View attachment 862898Hi all
This is the second crossover I have designed and I think I'm having some trouble. Could someone take a look at my design and give me some advice. Ill also leave the manufacturer frd. and zma. files if you would like to give it a shot. I'm just a beginner and am not looking for really Hi-Fi sound just something to make it sound good. Thanks.
This is the second crossover I have designed and I think I'm having some trouble. Could someone take a look at my design and give me some advice. Ill also leave the manufacturer frd. and zma. files if you would like to give it a shot. I'm just a beginner and am not looking for really Hi-Fi sound just something to make it sound good. Thanks.
Attachments
Last edited:
If you haven't actually purchased the drivers yet, then buy a kit designed by one of the experts instead. You'll be sure to get very good sound quality for the same amount of money.
If you bought the drivers from Parts Express and it's been less than 45 days, you can still return them for a full refund.
Lots of great 2-way kits available from several different suppliers.
If you bought the drivers from Parts Express and it's been less than 45 days, you can still return them for a full refund.
Lots of great 2-way kits available from several different suppliers.
Also Meniscus Audio, diysoundgroup, and Parts Express. Probably the best overall selection of all at Meniscus.
your component values look a bit off to me, the cap on the tweeter is too big, the inductor on the woofer too small.... I'd aim for a crossover frequency ~ 2.5-3KHz, and use a simple inductor on the woofer, 2nd order (Cap & Inductor) on the tweeter (reverse the phase of the tweeter)
re:'buy a kit' - you could do that, but you wouldn't learn much
re:'buy a kit' - you could do that, but you wouldn't learn much
He's not going to learn much by simply having someone else suggest different values for components either.
Bencarlson - we were all where you are now at some point. I am not an expert, far from it. but I may be able to get you on the right track.
As PeteMcK says, your values seem off. It is easy to get a simulation pointed down the wrong path, and then it is hard to un-sim your way out of it.
Try this quick and dirty circuit I just put together. This could point you in the right direction. It is not finished, it needs to be optimized... and that is for you to do. I modelled the on-axis response, and delayed the woofer response by 1.2 inches (a guess).
A couple of things to notice about this crossover:
(1) the woofer response drops about 6 dB between 100 Hz and 1000 Hz. This is a good feature, and it is called the Baffle Step Compensation (BSC). If you try tuning the 1.6 mH inductor on the woofer, you can see how the BSC changes.
(2) If you look at the response, you can see that the phase of the woofer and tweeter are approximately the same at the crossover frequency. This is a good feature.
(3) You can see there is a lot of resistor attenuation on the tweeter. This is needed because the tweeter is quite a bit more sensitive than the woofer. The woofer sensitivity is further reduced by the BSC. When it comes to fine tuning the response, you can tune both the driver side resistor and the amp side resistor. In general, the amp side resistor can be changed without affecting the rest of the circuit too much, while the driver side resistor will change the crossover characteristics.
Give it a go, and try varying the various components a little at a time, and see how they affect the response. Eventually you will find something that looks good.
After you get the on-axis response nice, then we can work through the off-axis curves.
As PeteMcK says, your values seem off. It is easy to get a simulation pointed down the wrong path, and then it is hard to un-sim your way out of it.
Try this quick and dirty circuit I just put together. This could point you in the right direction. It is not finished, it needs to be optimized... and that is for you to do. I modelled the on-axis response, and delayed the woofer response by 1.2 inches (a guess).
A couple of things to notice about this crossover:
(1) the woofer response drops about 6 dB between 100 Hz and 1000 Hz. This is a good feature, and it is called the Baffle Step Compensation (BSC). If you try tuning the 1.6 mH inductor on the woofer, you can see how the BSC changes.
(2) If you look at the response, you can see that the phase of the woofer and tweeter are approximately the same at the crossover frequency. This is a good feature.
(3) You can see there is a lot of resistor attenuation on the tweeter. This is needed because the tweeter is quite a bit more sensitive than the woofer. The woofer sensitivity is further reduced by the BSC. When it comes to fine tuning the response, you can tune both the driver side resistor and the amp side resistor. In general, the amp side resistor can be changed without affecting the rest of the circuit too much, while the driver side resistor will change the crossover characteristics.
Give it a go, and try varying the various components a little at a time, and see how they affect the response. Eventually you will find something that looks good.
After you get the on-axis response nice, then we can work through the off-axis curves.
Attachments
[ I'd disagree with that premise. Having the benefit of a design program gives the user exactly that option to futz around with component values and see what results. A recommendation for increasing the value of the woofer inductor and reducing the tweeter cap value is a good place to start. At least that teaches the user how those changes influence the final result.
It's not clear whether the OP even has the necessary tools to do his own crossover design. A simulator is not nearly enough.
