Lowering the tuning frequency will make it worse. The 1/4WL acoustic impedance peak is also frequency related.
so sorry - missed the "HT" part. A corner mounted 22Hz tuned 250 liter reflex doesn't look terribly shabby at ~120dB-20Hz with 150 watts input (excursion hits ~7.2mm around 33Hz) - it could be cool to make some sort of horn
Thanks Freddi, it seems like everybody keeps missing the HT part of the title, lol. It sounds like there's at the very least a lot of options for what to do, and I think they'll all produce satisfactory results. I think I'll have to grab at least one of these before they're gone, I'll update this thread along the way.
Hmm, how can the driver be positioned halfway outside the closed end of a pipe as the schematic implies? AFAIK, 1/2 the driver's width + a small amount = 24 cm min. offset each end.
Also, HR's 0.5 pi sims are pretty optimistic, with 1 pi for corner loading maybe a bit conservative in some cases, but short of mapping the room's acoustics, better overall to base designs on and best to use 2 pi [even 4 pi in some cases] to have plenty of tuning/tweaking options.
GM
how about a plain 6th order reflex? - here's 170 liters - either make/ adapt (EV Interface box) an underdamped 2nd order electrical highpass or fudge with parametric. DSP
smaller box can help with punch - woofer can take the boost at fb
that's a 2nd order HP cascaded with Q=2, gain = 6dB and F=22Hz filter in hornresp
smaller box can help with punch - woofer can take the boost at fb
that's a 2nd order HP cascaded with Q=2, gain = 6dB and F=22Hz filter in hornresp
That looks like it could work pretty well, assuming that curve remains constant at small-signal levels. I’m planning on buying something like a 200w mono amp module and using PLLXO to shape the curve as needed, that sound viable?
the amp probably has a boost/parametric function - dunno about the passive highpass - I made one in 1970😱
if the peak were set at the 23Hz tuning point, then 25Hz would be ~95dB and 20HZ ~93dB with that highpass setting (1 pi sim)
if the peak were set at the 23Hz tuning point, then 25Hz would be ~95dB and 20HZ ~93dB with that highpass setting (1 pi sim)
Last edited:
t-lines has a really neat tutorial on passive LL electronics, dunno if anyone’s posted results on the forum, though
meant to type "98dB" for the 170 liter 23Hz tuned reflex at 25Hz in 1 Pi with the highpass plus 23Hz tuned peaking filter - can't type (fingers are numb)
even tho qts is lower than recommended ~"0.312" for 6th order, I tjhink the lossy Le model says effective qts
acts higher than stated for larger coils
even tho qts is lower than recommended ~"0.312" for 6th order, I tjhink the lossy Le model says effective qts
acts higher than stated for larger coils
Last edited:
Hey Freddi, just saw the edit you made to the last post, good stuff there. I was concerned when I first looked at that rising response curve, but then I remembered I’m going to low-pass under 100 hZ anyway, probably 80 or so since we’re looking at a direct radiation scenario with the BR design. That’s about where I find I’m no longer sensitive to stereo imagining, so it’s my go-to for sub crossing. Who knows how it’ll actually behave in the real world, though, since we don’t have a published FRD available. Unless somebody happens to have one in their back pocket, that is.
its probably like some of the Eminence 18" speakers - I've used two spec of Eminence 18 in reflex (RJ vented) and Karlson
there were 4 RJ vented 18" reflex - the one you see ~6.4 cubic foot and the one below it, ~8.2 cubic feet
there were 4 RJ vented 18" reflex - the one you see ~6.4 cubic foot and the one below it, ~8.2 cubic feet
The driver has been ordered, so I'll need to decide on an alignment fairly soon. Going to be low-passed by a variable electronic preamp board, so I'll be able to mess around with that a fair bit to decide what works well for xover point.
I'm liking the 6th order curves for this, as well as the relative construction simplicity compared to a TH design, so I may end up going with that. I've got some 4" PVC drain pipe laying around from a previous TL project, would that suffice for the port diameter or would I want to double up on that?
I'm liking the 6th order curves for this, as well as the relative construction simplicity compared to a TH design, so I may end up going with that. I've got some 4" PVC drain pipe laying around from a previous TL project, would that suffice for the port diameter or would I want to double up on that?
Here's two 4" diameter ports - note the effect of the cascaded 2nd order highpass with 6dB 23Hz (Fb) peaking.
