I just finished my birch plywood monitors using a Dynaudio T330D and a Scan-speak 15wu8741.
It is an active monitor with a Hypex PSC2.400D plate amp(2 x 400 watt digital amps with a DSP) in the back. I use the AES/EBU digital input on the Hypex to stream music from my NAS.
The design is made in MS-Visio and exported it to Autocad DXF files which were used directly to steer the CNC router table. All elements are routed from one sheet of 30mm birch plywood and then glued together.
I am very pleased with the sound but still in the process of measuring and adjusting the DSP. The sound stage is stable and the bass very satisfying for such a small closed cabinet.
Just a few pics:
It is an active monitor with a Hypex PSC2.400D plate amp(2 x 400 watt digital amps with a DSP) in the back. I use the AES/EBU digital input on the Hypex to stream music from my NAS.
The design is made in MS-Visio and exported it to Autocad DXF files which were used directly to steer the CNC router table. All elements are routed from one sheet of 30mm birch plywood and then glued together.
I am very pleased with the sound but still in the process of measuring and adjusting the DSP. The sound stage is stable and the bass very satisfying for such a small closed cabinet.
Just a few pics:
Attachments
Last edited:
I attached the building plan in PDF. Net volume is appr. 16 liters
Crossover for Scan-speak is 6 db at 1250 Hz
Crossover for T330D is 12 db and about 1300Hz
In the DSP i have also made several adjustments to get a flat frequency response. I measure the response with REW and a behringer ECM8000 mic.
I will post some more info (DSP graph and freq. response) tonight.
Crossover for Scan-speak is 6 db at 1250 Hz
Crossover for T330D is 12 db and about 1300Hz
In the DSP i have also made several adjustments to get a flat frequency response. I measure the response with REW and a behringer ECM8000 mic.
I will post some more info (DSP graph and freq. response) tonight.
SPL and Phase graph
The Hypex sounds powerful and detailed. It's the all digital version but i use the analogue inputs for my Sonos system and for serious listening and use a PC as source.Hypex switches automatically between inputs.
PC --> USB --> Audio GD digital interface --> spdif/Coax --> Neutric impedance transformer --> EAS/EBU interlink --> Hypex AES/EBU.
DSP settings are a work in progress and the sound needs a little warmth. There is a 85u second delay for phase correction.
I attached a SPL and Phase graph but i'm having trouble to interpret the phase graph. Maybe someone can help me there?
The Hypex sounds powerful and detailed. It's the all digital version but i use the analogue inputs for my Sonos system and for serious listening and use a PC as source.Hypex switches automatically between inputs.
PC --> USB --> Audio GD digital interface --> spdif/Coax --> Neutric impedance transformer --> EAS/EBU interlink --> Hypex AES/EBU.
DSP settings are a work in progress and the sound needs a little warmth. There is a 85u second delay for phase correction.
I attached a SPL and Phase graph but i'm having trouble to interpret the phase graph. Maybe someone can help me there?
Attachments
Update on Monitor project
Just a small update on my speaker project.
I have been listening to my new monitors for some time now and it sounded ok but not great. No matter how many times i measured my setup and adjusted the Hypex DSP settings.
My approach was to get a flat response and, using the DSP, correct any 'hills and valleys' in the frequency response graph. But it all sounded a bit flat, boomy and restless. I guess all those DSP corrections messed up the phase behavior of the monitors.
So tried a different approach and i sort off emulated a passive crossover with the DSP. I decided to go for a cross-over frequency at 2Khz, using a 24db slope. A 24db slope is a bit tricky with passive components but not with a DSP.
The tweeter output was adjusted by -10db and i added 65 us delay for phase correction (could measure the result of the phase correction perfectly). This simple setup did the trick and they now sound absolutely great.
My lesson was that a DSP can really mess up your sound even if the result measures fine. Does anyone have the same experience with DSP's and more info on this like a 'DSP best practice guide'?
Just a small update on my speaker project.
I have been listening to my new monitors for some time now and it sounded ok but not great. No matter how many times i measured my setup and adjusted the Hypex DSP settings.
My approach was to get a flat response and, using the DSP, correct any 'hills and valleys' in the frequency response graph. But it all sounded a bit flat, boomy and restless. I guess all those DSP corrections messed up the phase behavior of the monitors.
So tried a different approach and i sort off emulated a passive crossover with the DSP. I decided to go for a cross-over frequency at 2Khz, using a 24db slope. A 24db slope is a bit tricky with passive components but not with a DSP.
The tweeter output was adjusted by -10db and i added 65 us delay for phase correction (could measure the result of the phase correction perfectly). This simple setup did the trick and they now sound absolutely great.
My lesson was that a DSP can really mess up your sound even if the result measures fine. Does anyone have the same experience with DSP's and more info on this like a 'DSP best practice guide'?
I used the exact same hypex plate in my project, a very nice piece of equipment indeed!
What you probably heard was distortion, a first order slope is very difficult on almost every driver. It may sum flat in the frequency domain but if you look at distortion you will be shocked by the results..
EQ a driver flat before XO is probably a good idea (i did it to) it also helps you with the phase response. And time alignment should definitely be done, it's benefits are obvious and the implementation with the DSP is very, very easy.
Just remember that DSP isn't magic, all the rules that apply to normal XO design also apply here, the DSP just makes it easier to implement (usually).
http://www.diyaudio.com/forums/multi-way/207159-trinium-my-latest-build.html
What you probably heard was distortion, a first order slope is very difficult on almost every driver. It may sum flat in the frequency domain but if you look at distortion you will be shocked by the results..
EQ a driver flat before XO is probably a good idea (i did it to) it also helps you with the phase response. And time alignment should definitely be done, it's benefits are obvious and the implementation with the DSP is very, very easy.
Just remember that DSP isn't magic, all the rules that apply to normal XO design also apply here, the DSP just makes it easier to implement (usually).
http://www.diyaudio.com/forums/multi-way/207159-trinium-my-latest-build.html
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.