Sorry, don't have off axis - might have reverse (not my speakers). In room tonal balance was excellent, so off axis probably was too. Will dig them up and post later today.
By targets I mean the goals expressed by the designer, low end extension top end response and slight mid dip. So yes there can be targets if done by ear. It's well beyond my capabilities, but I've known at least three men how could do it. Many, many years of experience to achieve those skills.
By targets I mean the goals expressed by the designer, low end extension top end response and slight mid dip. So yes there can be targets if done by ear. It's well beyond my capabilities, but I've known at least three men how could do it. Many, many years of experience to achieve those skills.
Jeez... relax. YES, of course it CAN be done, and the speakers WILL work. Add a moderate amount of information from spec sheets and that may be enough to make a builder very happy with his creation.
And what could be more important than that?
Please don't scare beginners away, many have to start with virtually nothing. That doesn't mean they should just give up because they don't have access to test equipment.
To be quite honest, I still don't use software. I have programs I've downloaded, but I don't even remember which they are and where they're located. Every now and then I'll make a spreadsheet (when I can't find my last one) to do repeated calculations, and that's enough for me. One day, I may go more in-depth, but I don't have to satisfy anyone but myself.
Be aware, you're suggesting that nothing before the advent of simulation software was properly designed. There are a few pioneers who would disagree.
The reason I reply to this is indeed that making such statements scaries beginners a lot and it is sad that interested people don't dig further because they think (reading on forums etc...) that they have to take measurements and have to use equipment that will cost some money just to build a simple 2 or 3 way traditional bookshelf speaker with passive crossover.
In the end they just give up what otherwise could be a nice and learning hobby.
I like the tutorial because it gives those people an easy approach to get decent results instead of putting them of saying it can't be done
I think instead of saying "don't work" it is better to say "not the most optimum"
In the end they just give up what otherwise could be a nice and learning hobby.
I like the tutorial because it gives those people an easy approach to get decent results instead of putting them of saying it can't be done
I think instead of saying "don't work" it is better to say "not the most optimum"
What did people do before calculators or computers? That's right - It was worked out on paper.
Right- on paper, not "imported into crossover design software".
Sorry to be so annoyed. In the 80's & 90's, I was a big proponent of computer use as a tool. I guess I still am as long as they are used as a "tool" and not as a substitute for your own brain. A fancy pencil and paper (or slide rule) they are.
The bottom line is a couple of simple formulas are enough for a beginner to make their first crossover, and it will work. If they understand those formulas, they can indeed tweak by ear and make some sort of predictable changes.
Vintage Audio, I agree entirely. That tutorial looks great, and is where newbies should be directed when they want to begin increasing their abilities.
Sorry to be so annoyed. In the 80's & 90's, I was a big proponent of computer use as a tool. I guess I still am as long as they are used as a "tool" and not as a substitute for your own brain. A fancy pencil and paper (or slide rule) they are.
The bottom line is a couple of simple formulas are enough for a beginner to make their first crossover, and it will work. If they understand those formulas, they can indeed tweak by ear and make some sort of predictable changes.
Vintage Audio, I agree entirely. That tutorial looks great, and is where newbies should be directed when they want to begin increasing their abilities.
My advice would be not to explain why something won't work, rather offer a way of achieving a goal in the manner that you find reasonable.
You could call it perhaps "A recipe on how to design a universally good sounding loudspeaker".
Crossovers tuned by ear do work for me since it's my ears that are doing the tuning! 
New posters on this sub forum are often encouraged to plug numbers into computer simulations whilst being left ignorant of what the terms and numbers mean.
The basic physics should come first (see the tutorial). Once crossovers are understood the computer may be introduced as a convenient tool for executing calculations.
New posters on this sub forum are often encouraged to plug numbers into computer simulations whilst being left ignorant of what the terms and numbers mean.
The basic physics should come first (see the tutorial). Once crossovers are understood the computer may be introduced as a convenient tool for executing calculations.
To replicate what I can do on my pc, on paper is obviously possible but it would be incredibly complicated and you would still need to take frequency response and impedance measurements. I very much doubt a newbie is going to do that when they could do it on a cheap laptop.
A couple of simple formulas is not enough to design a crossover and no the changes wont be predictable the large majority of the time.
I have measurement and design equipment (LspCad), so I'm usually able to come up with a good crossover design. Part of my design process involves listening. I find it useful (actually necessary) to listen in order to validate or discount the current configuration. Sometimes it takes days, or weeks, to iterate through to something that sounds good.
But I have a lot of friends that don't have the equipment, so they resort to "tuning by ear". Usually, they start with something that can be found on the web, an on-line calculator, or maybe a design for something "similar". The guys I'm referring to do not stop there, however. One guy I know loves to experiment, so he's constantly changing crossover components, or drivers, or even the enclosures to improve the sound. Often it takes months of listening and experimenting, and sometimes discussions with others, but he usually comes up with something that sounds quite good. Everything that he goes through in his diy effort gives him a great deal of satisfaction -- I think the journey is what it's all about for him.
I agree that a properly designed crossover takes some effort, and usually some good measurement and design equipment can make that journey easier, and in a lot of cases a better result. A simple off-the-shelf generic crossover may work for an "ideal" set of drivers. But there is no such thing as an "ideal" driver, much less an "ideal set". At best, it may provide a starting point.
