It would be interesting to see a sweep done with the mic off axis. I've noticed the super 12 is very beamy in some ranges.
I was considering either three filters (BSC, notch from 1.75-3, notch from 10-15) or just an Eq... I'm really interested in what you come up with for a passive filter.
Thank you very much this research.
Hi netbug
You are absolutely right. The Super 12 is very beamy off axis. This is one speaker you have to listen to head on. No toeing in or out. It will not tolerate even 15 degrees off axis.
Once I get the speaker corrected, I'll do some off-axis SPL for you. I'm sure the beaming will still be there. This is structural, there's nothing I can do about it.
It'll take me a while to smoothen the FR. I have to do it one step at a time. First is to zobel and install a BSC to bring the levels down. Probably about 3-6dB. After that, I'll work on the "hot spots" starting with the most severe ones. I think the Super 12 can sound pretty nice once corrected. It does have some good qualities.
You are absolutely right. The Super 12 is very beamy off axis. This is one speaker you have to listen to head on. No toeing in or out. It will not tolerate even 15 degrees off axis.
Once I get the speaker corrected, I'll do some off-axis SPL for you. I'm sure the beaming will still be there. This is structural, there's nothing I can do about it.
It'll take me a while to smoothen the FR. I have to do it one step at a time. First is to zobel and install a BSC to bring the levels down. Probably about 3-6dB. After that, I'll work on the "hot spots" starting with the most severe ones. I think the Super 12 can sound pretty nice once corrected. It does have some good qualities.
That is a little bit close to be in the farfiled of a 12" driver. Can you take one further out.
dave
Thankyou RJBOND for let me know the BVR boxes, seems they are a kind of simpler horn systems, very beautiful work of FrugalHorn guys.
Interesting stuff, may be usefull in the future.
Hi Mr.Michael Chua,
I sorry to read the Super12CF is a bad sound full range, I like too much big cones FR.
I wonder if you are using a bare wood(no stuffing) inside this sealed box for testing or fi you are using some kind of stuffing??
Do you think there is some kind of stuff that can decrease the Super12CF hi SPL response in the Mid and HIs??
Gustavo
Last edited:
Hi Gustavo
Yes, I do agree with you. A 12" is nice. A 12" Full Range is even better. Trouble is, its not easy to make a good one. A really good 10-12" FR is very costly. The Lowther is a good example.
The Super12 is much cheaper but very "rough". It is not the box, it is the speaker itself. My box is lined with R11 fiberglass wool. I first heard the Super12 without any wool inside. As expected, reproduction from vocals upwards is very bad. Then I added the wool. Cleared up a lot, but very beamy. Checked it with LMS and true enough, the tweeter needs to be looked into.
If nothing is done to correct the tweeter response, its simply not listenable. It may be acceptable to others but certainly not for me. The easiest way to resolve this problem is to use a 31-band EQ with a spectrum analyzer to even out the peaks. In professional work, this is normal. In Hi-Fi, some will not accept any electronics in the speakers. For myself, the most is an electronic crossover to a subwoofer. I do not use the electronic crossover High Pass for the Main speakers. Instead, I do it acoustically or with a passive network. So, the only option open is to try to EQ the speaker passively. I believe it can sound quite good after correction.
Yes, I do agree with you. A 12" is nice. A 12" Full Range is even better. Trouble is, its not easy to make a good one. A really good 10-12" FR is very costly. The Lowther is a good example.
The Super12 is much cheaper but very "rough". It is not the box, it is the speaker itself. My box is lined with R11 fiberglass wool. I first heard the Super12 without any wool inside. As expected, reproduction from vocals upwards is very bad. Then I added the wool. Cleared up a lot, but very beamy. Checked it with LMS and true enough, the tweeter needs to be looked into.
If nothing is done to correct the tweeter response, its simply not listenable. It may be acceptable to others but certainly not for me. The easiest way to resolve this problem is to use a 31-band EQ with a spectrum analyzer to even out the peaks. In professional work, this is normal. In Hi-Fi, some will not accept any electronics in the speakers. For myself, the most is an electronic crossover to a subwoofer. I do not use the electronic crossover High Pass for the Main speakers. Instead, I do it acoustically or with a passive network. So, the only option open is to try to EQ the speaker passively. I believe it can sound quite good after correction.
Hi dave
I've taken another sweep from 1M. This will the be distance I will be working on. The reason I took at 0.5M earlier was for the HF response. I wanted to see how much difference from 0.5M to 1M. Apparently, virtually identical.
The upper plot is smoothed 1/6dB. Lower plot is the actual measurement. I scaled the plot -12dB for easier viewing.
The sensitivity of the Super12 is actually slightly higher. I had to reduce the power from 1W to 0.3W. Firstly, its too loud, even with ear plugs on. Also, in my earlier sweep at 0.5M_1W, the tweeter distorted badly. Can hear it cracking from 2K-6K. In the final results, adding back 3-4dB will be quite close to 1W.
I've taken another sweep from 1M. This will the be distance I will be working on. The reason I took at 0.5M earlier was for the HF response. I wanted to see how much difference from 0.5M to 1M. Apparently, virtually identical.
The upper plot is smoothed 1/6dB. Lower plot is the actual measurement. I scaled the plot -12dB for easier viewing.
The sensitivity of the Super12 is actually slightly higher. I had to reduce the power from 1W to 0.3W. Firstly, its too loud, even with ear plugs on. Also, in my earlier sweep at 0.5M_1W, the tweeter distorted badly. Can hear it cracking from 2K-6K. In the final results, adding back 3-4dB will be quite close to 1W.
Attachments
Quasi Anechoic (1K-20K)
The left plot is a Gated response of the Super12 from 1K-20K. This would be what the HF would look like in an anechoic chamber.
