SB17NAC35 with SB26ADC build - Please help
I built my speakers from a kit 20 years ago and am a relative newbie to diy speakers. A little knowledge is a dangerous thing so I’m here to ask you for some advice.
I’m a theatre buff and looking to “refresh” my 20 year old speakers with a new tweeter, woofers and crossover. Currently, I have the Vifa d25ag tweeter and two P17WJ 6.5 inch woofers in very pretty 50 litre ported enclosures. The centre speaker is a different shielded kit that never really matched the L/R’s. The surround, rear and sub are Wharfdale and I’ll keep them for a while.
I was very happy with the Vifa’s but have since ordered the Anthem MRX 1120 and plan to put in 4 new ceiling speakers with enclosures. I was considering the RSL C34E ceiling speakers but they are expensive once imported into Australia. For the same price, I can build my own which has the advantage of using the same tweeters, drivers and effective enclosure volumes to match the main speakers, ensuring the same timbre from each.
I’m after an accurate, low distortion and a flat response curve that will carry audio down to 90 hz where my subwoofer crosses over. The fronts are 50 litres with holes for a 1 inch tweeter, two 6 inch woofers and 2 ports. My plan is to use:
Front speakers:
SB26ADC-C000-4 (recessed 18mm to time align to woofers)
SB17NAC35-8 (2 drivers run in parallel from a single crossover to appear as 4 ohms and increase power handling and sensitivity)
50 litres, 2 ports, 2000 hz / 2nd-order Linkwitz-Riley crossover
Ceiling speakers (in enclosures angled at 45 degrees down directed at my seat):
SB26ADC-C000-4 (recessed 18mm to time align to woofer)
SB17NAC35-4
25 litres, 1 port, 2000 hz / 2nd-order Linkwitz-Riley crossover
Reasoning:
I like the idea of these drivers since they reportedly have low distortion and a flat response curve at a good price and don’t seem to break up as badly as other aluminium drivers. I live in the tropics (hot and very humid) so paper drivers are out of the question.
Ceiling speakers: The sensitivity of the 4 ohm tweeter and 4 ohm woofer are the same so no attenuation is necessary.
Front speakers: These need more power than 60 watts so I’ll use 2 woofers. The 8 ohm version of the woofer has sensitivity about 3.5 db less than the tweeter. By running two 8 ohm woofers in parallel from a single crossover, it will increase sensitivity by 3 db so no other attenuation is necessary. These should sound the same as the ceiling speakers but handle more power.
Enclosure volumes: The SB site gives the main values of SB17NAC35-4 as:
Diameter: 12.26 cm, Fs: 32 hz, Qts: 0.32, Vas: 33 litres – giving an enclosure of 19 litres (per driver)
www.audioexcite.com >> SB Acoustics SB17NAC35-4 tested the SB17NAC35-4 as:
Diameter: 12.26 cm, Fs: 37 hz, Qts: 0.42, Vas: 21 litres – giving an enclosure of 26 litres (per driver)
Questions:
Are these drivers the most accurate reproduction across the range at this price and if not, which are significantly better?
Will these give me a big improvement over the 20 year old Vifa’s?
Will this go low enough to seamlessly meet the subwoofer at approx. 90 hz or should I go for a 3 way system?
Are the enclosure volumes okay? I’ll use an online calculator for the ports. I can stuff the speakers with solid blocks to reduce volume if it is too big.
Is a 2000 hz / 2nd-order Linkwitz-Riley crossover the best option?
Is using two 8 ohm drivers in parallel the best option for the mains and will it sound the same as a single 4 ohm in half the enclosure volume?
Is setting the tweeter back 18mm necessary to time align to the woofers (I read that it should be 18mm for a similar combination Yet another 6+1 construction: SB17MFC35-8 + SB26ADC-C000-4 )?
Am I supposed to invert the tweeter connections on the crossover to make sure phase is correct or this accounted for by the 18mm offset?
Any other advice? - Thanks!
I built my speakers from a kit 20 years ago and am a relative newbie to diy speakers. A little knowledge is a dangerous thing so I’m here to ask you for some advice.
I’m a theatre buff and looking to “refresh” my 20 year old speakers with a new tweeter, woofers and crossover. Currently, I have the Vifa d25ag tweeter and two P17WJ 6.5 inch woofers in very pretty 50 litre ported enclosures. The centre speaker is a different shielded kit that never really matched the L/R’s. The surround, rear and sub are Wharfdale and I’ll keep them for a while.
I was very happy with the Vifa’s but have since ordered the Anthem MRX 1120 and plan to put in 4 new ceiling speakers with enclosures. I was considering the RSL C34E ceiling speakers but they are expensive once imported into Australia. For the same price, I can build my own which has the advantage of using the same tweeters, drivers and effective enclosure volumes to match the main speakers, ensuring the same timbre from each.
