Clarity on Seas Thor Kit

The curved bit will not effect where BS occurs, and it is small enuff that it will impact the smoothness of the transition only a little. 2" wider baffle will push the transition frequency down a little bit.

dave

I guess I didn't realize it's actually 2" wider. My cabinets are 10" wide with 0.5" roundovers on the sides. I compared the baffle step response with the standard Thor/Odin 9.5" baffles and the difference was minute. I would try to come up with something closer to 10".

For the tweeter resistor values, I suggest getting a 3 Ohm and a 3.3 Ohm and use which one you like better.
 
Small Thor Driver excursion

After building Jim's last revised crossover, I liked my Thors much more, but they were still lacking in bass. So, I started building a curved small Thor. (Laminating 1/8 baltic birch is a real pain.) Anyway, I finally got one of the cabinets far enough along to put the drivers in and try them out. Definitely a lot more bass and I could feel the air pumping out of the port.

Two questions: First, there seems to be a lot more driver excursion when playing low notes. I was surprised and tried playing the same music on one of the originals and the drivers definitely did not move as much. Is that to be expected as a result of the cabinet design? I am interested if others who have built an alternate cabinet noticed the same thing.

Second, my port is not as low on the bottom of the cabinet as it is in the small Thor design drawings. The center of the port is about 4 inches from the bottom inside of the cabinet. Part of the reason is that I put a flare on the inside of the port on the theory that it would help smooth the air flow. Is that OK or should I have put the port lower as shown on the drawings?
 
I asked the same question when I built my curved Thors. :)
Dave responded with the port placement that was correct for the QWTL design.
I doubt that the flared interior of the port would make that much difference, I do remember modeling the differences when I was building mine. I chose not to flare the interior of the port. Flared the outside though.
My port is 3" in dia. and Center is at 4.5" from the bottom of the inside of the cabinet.
Yes, my drivers have tremendous excursion as well. Sounds GREAT!
I have a detailed build on my curved thors on this server, if you care to see how I built mine.
Ron
 
Thanks Ron. I can only hope my new cabinets turn out as nice as yours. WhenI get them done maybe I will post a picture.

Looks like I have the port perfectly placed. The driver excursion is interesting though. I guess the original enclosures must be damping the drivers somehow compared to the new version. Can any cabinet guru shed some light on that?

Now the next question is what do I do with the old cabinets. They are really very nice. Does anyone have a suggestion for some other drivers I can put in them that might work better than the Seas drivers given the sub-optimal size of the cabinets. Something cheaper like Usher 8945P or 8945a or Dayton RS 180's. Maybe something where I can use some of the original crossover components?
 
Odin box variant plus inductor Q's.

Build the LR4 crossed at 1500Hz, that's the best sounding version IMO. The baffle design should have minimal effect on the BSC verses what I have which is a 10" wide box with 0.5" roundovers on the sides of the baffle. You can also tweak the BSC some if needed by changing the value of resistor on the woofer circuit, but I doubt you'll need to do that. In the tweeter circuit, just use a 9uF cap and 3.3 Ohm resistor (instead of the 8.7 in and 3.59 in the picture).

I sealed mine because my boxes had a 3" port so I couldn't get the tuning frequency low enough to provide a satisfactory ported response. They are commercial speakers that I modded. The W18 woofers work well vented in a large box, 2 cu. feet. A smaller box like the Odin could also work if the tuning frequency is low enough, but the deep bass will not be there. You could always try it with the 2" port and plug the port and stuff the box if you can't get the bass response right. They will sound good sealed, but need a subwoofer for 50Hz and below.

(Background: bought mine used for a decent price and after discovering the crossover issues, am now trying to tune it up... I'm not experienced with speaker design at all)

Sounds like we have a similar box. The one I have is similar to an Odin and uses an Odin Mk 3 crossover, but is 10" wide (with 3/4" rounding) and 15" deep with a 3" port tube (the flared plastic cap takes a bit off down to 2.75") that is 10" long with no interior brace at the opening. The sides are all 1" MDF, which makes it almost an identical amount of interior space, if maybe slightly larger.

