Trachorn 400

[J.4knee]

By all means please continue your EQ inputs some of us are realy paying attention.

[mikebse2a3]

----------------

On 8/1/2005 5:06:24 PM J.4knee wrote:

By all means please continue your EQ inputs some of us are realy paying attention.

----------------

Guys This Was Kind Of Hijacking the Trachorn Thread by DeanG so I Hope its OK but I thought I would copy some of the EQ post to a new Thread:

Behringer DEQ2496 Measurement/Use Methods

(TECHNICAL QUESTIONS SECTION)

mike

[scriven]

Good idea Mike. For the lazy, it is here.

[Deang]

You know, this Trachorn can do some interesting things, and right now -- I'm definitely making it do some interesting things.

One of the things the RTA is consistantly showing me is some drooping response around 300Hz. It's annoying, because I can't seem to do anything about it. It's not much, but now that I know it's there it's bugging me. I had pretty much written it off as some of kind of nulling effect when I remembered some plots I'd seen of the Klipschorn's bass response -- including the one in the JAES paper on the Jubilee. The plots revealed that at least some of the effect is simply due to the natural limits of the bass bin. All of the plots show the same thing: A rapid roll off at approximately 300Hz, a sharp rise and last gasp at 400Hz, and then a down hill journey that basically corresponds to an 18db/octave roll off.

I went back and looked at Lee's comparison plot of the Trachorn, 511B, and K-400 -- and decided I was going to take a crack at seeing what the Trachorn could do with that bit of elevated response it has around 300Hz as opposed to the K-400 -- which is still spinning its wheels trying to get off the line. The attached plot won't win me any points for advanced engineering, but it was just something I put together to help me get an image of what was hoping to accomplish. I took a section of Lee's plot, scaled it the best I could, and then traced over the response of both the Trachorn and K-400. I then overlayed the image with the K-horn bass response plot from the Jubilee paper. It's sloppy, but probably pretty close. In the bottom right hand corner I inserted the section I cut out of the plot.

I went ahead and hacked together a simple 300Hz network, and as I started to insert the inductor for the bandpass I hesitated. I had remembered that earlier in the week I had thought about wanting to try the Trachorn without the inductor. My reasoning was that with the Trachorn's bigger throat, wider mouth, and smoother launching point for the sound -- that maybe it could handle it without choking/pinching. I still think this is a real struggle for the K-400, where I'm almost positive things are rattling around on top of each other in that 5/8" throat. So, I left the inductor out -- I was tired of soldering anyway. To annoy Al the best I could, I let the impedances run wild, and so this network is basically a 300Hz Type AA.

Crossover points are approximately 315Hz and 5800Hz. The tweeter filter was stolen from Al's 'Super AA' filter he designed for me -- which is a third order Chebyshev alignment. Values are as follows:

L1, 4mH

C1, 18uF

C2, 2.2uF

L2, .15mH

C3, 3.9uF

I didn't get the lift in response at 300Hz that I thought I would get, and I was disappointed -- but it sounds insanely good. The most audible thing I've noticed is a nice tight whack on kick drum and cleaner bass lines. The tubby effect with some material is gone now, and the sound is well balanced. I've been listening like this since Saturday afternoon, and so far so good. I think maybe it's just impossible to make this horn sound bad. If it can survive my handiwork and half baked schemes -- it must be pretty good.

[Deang]

Warble tone. Looks good here, but drops off with pink noise.

[Deang]

Pink noise.

What do you guys think about my suckout problem at 300Hz? Is this just a Klipschorn thing, a nulling effect, or maybe due to some loss from my false corners?

[ben.]

Dean, my WA guess would be a room null around 150Hz, as I see that is lacking as well. In any event, if I could pick one frequency to have that problem, it'd be 300-400Hz, especially with your music taste.

A common cure for boxy bass material is to cut 300-400Hz, and a common cure for tubby material is to cut around 160Hz.

[3dzapper]

Dean, Did you try a 100-150uF cap across the woofer? You might want to just for S&Gs.

