|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.194.172.154
first off, Happy Independence Day to All!
now down to business - I've been attempting to build this OB with the Eminence Alpha-15A as the base of build, swapping out mids and tweeters to find best combo. I'll put a combo on the board, then using various calculators figure out the Zobels, L-Pads and BSC and use in conjunction with PE 3-way XO 3K/375hz.
What I need help with is understanding why I can adjust, with component changes, for a nice balanced output from all three drivers using, in this example, a Stanley Clarke CD. So much so I sit back and say to myself, Damn, Job well done. I am happy with this iteration. Then I change the disc to say, again in this example, a John Cougar Mellincamp and it sounds like absolute horse shit at almost all volume levels. I then find myself beginning the viscous circle of struggling to correct for the Mellincamp CD to no avail, changing drivers included. Change back to Stanly Clarke CD and make changes to again get the sound back to an enjoyable state. Is this a recording quality issue, speaker issue or the combination of both? It's extremely frustrating and is almost becoming a throwing up the hands in the air and screaming F*CK IT! and go back to cookie-cutter box speakers?
What am I not understanding with this process? Should not a driver/filtering combo that is right at home with a piece of source material be equally at home with a large per cent age of other source material, all the rest of replay chain unchanged? Is this a miss-match of XO points as they relate to the driver combinations?
Can someone provide some links to better understand, in general, driver characteristics in layman terms? Also, study aids for understanding the various graphs associated with driver characteristics? I guess I'm looking for speaker design 100 level. I do not have or use and modeling software or the likes. My ears and soldering iron and my Mac/net are it as far as tools go.
As always, any and all advice welcomed.
"the only thing necessary for evil to succeed is for good men to do nothing"
Follow Ups:
yeah, what they-said. you're in good company and you'll find your way out, too. audio pursuits seem to not well-lend-themselves to task-based approaches, but they do provide the gifts of (a) a journey (whether we want that or not) and (b) hearing-education. double-edged stuff, both (i.e., you can't unhear something). listen, measure, wank, repeat. in the process, we get to meet and interact with a lot of great people. keep it fun and it will stay healthy. everything is data and it's all a gift. good luck!
S.
Posting the actual tweeters and mids you are using would help get you some practical advice, undoubtedly someone here has some experience with them. You don't say how you have chosen the crossover points (presumably from the manufacturers data sheets?) or how you have chosen the center frequency and Q of the Zobels. Guessing at these things will get you a box full of extra parts very rapidly.
Voicing a system solely by ear with one recording can be problematic, as you tend to wind up with a rig that sounds good only with that recording (I've been there, done that). Very well recorded pieces can sometimes sound good on an overly bright rig, while everything else does'nt (AKA the "audiophile screechies). Go over to Stereophile's website and read J. Gordon Holt's article "Space, the Final Frontier" from the March 1994 issue. In it JGH "spilled the beans" when he pointed out that speakers with a tilt towards the bass (for warmth) plus a treble peak (for detail) were getting rave reviews in the magazine, while comparable speakers with more of a flat response were getting luke warm reviews. Take a look at the frequency response graphs of most of the speakers tested in Stereophile and you will see the signature Holt described, or at least half of it. The manufacturers are aware of this, and now you know it too.
As to references, the late Ray Alden's "Advanced Speaker Systems" is hard to beat, it's out of print now but you can usually find used copies through Amazon. Vance Dickason's "Loudspeaker Cookbook" is also very good, though I have'nt seen the latest editions. You can download Room Eq Wizard (REW) to your Mac for free and with the mic Bill has linked to here, you'll have a very good test rig. Even using the internal mic in a Macbook is better than guessing by ear.
Good Luck! You are following a well worn path.
Paul
Do you have any kind of measurement setup?
If so, measure your frequency response and distortion. That may give you some hints.
If you don't have measurement gear, you'll end up wasting a lot of money.
> Is this a recording quality issue, speaker issue or the combination of both?
Both. Not all sources are created equally, so you can't rely on them and your ears alone to get the best possible result. There's no substitute for in-room response measurements. Then if a particular CD doesn't sound good you know it's not the fault of your system.
Just a few tips, listen to the drivers individually and adjust the crossover frequency to see if something nasty shows up. Sometimes a driver is crossed too low or too high.
Also, make sure that your amp, preamp, and source are all high quality top notch stuff. I've seen so many times people trying to design a DIY speaker and they are powering it with crappy class-D amps and a DVD player as a source!
I'm starting over and doing one driver at a time. This time I'm going to start with a Dire Straits disc as they do well recorded studio albums and see where this tack takes me.
my sources are Class A, P-P, UL EL34 dual mono amp and Denon DVD 2800 MK II as transport with heavily modded SVDAC05 DAC.
"the only thing necessary for evil to succeed is for good men to do nothing"
I am in the same boat as yours -- My only test instrument is a DVM. I have been seriously developing my system now for a little over 10 years, and I strongly recommend developing a group of different songs that can help you judge the effectiveness of your system. You mentioned two where one sounded great but the other highlighted problems. I now have about 20 that I will go through just to get a sense of where I am. That test set is as effective as the diversity of factors each song highlights. You don't develop this test set quickly, but it's very worthwhile.
As my system has becomes more refined, I also find I need a longer period of time (perhaps 2 months now) to get a handle on the next issue to address.
I do enjoy the process and all the music in the meantime.
Skip
> My only test instrument is a DVM
You already have a computer, there are a number of freeware programs available (I use HolmImpulse), so all you need is a USB mic, like this one:
http://www.parts-express.com/dayton-audio-umm-6-usb-measurement-microphone--390-808
There's really no excuse anymore for not being able to take precise measurements.
+1
I just started using a Dayton calibrated mic with the audiotool app, from which I can then transfer data to my pc if I want to do detailed comparisons. Total cost $30, and works marvellously. It beats seven shades out of manually plotting test tones.
that's patience my man!
I've only been at this particular project for a bit now and am starting to feel beat down. Two steps forward and one step back, all the while not being able to sit down, relax and enjoy some great sounding tunes as these will be my main system's speakers.......
time to throw a leg over the CrossBones and blast out into the country to unwind and clear my head.
"the only thing necessary for evil to succeed is for good men to do nothing"
Easy does it. :-)
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: