Speaker Asylum

General speaker questions for audio and home theater.

Return to Speaker Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Reverse Engineering thought,

174.62.20.102

Posted on December 22, 2010 at 10:52:01
430cruz
Audiophile

Posts: 253
Location: S. E. Pennsylvania
Joined: March 13, 2009
Hello Inmates,
I am fairly new here but have been an A-gon member since 1999(bryhifi). A musician and avid audio junkie for over 30 years. Several years ago I acquired a pair of DIY Zalytron built 2way speakers using Cabasse drivers. Elliot at Zalytron said they were a one-off by him and supposedly Dr. Joe D'Appolito and no info was around on the particulars. Curiosity started to get me the other day and started doing a search of older Cabasse models and found one that strikes my fancy for a project. Link below takes you to a model called the Chaloupe that uses the 21M18 woofer and DOM2 tweeter that I have in these DIYs. Sadly it is all in French so i can't make any sense of what was written about this speaker.

If any of you speak french could you give me an idea of what the consensus was? If it's good my thoughts were to try and reverse build these speakers. If I get dimensions of cabinet, port dia. and length, driver specs and go by the 4,000hz crossover point listed my brain tells me someone could figure out the crossover schematic, no?

If I am out of my mind just say so, but it seemed to make sense to me. If box dimensions, driver distance to center, port tuning ,thiele-small parameters and manufacturer claimed crossover point are all known I guess it could be worked out but I could use some help.

Any thoughts or translation of link would be greatly appreciated!

Regards,
Bruce

once you open link go down to model called "Boat" and click on it.

 

Hide full thread outline!
    ...
RE: Reverse Engineering thought,, posted on December 22, 2010 at 12:16:21
Godzilla.
Dealer

Posts: 1828
Joined: July 1, 2002
Why do you want to change what you have?

 

RE: Reverse Engineering thought,, posted on December 22, 2010 at 13:13:39
430cruz
Audiophile

Posts: 253
Location: S. E. Pennsylvania
Joined: March 13, 2009
I've looked inside these cabinets and don't like what I see. They appear to be double wall MDF, but random interior walls are baffle fronts from other projects(I guess) glued in rather than a solid panel like other interior walls. To my mind the interior volume can't be right. There are no plans available to me from the designer/builder. Crossover could use some upgrading as well.

The thing is, they are getting some things "right" to my ear. I think things are amiss in the midrange and midbass. Certain notes or registers seem to suddenly blur a bit but other times they have a complete clarity.

Oh Wait, you weren't referring to my maggies were you?
No change there! They're a keeper in my book.

here's what they look like.

 

RE: Reverse Engineering thought,, posted on December 23, 2010 at 10:58:30
Presto
Audiophile

Posts: 5957
Location: Canada
Joined: November 10, 2004
"If box dimensions, driver distance to center, port tuning ,thiele-small parameters and manufacturer claimed crossover point are all known I guess it could be worked out but I could use some help."

It is not "worked out" - it needs to be measured. Then the crossover is designed 100% based on those measurements. If you can't develop a crossover from measurements, you are largely wasting your time. You would need to stick with kits, or get the services of an accomplished DIYer / speaker designer. Many if not most DIYers think they know more than they do and trivialize many aspects of the design process.

Sorry, but this is just the truth here.

 

is that the zalytron box?, posted on December 23, 2010 at 13:06:28
Godzilla.
Dealer

Posts: 1828
Joined: July 1, 2002
?

 

What he said., posted on December 23, 2010 at 14:02:54
Duke
Dealer

Posts: 4429
Location: Princeton, Texas
Joined: March 31, 2000
Crossover design is considerably more complex than most people appreciate, and it all starts with good measurements.

Duke


Me being a dealer makes you leery?? It gets worse... I'm a manufacturer too.

 

RE: Reverse Engineering thought,, posted on December 23, 2010 at 18:49:42
430cruz
Audiophile

Posts: 253
Location: S. E. Pennsylvania
Joined: March 13, 2009
I totally understand what you and Duke are saying. My knowledge in that area is weak to say the least. I wish there was someone near me that offered these types of services.

Thank you both for steering me straight.

 

RE: is that the zalytron box?, posted on December 23, 2010 at 18:57:02
430cruz
Audiophile

Posts: 253
Location: S. E. Pennsylvania
Joined: March 13, 2009
Yes,
The previous owner lives near Zalytron and he had several other models from there. I have also had the Aria 5 from Zalytron.

 

info, posted on December 25, 2010 at 17:12:08
5uf cap .3 mh inductor simple network should work fine with your transducers.

 

I have the exact part specs and x\o schematic, only difficult part is going to be the tapped inductor, posted on December 27, 2010 at 14:12:24
cfb
Audiophile

Posts: 1130
Location: Midwest
Joined: October 19, 2002



as used in the high pass section.

