This is the freq. resp. of the fostex fe126e. I plan to use them in small closed enclosures down to about 250 Hz. I would like to use active filters around a chip amp to equalise the rising response, but I am not sure which would be the best way to go (i.e. if shelving up, down, or using other type of filtes (notch, etc).
Thanks for your help.
Sebastian
Thanks for your help.
Sebastian
Attachments
Have you been listening to them? It looks like they might sound pretty good with some toe-in. They look to be fairly flat off axis. This would certainly be the simiplest, least expensive method. Other than that I don't have any experience with active filters, but if it was a passive filter a simple inductor in parallel with a resistor would do the trick.
Joe
Joe
If you can trust the posted response graph, I advise a low Q notch filter centered about 6000 Hz. The main component of the rise begins about 2 octaves below and ends one octave above that frequency.
Looking at the off-axis response I see things that suggests the on-axis response includes a dose of wishful thinking.
Good designing and good building,
Mark
Looking at the off-axis response I see things that suggests the on-axis response includes a dose of wishful thinking.
Good designing and good building,
Mark
Thanks for your replyes.
I haven't measured the speakers yet; this is taken from the spec sheet from Fostex. I do not know how much they represent the true drivers response.
Mark: The notch filter was one of my alternatives, but I thought it was too wide to be taken care of that way. Would you suggest also raising the response after 15K? What Q would you suggest for the notch?
Thanks, Sebastian
I haven't measured the speakers yet; this is taken from the spec sheet from Fostex. I do not know how much they represent the true drivers response.
Mark: The notch filter was one of my alternatives, but I thought it was too wide to be taken care of that way. Would you suggest also raising the response after 15K? What Q would you suggest for the notch?
Thanks, Sebastian
Sebastian,
Honestly, my advice for the notch filter is to go speaker level passive and not active. This allows you to play with the complex impendance of the driver to shape the notch. If you go active, you do not have this option. Keep in mind that with active filter design bandwidth and depth of notch are somewhat independent and you can play with one or the other.
Without a design quality test result (instead of sale quality) I cannot make statements about appropriate filter Q. I fear the 6500 Hz peak may be more of a peak than it appears. I am also a little leary about what is contributing to the high end response of the driver. If it is resonance decay as I suspect, then equalizing the top end response will just give you more of a bad thing.
Fortunately, above 12 to 15 kHz, there are no musical tones and we do not hear resonance decay as easily as when it is altering a musical tone.
Lastly, you can also combine filters. Concentrate on the 6.5 kHz peak. Listen to the driver with only that controlled. You will likely be amazed at how much the sound is improved just by controlling that one cone vibration mode. Then you can add a bandpass filter to lower the rising frequency response. When you have controlled for those two problems, the highest frequencies will not be as far down compared to the midrange level and may be acceptable. If not, you can always bump the extreme top end.
Best of luck in your design work,
Mark
Honestly, my advice for the notch filter is to go speaker level passive and not active. This allows you to play with the complex impendance of the driver to shape the notch. If you go active, you do not have this option. Keep in mind that with active filter design bandwidth and depth of notch are somewhat independent and you can play with one or the other.
Without a design quality test result (instead of sale quality) I cannot make statements about appropriate filter Q. I fear the 6500 Hz peak may be more of a peak than it appears. I am also a little leary about what is contributing to the high end response of the driver. If it is resonance decay as I suspect, then equalizing the top end response will just give you more of a bad thing.
Fortunately, above 12 to 15 kHz, there are no musical tones and we do not hear resonance decay as easily as when it is altering a musical tone.
Lastly, you can also combine filters. Concentrate on the 6.5 kHz peak. Listen to the driver with only that controlled. You will likely be amazed at how much the sound is improved just by controlling that one cone vibration mode. Then you can add a bandpass filter to lower the rising frequency response. When you have controlled for those two problems, the highest frequencies will not be as far down compared to the midrange level and may be acceptable. If not, you can always bump the extreme top end.
Best of luck in your design work,
Mark
I think you want to listen to them before you get excited. Lots
of the smaller Fostex drivers have a fine top end without EQ
if you are a bit off axis, and some even directly on axis. You
can build a simple gentle RLC notch filter into the input to your
power amp, or you can do the same in series with the speaker,
nothing that some patience and experimentation won't get you.
of the smaller Fostex drivers have a fine top end without EQ
if you are a bit off axis, and some even directly on axis. You
can build a simple gentle RLC notch filter into the input to your
power amp, or you can do the same in series with the speaker,
nothing that some patience and experimentation won't get you.
Nelson: Thanks for answering! I already have the drivers built on small 10.5L BR "test" enclosures (as suggested by fostex) and have been hearing them for about two weeks. I must say I like them quite well. The upper and HF response seems nice and extended but they are a little "shouty". Now, I know this kind of enclosure is not ideal for the low Qts (0,25) of the driver, and surely on the suggested BH/BR or augmented with a sub they would sound much more balanced. But nevertheless I think some sort of EQ on could make their upper response even better. As I said, I intend to build sealed enclosures for them and then XO them actively at around 250-300Hz to H-frame dipoles. Now that I have the chance to ask: which of the smaller Fostex have you auditioned? How would you rate them against the fe166e? In the last moment I went for the 126, but the 166 had always been on my mind. The 126 has a higher sensibility that all other "smaller" fostex, high BL, low mass, and costs about 2/3 the price of the 166. I would be happy to try the 166, even more if you rank it above the 126.
Mark: Thanks for your helpful comments. I think I would prefer to keep the filters active as long as I can. I understand perfectly your worries about the resonance decay. I`ll try to come out with an active notch filter and then have a listen. My first try will be centered around 6.5K from 1.8k to 12.5k. I am also looking forward to buy a Panasonic Mic and make some measurements myself.
Sebastian
Mark: Thanks for your helpful comments. I think I would prefer to keep the filters active as long as I can. I understand perfectly your worries about the resonance decay. I`ll try to come out with an active notch filter and then have a listen. My first try will be centered around 6.5K from 1.8k to 12.5k. I am also looking forward to buy a Panasonic Mic and make some measurements myself.
Sebastian
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.