Hello all,
I have gathered enough speakers and amps to make my dream about a home 4 way home system come true. The description of the journey starts here: https://www.diyaudio.com/forums/multi-way/127334-horn-system-post5708525.html
The crossover will be realized using the t.racks DSP 4x4 mini - I have one for each side. the t.racks DSP 4x4 Mini – Thomann UK
Some pictures are here: https://www.diyaudio.com/forums/multi-way/127334-horn-system-post5853737.html
It is pretty flexible with EQ and crossover alignments, but no advanced FIR stuff.
The measurements of the cabinets in nearfield and at the listening positions are here: 4WayNFandLP.mdat 4WayOverall.mdat - ShareGadget.com Share Send files by air mail - Share your files - Upload your file here
The current crossover points are 200, 600 and 4000 Hz, all LR-24 and the delays measured by REW are set to compensate at the listening position. There is also the final measurement after level adjustments via pink noise and 1/3 octave frequency analyzer in REW. There is an intentional tilt decreasing the HF, no EQ but 6 db bass boost at 35 Hz.
My skills end at having a sort of flat FR, but have no idea about group delays and impulse responses, what is from the room and what is from the speakers, what is good, what is bad. So far, I am quite happy with how it sounds now. But I am sure there is some space for improvement. I would really appreciate any input on this case. And also thanks to all who helped me with my journey until now!
I have gathered enough speakers and amps to make my dream about a home 4 way home system come true. The description of the journey starts here: https://www.diyaudio.com/forums/multi-way/127334-horn-system-post5708525.html
The crossover will be realized using the t.racks DSP 4x4 mini - I have one for each side. the t.racks DSP 4x4 Mini – Thomann UK
Some pictures are here: https://www.diyaudio.com/forums/multi-way/127334-horn-system-post5853737.html
It is pretty flexible with EQ and crossover alignments, but no advanced FIR stuff.
The measurements of the cabinets in nearfield and at the listening positions are here: 4WayNFandLP.mdat 4WayOverall.mdat - ShareGadget.com Share Send files by air mail - Share your files - Upload your file here
The current crossover points are 200, 600 and 4000 Hz, all LR-24 and the delays measured by REW are set to compensate at the listening position. There is also the final measurement after level adjustments via pink noise and 1/3 octave frequency analyzer in REW. There is an intentional tilt decreasing the HF, no EQ but 6 db bass boost at 35 Hz.
My skills end at having a sort of flat FR, but have no idea about group delays and impulse responses, what is from the room and what is from the speakers, what is good, what is bad. So far, I am quite happy with how it sounds now. But I am sure there is some space for improvement. I would really appreciate any input on this case. And also thanks to all who helped me with my journey until now!
While my system is totally different, I too have 4 ways with active EQ. I have used REW so somewhat familiar. I recommend you choose your crossover points and type, then if you can, set the -3 db level at x-over for each speaker pair. Then apply the chosen room curve (target EQ) at the listening position. This should give you very good frequency response. Adjusting x-over and delays is more complex and too involved for me.
Anybody else?
Anybody else?
Your horns have sparse apparent termination. It's difficult imagining they cross physically. Without this you'd struggle to cross electrically.
I would begin by assessing polar plots. I can see by the way you have your horns oriented, you are trying to keep this 2D. However, doing a cross like this is a bit more involved. Sure, you might get lucky.
You'd want to know where your dispersion is controlled by walls, or by diffraction. You'll want to decide your bandwidths as a fundamental part of this cross.
I would begin by assessing polar plots. I can see by the way you have your horns oriented, you are trying to keep this 2D. However, doing a cross like this is a bit more involved. Sure, you might get lucky.
You'd want to know where your dispersion is controlled by walls, or by diffraction. You'll want to decide your bandwidths as a fundamental part of this cross.
Soldermizer: Thanks for reminding me the target curve EQ in REW. It will definitely help with adjusting the parametric EQ in tgr DSP, my trial results were not too good. Do you think it is a good idea to measure average response around the listening place?
AllenB: I am definitely shooting in the dark here. By polar plots you mean in room or outdoor, individual or whoke system? I still do not understand how to take out the in room reflections in REW. The physical arrangement is not too ideal,I am waiting for some new horns so that I could place the HF below the MF horn.
This means I will try to do more measurements today across the room.
By the way, the Beyma HF is invered in phase, otherwise there is a null at the XO point - I still have to find out if it is the amp or the way Beyma or JBL label positive terminals.
AllenB: I am definitely shooting in the dark here. By polar plots you mean in room or outdoor, individual or whoke system? I still do not understand how to take out the in room reflections in REW. The physical arrangement is not too ideal,I am waiting for some new horns so that I could place the HF below the MF horn.
This means I will try to do more measurements today across the room.
By the way, the Beyma HF is invered in phase, otherwise there is a null at the XO point - I still have to find out if it is the amp or the way Beyma or JBL label positive terminals.