At a minimum he needs a calibrated microphone and an analysis program like REW. Then there is the need for the learning curve in order to use them effectively.
If he had those tools and the experience, I suspect he would not have had to ask someone else how to fix the crossover.
At a minimum he needs a calibrated microphone and an analysis program like REW. Then there is the need for the learning curve in order to use them effectively.
If he had those tools and the experience, I suspect he would not have had to ask someone else how to fix the crossover.
Last edited:
All of what you said is true. But not everyone is looking to be the next Wilson Audio. Crossover design is probably the most challenging aspect of speaker building. There's a lot to learn. Sometimes trying to take it all at once can lead to frustration.
'At a minimum he needs a calibrated microphone and an analysis program like REW.' - you're making far too complicated, all that's required are manufacturer's spec sheets, a little math, and the capacity to think analytically
Almost every week you see posts like this one by someone who has just joined and is going to design their own speaker. I can understand it. The personal challenge can be very gratifying if you achieve success. But I think these people underestimate what it will take to create a speaker with really outstanding sound quality.
It gets down to what you enjoy more: the journey or the destination. If it’s the journey and you don’t mind taking many detours with the possibility still that you will never reach the destination, then this approach is fine.
But if the destination you seek is outstanding sound quality in order to enjoy your music, then there are much better routes to take. Namely, but a kit designed by one of the experts.
You certainly aren’t going to save any money by designing it yourself, and in fact, may very well end up spending more that way.
Even if the Thiele/Small parameters in the manufacturer's spread sheets are accurate, which by many reports here they are not, you still don't know what the speaker really is doing without measuring it.
The crossover simulators are only a first order approximation. They don't reveal anything about resonances or other performance anomalies that need correction.
Constructing speakers is a hobby. It is supposed to be fun. A person's first set of speakers is seldom their final set. I am assuming that anyone who posts here asking for help in designing a crossover is hoping to learn something. I am assuming that they are on a journey of pride toward self satisfaction and better sound.
Some people learn by reading and analyzing, and they do not begin a project without detailed study and planning. Others learn by doing, and they willingly make multiple iterations in their quest. When I led a team of engineers, I had both types on my team. The two types both achieve results, but they seldom see eye to eye. I have come to the conclusion that a successful design team needs both types.
I do not want to discourage Bencarlson from embarking on a very rewarding hobby... the pursuit of great audio.
It is a journey, isn't it. There is much to learn, but many rewards along the way so this is OK.
When I look at kits these days they are mostly the same as this. Choose some drivers and make them flat. Make it a box, cursory consideration of the room at best, keep it all on one axis.
To do something outstanding is within the realm of the DIYer who will begin with these first steps.
I think the kits these days by some of the very experienced designers are a lot better than you are giving them credit for, and probably not matchable in performance by most amateurs.
They are the products of some very smart and talented people who have learned the art of refining sound quality. An art that very few others are likely to master.
I do agree with you that room considerations are essential, but I doubt that many of the amateurs are starting there either. They should, but most likely they have started by picking a couple of drivers based on price or something they read somewhere.
So it gets back to my earlier point. Does someone want the journey or the destination?
They are the products of some very smart and talented people who have learned the art of refining sound quality. An art that very few others are likely to master.
I do agree with you that room considerations are essential, but I doubt that many of the amateurs are starting there either. They should, but most likely they have started by picking a couple of drivers based on price or something they read somewhere.
So it gets back to my earlier point. Does someone want the journey or the destination?
Last edited:
Hi Pete! Pardon my ignorance! What's "reverse the phase of the tweeter"? Did you mean reversed polarity?
Last edited:
Yes. Because the 2nd order xover adds a phase shift of 180 degrees, reversing the polarity of that driver brings its output into closer alignment with the input
for more : Phase, Time and Distortion in Loudspeakers
for more : Phase, Time and Distortion in Loudspeakers
Hi Ben,
As has been pointed out, the FRD files are not good enough as you need the measurements from the drivers in the cabinet. However for the sake of learning, I have a complete 2 way system simulated and ready for you to play with, but with XSim. If you'd like, I encourage you to grab the files and play with them. Try creating a new crossover from scratch for them and compare the results. I've also written a lot there about crossover basics so I hope you find it all useful.
A Speaker Maker's Journey: Introduction to The W. Marshal Leach Jr. Memorial Speakers
Best,
Erik
As has been pointed out, the FRD files are not good enough as you need the measurements from the drivers in the cabinet. However for the sake of learning, I have a complete 2 way system simulated and ready for you to play with, but with XSim. If you'd like, I encourage you to grab the files and play with them. Try creating a new crossover from scratch for them and compare the results. I've also written a lot there about crossover basics so I hope you find it all useful.
A Speaker Maker's Journey: Introduction to The W. Marshal Leach Jr. Memorial Speakers
Best,
Erik