Coax project using 6.5" car audio drivers ESB 4.6C

[ppataki]

Here is the next step in my coax journey

Coax drivers have been pretty popular in car audio, so I thought I would take a look around to see if there are any manufacturers out there that publish T/S parameters and/or (relatively) meaningful measurements
I have managed to find some manufacturers, like Focal that do all the above fairly well but their measurements look really really awful (at least in the coax department)
I kept looking and accidentally came across a brand called ESB. They are Italian (like Sica, Lavoce, etc.) and they have been on the market since 1969(!).

They publish full parameters and measurements (that actually look great) so I went ahead and ordered their top coax model, the 4.6C

ESB 4.6C

6.5"/165 mm Coaxial

www.esbcar.com

They sport paper (unbleached pulp) cones with silk dome tweeters.

Manufacturer published measurements:

If even half of this is true than I think I made a good deal
I plan to cross the tweeter at around 4kHz (using linear phase 24dB/oct. Linkwitz-Riley crossover, as usual)

Simulations in my sealed sphere cabinets:

Max. load is 70W at 80Hz (sub crossover point)

Where it shall produce 103dB SPL at 1m (109dB for stereo)

The drivers shall arrive later this week, stay tuned!

 

[voodooless]

Those are some thick surrounds! Motor is big, too!

I wonder how well the tweeter integrates. It sticks out. On-axis you can fix the path difference with delay. Off-axis though, it won’t add up.

 

[kemmler3D]

Those are some thick surrounds! Motor is big, too!

I wonder how well the tweeter integrates. It sticks out. On-axis you can fix the path difference with delay. Off-axis though, it won’t add up.

I guess it might do something funny way off-axis at the xover frequency, but I think the distance is such that it would only be a serious problem at high frequencies (over 3Khz) at which point the woofer is well out of the picture.

 

[voodooless]

Crossover seems to be at over 3kHz…

 

[smowry]

Nice looking transducers.
1. Nice lead out assembly
2. Robust Basket
3. Large spider
4. Large surround roll
5. Large Motor Assembly

*Seems to be designed for high relative Xmax. Manufacturer claims 4.5mm one way (9.0mm P-P).

 

[Audionaut]

Very interesting project
I wish you every success with the realization. Thumbs up
I've already installed quite a few coaxial speakers, from Tannoy, when they were easier to buy as single drivers, from RCF, BMS, B&C and also from Beyma.
This one from Beyma's car hi-fi range has appealed to me for a long time, but I'm a bit oversupplied with loudspeakers

Beyma Coaxial Neodymium Speakers - Beyma Pro8FX370Nd coaxial speakers - Beyma Pro8FX370Nd 500 watt 8" coaxial speaker with 1.4 inch exit HF driver for all 2-way applications. Beyma Pro8FX370Nd coxial speaker and other Beyma 10" coaxial speakers here.

Beyma Pro8FX370Nd professional lightweight neodymium 8 inch coaxial speaker with 1.4 inch exit HF driver for speaker replacement or upgrade. Beyma Pro8FX370Nd 8 inch coaxial speakers for all high quality 2-way cabinets. The Beyma Pro8FX370Nd delivers smooth sound with 340 watts program power...

usspeaker.com

 

[smowry]

Those are some thick surrounds! Motor is big, too!

I wonder how well the tweeter integrates. It sticks out. On-axis you can fix the path difference with delay. Off-axis though, it won’t add up.

A COMSOL Acoustic Simulation of this geometry could help to answer that question.

 

[kemmler3D]

Crossover seems to be at over 3kHz…

I should have looked closer before posting, you're right. Well, in that case it might cause some weird lobing... It does look like the woofer falls off pretty fast though, so I don't expect a total disaster.

 

[ppataki]

Why would you expect lobing?
As I mentioned above, I will apply digital crossover (linear phase, symmetrical, 24dB/oct. LR) at around 4kHz, or maybe 3kHz - I will see based on the measurements that I will perform (will share those here too)

This driver does not have any built-in passive crossover

 

[kemmler3D]

Why would you expect lobing?
As I mentioned above, I will apply digital crossover (linear phase, symmetrical, 24dB/oct. LR) at around 4kHz, or maybe 3kHz - I will see based on the measurements that I will perform (will share those here too)

This driver does not have any built-in passive crossover

When the physical distance between drivers is comparable or larger than a frequency inside the crossover band you get lobing due to interference, (hence the rules of thumb about c-c distance) so in this case just eyeballing it, you'd expect some lobing around/above 3khz or so. If at all possible you should cross lower, I think.