It can measure, do a lot of analysis stuff, windows on FFT - which should be equivalent to gating somehow, but I have not yet found out how to use that feature properly.
Just for fun, I made these measurements: 4WayBothPolar.mdat 4WayLeftPolar.mdat 4WayRightPolar.mdat - ShareGadget.com Share Send files by air mail - Share your files - Upload your file here
Across the room at the listening distance (250 cm), spaced 20 cm apart, both left and right. Height is ca 130 cm, which is my ear height on a chair.
I will try HolmImpulse inside and if that does not work, I will take the boxes out (not the bass horns
.
Just for fun, I made these measurements: 4WayBothPolar.mdat 4WayLeftPolar.mdat 4WayRightPolar.mdat - ShareGadget.com Share Send files by air mail - Share your files - Upload your file here
Across the room at the listening distance (250 cm), spaced 20 cm apart, both left and right. Height is ca 130 cm, which is my ear height on a chair.
I will try HolmImpulse inside and if that does not work, I will take the boxes out (not the bass horns
.
Why use horns? My current system is a collection of various PA horns, because I like how horn loaded PA sound with live bands. And then, I have heard a three way horn system at a friends's house - a copy of the Jealvox - Something about Jealvox speakers and I was amazed by what is possible - his one was active and full tube powered. The third reason is the directivity of the horns which should help in the non ideal cellar room (all concrete, almost zero treatment at the moment. And the fourth is the efficiency.
Even with the textbook crossover and some delay compensation, I get much better sound than with any other speaker I have or have had.
I want the bass horn to play 40 Hz and up
Then, the midbass horn kicks in for the extra kick (currently at 200 Hz, but maybe should be lower)
Then the large JBL CD from 600 Hz up and a horn tweeter when the JBL is loosing steam. Equalizing the JBL flat to the HF did not sound good at all.
Even with the textbook crossover and some delay compensation, I get much better sound than with any other speaker I have or have had.
I want the bass horn to play 40 Hz and up
Then, the midbass horn kicks in for the extra kick (currently at 200 Hz, but maybe should be lower)
Then the large JBL CD from 600 Hz up and a horn tweeter when the JBL is loosing steam. Equalizing the JBL flat to the HF did not sound good at all.
It doesn't have to be flat. Smooth is still good.
It's quite possible to use a 1" compression driver from 700Hz all the way up. This covers the fussiest part of the spectrum to build for without a crossover. Horns large enough to do this well, are difficult to buy. I don't know if you are a builder. Maybe you would modify some you bought to cover a wide range.
It's quite possible to use a 1" compression driver from 700Hz all the way up. This covers the fussiest part of the spectrum to build for without a crossover. Horns large enough to do this well, are difficult to buy. I don't know if you are a builder. Maybe you would modify some you bought to cover a wide range.
1" drivers that play from 700 Hz are not very common. The JBL horn with a 2" to 1" adapter will load down to 500 Hz and should be good at least for a test.
I am actually more concerned about the other crossover points than the one for the tweeter. I could even live without tweeters. The very necessary components are the two lower horns. There may be also a "unity" type horn as a replacement for the top two. I will definitely try one of the designs from the forum.
I am actually more concerned about the other crossover points than the one for the tweeter. I could even live without tweeters. The very necessary components are the two lower horns. There may be also a "unity" type horn as a replacement for the top two. I will definitely try one of the designs from the forum.
Your 40Hz horn might have low loading but little pattern control. The room and its modes will dictate its performance.
Your 200Hz horn is the same size as my tweeter. This should make it a good match to the lower horn.
The 200Hz horn should be holding at 600Hz, so whether it crosses well above depends on whether the midrange horn is holding there.
There are many 1" drivers that will go to 700Hz, just not officially.
Your 200Hz horn is the same size as my tweeter. This should make it a good match to the lower horn.
The 200Hz horn should be holding at 600Hz, so whether it crosses well above depends on whether the midrange horn is holding there.
There are many 1" drivers that will go to 700Hz, just not officially.
No new measurements unfortunately, but I have a question to the spectrogram in the attachment.
What does the encircled step mean (if anything)? 220 Hz is the XO point between the midbass and bass horn. It looks like it is less than 10 ms, which would correspond to half the wavelength, so may this indicate an out of phase connection? Or is it just a coincidence?
By the way, the lower step can be explained as the phase shift around Fs of the bass horn. The delay length corresponds also to half wavelenght.
What does the encircled step mean (if anything)? 220 Hz is the XO point between the midbass and bass horn. It looks like it is less than 10 ms, which would correspond to half the wavelength, so may this indicate an out of phase connection? Or is it just a coincidence?
By the way, the lower step can be explained as the phase shift around Fs of the bass horn. The delay length corresponds also to half wavelenght.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.