Crossover for open baffle: please need help!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I fell in love with this open baffle project and built these gigantic speakers with vintage Saba greencones plus Eminence Delta 15'' woofers (please see pic.).
The crossover was included in the project so I didn't worry about it. But: midrange speakers crackle, I think they have been little damaged or simply they are too old.
I don't like the tweeters too... The woofers just whisper. Don't know what :scratch: but of course something's wrong... I tried to listen to greencones only and crossoverless, just with 2uF PIO capacitor in series with the tweeter.
HOW BAD.
Brief, I reverted to my beloved 80's british speakers and just bought -as a replacement- Celestion tweeters (from DL4) and glorious Kef B200 (obviously the Eminence woofers stay).

Now: first of all I hope these drivers could sounds well on open baffle :hot:
Then: I'm veryvery scared by making crossover as I definitely haven't necessary skills! :eek: (lot of smilies tonight...)
Speakers have different impedance: what this requires?
Maybe someone here will be so kind to help me... :djinn: And I miss some important specifications too...
Thank you very much in advance for any suggestion.



*******************************************************

Baffle: 140 cm x 50 cm / thickness 4 cm / bass box w 45 cm, h 80 cm, d 50 cm - the box is open on the back
Tweeter mounting height 100 cm, midwoofer just above, woofer have open box

*Tweeter Celestion: shows 5 ohm resistance

*Midwoofer Kef B200G sp1069: this version (triangular shape) have 16 ohm resistance, 35-3000 Hz . It was used in Kef 304, Ref 105.2 and 105.4.
http://www.kef.com/history/1970/model304.asp
http://www.t-linespeakers.org/drivers/b200.html

*Woofer Eminence Delta 15 PA:


* Impedance: 8 ohms
* Power Rating:
o Music Program: 800 Watts
o Watts: 400 Watts
* Resonance: 40 Hz
* Usable Frequency Range: 48 Hz - 4 kHz
* Average Sensitivity 1W @ 1m: 101 dB
* Magnet Weight: 56 oz
* Magnet Gap Depth: 0.375 inches, 9.53 mm
* Voice Coil Diameter: 2.5 inches, 63.5 mm

Thiele-Small Parameters

* Resonant Frequency: 40.27 Hz
* Impedance (RE): 6.99 ohms
* Coil Inductance (Le): 0.86 mH
* Electromagnetic Q (Qes): 0.537
* Mechanical Q (Qms): 6.56
* Total Q (Qts): 0.49
* Compliance Equivalent Volume (Vas): 293.11 litres, 10.4 cu.ft
* Peak Diaphragm Displacement Volume (Vd): 231.38 cc
* Mechanical Compliance of suspension (Cms): 0.2846 mm/N
* BL Product (BL): 13.59 T-M
* Diaphragm Mass inc. Airload (Mms): 54.89 grams
* Equiv. Resistance of mechanical suspension loss (Rms): 2.1166 N*sec/N
* Efficiency Bandwidth Product (EBP): 76.6
* Voice Coil Overhang (Xmax): 2.7 mm
* Surface Area of Cone (Sd): 856.34 cm2
 

Attachments

  • ob.jpg
    ob.jpg
    26 KB · Views: 1,217
Oh, how wonderful answer!
I'm really happy you think I choosed good replacements :radar:

I'm still regretting the lovely midrange of my old Linn Index Plus (yes they had Kef B200 but another version, I think this SP1069 is little known and rare, found almost no informations on the web).

My amp is an old Nad 3020 series 20 recapped; and this is the crossover as I found it in that ( wonderful ) speakers project:
*tweeter: 4uF series
*midrange: 0,39mh + 80 uF series
*woofer: 4,7mH series + 68uF parallel - this driver is in opposite polarity (please don't ask me why).
 
Hi There,
If I were you, I'd invest in 2 more additional power amplifier for 3 separate amps to the 3 drivers (perhaps you can build the gainclone?), get a dcx 2496 and connection cables to interface between RCA and AES/EBU connection; not forgetting a simple preamp. This will get you up several notches and get you going much quicker to audio nirvana instead of harping on passive x-over fillding components one at a time which is slow and without measurement qeuipment you're basically shooting in the dark. With the DCX you can dial in the x-over numbers, play with the relative phase of the drivers and play with the relative sensitivity offset of each drivers to obtain a seamless integration.

cheers.
 
Will, you open a new world to me. This sounds very interesting, I definitely will investigate this way too.
A digital solution :wave2s: Also if I'm a really analogue man....:knight:
Peanut, yes to control everything is my dream!!!
Do you think I could stay with my Nad as a preamp plus 3 Gainclones and the DCX? Or take just 2 more vintage amps (as I have a strange addiction to old devices) plus DCX?
 
"necessary skills!" - get them here (before you waste any more money):

Getting started in Crossover Design

The basics (You've got to understand the rules before you can break them :)
http://sound.westhost.com/lr-passive.htm
choosing xover point:
http://www.speakerbuilder.net/web_files/Articles/xover article/xpointmain.htm
Baffle Step Compensation: http://sound.westhost.com/bafflestep.htm

Getting it done:
Jay has a page on designing xovers using manufacturers specs without measuring:
http://www.geocities.com/woove99/Sp...DesigningXO.htm

Calculators:
http://ccs.exl.info/calc_cr.html#second
http://www.sengpielaudio.com/calculator-Lpad.htm

Measurement freeware:
Synrta - http://libinst.com/SynRTA.htm
ARTA http://www.fesb.hr/~mateljan/arta/download.htm
ARTA Jig - http://zobsky.blogspot.com/2008/01/simple-loudspeaker-measurement-jig-for.html

Examples from the designs of others can be quite instructive:
http://www.zaphaudio.com/
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
http://www.rjbaudio.com/projects.html
http://www.humblehomemadehifi.com/
 
Calamaro said:
midrange speakers crackle, I think they have been little damaged or simply they are too old.
I don't like the tweeters too... The woofers just whisper.
[/I]
Could it be that the connections to the crossover are mixed up?
The midrange crackles, because it is taken to its mechanical limits. Probably connected to the woofer output.
The woofer whispers, because it is connected to the midrange output.
And you don't like the tweeters, because the rest is not working correctly.
You should also check, if both speakers are in phase.
 
Hi Calamaro,

Can I suggest you look into pacificblue's check before you move on. Get something up and running first before you commit something much more expensive.

Start with the mid/widerange driver first, and, on its own.
Then add the other drivers in with any necessary crossover circuitry as you extend the audio bandwidth.

Another document worth scanning -
http://www.pispeakers.com/Speaker_Crossover.doc

I tried to e-mail you but that function is not open via your 'user cp'.

Cheers ....... Graham.
 
Pacificblue and Graham,
thank you for your suggestions.

Drivers are biwired (mid+tweeter / woofer) so mid and woofer are not connected. All drivers was in parallel. I checked the links and all was correct. Woofer was in reverse polarity.
I tried to listen just midrange directly connected to the amplifier+tweeter connected to midrage via 2uF capacitor in series but
-one midrange works well, the other one crackle (that's why I thought it's damaged)
-tweeters sound like dog whistles (ultrasonic????) :yikes:
Anyway, I think this is the right way, to begin with mid/fullrange and then add the others speakers. I'm reading lot of papers about crossover making, it seems that (also in this field) 'simpler is better', so I'll try to tune a good first order. I'm still trying to understand the following

******************************
from THIEL ws:

"There is a type of crossover system that does not introduce any phase shift or time smear, although it is expensive and difficult to execute. Well known as a technically perfect solution, it was often considered impractical. This is the first order (6 dB/octave) system that THIEL has perfected for practical use and employed in THIEL's Coherent Source systems since 1977.

A first order system achieves its perfect results by keeping the phase shift of each filter to less than 90° so that it can be canceled with a filter that has an identical phase shift of the opposite direction. The phase shift is kept low by using very gradual (6 dB/octave) roll-off slopes which produce a phase lag of 45° for the low frequency driver and a phase lead of 45° for the high frequency driver at the crossover point. Because the phase shift of each driver is much less than 90° and is equal and opposite, their outputs combine to produce a system output with no phase shift and perfect transient response. For any other type of crossover system it is not possible to completely eliminate time smear and phase shift."

http://www.thielaudio.com/THIEL_Site05/Pages/FAQs/faqtimephase.html

******************************

Graham, sorry I didn't know private messages option was disabled.
Thank you for the document.

Cheers :emoticon:
 
The Thiele info is just marketing BS - actually first order crossovers are cheapest to impliment, but have two problems - they don't protect low power rated drivers enough (which I think you've found -as Pacificblue has said), and they need drivers that have a very flat response outside the passband
 
Today I've listened to the greencones after a week. Separately, just one for time. With no filters.
Saba 8': one of the drivers is ok, good sound but as pacificblue has said it show difficulty especially on bottom end. The other one crackles so I definitely think it has a problem.
Saba 4': the tweeters are ok. The only problem is (will be) to tune properly filter capacitors.
In the meantime the new drivers are arrived: Kef B200 and Celestion tweeters. I like them. I really do. But I want to end this project keeping vintage speakers so maybe I'll just change midrange speakers with 2 others old ones, something less expensive than Saba (say Graetz, Loewe, or even RFT).
And the problem persists... the crossover! :bawling:
 
Will said:
Hi There,
If I were you, I'd invest in 2 more additional power amplifier for 3 separate amps to the 3 drivers (perhaps you can build the gainclone?), get a dcx 2496 and connection cables to interface between RCA and AES/EBU connection; not forgetting a simple preamp. This will get you up several notches and get you going much quicker to audio nirvana instead of harping on passive x-over fillding components one at a time which is slow and without measurement qeuipment you're basically shooting in the dark. With the DCX you can dial in the x-over numbers, play with the relative phase of the drivers and play with the relative sensitivity offset of each drivers to obtain a seamless integration.

cheers.

Hi Will,

This pragmatic approach sounds about right for me. The DCX2496 looks nice but is not that cheap. I've seen this auction 280163495024 on ebay and wonder that it might do a similar if less refined/refinable job?

Could you expand on what cabling is required from the AES/EBU to phono adaption? I thought AES/EBU was digital over XLR in which case a simple cable wouldn't do it...

Thanks,

Dan
 
Will said:
Hi There,
If I were you, I'd invest in 2 more additional power amplifier for 3 separate amps to the 3 drivers (perhaps you can build the gainclone?), get a dcx 2496 and connection cables to interface between RCA and AES/EBU connection; not forgetting a simple preamp. This will get you up several notches and get you going much quicker to audio nirvana instead of harping on passive x-over fillding components one at a time which is slow and without measurement qeuipment you're basically shooting in the dark. With the DCX you can dial in the x-over numbers, play with the relative phase of the drivers and play with the relative sensitivity offset of each drivers to obtain a seamless integration.

cheers.

Does anyone know if the DCX2496 offers the kind of complexity that could match something like this:

http://www.linkwitzlab.com/orion_asp.htm

If it can, then it would create a comparitively easy way to set up such a system.

Thanks,

Dan
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.