The build of the Heissmann Acoustics DXT-MON

[EXIF68]

No. i mean a 3rd way for DXT MON with e.g. Wavecor WF223BD01 in 50 Liter BR with passive crossover to the DXT MON.
Take a look to the pipeline:
https://heissmann-acoustics.de/en/in-der-pipeline/

 

[EXIF68]

What's the internal volume of the cabinet approximately?

Internal volume is 9.5 liters.

 

[andreasmaaan]

Internal volume is 9.5 liters.

Oh, sorry. I also should have asked which passive radiator it is?

I thought it said it on the Heissmann website, but it doesn't..

 

[EXIF68]

the passive Radiator is from Seas, Modell SP18R.

 

[andreasmaaan]

the passive Radiator is from Seas, Modell SP18R.

Thanks. My sim shows that the predicted saddle in the impedance respopnse for that woofer/passive radiator in a 9.5L enclosure is arond 60Hz, which is a little higher than measured.

Adding 5g to passive radiator drops the saddle 5Hz to about 55Hz.

If you're measuring 55Hz now, adding 5g should bring you down to about 49Hz-50Hz, IMO.

 

[EXIF68]

Thank you. I will try this. On the backside of the passive Radiator is a screw mounted in the middle of the membran usable for adding mass. I have an old ladder scale from the rummage sale. This thing is perfect for such less weight.

 

[Kustomize]

Try measuring at 1 meter. Gated with REW. With 5db scale instead of 10. Everything looks good.

 

[digitalfrost]

No. i mean a 3rd way for DXT MON with e.g. Wavecor WF223BD01 in 50 Liter BR with passive crossover to the DXT MON.
Take a look to the pipeline:
https://heissmann-acoustics.de/en/in-der-pipeline/

Oh wow, they're also working on

DXT-MON-RNX
Seas-DXT & Seas ER18RNX in ~ 17l BR. -3dB around 48Hz. Baffle with bevels.Development largely completed. Release timely!

So bascially the same setup as Mark Ks ER18DXT. This will be interesting.

 

[EXIF68]

Try measuring at 1 meter. Gated with REW. With 5db scale instead of 10. Everything looks good.

What is REW?
My measurement system allows the following window-types:
rectangular
hamming
half hamming
hanning
half hanning
bingham
half bingham
all measurements above are done with "rectangular" window.
5db-scale is not supported by this measurement system.

This measurement was done with mic-distance 1m. Mic on axis, between tweeter and bass.
The problem is the room resonances will become more influence i think...

 

[Frank Dernie]

Greetings from Austria,

Walter

P.S. apology my bad English. Im not using this language very often ;-)

Super post, most enjoyable and interesting, I would bet your English is very much better than most people here's German.
Bravo

 

[EXIF68]

And here is a comparison between the DXT-MON and his big brother, the AOS Studio 24 XL, also an early DIY-Speaker from Germany. Chassis are the ScanSpeak 15W9530-K00 and Tweeter R2904/700000. 1st order crossover.
violet curve: DXT_MON
blue curve: AOS Stuido 24XL.
Measured under the same conditions as mentioned above.
And as shown in the graph so it sounds. The DXT-MON uses much cheaper chassis and is sounding in the middle heights much better than this very expensive system. Only from the costs of the R2904 tweeter you get a complete DXT-MON system without cabinet. Maybe i will try to change the crossover in the S24, but this is a long time job....

The AOS Studio 24 is this part: closed box.

Here you can see the comparison of the step response of the two systems. yellow is S24 and violet is DXT-MON
as you can see, the S24 is a typical 1st order step responce. Why the amplitude are so different i don´t konw exactely. Settings are completely the same. Maybe it depends on the efficiency of the speakers.

 

[andreasmaaan]

What is REW?
My measurement system allows the following window-types:
rectangular
hamming
half hamming
hanning
half hanning
bingham
half bingham
all measurements above are done with "rectangular" window.
5db-scale is not supported by this measurement system.

Windowing the measurement will allow you to cut out the reflections from within the room, which will leave you with only the direct sound. The important thing is not so much the window type, but its length. It needs to start before the direct sound arrives at the microphone, and then end before the first reflection arrives.

