Help with filter for Scan Speak 18W/8545 and D2905/9500 combo.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi...I’m working on a speaker who uses Scan Speak 18W/8545 and D2905/9500. But I have little or no knowledge about crossovers. I have found some crossovers on the net and I don't know which to choose. Number 1, 2 or 3? This filters doesn’t use the same drivers as I do so I wonder how I should modify them to make them fit my 8545 and 9500? Or maybe somebody have crossover that already fits my combo.

1.
http://www.audiocomponents.nl/speakers/scanspeak/tekening/ref-mon_4.htm

2.
http://members.home.nl/edgar.beers/X-over.htm

3.
http://www.speakerbuilding.com/content/1021/net_draw.gif

I will try to describe my cabinet due my bad English. It is 995mm high and 247mm wide (30mm radius+5mm flat+177mm woofer+5mm flat+30mm radius). The baffle is angled 6 degrees to obtain a correct mechanical phase. The cabinet walls is made of 2x19mm MDF-board + 4 mm oak glued together with a resonance-damping glue called HQ-box, the baffle is made of 32mm MDF-board +32mm oak. Inside there are a matrix design like in Thiel's speakers, all internal walls have 4mm bitumen plates glued with HQ-box, the side an top walls are slightly angled (the wall gets thinner towards the back of the speaker) just to guide the remaining resonances to the back of the speaker like some of Wilson Audio's concepts. There are no parallel valls inside the speaker. I use pure wool as damping material inside the cabinet, and a piece of 12mm thick bondex damping material on the back wall behind SS 18W/8545.The free volume inside is 19 litres + 4 litre off free (closed) space filled with sea-sand. I use an aerodynamic port made by Sonic Design to vent. the closure. I have bi-amping terminals from WBT. The top of the speaker is decorated with 8mm thick black glass, glued with HQ-box. I use rubber damping feet instead of spikes. The total weight of a complete single speaker is about 48kg.
 
diyAudio Moderator Emeritus
Joined 2001
Well, let's take a look at the charts. The crossover in #3 contains a D2905/9300, while you have a D2905/9500. Both ahve the same VcL and both have the same DC resistance. The 9500 has an Fc of 550, the 9300 has an Fc of 650. Electrically, everything seems either identical or very close.

Here is your tweeter, the 9500:
 

Attachments

  • scanspeak2905.9500.gif
    scanspeak2905.9500.gif
    18.9 KB · Views: 1,632
diyAudio Moderator Emeritus
Joined 2001
The crossover frequency of choce #3 is around 2K Hz. Therefore, the region of most interest for our purposes is 1K to 4K Hz.

In the 9300 tweeter, the impedance is 9.5 ohms at 1K, swings down to 6 ohms at 2K and is at 5.5 ohms at 4K.

In your tweeter, the 9500, the impedance is at 8 ohms at 1K, swings down to 6 ohms at 2K and is at 5.5 ohms at 4K.

Both tweeters have identical inductance and DC resistance. I would be very, very surprised if there is any big difference in phase with so much being identical or very similar.

I therefore suggest you that you plug your tweeter into the crossover of #3 and you should be all set.
 
diyAudio Moderator Emeritus
Joined 2001
One more thing. Spacing of the drivers is very important in crossovers. If you are going to use the crossovers, the space between the centers of the drivers should be the same. By looking at the cabinet drawing on the website of crossover#3, I calculated the space between the woofer's center and the tweeter's center as 152.4 mm, or about 6 1/8". You don't have to copy the tweeter's offset like the author's project, but you should keep the distance between the drivers' centers the same.

I do not think that the 6 degree slope you put on the front baffle is going to make that much of a difference, crossoverwise. With a spacing between centers of 152 mm, the center of the tweeter will be set back only about 16 mm from where it would be if the front baffle were cut square. That is about 5/8".

These things run on wavelengths. The wavelength of a 2K Hz frequency is about 6 1/2". That is about 162 mm. So the offset is about 10 percent of the crossover frequency wavelength.

If the offset were closer to 25 percent of the corossover frequency wavelength, we would have a problem. However, there is a rule of thumb in electronics and sound that 10 percent often is a negligible percentage. It doesn't always hold true, of course, but it frequently does. In this case, the "Ten Percent Rule" would hold. There should be a negligible difference in your crossover performance because of the 6 percent slope of your front baffle.
 
Grateful...

Kelticwizard, thanks for sharing your knowledge. I will try to follow your tips and advices.
Some people on this forum use all their energy just to criticize other peoples work or make them feel uncomfortable… :mad: Kelticwizard, you are NOT one of them. I have done some serious woodwork on my speakers but I have little or no knowledge about crossovers, that’s why I subscribed to this forum. Thank you all for contributing posts on my thread.

/The Swede…Softman :)
 
welcome mike,

for those of you who dont know mike he is good designer and good person. mike, welcome to this forum. the odd thing is that they have listed you as a neophyte (along with a lot of other talented people out there) just on the count of your posts. dont take that definition seriously we dont want to loose you.
 
hi softman,

you should check the forum there are a lot of posts on this driver combo.

i remember posting before on this topic.

since we are diyers here w have a little advatage over the run of th mill manufacturer in that we can tailor the speaker to our taste and our rooms.

i have used the 8546 and 9500 and posted that XO on the foums before. in case you want it and cant find it email me. the box i used was a hexagonal pyramid each side's base being 9".
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.