Faital HF 144 + LTH142 measurements

[eternal_intro]

I recently picked up a set of Faital HF144 compression drivers very cheaply and paired them with the LTH142 tractrix horn, also from Faital.
Since I couldn't find any measurements of the pairing available on the WWW I thought I would share a quick and dirty on axis, nearfield measurement I did yesterday.

Measurement details:
Position: Centered, at horn mouth.
Input: Umik 1 into MacBook
Output: Topping E30 Dac rca into Hypex Fusion122 @-28db
REW: -20dBfs, no smoothing, gated at 12ms.



SPL

Distortion

Spectogram

I'm planning to implement these in a active two way 15 inch with the Faital 15PR400 but I'm a bit worried about the dip @2.4khz and the breakup above 10khz.
The combination is a lot less linear than the published measurements from Faital. What's going on?

 

[voodooless]

There really isn't that much of a difference:

I think you can easily EQ that away with a few of the other larger ripples. You may want to do an off-axis measurement as well to see how they move with axis change.

Here is the 142 on the same horn: https://audioxpress.com/article/tes...ession-driver-coupled-with-lth142-60-o50-horn . It shows a dip a 3 kHz, but off-axis it's almost gone.

 

[eternal_intro]

I agree. The characteristics are quite similat but the amplitude of the dips are higher in my measurements it seems. The dip @2.3 is almost 8db.
My initial plan was to cross them with the woofer at around 850Hz. Before that I will try to flatten them slightly with EQ.
I was thinking to attenuate the first octave from 900 to 1.8 and then boost the dip at 2.3K.

Would that be reasonable? It's my first time working with EQ in speaker design.

 

[voodooless]

Also, check the gasket for air leaks and make sure the transition to the horn is smooth (alignment).

I would just add the LR4 at 850 first, then try to shelve it down whatever is too much (you can aim for a mid shelve of 2 kHz). That should already make the dip far less pronounced. Finally, EQ it as flat as possible. But there may be other ways as well. There is never a single good solution.

And if you're not planing on listing to them toed in, make sure it's flat off-axis.

 

[eternal_intro]

Cheers! I'll look into that over the weekend.
Thanks a lot ^^

 

[Trdat]

Mate, I am sure you can find a polar plot for that driver and with off axis measurements with your chosen woofer you can get a good directivity match.

Once that is done, you will need to work out how to Eq with the Hypex amp, this link might help. Also check out the Audio Nirvana Forum(https://www.avnirvana.com/) and search Jtalden he has a lot on how to measure, and time align in general in particular using REW.

Eq-ing Concepts and Biquad Understanding

I just want to get a few things down pat with using biquads, well with any Eq-ing. I understand the goal is to attenuate peaks and dips and most of the time we get these around the crossover over region. But, lets say you get good SPL support across the frequency response around the crossover...

www.audiosciencereview.com

Lastly, I would just purchase a software DSP like Audiolense and tie it all in. Audiolense is like magic, but of course if all things are correct from the get go.

I am just curious at which frequency the loading ends on that horn, if it goes down to 850herts thats great. I personally love horn speakers.