I don't have your software, but in my software it looks like this. I've put arrows showing the setting for the window on the left, and also the impulse response at the bottom, with the arrow pointing at the first reflection. You can see that the window in this case start just before the direct sound hits the microphone, and then ends just before the first reflection, cutting out everything after the direct sound.

For a speaker the size of yours, the method would be to put it in the middle of the room, as far from reflective surfaces as possible, and then to measure from a distance of at least one metre. You would then use the software to window the measurement, which will result in you seeing the "quasianechoic" frequency response.

There's no need to do this, ofc But in case you're interested, that's how it works...

 

[EXIF68]

Thanks for the info.
if i do this in that way than you can see the following:
A line as flat as it gets...but i think the information below e.g. 2kHz is missing.

 

[EXIF68]

And if i change the sampling rate of the system to 96kHz you can see the resonance of the alu dome of the tweeter in the area of e.g. 27kHz.

 

[andreasmaaan]

Looking very nice @EXIF68.

 

[EXIF68]

Here the configuration of the measurement setup.
Speaker in the middle of the room, mic 1m in front of the middl beween tweeter and bass. I think the problem of measuring here is the roof-influence of the early reflections.

 

[EXIF68]

Thanks. My sim shows that the predicted saddle in the impedance respopnse for that woofer/passive radiator in a 9.5L enclosure is arond 60Hz, which is a little higher than measured.

Adding 5g to passive radiator drops the saddle 5Hz to about 55Hz.

If you're measuring 55Hz now, adding 5g should bring you down to about 49Hz-50Hz, IMO.

Hello Andreas,

your simulation was right:
passive radiator resonance now at 49Hz, Woofer resonance at 51Hz.
The curves yellow and light pink without additonal mass
the oters with exact 4.5g. (in the photo below is shown 4 pcs of larger shims and a smaller one. This was a little bit more than 5g. So i removed one of the lager shims.
Additional you can see that the level of the passive radiator is a little bit less than bevor (e.g. 2db), this is due to the additional mass? Measuring level is the same.

 

[andreasmaaan]

Additional you can see that the level of the passive radiator is a little bit less than bevor (e.g. 2db), this is due to the additional mass? Measuring level is the same.

Yeh, that's to be expected. The added mass reduces both the frequency and SPL of the output passive radiator's output.

There's no right or wrong here, just different system Q.

One thing, though, is that if you're measuring in the nearfield, you need to take into account that the passive membrane has a larger radiating surface than the woofer. So you need to scale the measurements by the square roots of the radiating diameters.

In this case, your woofer's Sd is 93cm^2 and your passive radiator's Sd is 130cm^2, which implies that their diameters are approximately 10.9cm and 12.9cm, respectively.

If we take the square roots of the diamters we get about 3.3 for the woofer and 3.6 for the passivate radiator, which means that, for the graphs to be more accurate, you should increase the plot amplitude of the passive radiator by a factor of 3.6/3.3 = 1.09 (approxmitately).

Of course, that will not take into account baffle step. But it will at least allow you to display the measurements for the passive radiator and the woofer more accurately in relation to each other (and should bring the passive radiator's peak up closer to the woofer's output level).

Having said all of that, you're going to be listening to this speaker in your room, and the room will be the main factor in determining the bass response. So you may as well choose the passive radiator tuning that works best in your room.

 

[Deleted member 65]

@EXIF68 love that scale ...

 

[EXIF68]

First listening to this setup is very exciting. Bass response is exceptional for such a small speaker. My wife did‘t belief that bass coming out from this speaker and not from an additional subwoofer. But yes, really deep bass is not there, but for music of all genres more than enough and this to really high leves (but no discolevel at all). the difference in the bass response to the situation before (no additional mass) is not very pronounced, to check this you need really special sources and also a lot of time for evaluating this.
placement of the speakers are now e.g 80cm in front off backwall and e.g. 1m beside side wall. Listening distance is about 2 m from the speakers. Distance between the speakers 2m.
Reproduction off middle and treble, room imaging: there is nothing to worry about. Perfect.