I’m after an accurate, low distortion and a flat response curve that will carry audio down to 90 hz where my subwoofer crosses over. The fronts are 50 litres with holes for a 1 inch tweeter, two 6 inch woofers and 2 ports. My plan is to use:
Front speakers:
SB26ADC-C000-4 (recessed 18mm to time align to woofers)
SB17NAC35-8 (2 drivers run in parallel from a single crossover to appear as 4 ohms and increase power handling and sensitivity)
50 litres, 2 ports, 2000 hz / 2nd-order Linkwitz-Riley crossover
Ceiling speakers (in enclosures angled at 45 degrees down directed at my seat):
SB26ADC-C000-4 (recessed 18mm to time align to woofer)
SB17NAC35-4
25 litres, 1 port, 2000 hz / 2nd-order Linkwitz-Riley crossover
Reasoning:
I like the idea of these drivers since they reportedly have low distortion and a flat response curve at a good price and don’t seem to break up as badly as other aluminium drivers. I live in the tropics (hot and very humid) so paper drivers are out of the question.
Ceiling speakers: The sensitivity of the 4 ohm tweeter and 4 ohm woofer are the same so no attenuation is necessary.
Front speakers: These need more power than 60 watts so I’ll use 2 woofers. The 8 ohm version of the woofer has sensitivity about 3.5 db less than the tweeter. By running two 8 ohm woofers in parallel from a single crossover, it will increase sensitivity by 3 db so no other attenuation is necessary. These should sound the same as the ceiling speakers but handle more power.
Enclosure volumes: The SB site gives the main values of SB17NAC35-4 as:
Diameter: 12.26 cm, Fs: 32 hz, Qts: 0.32, Vas: 33 litres – giving an enclosure of 19 litres (per driver)
www.audioexcite.com >> SB Acoustics SB17NAC35-4 tested the SB17NAC35-4 as:
Diameter: 12.26 cm, Fs: 37 hz, Qts: 0.42, Vas: 21 litres – giving an enclosure of 26 litres (per driver)
Questions:
Are these drivers the most accurate reproduction across the range at this price and if not, which are significantly better?
Will these give me a big improvement over the 20 year old Vifa’s?
Will this go low enough to seamlessly meet the subwoofer at approx. 90 hz or should I go for a 3 way system?
Are the enclosure volumes okay? I’ll use an online calculator for the ports. I can stuff the speakers with solid blocks to reduce volume if it is too big.
Is a 2000 hz / 2nd-order Linkwitz-Riley crossover the best option?
Is using two 8 ohm drivers in parallel the best option for the mains and will it sound the same as a single 4 ohm in half the enclosure volume?
Is setting the tweeter back 18mm necessary to time align to the woofers (I read that it should be 18mm for a similar combination Yet another 6+1 construction: SB17MFC35-8 + SB26ADC-C000-4 )?
Am I supposed to invert the tweeter connections on the crossover to make sure phase is correct or this accounted for by the 18mm offset?
Any other advice? - Thanks!
Attachments
Last edited:
This may be helpful: SBAcoustics-61-NAC - but the actual xo details are not included and that's really the hardest part.
Awesome theater space! I have no experience with the new drivers you are proposing, but your existing drivers are respectable. I wouldn't be surprised if you found the new speakers very similar to the old ones, and perhaps worse with the challenge of properly designing a crossover. The xo freq and details will have to be simulated before ordering any parts. If I were you, I would set out to build a center channel with some controlled directivity (horn, waveguide, short line array). The improvement in hearing the voices clearly is immense, plus you can improve dynamic capabilities. Then if you like the sound, you can build matching LR. If not, you don't have much cash into the project and you've learned something. I would not try to use the same drivers in the ceiling and surrounds, theoretically ideal but the advantage of directivity up front outweighs that imo. I have very capable theater speakers, but after living with horns in my 2 channel room, I am planning something like this for my theater. For atmos in the ceiling, I think it would be difficult to improve greatly on a decent pre-made coax at the end of the day. There's not a lot of sound coming from them in my theater that I think I'd notice a huge difference. I could be wrong.
+1 on using the many decades long proven co-ax ceiling PA systems, so FWIW I'd use the '60s era Altec/EV or similar 'drop-in' speaker: http://av.loyola.com/products/audio/pdf/ev-409.pdf
GM
GM
Last edited:
Thanks for the advice. But am I at the right site? DIY'ers advocating buying pre-made coaxial speakers? That just seems wrong. I'm going to persist with my DIY plan to expand to an 11.2.4 system but I will start by just building a single enclosure in case it doesn't work as good as my current speakers. I like the DIY approach and tend to do it with a lot of things and take pride in knowing "I made that".
Yes, I'd seen the SBAcoustics-61-NAC but as you say, without the crossover components it isn't super helpful. It is interesting to see the measured speaker parameters which are different again, with the 16 litre enclosure used is lower than either of the other calculations. I suppose I can always stuff some blocks into my enclosures if necessary.