Sounds like the 3" (2.75" opening) port should probably be sealed up then?

My tentative plan so far has been to build your 1500Hz crossover, and am about to order parts for it.

When looking at sources of air-core inductors, the DC resistance of the available parts seems to vary non-trivially. Specifically, mH size + AWG rating doesn't seem sufficient to select a matching part.

So, do you have recommended part types for the air-core inductors, to be sure to match the design in question?

(Edit: After checking Parts Express, I can see that the sizes/DCR values in question can be had from Erse and Jantzen ... still curious about any comments on the Port or Box size)

Thanks!

Erich Boleyn
 
Last edited:
Hi Erich,

I recently had new cabinets built for a 2.1 cubic foot ported enclosure for these drivers. Once installed, I immediately noticed that I needed to reduce the baffle step compensation by about 2 dB. I then designed a reduced baffle step crossover. This crossover will probably sound good in your ported box so you wouldn't have to convert it to sealed. Along with the crossover parts I would order 2 lbs or so of acousta-stuf so you can add as needed to adjust the port output level once you hear the new crossover.

The reduced baffle step crossover schematic is attached.
 

Attachments

  • reduced baffle step crossover.JPG
    reduced baffle step crossover.JPG
    41.2 KB · Views: 646
Ignore the schematic in the previous post. This is the correct one.

(Been there before!)

Thanks for the updated crossover concept! I take it you think that the difference between the interval volume from yours to mine (which is about 1.2 cu ft.) won't require adjustments, or that it would be too tiny to care about?

In any case, your adjusted version is less expensive parts-wise, so I won't argue too hard. :D

...and BTW, what is the SW you use for the simulations? I'm kind of curious.

Erich Boleyn
 
Depending on the length of the port, the 1.2 cu ft box may have a peaking due to the higher tuning frequency. This possible peaking is why I recommend you get the acousta-stuf to damp the reflex response with.

I use Matlab at work and the free clone Octave at home. I do all in-box acoustic measurements myself using ARTA software and an ECM8000 mic with Xenyx 802 preamp into M-audio 2496 soundcard. I use LIMP for impedance measurements. You can download octave and run the simulation yourself if you wish. PM me with your email address and I'll send you all the relevant files.
 
Last edited:
Depending on the length of the port, the 1.2 cu ft box may have a peaking due to the higher tuning frequency. This possible peaking is why I recommend you get the acousta-stuf to damp the reflex response with.

I use Matlab at work and the free clone Octave at home. I do all in-box acoustic measurements myself using ARTA software and an ECM8000 mic with Xenyx 802 preamp into M-audio 2496 soundcard. I use LIMP for impedance measurements. You can download octave and run the simulation yourself if you wish. PM me with your email address and I'll send you all the relevant files.

I found a "port calculator" on the web which suggests the port on this box may in fact be correctly tuned to the "Free-air resonance" of the Excel W18 driver (34 Hz).

The box was already (likely over-)stuffed with dampening material, rather tightly. I'm going to start by removing most of it and go from there.

Erich Boleyn
 
Depending on the length of the port, the 1.2 cu ft box may have a peaking due to the higher tuning frequency. This possible peaking is why I recommend you get the acousta-stuf to damp the reflex response with.

One more question on ports that might be interesting for the general forum:

The port outlet has a plastic ring that serves to round the edges on the hole a bit. I.e. my port tube is about 10" long and 3" wide, but it has a short constriction at the end (due to the plastic ring) that reduces it to 2.75" for a very short distance.

So, the question is: Does having a short constriction at the end of the port change the port size calculation?

For example, the calculation for a 2.75" port would give a different tuning.

Erich Boleyn
 
I just ran your numbers in Unibox and it looks like you get a pretty good response with heavy fill. You get an F3 of about 50 Hz. I would leave the fill in there, just don't cover the port entrance and try to allow a path from the woofers to the port entrance.

Once the new crossover is in place, then mess around with the fill until you get the balance you desire.

The 2.75" port end will cause additional resistance to the air flow which will lower the tuning frequency slightly. In your case, this is a good thing because you want a low tuning frequency for these woofers.
 