With EQ the input power right at tuning would be close to 500 watts for a nominal 150 watt midband drive.
IIRC your amp will be 200 watts so that's the real limit on port velocity.
my RJ18 above used 61.75 sq.in. of port area formed by the upper and lower duct spaced with 2"x2" cleats that terminated out that aperture.
With EQ the input power right at tuning would be close to 500 watts for a nominal 150 watt midband drive.
IIRC your amp will be 200 watts so that's the real limit on port velocity.
my RJ18 above used 61.75 sq.in. of port area formed by the upper and lower duct spaced with 2"x2" cleats that terminated out that aperture.
I haven't decided on a specific amp yet, but 200W is a ballpark goal. Looks like there shouldn't be too much port noise with that tuning and two 4" ports, and they don't end up being too long to fit in the cabinet. Plus, I'm a fan of the dual-port look so I think I'll go with that. Which leaves me with a 170L cabinet with two 14.5" long 4" ports. That's definitely manageable in terms of space use for the kind of output it models with. You mentioned multiple times the cascaded 2nd order highpass, and forgive my ignorance but that's not something I'm familiar with. Could you explain, or point me in the direction of something that does?
David McBean's hornresp has highpass and equalizer functions separate from each other for flexibility. There's two bands of equalizer independent from each other) (Download hornresp, run a sim then see those can do)
EQ for a 6th order vented box is generally done with a one op-amp highpass filter with 6dB peaking ("Q=2")
To get that characteristic in hornresp, I first set a second order highpass filter to 16Hz corner - then added 6dB/Q=2 peaking at 23Hz (our box tuning frequency)
LOOK at the "ACTIVE FILTER WIZARD" & "EQUALIZER FILTER WIZARD" graphs
There should be a circuit for a tuneable highpass filter as SpeakerLab used one back in the 1970's for vented boxes. I'm ignorant on most subjects so ask around. Also in the 1970's, Electro-Voice marketed a series of 6th ordered electrically assisted bass reflex speakers under "Interface Series". those came with an EQ box made with discrete transistors and sometimes can be found cheap at Ebay. IIRC you would have to make some resistor changes to tune one to 23Hz) - also, their power supply should be re-capped
EQ for a 6th order vented box is generally done with a one op-amp highpass filter with 6dB peaking ("Q=2")
To get that characteristic in hornresp, I first set a second order highpass filter to 16Hz corner - then added 6dB/Q=2 peaking at 23Hz (our box tuning frequency)
LOOK at the "ACTIVE FILTER WIZARD" & "EQUALIZER FILTER WIZARD" graphs
There should be a circuit for a tuneable highpass filter as SpeakerLab used one back in the 1970's for vented boxes. I'm ignorant on most subjects so ask around. Also in the 1970's, Electro-Voice marketed a series of 6th ordered electrically assisted bass reflex speakers under "Interface Series". those came with an EQ box made with discrete transistors and sometimes can be found cheap at Ebay. IIRC you would have to make some resistor changes to tune one to 23Hz) - also, their power supply should be re-capped
Last edited:
Hmm, I'll have to look into that deeper when I'm not at work. I recall a thread recently about using a large capacitor with a high-pass function at the tuning frequency to boost output at Fb and high-pass below that, could that provide a similar but passive solution in this case? Could also provide some protection from DC hits.
a large capacitor usually works best with undersized sealed box alignments where it can reduce input impedance at system resonance. It could go very bad with a vented box. For sealed boxes and 8 ohm drivers, 450uF can be a good starting value.
A sim says you could use 1200uF to 1400uF of good quality capacitor but its interaction with the lower reflex Z peak will increase excursion in the 10Hz area vs a straight 4th order reflex with no filter whatsoever.
A sim says you could use 1200uF to 1400uF of good quality capacitor but its interaction with the lower reflex Z peak will increase excursion in the 10Hz area vs a straight 4th order reflex with no filter whatsoever.
Here's some notes regarding 6th order bass reflex from the late and great DJK (Dennis Kleitsch)
Low Qts
Posted by djk (M) on September 29, 2007 at 03:03:05
In Reply to: RE: JBL Scoop posted by Hornlover on September 28, 2007 at 08:33:11:
The optimum Qts for minimum box size and maximum bass extension is 0.312
Qts in the range of 0.25~0.40 is OK.
If much higher than 0.40 a sealed box is indicated, if much lower than 0.25 a horn is indicated.