For those who might take issue with this crossover, I didn't design it, aim that towards Cabasse.

Crossover is at 4,000hz. There are only five components to the whole crossover.

The low pass for the 21M18 woofer is straight forward; inductor L1 is 0.9mH and capacitor C1 is a 10 uF metallized polyester film ERO MKT 1822 series, owned by VISHAY, see pdf at
http://docs-europe.electrocomponents.com/webdocs/00a3/0900766b800a34cd.pdf

Now, the high pass for the DOM2 tweeter is as follows; capacitor C2 is 3.3uF, ERO MKT 1822 series again, inductor L2 is 0.37mH (tapped), and capacitor C3 is 4.7uF, MKT once more. Note that the tweeter polarity is reversed.

The tapped inductor is going to be a bit of work, first they used a transformer core for the inductors not your standard off the shelf iron or air core types and the high pass was wound with a fine gauge wire for some intentional series resistance to match sensitivity between the tweeter (94db) and woofer (91db). You'd think they would reduce -3db but looking at the frequency response graph it appears that it is only a -1.5 to -2db reduction so the highs are going to be slightly elevated.

Now the tapped inductor L2 will require some experimentation. Not having the tweeter I do not know where they tapped, at the 50% point or less. I would start with one and one-half ohms of winding before the tap and then measure the tweeter response. At least I have left you with only one variable to deal with in the crossover. You might try starting with a Jantzen C-Coil Toroidal Inductor available from Parts Express or the Sledgehammer steel core inductor from Madisound. Use it just for its core as neither are available in a 0.37mH value and they have the wrong wire gauge anyway. Buy some 26 or 28 gauge magnet wire and an inexpensive LCR meter such as the Tenma 72-8155 available from MCM electronics for under $50. Next, completely unwind the inductor then start winding with the new magnet wire, remembering to keep track of the turns count so at the half-way point you can get your center tap or drop several taps for easier experimentation for when you proceed to measure the tweeter response.

The speaker cabinet was built of veneered particle board not even MDF; standard over there is 19mm that is equivalent to 3/4” thick MDF here in the States. Outside dimensions are 33-1/2 inches high by 10 -1/4 inches wide by 15-5/32 inches deep with some minimal cross-bracing and Acousta-Stuf polyfill, best start with about 16 oz of polyfill and adjust (may need more) after you have finalized the crossover. It's a bass reflex tuned to 55hz, single port 2-7/8 inches in diameter, cutout may be 3” but the plastic port insert restricts the diameter.

You should at least read the Loudspeaker Design Cookbook by Vance Dickason for some needed basics and background.

On the software side Speaker Workshop by Audua is freeware, or go with LspCAD 6 standard, it runs $235.

My apologies for the rough hand drawn schematic,

 

RE: I have the exact part specs and x\o schematic, only difficult part is going to be the tapped inductor, posted on December 28, 2010 at 07:32:19
430cruz
Audiophile

Posts: 253
Location: S. E. Pennsylvania
Joined: March 13, 2009
Hey cfb,
Thanks for your response! Was some of this info in the link?

I had and read about an eon ago the LDC by Vance. Even had some floppy discs with TopBox and another program that ran on the old Tandy computer. When I saw the picture in the link of the crossover it looked kind of minimalist.

From what you have provided me it looks pretty straight forward until L2, which I think I understand and will read up on a bit more.

Having some Waterfall Bubinga and Burl Walnut sitting here is what's driving me to do this project.

Thanks,
Brue

 

That sounds like some beautiful wood for a project. As far as the background on the crossover,, posted on December 28, 2010 at 21:03:13
cfb
Audiophile

Posts: 1130
Location: Midwest
Joined: October 19, 2002
I don't recall where exactly I ran across the schematic. It's just that I have acquired a lot of information and bits and pieces pertaining to the audiophile hobby over the years and sometimes I have something and many times not. From what I recall, while the Cabasse speakers were praised for their efficiency and clarity allowing low power tubes to work well; there was some slight criticism that they were a touch on the bright side with a slightly hot tweeter. This seems to be evident in the frequency response graphs where the tweeter is slightly elevated in relation to the midbass region. The crossover lacks any L-pad or series resistor to lower the tweeter level; apparently they are using the internal wire resistance of the L2 tapped inductor to achieve the driver matching and it looks like they are only attenuating the tweeter level by about -2db. You might want to experiment with a low value series resistor in the tweeter circuit just to cut the output by another -1db if you want a more flat frequency response. The resistor will however affect the sound; the Duelund brand of resistors has high recommendations for use in crossovers. See link; http://www.duelundaudio.com/Resistors.asp

If you proceed with the project you might want to post over at diyaudio.com as you'll probably receive more useful input over there considering that the Asylum forum does not seem to favor much in the way of DIY compared to the diyaudio boards. Hope this helps some.

 

Page processed in 0.031 seconds.