Im rebuilding boxes for an old set of polk speakers. Each speaker has a tweeter and two 6.5 inch woofers. I dont know the impedance of each driver, only the system rating that Polk gave them as 8 Ohms nominal. I am probably not going to go through the trouble of getting a mic, a preamp, a jig and learning Speaker Workshop just for these speakers, so given that, I have a crossover question.
The crossovers that came out of them are fairly simple. The only thing connected to the positive lead to the woofers (which are wired in parallel) is a 1.45mH iron core inductor. The tweeter has a 1ohm resistor and a 5.8mF cap in series, and a .4mH air core inductor in parallel.
So I have a 1st order xover on the woofers rolling off at 878 hz at 8 ohms, or 439 hz at 4 ohms. The 2nd order highpass is harder to figure out. The closest figures I can get using online crossover calculators for these component values, are 3400hz figuring a 4 ohm tweeter, Linkwitz-Riley calcs. Figuring 8 ohms on the tweeter I cant match up the cap they used with the coil at any frequency. Im assuming the resistor is just for tweeter attenuation.
So even if the highpass is really rolling off lower, say 2000, doesnt that leave me with a fairly good sized response hole? Or do you think that using a 1st order on the woofers is mild enough so that its just bringing down some peaks that these drivers may have?
I know I could answer all of these questions by properly testing my drivers, but beings that that isnt gong to happen then I thought this would be something the experts here could mull over and have some ideas about.
Thanks
The crossovers that came out of them are fairly simple. The only thing connected to the positive lead to the woofers (which are wired in parallel) is a 1.45mH iron core inductor. The tweeter has a 1ohm resistor and a 5.8mF cap in series, and a .4mH air core inductor in parallel.
So I have a 1st order xover on the woofers rolling off at 878 hz at 8 ohms, or 439 hz at 4 ohms. The 2nd order highpass is harder to figure out. The closest figures I can get using online crossover calculators for these component values, are 3400hz figuring a 4 ohm tweeter, Linkwitz-Riley calcs. Figuring 8 ohms on the tweeter I cant match up the cap they used with the coil at any frequency. Im assuming the resistor is just for tweeter attenuation.
So even if the highpass is really rolling off lower, say 2000, doesnt that leave me with a fairly good sized response hole? Or do you think that using a 1st order on the woofers is mild enough so that its just bringing down some peaks that these drivers may have?
I know I could answer all of these questions by properly testing my drivers, but beings that that isnt gong to happen then I thought this would be something the experts here could mull over and have some ideas about.
Thanks
Whats the objective Joelives? Are you wanting to design a new XO, or just curious about the old one?
How did you find the value of the inductors?
If there was a hole as big as you say, you'd certainly hear it. Sounds like maybe the inductor value is not correct.
If you are planning to design a new XO, better plan on some measuring equipment, or spending a lot of time...
Mick
How did you find the value of the inductors?
If there was a hole as big as you say, you'd certainly hear it. Sounds like maybe the inductor value is not correct.
If you are planning to design a new XO, better plan on some measuring equipment, or spending a lot of time...
Mick
hi joe. i've done a sim of your xo and the frq. is 1.5khz (6db/octave for low-pass and 12db/octave for high-pass) whith no gap. the speakers are 8ohm for the tweeter and 16ohm each for the woofer. i hope this helps😉
That makes the point exactly!!
Without actual measurements of driver responses (impedance, amplitude, phase) everything else is even less than educated guessing.
Without measurements maybe You can avoid big mistakes by hearing but forget about to exhaust the possibilties of Your drivers only to some reasonable extend.
I just don`t get it how You guys can simulate speaker xovers without having confirmed datas of impedance versus frequency not to talk from the acoustical drivers behavier including acoustical phase response.
Speaker drivers just don`t behave like resistors.
thanks, raindog, I guess thats it makes sense that the woofers are 16ohm, I didnt think of that.(Very NewB here) Thanks for the input.
Kanga, the coil value was written on the back side of the board it was mounted on, and I was definately thinking that the value was wrong, except that its a fairly big iron core inductor, so it seems like it should have a higher value.
Actually what I was thinking of doing was using the same value of components, but replacing them with better quality ones. I might possibly play with a few things just for the fun of it and listen to he results. The reason I am reluctant to get measurement gear is that I already know what sounds OK (the current XO) so I dont have to worry about designing a new xo from the ground up.
Cocolino- Hum, not sure what point you are refering to but sorry to upset you so. If Im not letting my speakers live up to your expectations I apologize.
Kanga, the coil value was written on the back side of the board it was mounted on, and I was definately thinking that the value was wrong, except that its a fairly big iron core inductor, so it seems like it should have a higher value.
Actually what I was thinking of doing was using the same value of components, but replacing them with better quality ones. I might possibly play with a few things just for the fun of it and listen to he results. The reason I am reluctant to get measurement gear is that I already know what sounds OK (the current XO) so I dont have to worry about designing a new xo from the ground up.
Cocolino- Hum, not sure what point you are refering to but sorry to upset you so. If Im not letting my speakers live up to your expectations I apologize.
joelives,
How could I be "upset" about this
...... but by all means I want to apologize if my post gave You that impression.I just intended to point out that without measure the acoustic effects, modify existing or building new crossovers for speakers cannot be very succesful IMO.