I just ran your numbers in Unibox and it looks like you get a pretty good response with heavy fill. You get an F3 of about 50 Hz. I would leave the fill in there, just don't cover the port entrance and try to allow a path from the woofers to the port entrance.

Once the new crossover is in place, then mess around with the fill until you get the balance you desire.

The 2.75" port end will cause additional resistance to the air flow which will lower the tuning frequency slightly. In your case, this is a good thing because you want a low tuning frequency for these woofers.

Thanks again for your help.

I've ordered the various parts for the lower baffle-step compensation version of the 1.5K crossover.

I also have some port covers (thick plastic cover which is open to the sides and has a spike in the center to minimize port noise) which should have a similar effect of tuning the frequency a bit lower yet that I can try out as an alternative to stuffing the box very much.

Anyway, I'll know more in another week or two when the parts are here.
 
Crossover versions

Sorry, I am a little confused about crossover versions. One has an 8.7 cap and 3.6 ohm resistor in the tweeter circuit. Another has a 5.7 cap and 4 ohm resistor in the tweeter circuit.

Which is the latest and best? What is the difference in sound character?

I had thought the 5.7 / 4 was the latest based on this:

http://www.diyaudio.com/forums/multi-way/155887-seas-thor-index-threads.html
 
Hey guys .Been reading through this massive thread and some others as well looking for info on the Thor kit.

In the shop i work in we sell hifi and home cinema equipment amongst other and one of the brands we sell uses the W18E001
drivers (XTZ speakers), so i can get some good prices on these, and i have some of their speakers in my HT and like the sound of them. and on another totally useless point i am from Norway and live only an hour away from the factory making these drivers :p. This made me look at the Thor DIY kit for my first speaker build.

I also got an bargain on a pair of
T29MF001 magnum tweeters ( 150$ instead of 770$ ) that i just couldn't resist taking.
From what i have read here i have decided to build a pair of FAT THOR but i need some help remodeling the xo for the different tweeters.

According to seas, i mailed them and asked, they say i have to adjust the xo for the higher sensitivity of the magnum tweeters.

This is my first Diy hifi speakers, I have made a few sub enclosures for car stereo and so on, and first attempt to make my own xo. so all help is appreciated

Data sheet for the T29MF001
http://www.seas.no/images/stories/excel/pdfdataheet/e0047_t29mf001_magnum_datasheet.pdf

f_seas_excel_loudspeaker_tweeter_e0047_t29mf001.jpg


An externally hosted image should be here but it was not working when we last tested it.

 
I recommend keeping the "common" of the high pass and low pass XO filters separate -- easily done if you're using the Madisound XO printed ckt boards. I checked and double checked the wiring on my XO's before doing this == they were wired according to the schematic in AudioXpress == (the bass and treble units wired in the same polarity) but when I reversed the polarity on the tweeter the staging opened up like never before. YMMV, but perhaps this is why one of the schematics floating around on the web has the phasing opposed.
 
Last edited:
Member
Joined 2004
Paid Member
YES!!!

YES!!! Thank you for reminding me. I had thought about this quite sometime ago then never got around to trying it. I would sometimes get a slight image shift when moving my head a foot or so left-right. You are right on, staging has opened up and depth of field.
I recommend keeping the "common" of the high pass and low pass XO filters separate -- easily done if you're using the Madisound XO printed ckt boards. I checked and double checked the wiring on my XO's before doing this == they were wired according to the schematic in AudioXpress == (the bass and treble units wired in the same polarity) but when I reversed the polarity on the tweeter the staging opened up like never before. YMMV, but perhaps this is why one of the schematics floating around on the web has the phasing opposed.
 
Member
Joined 2004
Paid Member
Jackinnj and Renron, after the change I listened for three hours and noticed a fine graininess to the sound. :( Switched out the F5 for a Borbley DC100 and the grain was gone. Now I'm wondering if maybe the current limiting of the F5 was kicking in and causing this. :confused:

Perhaps the crossover with the dip in it was filtering out the distortion products at around the 1.5k notch? With the notch removed the grain/distortion is exposed.

I believe both of you also have the F5 and would be interested in your observations.