The concept of the speaer requires that the mid is filtered as it's lower crossover should be in the 250-500Hz range. The woofer needs to play the higher bass/mow mid mostly for being a classic monkey coffin. Otherwise you could just build bookshelfs with a pair of subwoofers. But that is done already a lot down here and not the goal i thought.
Here is shopping cart for Gema speakers:
Looks good to me. Only problem i see with Gema is that there is not a single measurement of it.
Looks good to me. Only problem i see with Gema is that there is not a single measurement of it.
I heared (no confirmed info) that the crossover is not really good, but that the box design allows for a much better crossover that makes it a great speaker. But the standard kit has crossover faults. Never saw measurements to confirm that altough.
If i would build it, i would change the tweeter to the SB26ADC that I like a lot more than that SB26STAC and i would redo the crossover on my own measurements But the rest of the kit looks good on the first sight.
But for this project, it's over budget if we are strickt, so we choose another woofer i thouth (the SLS12)
If i would build it, i would change the tweeter to the SB26ADC that I like a lot more than that SB26STAC and i would redo the crossover on my own measurements But the rest of the kit looks good on the first sight.
But for this project, it's over budget if we are strickt, so we choose another woofer i thouth (the SLS12)
@A4eaudio Yes, a HP filter is absolutely needed on a mid in a 3 way. Maybe some people got confused when I said you'd only need a 1st order electrical filter on a mid which is operating in a sealed enclosure that rolls it off right at the intended xover. You're just adding to the natural 12dB/oct rolloff of the sealed enclosure, so you end up with a 3rd order filter with a single cap and a 4th order with a cap and coil. Of course these are simplistic analogies, so the actual rollof rate would be governed by other things like BSL and resonance peak affecting the passive network response.
I was on the Peerless site and it stated they were discontinuing the NE149W-04, so that one is definitely off the table. The NE123W-04 is honestly the best mid for the price. It has the SS Revalator sound for a fraction of the price. The SB12MNRX2 is the only other one I can recommend for several reasons, especially the price, performance and build quality.
I was on the Peerless site and it stated they were discontinuing the NE149W-04, so that one is definitely off the table. The NE123W-04 is honestly the best mid for the price. It has the SS Revalator sound for a fraction of the price. The SB12MNRX2 is the only other one I can recommend for several reasons, especially the price, performance and build quality.
The SB26ADC barely has enough efficiency to keep up with the other drivers, unless you want to voice the speaker for the Harman shaped curve. The SB26STAC is an excellent tweeter, but its a soft dome, so it will sound a little more tame, but its sensitive enough for some flexibility and won't go up in smoke if pushed hard with a padding network.
@Zvu That 3k peak is going to be an issue with a passive xover. It will be hard to make inaudible. The decay time is long enough on the CSD to be noticed, especially at higher SPL. Yes, the HD is low, but the aluminum dish diaphragm with its hard 1st radial mode breakup peak will be a challenge for use with a passive network.
@waxx The SB Gema is a 3rd party design, as stated on SB's site. Thats probably why its not as good as it could be. Large 3 way speakers aren't a one size fits all design in terms of rhe crossover. There are alot of variables which can make it sound considerably different when placed in various location in a room. Most larger speakers don't sound that great placed closer to a boundary. The lower mids become too overbearing.
I know. I saw that midwoofer mentioned earlier as possible midrange so i posted measurements. The distortion is not particularly low imo. I wouldn't use it above 500-600Hz or so.
Sorry everyone, there was supposed to be an image attached. The good news is I DO know the mid needs a high pass! 🙂
So, @profiguy , I was basically asking how (more precisely) do you determine the enclosure and high-pass filter to achieve your goal that you were stating about the Fs peak? The enclosure choice of 0.08cf was just to give the curve below where 62Hz is still within the natural roll-off in the image as a reference, in case this helps you explain your point. Then the question is, is this image helpful, or do I need to add a filter to the curve in the image, for you to make your point? (Knowing, that we WILL need to add a filter when we actually design the speaker.)
We can discuss chamber size later, once we decide the drivers. I too prefer larger but I think that will benefit greatly from a group discussion, especially the technical merits.
Note, a 15L enclosure for the SB12MNRX would give a Q of 0.379. Usually getting 0.5 requires too large of an enclosure and 0.70 nis more reasonable. But, for the SB12 Q=0.5 requires 3.8L and Q=0.7 requires 1.25L
It is hard to tell, but based on some stuff I find on the internet, it is only the 4-ohm version discontinued. However, I don't even see the 8-ohm version on the product list (active or discontinued). The sensitivity shown for the 4-ohm is 85.6dB but the sensitivity claimed for the 8 ohm is 86dB, although several of us seem to have this driver and can actually measure it.
Regardless though, I think we will select something else due to price.
