Oh gosh. Thanks for posting those LEAP screens. I was all wrong when I said it took loosey-goosey inputs (like all the measurements except maybe for DC resistance) and makes three decimal places. It makes FOUR decimal places.
OK, maybe not everybody is laughing as loud as I am. But that kind of phony-looking preposterous precision is misleading if not fraudulent.
BTW, anybody who doesn't know the significance of "160" and "6.3" has no business entrusting their crossover design to a sim. Do you know what Fermi-izing is?
Fermi problem - Wikipedia
As I posted earlier, the practices of manufacturers are necessarily quite different than DIY needs. Even so, how many times have you heard of fine store-bought speakers sounding lousy in some rooms?
B.
Last edited:
... and a reason LEAP creates mega-element designs is because it is sticking in bits and pieces so as achieve the illusory precision it calculates using standard value resistors etc. At least as far as I can read the diagrams.
B.
B.
Ben,
LEAP doesn't create "mega-element designs". Like every other SPICE based network modeling program, the user defines the network and the software analyzes its behavior based on the values.
LEAP doesn't create "mega-element designs". Like every other SPICE based network modeling program, the user defines the network and the software analyzes its behavior based on the values.
But any of the circuits I saw had a great many elements. Silly illusion of precision.
B.
From the LEAP manual. I can read it just well enough to guess it is nuts.
I think the burden of proof is on those who say this level of complexity is audibly better than simple circuits because it just might be worse.
B.
Attachments
Last edited:
IMO Off axis response is just as important as on axis response as there is more of it and needed for correct reflections or total sound in the room.
Ben I think you have answered the question. I'm sorry if you took my comment as a put down. Perhaps I should have asked the question directly.
Have you ever used a tool like Leap (I haven't because it is for professionals), or even some of the excellent free software like Xsim, PCD, VituixCad or even speaker workshop to design a passive crossover? Using properly taken measurements that include relative phase data. Shock horror even if you have to use measurements where minimum phase has been extracted you can put in all of the offsets and still get accurate results.
I think your responses clearly indicate the answer is no. So therefore my comment that your idea of desiging a passive crossover is stuck in the 70's I think is pretty accurate.
Apparently you have looked at these tools and decided that they are useless. Well that's your loss 🙂
If you ever bothered to use them and then measure the actual result you got and compared it to what was simulated (assuming your measurements were good in the first place) then you might be shocked that yes it does actually work!
As to your last post about the complexity of a crossover designed with leap. It's just a tool, the crossover designed is simply the result of what the designer decided was best, not the tool. You can use it to design the best possible two component crossover if you so desire. There is nothing that says that a crossover designed with leap (or any other tool for that matter) has to be complex. You are clutching at straws (or trolling) 🙂
Tony.
Have you ever used a tool like Leap (I haven't because it is for professionals), or even some of the excellent free software like Xsim, PCD, VituixCad or even speaker workshop to design a passive crossover? Using properly taken measurements that include relative phase data. Shock horror even if you have to use measurements where minimum phase has been extracted you can put in all of the offsets and still get accurate results.
I think your responses clearly indicate the answer is no. So therefore my comment that your idea of desiging a passive crossover is stuck in the 70's I think is pretty accurate.
Apparently you have looked at these tools and decided that they are useless. Well that's your loss 🙂
If you ever bothered to use them and then measure the actual result you got and compared it to what was simulated (assuming your measurements were good in the first place) then you might be shocked that yes it does actually work!
As to your last post about the complexity of a crossover designed with leap. It's just a tool, the crossover designed is simply the result of what the designer decided was best, not the tool. You can use it to design the best possible two component crossover if you so desire. There is nothing that says that a crossover designed with leap (or any other tool for that matter) has to be complex. You are clutching at straws (or trolling) 🙂
Tony.
Last edited:
That partly explains why old loudspeakers sounded the way they did. Glad that era is over though.
Last edited:
Hello Ben
For you to enter the DCR for an inductor you need to switch from ideal mode to parasitic mode when you do that the program calculates what the parasitic parameters effects are for that part.
I don't agree at all. You need to learn to use as many of the same tools or you will be chasing your tail. You also need the same commitment to good engineering practices or you will never be able to successfully apply what tools you have.
As far as the second part I am from the Floyd Toole /JBL /Revel camp so you can minimize room issues up front. Many don't agree and there are lots of designs out there that don't put enough emphasis on off axis polar response and the overall power response IMHO. This makes room integration more difficult.
Rob🙂
to succeed with a new loudspeaker design you must first have a specific goal in mind, you must think throu how the loudspeaker should behave in all dimensions, in and out, then choose the right drivers for it, do not be tempted to use new and cool drivers just to show off! At third you need a good measurement system, not a cheap toy-like crap. for simulators choose the ones that simulates the least! For crossovers you only need xsim, for boxes you only need a early version of unibox and then edge for baffle shapes
Yes, and that's why there's no way to design the correct crossover, in theory or in practice... at least until the drivers are installed in the room (listening in anechoic chambers excepted).
Think about it.
Footnote for acoustic intellectuals: perhaps like weather forecasting, the room response could be predicted but that may await delivery of your quantum computer; but you'd still have the "Brazilian butterfly" issues to consider if somebody moved your rug a few inches and upset your analysis.
B.
Last edited:
What non-DIY audiophiles do is changing the angle of the speaker, and manufactures and shops are recommending it. It's certainly not a perfect solution, but it should work OK if the dispersion is not bumpy and the room is reasonable enough.
Yes, and that's another reason to think passive crossovers are just crude approximations (even if calculated to an illusory four decimal place accuracy) awaiting fine-tuning with a mic and DSP.
B.
It depends how experienced you are, but for newbies i would rec'd clio pocket. For those with experience there is no need for any recommendation, they usually know what the need and how to put a proper system together
Is it speaker toe in you write about? Hefty toe in can be used to reduce that the sound capsizing to the nearest speaker if you sit slightly off center