I've been playing around with the Xsim program but without actual .frd and .zma files, I've had to try to make my own from the response and impedance charts. I've managed to simulate quite a flat response using a 2nd order crossover (red line on the attached image). It crosses at 1800 hz. Is this too low? I don't plan on very high volumes but don't want to burn out the tweeters.
Does anyone have these files or know where I can get the .frd and .zma files for these drivers? The Anthem receiver comes with a microphone for the room correction. Is there a piece of (free) software around that I can use to make my own files once I buy the drivers?
Yes, I'd seen the SBAcoustics-61-NAC but as you say, without the crossover components it isn't super helpful. It is interesting to see the measured speaker parameters which are different again, with the 16 litre enclosure used is lower than either of the other calculations. I suppose I can always stuff some blocks into my enclosures if necessary.
I've been playing around with the Xsim program but without actual .frd and .zma files, I've had to try to make my own from the response and impedance charts. I've managed to simulate quite a flat response using a 2nd order crossover (red line on the attached image). It crosses at 1800 hz. Is this too low? I don't plan on very high volumes but don't want to burn out the tweeters.
Does anyone have these files or know where I can get the .frd and .zma files for these drivers? The Anthem receiver comes with a microphone for the room correction. Is there a piece of (free) software around that I can use to make my own files once I buy the drivers?
Attachments
I'm also in process of building SB 61-NAC.
I've read carefully the Troels Gravesen design on his website. He has provided pics of the frequency response of both drivers measured in speaker boxes.
You can trace them and create the .frd files.
The crossover
I added the frd files in VituixCAD
Don't forget to offset the tweeter by -19mm as Troels recommended if you going to follow his design.
Take a look carefully the crossover photos on the same page.
L3 coil - 13 layers x 17 winds ~= 1mH (Jentzen 000-0018)
L5 coil is labelled as Jentzen 000-0009 which is 0.39mH
L1 coil is labelled with marker ...15. I suspect it is unwinded from a coil with greater inductance.
Having the inductances is a good starting point to unveil the whole crossover schematic.
The tweeter network contains two caps one of them larger than the other. The smaller one should be 10 or 12uF, as the larger is 15, 18 or 22uF according Jentzen specsheet. I played with the different values and C1=12uF and C7=18uF gave me a closer match to desired slope.
In the woofer network capacitance includes multiple caps. The best result I got with values of C8~16-17uF and C3~13uF.
The resistors values are R1 (tweeter attenuation)- 1.5 to 2.7 and R4 is around 1 ohm. For R4 I used R1.2 and it works well.
Troels has provided very detailed information about his design even w/o the component values. Seems that everything you needed has been in the front of your eyes
Hope that info was helpful for you.
Greetings
I've read carefully the Troels Gravesen design on his website. He has provided pics of the frequency response of both drivers measured in speaker boxes.
You can trace them and create the .frd files.
The crossover
I added the frd files in VituixCAD
Don't forget to offset the tweeter by -19mm as Troels recommended if you going to follow his design.
Take a look carefully the crossover photos on the same page.
L3 coil - 13 layers x 17 winds ~= 1mH (Jentzen 000-0018)
L5 coil is labelled as Jentzen 000-0009 which is 0.39mH
L1 coil is labelled with marker ...15. I suspect it is unwinded from a coil with greater inductance.
Having the inductances is a good starting point to unveil the whole crossover schematic.
The tweeter network contains two caps one of them larger than the other. The smaller one should be 10 or 12uF, as the larger is 15, 18 or 22uF according Jentzen specsheet. I played with the different values and C1=12uF and C7=18uF gave me a closer match to desired slope.
In the woofer network capacitance includes multiple caps. The best result I got with values of C8~16-17uF and C3~13uF.
The resistors values are R1 (tweeter attenuation)- 1.5 to 2.7 and R4 is around 1 ohm. For R4 I used R1.2 and it works well.
Troels has provided very detailed information about his design even w/o the component values. Seems that everything you needed has been in the front of your eyes
Hope that info was helpful for you.
Greetings
Last edited:
You're welcome! Yes, but having long experience in cinema, HT systems since '64 with DIY Altec A7-500 small cinema horns, speaker design, this prosound coaxial speaker and not having a clue about your choices, I'd be surprised if you could surpass it except in the extreme HF, which film sound tracks don't have.
Regardless, best of luck and curious if you solved your CC speaker issues.
Nice room
If your centre speaker doesn't match your FR & FL then perhaps you should start there.
Get those right and working well together might be your biggest improvement. We went from 5.1 to 7.2 and apart from much better explosions we didn't hear much benefit, some but not much. Your room would most probably benefit more but the most important speakers are those front three
If your centre speaker doesn't match your FR & FL then perhaps you should start there.
Get those right and working well together might be your biggest improvement. We went from 5.1 to 7.2 and apart from much better explosions we didn't hear much benefit, some but not much. Your room would most probably benefit more but the most important speakers are those front three