***********************************************************************
A little known fact is you can 'burn' Qts if it is lower than what you need(no more than 20%).Another tactic would be to use the low Qts to reduce the box size and raise the Fb to keep it flat.
The optimum Qts for an equalized 6th order Butterworth design(not to be confused with a bandpass design)is .312 .The formula for box size is Qts*Qts*4.1*Vas=Vb.The formula for Fb is (Fs/Qts)*.312 .For a Qts of .312 the Faux(eq frequency)=Fb with a Q of 2 .
***********************************************************************
Box size is roughly proportional to the square of the Qts, so the lower the Q the smaller the box. Also the lower the Q, the higher the efficiency. Bass rolls off though if the Q is too low, the magnet can be too big for its own good.
For a vented box, the optimum Qts is 0.312
0.312 gives the best bass extension, minimum box size, and highest efficiency.
Posted by djk ( M ) on July 25, 2005 at 00:55:59
A 2 cu ft 2nd order sealed box with an F3 of 30hz can only be 0.18% efficient (84.55dB).
A 2 cu ft 4th order vented box with an F3 of 30hz can only be 0.36% efficient (87.56dB).
A 2 cu ft 6th order vented box with an F3 of 30hz can only be 0.90% efficient (91.54dB).
The correct box size for a Qts=0.312 woofer for the 6th order vented design is Vas*4.1, Fb=Fs=Faux, Q=2 for Faux.
Without the Q=2 filter the box is an SBB4, with the best transient response of all the standard vented alignments, with the filter it is a B6(maximally flat).
The link describes a couple of DIY 6th order sub designs from JBL.
1983 JBL SUBWOFFER DESIGN
*****************************************************************
The Qts of the old JBL drivers is way to low to be optimum for bass reflex.
Will they work?
Sure.
But they will sound more balanced in a BLH
Low Qts
Posted by djk (M) on September 29, 2007 at 03:03:05
In Reply to: RE: JBL Scoop posted by Hornlover on September 28, 2007 at 08:33:11:
The optimum Qts for minimum box size and maximum bass extension is 0.312
Qts in the range of 0.25~0.40 is OK.
If much higher than 0.40 a sealed box is indicated, if much lower than 0.25 a horn is indicated.
***********************************************************************
A little known fact is you can 'burn' Qts if it is lower than what you need(no more than 20%).Another tactic would be to use the low Qts to reduce the box size and raise the Fb to keep it flat.
The optimum Qts for an equalized 6th order Butterworth design(not to be confused with a bandpass design)is .312 .The formula for box size is Qts*Qts*4.1*Vas=Vb.The formula for Fb is (Fs/Qts)*.312 .For a Qts of .312 the Faux(eq frequency)=Fb with a Q of 2 .
***********************************************************************
Box size is roughly proportional to the square of the Qts, so the lower the Q the smaller the box. Also the lower the Q, the higher the efficiency. Bass rolls off though if the Q is too low, the magnet can be too big for its own good.
For a vented box, the optimum Qts is 0.312
0.312 gives the best bass extension, minimum box size, and highest efficiency.
Posted by djk ( M ) on July 25, 2005 at 00:55:59
A 2 cu ft 2nd order sealed box with an F3 of 30hz can only be 0.18% efficient (84.55dB).
A 2 cu ft 4th order vented box with an F3 of 30hz can only be 0.36% efficient (87.56dB).
A 2 cu ft 6th order vented box with an F3 of 30hz can only be 0.90% efficient (91.54dB).
The correct box size for a Qts=0.312 woofer for the 6th order vented design is Vas*4.1, Fb=Fs=Faux, Q=2 for Faux.
Without the Q=2 filter the box is an SBB4, with the best transient response of all the standard vented alignments, with the filter it is a B6(maximally flat).
The link describes a couple of DIY 6th order sub designs from JBL.
1983 JBL SUBWOFFER DESIGN
*****************************************************************
The Qts of the old JBL drivers is way to low to be optimum for bass reflex.
Will they work?
Sure.
But they will sound more balanced in a BLH
IIRC, the B-52 pro-line consisted mainly of dual-opposed 4th order bandpass designs, or at least a wedged-pair bin where they were mounted basket exposed angled at each other. They also stated plainly in their literature to make sure and reinforce the flooring in which these would be used.
Later,
Wolf
Later,
Wolf
Last edited:
- Status
- Not open for further replies.