We can see there is a major peak at 2K5. Next peak at about 5K4. A smaller one at 3K8 and lastly, Mount Everest at 14K. The one that is the most objectionable is the 2K5. This is where our ears are most sensitive. Vocals will screech with peak like this.
The adjacent display is the Gated Sweep superimposed with the Normal (ungated) sweep. Comparing the two, we can clearly identify how much the room has on the actual response.
All present and future measurements will be at 0.3W at 1M.
The left plot is a Gated response of the Super12 from 1K-20K. This would be what the HF would look like in an anechoic chamber.
We can see there is a major peak at 2K5. Next peak at about 5K4. A smaller one at 3K8 and lastly, Mount Everest at 14K. The one that is the most objectionable is the 2K5. This is where our ears are most sensitive. Vocals will screech with peak like this.
The adjacent display is the Gated Sweep superimposed with the Normal (ungated) sweep. Comparing the two, we can clearly identify how much the room has on the actual response.
All present and future measurements will be at 0.3W at 1M.
Attachments
Hi Mr.Chua,
Many Thanks for your honest opinion, you save me a lot of money and time, here in Brazil there is 100% of taxes over the raw price(product+ship+insurance) at the Custon House for imported electronics. I look forward for your future corretive Super12CF tests in a passive way(stuffing), maybe it could have a good sound after it.
I read good comments about the Nirvana SUPER15CF(unconfirmed), it fit to my taste too even with the treble limit at 14/15khz, I have a Beyma 5MP60/N, 5 inches fullrange specified at only 12Khz and the treble performance is good.
http://profesional.beyma.com/ingles/pdf/5MP60N.pdf
I could use two Super15CF units in parallel for 4ohms and 98dB SPL. Do you think it is a good fullrange(direct, no filtering) for OB or a big BassReflex(two units in a 447 Litres BassReflex box)??
Thankyou for your input, Gustavo
Last edited:
Step 1 - Impedance EQ
Impedance sweep at 2Vrms with Raw Super12 and with Zobel network in place. Zobel is manually tuned for flatest response in the critical bandwidth (1K-10K). From 1K-6K, the Super12 is basically resistive. Any tuned networks that I apply now will conform to know physics.
Impedance sweep at 2Vrms with Raw Super12 and with Zobel network in place. Zobel is manually tuned for flatest response in the critical bandwidth (1K-10K). From 1K-6K, the Super12 is basically resistive. Any tuned networks that I apply now will conform to know physics.
Attachments
Aha! FRM I have been looking at that Beyma driver for some time now. It looks like quite the little beast and the provided plot looks decent. Could I have your impressions of it plz?
Edit: Whoops, I should have used your link first. I was referring to the neodymium 5".
Edit: Whoops, I should have used your link first. I was referring to the neodymium 5".
HI Inclined Plane,Aha! FRM I have been looking at that Beyma driver for some time now. It looks like quite the little beast and the provided plot looks decent. Could I have your impressions of it plz?
Edit: Whoops, I should have used your link first. I was referring to the neodymium 5".
The Beyma 5MP60/N is great performance for the price, just 160reais today(90 USdollars each+ship) it is ferrite magnet, the Beyma quality control was poor, cause I receive it with the rubber suspension bad glued in 3 position around the basket and with a voice coil loose.
I pay return ship to the brazilian distributor in São Paulo and ask a new driver, but I receive the same driver just with the VC fixed and the suspension in the same condition(untouched), I ask a report about the defect but never receive a word.
Iam using it as oneway in a cylinder, automotive type sealed box(7mm plywood) 16 litres, just with the usual 20mm white poliester blanket, so the bass is good, well defined and firm, just a bit small, but If I put some foam in the botton of the box the bass will be excelent.
The Mid and Hi are good too, musical, warm, but not as Transparent as a hi quality paper come FR. They are transparent but in the Polipropilene language.
I run it direct, no filter of any kind, it do not need correction. The power handling is very good, it do not ring in Hi SPL, sometimes I do not believe it have just 50Wrms. I recommend it, maybe buy it again If Beyma can supply it in a 16 ohms version.
Regards, Gustavo
Last edited:
Gustavo
I strongly advise against 2 Super15 in one box. The main advantage of a FR is a point source. If you have 2 FR to increase SPL, they will interfere with each other. The end result is you will lose clarity.
I do not know because I never tested a Super15CF. But judging from the Super12CF, I expect the 15 to have the same problems.
Michael,
Pardon my ignorance, are these values appropriate for a sealed box (if so, what volume) or do they apply generally?
Cheers,
Ryan
Hi Ryan
The Zobel network is unaffected by the type of box (sealed, bass reflex, etc). The Volume of the box has no effect too.
What affects the Zobel are the speaker's inductance and voice coil resistance. Due to its inductive nature, a speaker is not resistive throughout the audio bandwidth. As the frequencies go higher, the resistance increases accordingly, hence the term "impedance". The net result is an upward tilt towards 20kHz. This makes it quite impossible to apply any networks with certainty.
The Zobel is wonderful. Only a resistor and a capacitor but with the right values, it converts a reactive device to resistive. Once its resistive, the speaker will behave as predicted.
The Zobel network is unaffected by the type of box (sealed, bass reflex, etc). The Volume of the box has no effect too.
What affects the Zobel are the speaker's inductance and voice coil resistance. Due to its inductive nature, a speaker is not resistive throughout the audio bandwidth. As the frequencies go higher, the resistance increases accordingly, hence the term "impedance". The net result is an upward tilt towards 20kHz. This makes it quite impossible to apply any networks with certainty.
The Zobel is wonderful. Only a resistor and a capacitor but with the right values, it converts a reactive device to resistive. Once its resistive, the speaker will behave as predicted.
- Status
- Not open for further replies.