During all the discussion that happened with your choice of FE108e∑ you did not glom onto the 1 KHz dip. Never bothered me.
Did you listen or are you judging only on yhe measure?
dave
FF105wk is a nice driver. Soe solid mods can be done to improve them.
FE108e∑ stuck around for so long because people like how it sounds. It has a romantic midrange and i expect the dip is part of that.
dave
Last edited:
I did gloom on the 1khz dip
Ill trust the measurements next time, no prob Dave. I really dont blame you. I only have myself to blame. This is a great experience situation for me. Next time, ill seek drivers that measure good enuff
Attachments
I'm not suggesting you should...
However if you crossed at 1200Hz LR2, you could incorporate a filter bump at 1kHz to EQ the Fostex. Here I show an axial plot over the ideal curve, and the next plot shows the filter response with a peak at 1kHz.
Interesting
i could also xo at 1khz, first order. That could help fill the dip.
as i have the drivers, id love to end up with a good sounding speaker.
i could also xo at 1khz, first order. That could help fill the dip.
as i have the drivers, id love to end up with a good sounding speaker.
Ah very clever. Question: what if the LR2 HPF is paralleled by a gentler LPF to reclaim below ~600hz? Like the parallel notch filter (first-order LC) but with a second-order pinch. Might it work?
My 'ez just came, new-in-box so response going to be very preliminary. I propped one up ~20cm over desktop facing up and listened standing right over it, tone-sweep and contralto. HF voice shrill on top (brand new). But otherwise the dip 800-1000 (max 770) was barely discernable then a quick rise into HF shelf and stayed there not particularly uneven (narrow valley 9700, recovered then dropped ~12k). So I think I got the "desired" boost to neutralize the drop, and a flat-ish step function above ~200hz (for cab and BSC to handle). So perhaps the half-OB not-BR front-vented cab ~20cm-deep (like Diatone 610). OP can make a test box with holes on the baffle. Have to run....
(Contralto voice better than I expected when played low-volume -- loudness compensation built-in for small homes.)
p.s. the other unit showed clearer valley at 910 but not deep, then very sharp rise 1100-2000; small bump 6k-something and wider mount around 8k. Maybe amp and ear warmed up a little more sensitive....
Last edited:
I think you're asking if the response can be flattened around the dip. Yes it could. It won't be as simple to do as you suggest. Since other factors should be taken into account for it to be worthwhile, I thought I'd look at the baffle step.
I don't usually like to overstate baffle responses because they should be taken in the bigger picture, over a range of angles. However it's interesting that a baffle can cause a peak right where you're looking for one. Might be worth experimenting with...
I don't usually like to overstate baffle responses because they should be taken in the bigger picture, over a range of angles. However it's interesting that a baffle can cause a peak right where you're looking for one. Might be worth experimenting with...
Quickly looked at the accompanying paper chart (after making above post) -- remarkably consistent with what I had heard given lack of baffle and 2*20cm bounce etc.
black = chart accompanying driver (anechoic 1m)
green = naked driver without box/baffle 2ft from wall
red = open-back small paper box 2ft from wall, parallel notch-filter (1.1mH || 2.7uF)
Frequency Sound Generator APP sine wave, measured by ear positioned for maximum sensitivity, and red curve adjusted for loudness curve 2-4khz.
Recommendation to OP: 'ez dipole/OB-U with notch-filter (or BSC) starting cut ~1.1khz, then reclaim HF above ~7.3khz.
green = naked driver without box/baffle 2ft from wall
red = open-back small paper box 2ft from wall, parallel notch-filter (1.1mH || 2.7uF)
Frequency Sound Generator APP sine wave, measured by ear positioned for maximum sensitivity, and red curve adjusted for loudness curve 2-4khz.
Recommendation to OP: 'ez dipole/OB-U with notch-filter (or BSC) starting cut ~1.1khz, then reclaim HF above ~7.3khz.
Attachments
Last edited:
Too preliminary. I did check for phase alignment at 1khz and (with one inverted) they happen to be near enough. This is with both 4" and 12" running straight-through, the guitar/bass driver polluting the upper mid-range so must be filtered somehow. Placed over the iron rice bowls they were very listenable, just no bass. I still think the 1k dip isn't the problem; the very steep rise to 2k is. A broad notch-filter/BSC (1.1mH || 2.7uF) is the best candidate for mid-to-high; but low still needs help from woofer or cab..
Tried the "eazy" route -- 5-knob 2.1 amp -- but spent ~3hr flattening the combined response to near-enough flat. Tonality and clarity are just fine... but I'm listening to opamps where did the magic go?
$10 TPA3116 2.1 amp: LO vol, LPF freq; HI vol, HI tone-control; main vol.
$30 (single, disadvantaged) vintage ferrite 12" fullrange driver over $10 steel can.
$10 TPA3116 2.1 amp: LO vol, LPF freq; HI vol, HI tone-control; main vol.
$30 (single, disadvantaged) vintage ferrite 12" fullrange driver over $10 steel can.
Attachments
I dunno
the 1khz dip is 8 db deep, the peak at 1.8 khz is about 2 db.
I truly am puzzled how im supposed to work with such a driver
maybe treating the cone would help?
the 1khz dip is 8 db deep, the peak at 1.8 khz is about 2 db.
I truly am puzzled how im supposed to work with such a driver
maybe treating the cone would help?