 

[voodooless]

Why would you expect lobing?
As I mentioned above, I will apply digital crossover (linear phase, symmetrical, 24dB/oct. LR)

Maybe not lobing, but you’ll get something… that linear crossover can only fix one directivity perfectly. On all other angles the path difference between tweeter and cone will have some kind of effect. It’s not going to be super huge, but it might be noticeable enough. It will be interesting to see how it manifests.

 

[ppataki]

Maybe not lobing, but you’ll get something… that linear crossover can only fix one directivity perfectly. On all other angles the path difference between tweeter and cone will have some kind of effect. It’s not going to be super huge, but it might be noticeable enough. It will be interesting to see how it manifests.

Theoretically, would it be visible on any of the MLP measurements?
I will be able to perform MLP measurements with crossover set to 2kHz, 3Khz, 4kHz, etc. like with one of my previous coax projects (Sica) but there was no difference whatsoever there, at least with those Sica drivers (in the MLP)

I am afraid I will not be able to perform proper off-axis measurements (polar or globe charts) - hence I am bringing up the above question about MLP

 

[voodooless]

Theoretically, would it be visible on any of the MLP measurements?

Well, the point is that you will optimize the filter for the MLP, so no, it will not. It will, however, influence the reflected sound and off-axis response. With FIR, you can fix the reflection on the MLP as well (partly), but not the latter.

 

[ppataki]

Well, the point is that you will optimize the filter for the MLP, so no, it will not. It will, however, influence the reflected sound and off-axis response. With FIR, you can fix the reflection on the MLP as well (partly), but not the latter.

Ok, so unfortunately, I will not be able to find that out then
Neverheless, I will provide my MLP measurements here + my subjective findings, might be useful still

 

[voodooless]

might be useful

It is very useful. There are of course more potential pitfalls here, particularly the tweeter stem might produce diffraction artifacts.

Fun experiment could be to add a ring of melamine foam around the tweeter to see if that works, similar to what KEF did with their flagship in-ceiling speaker.

 

[laserluxxer]

With a coax like this, there will be almost definetely a dip in the DI at about 6k.
Its edge diffraction from the tweeter rim amplified by the cone behind it.
The Geithain MO-1 uses a similar design and the one can see the effect in the 45° meassurement.
No way of getting arround this with a symmetrical coax with offset tweeter.

 

[ppataki]

And now the pictures, finally

I guess this is the high-pass filter for the tweeter (which I am not going to use)

I hope I will be able to perform some measurements later today, stay tuned!

 

[ppataki]

Now let's see the measurements!

And with that I am closing this thread and sending the speakers back immediately; you will see below why

Frequency response of the left woofer:

OK, this is not bad, however.....let's see the distortion:

Holy sh*t!

The root cause of this is that the woofer is really far from being airtight.....there is air leaking heavily around the tweeter pole
See some videos about it:

New video by Peter Pataki

photos.app.goo.gl

New video by Peter Pataki

photos.app.goo.gl

New video by Peter Pataki

photos.app.goo.gl

I continued nevertheless.....

Frequency response of left and right tweeters:

There is a dip around 3kHz and the distortion curve confirms the same:

I measured the acoustic delay between the woofer and the tweeter (0.08ms) and applied the crossover (at 4kHz, 24dB/octave LR, linear phase, tweeter inverted)

L+R

Actually not that bad but it is a bit far away from the published response graph

With a low shelf it looks like this:

That 3kHz dip and the 6kHz peak could be fixed with two bell filters but I did not bother
Distortion is horrible, it is very clearly audible, even with music

I am attaching the mdat file for your scrutiny too:

20250305 ESB 4.6C.mdat

drive.google.com

Needless to say, strongly not recommended!

 

[bmc0]

The root cause of this is that the woofer is really far from being airtight.....there is air leaking heavily around the tweeter pole

Do you get the same kind of noise when it's not in a cabinet? I'm wondering if it's actually that felt ring dragging on the bit the tweeter's attached to (seems to double as a heatsink). Poor design in any case.

 

[kemmler3D]

So THD was almost 1% for the whole woofer spectrum and they charge $400 per driver??