In the end, what sounds good to the ear is really what matters.
just my
But I have a lot of friends that don't have the equipment, so they resort to "tuning by ear". Usually, they start with something that can be found on the web, an on-line calculator, or maybe a design for something "similar". The guys I'm referring to do not stop there, however. One guy I know loves to experiment, so he's constantly changing crossover components, or drivers, or even the enclosures to improve the sound. Often it takes months of listening and experimenting, and sometimes discussions with others, but he usually comes up with something that sounds quite good. Everything that he goes through in his diy effort gives him a great deal of satisfaction -- I think the journey is what it's all about for him.
I agree that a properly designed crossover takes some effort, and usually some good measurement and design equipment can make that journey easier, and in a lot of cases a better result. A simple off-the-shelf generic crossover may work for an "ideal" set of drivers. But there is no such thing as an "ideal" driver, much less an "ideal set". At best, it may provide a starting point.
In the end, what sounds good to the ear is really what matters.
just my
Just my point. My hearing is a mess and I have tinnitus on both ears. A flat measuring speaker will not sound good in my ears.
Thats why I tweak by ears, just using simple clacs as a starting point to work from.
That is the point I wanted to make. When I tell people I make my own speakers then they immediately say "yeah k but they will not be perfect like a speaker you can buy...." Ok I am completely convinced that measuring can help the designing a lot. You can see what you are doing etc...and with havng software like REW and measuring microphones that are affordable these days it's something A diy'er should consider.
On the other end if someone just wants to make a speaker that sounds good for him it's not that it will be "bad" or "crap" it al depends on drivers and woodworking skills and some basic logic thinking (no a 4" woofer is not going to give you earthshaking sub for example.....) It can be quite good, but ok if you measure the combination you can make it most of the time even better.
On the other had doing only measurements and no listening tests is not always good either....
For example we had a PA line array system that was completely flat with loads of filters/EQ/delay correction and measured good.
Now I didn't like the sound of it it sounded lifefless.
now we made an other setting using our ears and also measuring the response. We used less correction EQ also and let the response a little bit less flat. It was way better overall sounding.
It's just an example
Saying you can't make a good speaker with conventional crossover without taking measurements is a step to far I think. Using the datasheets of the drivers with sometimes the response of the driver and carefull thinking you can achieve good things.
Saying you can make that same speaker even better by measuring it and make small adjustments when needed I think is completely true but if you go to the extremes and start using a lot of EQ correction etc..... it will not always be "better" Using an active speaker processor to test crossover point ideas is also nice to have. afterwards you can make a passive version (taking into account the thiele/small parameters of the driver).
Just to say I don't want to offend someone he
+1
Crossovers tuned by ear do work for me since it's my ears that are doing the tuning!
Yes, at the grossest level only. But Toole showed you can't tell a cranked-up treble from a cranked-down bass and vice versa. You'd think nobody would be so naive about perception today as to utter that kind of conceited opinion.
+1. In days of yore, all you needed to know was "6.3" and "160" and no calculator was needed. If you don't know what that means, you better stick with paint-by-number simulations.
B.
A story from "way back when". In the early '70s, I worked for what became a major speaker manufacturer. At that time, measurement equipment was far more expensive than we could afford, so all tuning was done by ear. We had a range of speakers selling from $149 each to over $2000 for a set, all tuned by ear.
I eventually left my position as foreman of the small speaker production line, and went to work at a dealer for this brand in another city. a new model was shipped to us, and our first impression was, "These must be defective."
We were asked about the phone cart used, and told to use a Decca. We were asked about what amplifier, and told which one to use. Finally, we were told what album to play. On that album, they sounded rather OK...... Not on anything else. This is one of the likely failures of tuning by ear.....
I eventually left my position as foreman of the small speaker production line, and went to work at a dealer for this brand in another city. a new model was shipped to us, and our first impression was, "These must be defective."
We were asked about the phone cart used, and told to use a Decca. We were asked about what amplifier, and told which one to use. Finally, we were told what album to play. On that album, they sounded rather OK...... Not on anything else. This is one of the likely failures of tuning by ear.....
Great story. Yes, a whole variety of fallacies and artefacts confound our auditory judgments... and nobody should be blissfully confident in their subjective judgments. OK to say "I like this" and quite another thing to say "This is flat sound". We all think we are soooooo smart and can hear sooooo absolutely perfectly.
Another illusion, stuff we make DIY sounds great. Here's my "research" into hearing and DIY evaluations (see post #9):
Krell KSA clone, Aleph J or AMB for Quad 57?
B.
Last edited:
Sure it is, I've done it. Enjoyed a couple of them for years in when I was in college. I worked for audio retailers, and my speakers performance compared to "mid-fi" brands. (edit: I should say "casually compared" -simple listening, no fancy comparison tests)
Measurement tools were far out of reach, but that doesn't mean I shouldn't have done it.
I'm not suggesting that complexity is pointless, only that it is NOT necessary to make a pair of working speakers. Although spec sheets are generic examples of performance, it IS enough to build a working filter. Measurements are helpful, but NOT necessary for a beginner to build a working filter.
How about this:
You build yours your way, I'll build mine my way. We will both enjoy our hobby together and be happy for each other doing things the way that we each see fit. There is room for many perspectives here.
Last edited: