I wasn't talking about designing the entire crossover by ear.You can say something to 10 differant people and it will be understood 10 differant ways.Don't cha love it.
Absolutely correct !Originally Posted by sba2
Bfish was talking of systematising/mathematising the whole thing so that one need not go to the dumb old world of trial&error method when all the datas/helps are available almost for free today.
Pano was saying that at the end of the day one has to finally listen to the result....and it will let one know where the whole thing is !
The OP was about a novice like person's approach to the world of Xover designing. And here Bfish's approach is more appropriate. Otherwise it may make the person's learning curve wayward.
Pano's method calls for years of exposure and thus the development of "that" keen sense/skill which can make a person go straight for the solution without any further wastage of time. Yes, it is true. The human brain is very very capable.
@Altec Best
If you are refering to my post, then I am really sorry if I have offended you somehow. It was not my intention. But if it has sounded like that, I am really sorry. It will be nice if the reason is pointed out for me.
Aditya
I'd like to see you hold the soldering iron with your ear :rotfl:I wasn't talking about designing the entire crossover by ear.You can say something to 10 differant people and it will be understood 10 differant ways.Don't cha love it.
Audio_by_Goodwill
Michigan, USA
Not at All ~ Have a Great Day ~ Best Wishes@Altec Best
If you are refering to my post, then I am really sorry if I have offended you somehow. It was not my intention. But if it has sounded like that, I am really sorry. It will be nice if the reason is pointed out for me.
Aditya
John
The complexity comes from two sources. The simplest is the difference between resistance and impedance - which is why I like Linkwitz-Riley 24 dB / octave XO's. The other has to do with those odd looking things attached to the sides of our heads - sometimes referred to as "head handles".
---
You can hear stereo just fine without them. They are acoustic, frequency dependent, phase distorters that our brains use to determine source positioning.
---
Just as musicians get their ears trained, so do audio engineers.
---
FYI the other thing these things do is signal delay for speaker stacks, where the diaphragms are not the same distance from the listener's ears - common with horn installations.
Thought I would add something useful.
---
Processor tuning (crossover, shelving, delay, etc) be it passive or active goes something like this:
- figure out gross settings with calculations - spreadsheet using SPL at reference.
- Refine with SPL meter.
- Refine by ear. I use Bach's Brandenburg Concertos and Vivaldi. This is all about "locating" the instrument. This stuff is produced by craftsmen, skilled in the art of blending sound. Settings that improve your ability to locate the positions of the instruments are better than those that don't.
---
Altec was pretty hard over on 2 way systems, focussing on 500 Hz for their high end systems. Middle C is 556 Hz. It is also, comfortably far away, from 3 kHz, where the human ear is most sensitive. Very few people have trouble locating a crying baby.
---
If you are using passive crossovers, cheap is not good. They come in 6 and 12 dB per octave. Low cost 6's are better than low cost 12's. May I suggest an approach, that I stole from Henry Kloss. Let the high end have a nominal impedance of 1/2 of the low end (can use two HF drivers in parallel) and use an appropriately sized, blocking capacitor. The initial roll off from the crossover point is 5 dB / octave and it doesn't suffer the problems of cheap inductors.
---
Note: the "problems" referred to are frequency dependent phase alterations that "confuse" the outer ear's frequency dependent, phase alterations that the brain uses to locate the source of the sound - especially in the vertical dimension.
---
L pads provide shelving (attenuation) without altering energy distribution. Translation - they have minimal effect on crossover characteristics and let you balance the output of the HF and LF drivers.
---
For those using Linux, there are two RPM's xoscope and siggen (University of Leeds), which combined with a microphone and your sound card, make quite a good set of test instruments.
A few more thoughts about crossovers and my approach to them.
I do like to use my ears when building crossovers, but make heavy use of measurements too. The reason I don't like to use measurements only is not because I don't trust the measurements - it's because I don't trust my use of them.
Unless you have access to an anechoic chamber and lot's of good measurement gear, getting good measurements is hard. Being able to go outside to a still and quiet place can help a lot, too. I don't have that.
So I find it too easy to be led astray by measurements only. I measure and tweak like crazy, but sit down and listen as a sanity check. If something doesn't sound right, or even "feel" right, I go back and measure again. I can't do it all by ear or measurement alone, I need both.
In the end, tho - the final little tweaks will be done by ear. Moving the measurement mic by a foot can make noticeable changes. Moving my ears by that amount rarely does.
One nice thing these days is that for about $300 you can have a great measurement rig. Plenty good for the home audio builder. Software is free or inexpensive.
Crossovers are hard. But they will make or break a speaker.
Now that's more like it. A perfectly logical, and not that uncommon approach.
I am a bit disappointed that some of the techs and degreed engineers that I know lurk here haven't chipped in, but I assume, like me, they've grown tired of posting when it seems the majority of members fall in the "if you like the sound of it, it's right" category, and don't care to hear the scientific method. Having been a tech for my adult life, I know that definitive answers can only be had by methodical testing and objective analysis of the results. Trial and error might eventually get you close, but more often results in the person just "calling it good".
The recent availability of software to the populace has been both good and bad. Until one is skilled enough in "reading" what the test tells you, incorrect conclusions are often drawn. I don't post my own results for this reason, as well as the fact that my acoustic test setup has some (common) flaws.
When someone says "I could hear a difference, but the measurements were the same" it tells me either the test resolution is too low, or the user can't properly interpret the results. Nothing wrong with that, as long as you keep progressing on your learning curve. I really believe (I want to say know) that anything that can be heard can be quantified by proper measurement technique.
Take poor Seadweller's case for example. The array of well-intended trial-and-error suggested fixes run the full gamut, but none yet have solved anything. By running a voltage trace of the crossover transfer function and comparing that to an acoustic output sweep, one could easily determine the problem in less than an hour. It might be electronic, it might be acoustic, or it might be the OP. Without some actual parameters, curing his shout/beaming complaint remotely, via forum, is unlikely to happen. In the meantime, posting speculations on the various XOs and drivers might have a lasting, hurtful, and unnecessary effect on the producers of such items.
So where does that leave the guys that aren't interested/can't manage proper testing? My suggestion (and echoed by others) is to go active. The advantages are multitude, and the user still has the option of tweaking to suit his ear (though even there, I contend the best setup is done with measurements). There's really no reason to go beyond that, I know some just use active to design a passive crossover for permanent use, but I feel that's a step backward.
This site has long been home to discussion of the science and physics of audio, and has, for the best part, remained uninfected by the "if it sounds good, do it" crowd that seems to control other such forums. I hope it will remain that way...
"[I]We're going all the way, till the wheels fall off and burn[/I]!"
Bob Dylan, from [I]Brownsville Girl[/I]
[I]"Time wounds all heels"[/I]
John Lennon, referring to the Nixon/Hoover deportation fiasco.
I couldn't say it better myself in fact I didn't, but I think most people understood what I was trying to say.All the theory in the world is no replacement for experiance.And in Seadwellers case if it is never going to sound good to his ears "What's the Point" (White Flag) so he is just trying to get it to Sound good to him (His Ears) and I think he will, he has gotten alot of good suggestions now it is just a matter of time. Now that is all we all are trying to do with our speakers is to get them to sound good to our own liking that is the only point here I was making, I also said you use measurement to get there and if you encounter sound quality issue you use your ear to help get there,final Minor,Minor adjustments correct. Now how does one assume that I said to do the entire crossover by earlisten to your work to see what you have and if more adjustments are necessary.
You build a set of crossovers for someone in your workshop and on your test equipment they are dead on you send him the crossovers and he installs them and they sound like crap to Him something a miss. Then what do you do ????????(Ear) Say the measurements say they are dead-on but they just don't sound good to you then What do you Do, WHAT DO YOU DO ???? Go back to your test equip. or try to make final adjustment with your ear I choose both.
Theory is good to a point but you also need experiance,hands on or should I say ears onEvery situation will be differant,therefore differant results.When your working with sound how can you only listen to your eyes(Computer screen measurements)and not your EARS
Posting Permissions
Bookmarks