Hi , guys
I am building a 2-way loudspeaker.
I have two Peerles 831882
and two Seas 27TFFC
The box will be 8.6L BR.
Tuned to 57hz.
Can you give me a advice about the crossover network?
Could I use a crossover design from here
I found this project , and it uses almost the same driver.
The woofer is the same one , but the tweeter is the one without ferroflouid and it has a soft polymer surround , while mine has
sonolex surround.
I am building a 2-way loudspeaker.
I have two Peerles 831882
and two Seas 27TFFC
The box will be 8.6L BR.
Tuned to 57hz.
Can you give me a advice about the crossover network?
Could I use a crossover design from here
I found this project , and it uses almost the same driver.
The woofer is the same one , but the tweeter is the one without ferroflouid and it has a soft polymer surround , while mine has
sonolex surround.
Good choice of drivers.
Someone may design a crossover for you here, but why not learn how to do it yourself?
http://www.rjbaudio.com/Audiofiles/FRDtools.html
Also this site will be useful - read the driver test pages:
www.zaphaudio.com
Someone may design a crossover for you here, but why not learn how to do it yourself?
http://www.rjbaudio.com/Audiofiles/FRDtools.html
Also this site will be useful - read the driver test pages:
www.zaphaudio.com
Thanks for the advice.
Zaph pages ware the guideline in chousing the woofer , and the looks of the woofer had to do somethig with it.
It has low distortion , and a nice response.
I wanted only an advice where to cross , and what type of a crossover should I use.
Well , I know that 6db per octave wouldn"t work.
And I know about the woofers response it has a bump after 2kHz.
Zaph pages ware the guideline in chousing the woofer , and the looks of the woofer had to do somethig with it.
It has low distortion , and a nice response.
I wanted only an advice where to cross , and what type of a crossover should I use.
Well , I know that 6db per octave wouldn"t work.
And I know about the woofers response it has a bump after 2kHz.
zega55 said:
That's the result of picking a driver with a pole piece, i.e. extended response inc. the cone breakup. Makes the crossover more difficult/complicated now. IMO concave cone shapes allows use of combined acoustical rolloff with the electrical response much simpler.
compare to http://www.tymphany.com/files/products/pdf/832873.pdf
lower Fs as well
I guess it's not as sexy looking tho.
zega55
I would suggest 2Khz.
I would also use a more aggressive slope than 12db/octave.
Since this is an untested design(?), I would be tempted to outboard the xover to allow for experimentation and evaluation.
One of the advantages of active crossovers is more aggressive filter rate 24 - 48db/octave which allows for a lower xover point and easy of experimentation.
I would suggest 2Khz.
I would also use a more aggressive slope than 12db/octave.
Since this is an untested design(?), I would be tempted to outboard the xover to allow for experimentation and evaluation.
One of the advantages of active crossovers is more aggressive filter rate 24 - 48db/octave which allows for a lower xover point and easy of experimentation.
Thank's Jay
The baffle size 6.6in(width) by 13.3in(high).
The woofer is placed 0.3in from left and right edge and 1in from the bottom edge of the woofer.
The tweeter is placed to be 0.3in from the right edge on one speaker, and from the left on the other speaker.
I'm still playing with distance from tweeter and woofer.
The speakers will be placed in my room that is 4m by 3m , and about 15in from the wall.
The baffle size 6.6in(width) by 13.3in(high).
The woofer is placed 0.3in from left and right edge and 1in from the bottom edge of the woofer.
The tweeter is placed to be 0.3in from the right edge on one speaker, and from the left on the other speaker.
I'm still playing with distance from tweeter and woofer.
The speakers will be placed in my room that is 4m by 3m , and about 15in from the wall.
Since the vast majority of DIY speakers are basically rectilinear, adding a baffle step compensation is a requisite, to overcome the inherent problems with a box. (IMHO)
In the 80's I purchased a pair of Focal Eggs; This effectively ended my days of 6 sided boxes.
I had been aware of Olson's work on the intrinsic response characteristics of the various geometric shapes, and had seen the work of Linkwitz and others to implement compensation on a flat 2 dimensional baffle. This seems at best a compromise since the baffle is not circular with the woofer centered, and in view of the wavelengths involved, simple baffle round over was only marginally effective
Hearing the differences of a non-rectilinear cab was ultimately what did it for me.
In the 80's I purchased a pair of Focal Eggs; This effectively ended my days of 6 sided boxes.
I had been aware of Olson's work on the intrinsic response characteristics of the various geometric shapes, and had seen the work of Linkwitz and others to implement compensation on a flat 2 dimensional baffle. This seems at best a compromise since the baffle is not circular with the woofer centered, and in view of the wavelengths involved, simple baffle round over was only marginally effective
Hearing the differences of a non-rectilinear cab was ultimately what did it for me.
Ivan,
Don't bother to build them again if you already did. Edge roundover is beneficial only to the tweeter response. So, you can make only top and side edges (only the portion above the point where the woofer touches the side edges) round. Or in case you round over the enitire side portion, the woofer's frame sticking out a tiny bit on sides won't cause any problem functionally or aesthetically.
