton Audio chart.

I'm not liking the notches at 260 and 380, but they are acceptable? I am really hoping the version2 cement enclosure with it's asymmetric spiral rib interior will be even better. But overall I am satisfied with my progress. Suggestions welcome!

 

[Winkleswizard]

Try a ground plane measurement or some other significant change in the speaker position. If the notches move, it is likely a measurement artifact and can be ignored. If they stay, they are inherent to the design and will need to determine how to address...

Ww