When I began DIYing speakers about 25 years ago I did not know this and I had to learn it the "hard" way (and there was nobody who could have told me) after a lot of frustration and waste of lots of efforts in time and money.
Although learning from own mistakes often is the best teaching, others don`t necessarily need to make the same mistakes as I did, therefore I only wanted to offer some help but obviously I have somewhat overreacted 🙁 .
I have to agree with Cristoph here, there is no point in upgrading an incorrect crossover, you might as well just pour your money down the drain.
As for measurement, you obviously have a PC, so all you need to do is download one of the many FFT programs on the net, and use a cheap computer headset mic, that will give you good enough data to work on. It's not difficult, after all if I can do it, anyone can😉
As for measurement, you obviously have a PC, so all you need to do is download one of the many FFT programs on the net, and use a cheap computer headset mic, that will give you good enough data to work on. It's not difficult, after all if I can do it, anyone can😉
Crossovers
Hey joelives,
Just because it appears that it may cross there electrically does not mean that it will cross there acoustically.
There will be about 4 to 6db worth of baffle step loss that will come in at about 800 to 1,000Hz and down.
So the first order network starting its reactance there is common. It will have to pull the response level down 4 or 5db just to flatten it out with what is at 800 or so and down.
So where you are thinking it to be nearly 10 or 12db down it is really only about 6. In other words it will really "cross" about 1000Hz or more higher than you think.
Post the question for help on the Mad board or other forums. Someone in your area will have measuring equipment and will measure it for you for free.
I have sent a lot of speakers out to customers in the Phoenix area, all big hobbyist for the most part, but I do not recall if any have measuring equipment.
And yes there is certainly room for improvement by upgrading those parts. Iron core inductors? Eeeeek!
I good quality cap and resistor will go a long way too.
Good luck.
Oh yea, if you don't mind the shipping charges both ways send them to us to be measured. We measure speakers for free all the time, brought in and sent it.
Hey joelives,
Just because it appears that it may cross there electrically does not mean that it will cross there acoustically.
There will be about 4 to 6db worth of baffle step loss that will come in at about 800 to 1,000Hz and down.
So the first order network starting its reactance there is common. It will have to pull the response level down 4 or 5db just to flatten it out with what is at 800 or so and down.
So where you are thinking it to be nearly 10 or 12db down it is really only about 6. In other words it will really "cross" about 1000Hz or more higher than you think.
Post the question for help on the Mad board or other forums. Someone in your area will have measuring equipment and will measure it for you for free.
I have sent a lot of speakers out to customers in the Phoenix area, all big hobbyist for the most part, but I do not recall if any have measuring equipment.
And yes there is certainly room for improvement by upgrading those parts. Iron core inductors? Eeeeek!
I good quality cap and resistor will go a long way too.
Good luck.
Oh yea, if you don't mind the shipping charges both ways send them to us to be measured. We measure speakers for free all the time, brought in and sent it.
Danny- very good point, I never considered an acoustic crossover point. That explains alot. Thanks for the considered response, and very nice looking stuff on your website BTW. You must really love your job.
Pinkmouse- Since I am just rebuiding the cabinets on (using the same Vb) speakers that I have been fairly happy with for 10 years I know that the existing crossover is fine, its not like its incorrect because it sounds ok to me. Though if all that I need is an inexpensive mic plugged into my soundcard to get rudimentary measurments then great, I can do that. I was under the impression that I needed a preamp and a jig and a month to invest into Speaker Workshop. Could you recommend an easier to use, simple measurement program? That would be great.
Cocolino- Sorry I snapped, sleep deprivation makes me cranky.
Thanks all for your help!
Joe
Pinkmouse- Since I am just rebuiding the cabinets on (using the same Vb) speakers that I have been fairly happy with for 10 years I know that the existing crossover is fine, its not like its incorrect because it sounds ok to me. Though if all that I need is an inexpensive mic plugged into my soundcard to get rudimentary measurments then great, I can do that. I was under the impression that I needed a preamp and a jig and a month to invest into Speaker Workshop. Could you recommend an easier to use, simple measurement program? That would be great.
Cocolino- Sorry I snapped, sleep deprivation makes me cranky.
Thanks all for your help!
Joe
Joe
I can't recommend any particular PC software as I run a Mac, but just do a google search for say, audio and FFT, and you will find lots out there. Some even output pink noise, and this is very handy, otherwise you will also need a program to do this ( or burn a couple of minutes onto a CD).
Then just put the mic as close to the driver as you possibly can, within a cm is good, and play your pink noise. This will give you a basic idea of the response curve of your driver/box/crossover. Then do the same with the tweeter and if you look at the graphs side by side you will be able to see if you have any problems.
Good luck😉
I can't recommend any particular PC software as I run a Mac, but just do a google search for say, audio and FFT, and you will find lots out there. Some even output pink noise, and this is very handy, otherwise you will also need a program to do this ( or burn a couple of minutes onto a CD).
Then just put the mic as close to the driver as you possibly can, within a cm is good, and play your pink noise. This will give you a basic idea of the response curve of your driver/box/crossover. Then do the same with the tweeter and if you look at the graphs side by side you will be able to see if you have any problems.
Good luck😉
- Status
- Not open for further replies.