Fullrange speaker project based on Markaudio CHR-120

[ppataki]

Dear ASR Community,

I am starting this thread to document one of my new projects. Last year I built a full range project in my home office; now I managed to sell those Morel drivers and bought a pair of Markaudio CHR-120 drivers to replace them - obviously the cabinet will also be replaced.

So this new project will not start immediately, since there is another one that needs to be finished first, but I thought that I would share whatever I know already at this stage.

So first of all, the drivers have already arrived:

The cabinet will be a sealed one and I will use heavy DSP correction to flatten the response (Dirac Live 3, CurveEQ and Saturn 2 for the final touch)

The cabinet walls will probably be 2x9 mm of birch plywood with a layer of Green Glue in between (hence the 19mm above).
Q=0.707, Vb= 11.35 liters
The driver will be rear-mounted with a chamfer on the outside of the cutout
Coating will be done using Raptor paint

Below you will see some simulation curves:

Showing here the F15 point since this will be my F0 point when adding the room gain and the DSP

Xmax reached around 100W (in real life this will be much lower due to the DSP compensation but since I will listen to the speakers from 70cm I will not need a lot of SPL anyway)

Building the cabinet will start hopefully in 6 weeks' time (due to the other project I mentioned above) - stay tuned!

 

[abdo123]

You need to take roughly 6 db away for baffle step compensation too.

 

[Ericglo]

Slow down. You are making the rest of us DIYers look bad.

 

[ppataki]

After a long time, I am happy to report that this project will continue in the coming weeks; currently in the process of ordering the plywood
Stay tuned!

 

[ppataki]

A new update, at last!
I finally managed to get the 9mm baltic birch plywood and also managed to finalize the constrained layer design:

I knew that those Atacama Atabites will come in handy one day!!!

Next update hopefully next week when we will actually put the cabinet together with the best carpenter of the world (my neighbor )
Have a nice weekend everybody!

 

[PatentLawyer]

I use Markaudio drivers on my desktop with Dirac. My speakers use their cheapest drivers (~$19/per or so) and yet they are sounding great with DSP.

 

[ppataki]

Yet another quick update: the cabinets are now assembled, puttied and fully sanded (forgot to take that picture but will follow soon)
Next step is to do the cutouts and the rounding then I will take them home and start working on the inside
Still debating with myself if any additional bracing is needed or not; will see...

As always, Thanks to the best carpenter in the world (my neighbor)

 

[ppataki]

Actually things started to pick up pace here

Carpentry is fully done:

And the 'simulation' also - still sounds a little bit 'boxy'....

Next steps: will work on the inside of the cabinets; will share pictures hopefully in a few days

 

[ppataki]

Some new updates: inside of the cabinets is done now

And some videos too:

3 new items by Peter Pataki

photos.app.goo.gl

Cabinet is now airtight, next step is the Raptor coating

 

[ernestcarl]

Actually things started to pick up pace here

Carpentry is fully done:


View attachment 232484


And the 'simulation' also - still sounds a little bit 'boxy'....


View attachment 232483


Next steps: will work on the inside of the cabinets; will share pictures hopefully in a few days

Can you please also tell us how heavy your fully assembled speaker is? Hmmmn… can’t help but wonder. It “looks” pretty hefty despite the small single driver concept.

 

[ppataki]

Can you please also tell us how heavy your fully assembled speaker is? Hmmmn… can’t help but wonder. It “looks” pretty hefty despite the small single driver concept.

They weigh 7 kgs each
They are not made of MDF hence the lighter weight

 

[ppataki]

Coating is done; will post pictures tomorrow

In the meantime I would like to share the results of an experiment that I was doing with these cabinets
One of the cabinets I left unfilled, like this:

Whereas the other got filled fully like this:

Then I measured them both from 1cm and here is how the FR looks like:

As you can see the FR curve of the cabinet fully filled looks smoother and even more importantly it extends lower by 4Hz! - that is a substantial improvement I would say

Waterfall also shows quite a difference:

So overall it seems that it is worth stuffing the cabinet fully (there can be a concern of the driver overheating but I don't think it will be an issue in my case)

More to follow!

 

[ppataki]

And the final pictures!

Measurements will follow either today or during the weekend - stay tuned!

 

[ernestcarl]

Measurements will follow either today or during the weekend - stay tuned!

Nice! Rougher paint texture than I expected, but still pretty cool. Hmmmn... I wonder if you will be using Dirac for this set... or opt for manual EQ.