@A4eaudio I have some NE149W-08 which are about 87 dB/2.8V but at 2 pi that changes a little on the bottom. Realistically, you gain a little with staggered filters in the bandpass and also some filter Q if its over 0.7 Qtc. Staggering HP with the rolloff Q (if over 0.5) you get some extra peaking depending on where you place the knee point of the HP. Butterworth filters add to a higher Qtc if one of the overlapping filters is higher. This can be softened with a smaller cap but it will interact with the electrical HP.
@A4eaudio To simplify the answer to your filter question with the chamber - essentially all you're doing is placing the electric HP filter knee point on top of the chamber rolloff knee point, so they add up to a steeper slope. This places the resonance slightly below the actual HP filter frequency knee (-3 to -6 dB down, depending on LR or BW), so everything below the knee is in the impedance peak. The higher impedance translates into less current from the amp, ergo less power dissipated by the VC. The extra output can come from the staggered filters (electrical and mechanical) when they peak from a Q higher than 0.7.
That means, if you opt for a target HP of 350 hz, you size the enclosure to roll the driver off mechanically -3 to -6 dB down at 350 hz. The electrical HP is then set at 350 hz as well, in order for for both xover HP and enclosure rolloff to coincide. I hope that makes sense.
That means, if you opt for a target HP of 350 hz, you size the enclosure to roll the driver off mechanically -3 to -6 dB down at 350 hz. The electrical HP is then set at 350 hz as well, in order for for both xover HP and enclosure rolloff to coincide. I hope that makes sense.
Is the 8 ohms version of the NE149W with 88/2.83V could be enough. It can be sourced, at least at Ali, I bougth mine at Soundimports two 2 years ago.
@waxx Also why we should choose high cutr off when one say mid filter works better at less than 300/350 hz and have 2.7 to 3 decade before the tweeter !
We cope the look of Classic 3 Ways not the default theey made in the 60s' ?! Better to keep the mid low pass the nearest to the Schroeder room frequency, no ?
And 300 hz is a big classic if you read Dickason
@waxx Also why we should choose high cutr off when one say mid filter works better at less than 300/350 hz and have 2.7 to 3 decade before the tweeter !
We cope the look of Classic 3 Ways not the default theey made in the 60s' ?! Better to keep the mid low pass the nearest to the Schroeder room frequency, no ?
And 300 hz is a big classic if you read Dickason
I mean more or less the same as you, just otherwise said. The woofer needs to be arround schroeder frequency of the room idealy indeed. That was the big trick of those monkey boxes, a big woofer below schroeder, and smaller mid(bas) and tweeter to cover the rest. It's something that works well in reality and that is why this type of speaker is still so popular.
Yup that's why I highligthed a 3 or a 4" will not go low enough for that and it ask like in the Gema a clean 12" there, so the more expensive L version of the SB34RNX.
I am also intrigued by the cone wooven glass fiber of SCANSPEAK 15M4624G with sensivity 92 dB and being a pure mid ! >= 220 hez needed and more in the 80 euros, certainly more in US. Used by J Gherard from Suesskind Audio ! It is a 5" true mid which is rare. foam surround, made for details. 220 Hz emectrical should hit 300 hz acoustical ?
Also some like the Emminence alpha or beta 6 which is cheap almost everywhere ?
SB15Nbac is said good but breakups rise the cost of the filter ! the true bargain is maybe the FSL used in the KII 20 euors peereless , but doesn't if the not so flat behavior while noo severe break up (relativly flat within 2 or 3 dB).
I though also to the Hivi à la Dynaudio like drivers, but their efficienty is very low ! I heard the 2 ways Totem Forest times ago and liked its sound !
I am also intrigued by the cone wooven glass fiber of SCANSPEAK 15M4624G with sensivity 92 dB and being a pure mid ! >= 220 hez needed and more in the 80 euros, certainly more in US. Used by J Gherard from Suesskind Audio ! It is a 5" true mid which is rare. foam surround, made for details. 220 Hz emectrical should hit 300 hz acoustical ?
Also some like the Emminence alpha or beta 6 which is cheap almost everywhere ?
SB15Nbac is said good but breakups rise the cost of the filter ! the true bargain is maybe the FSL used in the KII 20 euors peereless , but doesn't if the not so flat behavior while noo severe break up (relativly flat within 2 or 3 dB).
I though also to the Hivi à la Dynaudio like drivers, but their efficienty is very low ! I heard the 2 ways Totem Forest times ago and liked its sound !
Last edited:
Member
Joined 2003
Here's a comparison I completed some time ago. Wavecor WF120, good 4" midbass, not exactly a budget driver. I tested with and without a 300Hz LR-12 high pass. Comparison of fundamental response looks like this:
...and comparison of total distortion produce from the driver looks like this:
If that doesn't show the reduction in modulated components by high passing a midrange driver, I don't know what does. I suppose you could always listen to it... Attached MDAT file, open in latest REW beta, go to distortion tab - controls - play FSAF residual.