And as for the internal volume, augerpro's advice is good. But since you already have the boxes built, don't bother, either. According to my unibox sim (using Zaph's measured T/S params), 8.6 liters is not bad. 50 Hz Fb seems right, too.
I'll model a crossover and get back to you. You haven't cut out driver holes, have you? If not, I'll decide on their positions and use them in the modeling.
Don't bother to build them again if you already did. Edge roundover is beneficial only to the tweeter response. So, you can make only top and side edges (only the portion above the point where the woofer touches the side edges) round. Or in case you round over the enitire side portion, the woofer's frame sticking out a tiny bit on sides won't cause any problem functionally or aesthetically.
And as for the internal volume, augerpro's advice is good. But since you already have the boxes built, don't bother, either. According to my unibox sim (using Zaph's measured T/S params), 8.6 liters is not bad. 50 Hz Fb seems right, too.
I'll model a crossover and get back to you. You haven't cut out driver holes, have you? If not, I'll decide on their positions and use them in the modeling.
Ivan,
Here's my modeling result. This is from full modeling, including box, port, and baffle step loss/diffraction effects. I expect that these predicted responses will be very close to reality so you will only need to tweak the tweeter padding resistance to find a value that suits your taste.
The 27TFFC's impedance is identical to the 27TDFC's, and its FR is also very close to the 27TDFC's except that it's 1 dB more sensitive. So, I used my 27TDFC files by scaling its FR for convenience. The 27TFFC has the same, excellent motor as the 27TDFC. This means it should have no problem being crossed at 1.6 kHz with a LR4 filter for a moderate SPL goal. But for extra headroom and peace of mind, I crossed it at 1.85 kHz LR4 in this design.
Driver positions on your 6.6"x13.3" baffle I used for modeling are as follows. The tweeter is 3 inch away from baffle top with a horizontal offset of 0.75 inch. The midwoofer is located 8.25 inch away from the baffle top.
This crossover has about 4.5 to 5 dB baffle step compensation, which should be about right for your speaker placement.
I could use a simpler crossover topology, but this time I chose a no-compromise approach because I found a few more components made the driver phase aligned better and the system frequency response more linear. This crossover won't cost much to build, though.
Everything you need to know to build the crossover is in the schematic. If Jantzen coils are available, I recommend using the P-core for the primary coil (2.2 mH) in the woofer net to meet the low DCR I used in modeling.
As for the port tuning, 50 to 55 Hz Fb will be suitable. I used 55 Hz in the above modeling.
Let me know if you have questions.
-jAy
Here's my modeling result. This is from full modeling, including box, port, and baffle step loss/diffraction effects. I expect that these predicted responses will be very close to reality so you will only need to tweak the tweeter padding resistance to find a value that suits your taste.
An externally hosted image should be here but it was not working when we last tested it.
The 27TFFC's impedance is identical to the 27TDFC's, and its FR is also very close to the 27TDFC's except that it's 1 dB more sensitive. So, I used my 27TDFC files by scaling its FR for convenience. The 27TFFC has the same, excellent motor as the 27TDFC. This means it should have no problem being crossed at 1.6 kHz with a LR4 filter for a moderate SPL goal. But for extra headroom and peace of mind, I crossed it at 1.85 kHz LR4 in this design.
Driver positions on your 6.6"x13.3" baffle I used for modeling are as follows. The tweeter is 3 inch away from baffle top with a horizontal offset of 0.75 inch. The midwoofer is located 8.25 inch away from the baffle top.
This crossover has about 4.5 to 5 dB baffle step compensation, which should be about right for your speaker placement.
I could use a simpler crossover topology, but this time I chose a no-compromise approach because I found a few more components made the driver phase aligned better and the system frequency response more linear. This crossover won't cost much to build, though.
Everything you need to know to build the crossover is in the schematic. If Jantzen coils are available, I recommend using the P-core for the primary coil (2.2 mH) in the woofer net to meet the low DCR I used in modeling.
As for the port tuning, 50 to 55 Hz Fb will be suitable. I used 55 Hz in the above modeling.
Let me know if you have questions.
-jAy
Thanks Jay.
It will be hard to get coils with low R , could I use ferite core coils for the woofer.
Can I use audyn MKP caps for tweeter , and MKT for woofer?
I'm modeling passive crossover design calculator 6.2 right now.
I found that it is quite usefull.
I found the FRD i ZMA files for my drivers and I'm kind of playing with the PSC 6.2 to learn how to inplement drivers.
It will be hard to get coils with low R , could I use ferite core coils for the woofer.
Can I use audyn MKP caps for tweeter , and MKT for woofer?
I'm modeling passive crossover design calculator 6.2 right now.
I found that it is quite usefull.
I found the FRD i ZMA files for my drivers and I'm kind of playing with the PSC 6.2 to learn how to inplement drivers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.