 

[ppataki]

Nice! Rougher paint texture than I expected, but still pretty cool. Hmmmn... I wonder if you will be using Dirac for this set... or opt for manual EQ.

Will use manual EQ:

- CurveEQ to flatten the FR
- Convolution (using REW filters and ConvologyXT) to correct the low-end that CurveEQ cannot do
- CraveEQ to add the final 'target'

I can achieve a flatter FR manually vs with Dirac and since this is a 1-way system I don't need time alignment between the drivers (that Dirac does really well btw)

 

[UltraNearFieldJock]

Great! I have a similar DIY-setup in my office (alu closed Visaton B80), but with a sub (closed Mivoc AVM124) under the main monitor in the front. Filters, time-delay and PEQ with MiniDSP. What is the listening distance? I have now 25 cm. What about bass-section?

 

[ppataki]

And now the measurements
Room is totally acoustically untreated!

Let's start with the FR with zero DSP:

Not too bad for a 1-way system actually; my MCLA looks waaaay worse than this

Now let's see what I could make out of it:

I have applied a HP at 27Hz so my -3dB point is at 28Hz (-4dB at 27Hz)
The dips between 40 and 80Hz are cancellations caused by the room - I can correct them easily with filters but more on that later

Distortion is pretty bad - not in the very low-end though but where the room is causing the dips (awesome values between 26 and 40Hz):

Step response look great to me, very slight pre-ringing, definitely not audible

IR shows no major reflections thanks to the beaming of the fullrange drivers (I know for a lot of people here beaming is a bad thing but it is a great thing for me)

Wavelet shows virtually no energy delay:

Lack of room treatment is clearly seen in the waterfall diagram:

Finally phase (with excess phase) and group delay (with excess group delay):

Now a few notes:

- I am driving these speakers with a Loxjie A30 amp that can deliver 2x18W at 8 ohms (2x30W if 1% distortion is allowed) - now I am applying a whopping +16dB!! low shelf below 100Hz so I am definitely running into clipping with this amp, even at lower listening levels - actually I am hearing the same issues I was hearing when I was testing the 30W Aleph 3 clone amps with my MCLAs. Next week I will bring my NC250MPs upstairs with my Topping DM7 DAC and re-measure the whole thing cause I feel that it is the amp being the limiting factor and not the speaker drivers (I am far away from their 9mm one-way xmax!)
So more to come next week about that

- for the same reason I have left the room modes uncorrected, those would bring even more strain on the amp - however I will try that next week with the NC250MPs
I suspect that ringing might become an issue then but let's see

- apart from the above mentioned issue the sound is what can be expected from a corrected 1-way system: very 'live' and very 'being there' sensation. Tons of details. Big smile on my face

 

[ppataki]

Great! I have a similar DIY-setup in my office (alu closed Visaton B80), but with a sub (closed Mivoc AVM124) under the main monitor in the front. Filters, time-delay and PEQ with MiniDSP. What is the listening distance? I have now 25 cm. What about bass-section?

Thank you!
My listening distance is just below 1 meter. I do not have a subwoofer

 

[ppataki]

So I have tried the speakers with my Topping DM7 and Hypex NC250MP amps:

My immediate reaction was: WOOOOOOOOOOOOW

The difference is night and day.....better soundstage, much more controlled low-end, actually the lows are way more prominent and precise now, I can no longer hear the amp getting strained + I hear even more details

I have re-calibrated the system accordingly

There is not much of a difference in terms of the FR (obviously, since it is forced to flat as much as possible)

Phase looks much better with the Hypex though:

Huge difference in the IR!

Step response shows a bit more pre-ringing with the Hypex but less oscillation and more controlled around 0ms:

So the conclusion is that it is high time for me to upgrade my Loxjie A30 to something more beefy....
Since I do not want two devices (DAC+Amp) sitting on my desk, I would probably go for something like the Sabaj A30A full digital amp (FDA) which is kind of the same as the SMSL A2 if I understood correctly

 

[ernestcarl]

For a more one-to-one comparison of the impulse and step, identical EQ settings should have been used. Some of the difference in the phase response could possibly be because of the impulse centering and timing reference -- if you are using a USB mic, the clock timing can still drift even with a reference "chirp" -- repeat measurements may be necessary to check for consistency. It seems strange for your other amp to shift the phase by that much.