Rick

PS: Did you see the response plot for the Wilson MAXX 2 in the latest Stereophile? Your graph looks great by comparison and the MAXX got raves.

[ZAKO]

Placing the mic in the far field without using a anechoic chamber will befuddle the results. Do you want to measure the speaker or the room? Its best to separate the two for accurate results. Now get back to work & do it over. You guys are going over the results done 30 yrs ago. but with poorer equipment.

[Deang]

Consistent -- the only high quality polypropylene that you will be able to fit on the AK-4 PCB will be the Reliable Capacitor Multicap. The AK-4 board is very tight, and any round shaped polypropylene just isn't going to work. The MultiCaps are oval, and I think they will squeeze in nicely. Use the PPFXS for the smaller values (1.5uF, 4uF, 10uF), and the PPMFX for the larger values in the squawker circuit (30uF, 50uF). All of the 50uF Mylars in the woofer circuit are fine, leave them in. The inductors are fine, leave them in. If it were me, I would probably pull all of the resistors and drop in some Mills 12 watters -- unless you like dumping over 100 watts into your Klipschorns on a regular basis.

Thanks Ben, I think you're right -- I'm pretty sure now it's the room.

Rick, that small drop in response was there with the AK-3 network I was running too -- which has the 100uF across the woofer. It's kind of tricky trying to figure out how that thing works. I think maybe it only lifts the sagging response of the bass bin if you use a solid state amp. In the area it's doing its work, impedance is low -- where a solid state amp will double its output. I only say this because I noticed a marked difference in response between the Teac and my Quicksilvers in the lower end of the Klipschorn's bass response (3db). At any rate, I think that cap works on the response below a 100Hz.

Maron, I've been working with both the speaker and the room. Initially with the speaker somewhat just to see what I could see. I had fiberglass batting hanging from the walls and ceiling around one Klipschorn, and laid a layer on top as well -- mostly to see how it effected what the microphone saw. I'm just trying to put a little dent in my learning curve. I did hang all the crap around the speaker again after I changed the crossover point to see if anything obviously bad was showing up -- but it checked out O.K. It's not like I'm trying to fix anything -- the sound is great.

[mikebse2a3]

First Maron Horonzak is correct there isn't any real good way to measure the drivers pure response because the measurements will always have the room interference to some degree unless you can use some time gated technique like the ETF program I spoke of in the past and even that is limited in the lower frequencies for smaller rooms (THERE IS D.B. KEELE's close Microphone Measurement Technique for looking at very low frequency response in a room). Also if you move the MIC close to the speaker then you will not see the integration of the drivers responses properly so Again if you have the Heyser Audio review then you will see this form of measurement is the only way to begin to see what all is going on with the Khorn(OR ANY SPEAKER) as far as Anechoic response combined with real world use/measurements and then the real talent is how to interpret the results to what we HEAR.

I guess the way I see things, is that using an RTA is always going to be tricky in the fact of it isn't always measuring what you might think it is because it can't show us what is happening to the frequency response in time like a TEF or ETF type of measurement can. I do believe it does give us some good insight into what we are hearing at our listening position because it is a combination of the actual speakers response combined with the influences our individual rooms create which our ear/brain combines in some instances but it is limited in its uses.

Some things you can do to see some of what is going on is to move the MIC around some because this can reveal say room mode issues in the measurements. Also you might want to use proper sized resistors to install in place of a driver "lets say Squawker for example so that you can see what happens to the response of the Woofer/Squawker frequency region without the actual acoustical influence of the squawker in this example and then compare that to the actual Woofer and Squawker in operation together measurement in the same region this way you might discover that a dip could be the result of a phase interference between drivers due to path length differences in combination with crossover phase shifts that reach the MIC Position versus say a room mode anomaly".

Anyway you can have fun and learn while it drives you crazy doing these kind of measurements.

Mike

[Deang]

Nice post.

It's only frustrating if it sounds bad. When it sounds good -- it's just plain fun.