Hello,
I bought a couple of 5¼" fullrange drivers from McBride, model Linear A 5.5, with the following T/S Parameters:
Fs: 70 Hz
Qms: 4.04
Qes: 0.49
Qts: 0.44
Vas: 8.72 liters
So I set out to build a ~54 liter sealed enclosure with the following inner dimensions:
h: 47.5"
w: 7"
d: 10"
I was going for infinite baffle with such a big enclosure, ~6 to 7 times Vas, that being my first build and this design looking pretty forgiving. I added series resistance of 4.7ohm to each driver to raise Qts and bring Qtc ~0.73. WinISD simulated an f3 ~72Hz.
I positionned the driver at a height of 36" to have it around ear height and stuffed the middle of the column, from just under the driver to 16" lower, with ½ a pound of polyester fiber to dampen the antinode area of the ~143Hz standing wave in the 47.5" dimension.
So that's what I build and here are my impressions now. This system is simply unlistenable. Not one frequency range sound right. Bass is muffled and boomy, mids are telephone-esque and the highs are kinda shrilly.
What could have happened here and what could be done about it?
I know my drivers are not high-end, but I was still expecting a bit better, something decent at least... I wouldn't think driver break-in would solve any of that, what do you think?
Could the stuffing be too dense? I have ½ a pound in the middle section, which translates to a density of ~0.75 pound per cubic feet, what do you think? Would too high a density fool the driver into seeing a much smaller Vb thus causing my boomy bass?
What about the telephony mids? I cannot EQ them out any way I tried. They just seem to be a "quality" of this driver, especially on the snare drum.
So I don't expect to make these sound like my Pierre Étienne Léon ML2's, but at least a bit better than my crappy Altec Lansing satellite+sub computer system, that would be a good start...
Thanks!
I bought a couple of 5¼" fullrange drivers from McBride, model Linear A 5.5, with the following T/S Parameters:
Fs: 70 Hz
Qms: 4.04
Qes: 0.49
Qts: 0.44
Vas: 8.72 liters
So I set out to build a ~54 liter sealed enclosure with the following inner dimensions:
h: 47.5"
w: 7"
d: 10"
I was going for infinite baffle with such a big enclosure, ~6 to 7 times Vas, that being my first build and this design looking pretty forgiving. I added series resistance of 4.7ohm to each driver to raise Qts and bring Qtc ~0.73. WinISD simulated an f3 ~72Hz.
I positionned the driver at a height of 36" to have it around ear height and stuffed the middle of the column, from just under the driver to 16" lower, with ½ a pound of polyester fiber to dampen the antinode area of the ~143Hz standing wave in the 47.5" dimension.
So that's what I build and here are my impressions now. This system is simply unlistenable. Not one frequency range sound right. Bass is muffled and boomy, mids are telephone-esque and the highs are kinda shrilly.
What could have happened here and what could be done about it?
I know my drivers are not high-end, but I was still expecting a bit better, something decent at least... I wouldn't think driver break-in would solve any of that, what do you think?
Could the stuffing be too dense? I have ½ a pound in the middle section, which translates to a density of ~0.75 pound per cubic feet, what do you think? Would too high a density fool the driver into seeing a much smaller Vb thus causing my boomy bass?
What about the telephony mids? I cannot EQ them out any way I tried. They just seem to be a "quality" of this driver, especially on the snare drum.
So I don't expect to make these sound like my Pierre Étienne Léon ML2's, but at least a bit better than my crappy Altec Lansing satellite+sub computer system, that would be a good start...
Thanks!
It took a bit of searching...
The Linear A 5.5 is on pg 42 of the Q-components catalog - download from McBride http://www.loudspeakers.ca/products.htm
Repeating the specs:
LINEAR A 5.5 $20.31
5-1/4” foam edge
dual cone fullrange
Outside Dim................134 x 134 mm
Impedance...............................8 ohm
Power ........................................35 W
Freq. Resp ................65 Hz - 18 KHz
Magnet Wt ................................10 oz
Fs .............................................70 Hz
Re .......................................6.7 ohms
Qts ..............................................0.71
Qes .............................................0.86
Qms............................................3.91
Vas .....................................8.82 litres
Voice Coil Dia .........................25 mm
SPL...........................................89 db
Depth ......................................61 mm
Baffle Opening......................123 mm
Mtg. Cntr...................139 mm BC x 4
Cheers
John
The Linear A 5.5 is on pg 42 of the Q-components catalog - download from McBride http://www.loudspeakers.ca/products.htm
Repeating the specs:
LINEAR A 5.5 $20.31
5-1/4” foam edge
dual cone fullrange
Outside Dim................134 x 134 mm
Impedance...............................8 ohm
Power ........................................35 W
Freq. Resp ................65 Hz - 18 KHz
Magnet Wt ................................10 oz
Fs .............................................70 Hz
Re .......................................6.7 ohms
Qts ..............................................0.71
Qes .............................................0.86
Qms............................................3.91
Vas .....................................8.82 litres
Voice Coil Dia .........................25 mm
SPL...........................................89 db
Depth ......................................61 mm
Baffle Opening......................123 mm
Mtg. Cntr...................139 mm BC x 4
Cheers
John
there are others here are much more technically proficient on this stuff than I am. It sound like you have a good grasp of the concepts involved. Just curious, did you try the driver in other enclosures? I'm really fond of the 2 pairs of TQWTL speakers I made. I have also built sealed box as well as vented box single driver speakers. Are you perhaps expecting too much from a single driver of this size for this large an enclosure. I wonder if you wouldn't be much more impressed with these mounted in a simple tapered pipe speaker. I usually think of open baffle units being larger dia and perhaps multiple drivers. Could you or have you experimented with reducing the volume?
plugging your number into WIN isdPro...
gives an Efficiency Bandwidth Product of 142.9, which indicates a 6th order bandbass would be best, (maybe I'm wrong , but I just added 4.7 Ω to the Re of 6.7 Ω , becoming 11.4 Ω ). This results in a Qtc of .475 with a 54 litre cabinet. Also your stuffing is making the box seem even larger, so the effected range will be greater than 135Hz-1000Hz
An EBP of 50 or less usually indicates a closed box, while one above 80 suggets a vented enclosure
F3 is at 135 Hz, but not smoothing out until 1000Hz. So any possible bass is too shallow, and everything up to 1000Hz is pulled back or "cupped"
now if making a BR or tuned enclosure, the results are much better, F3 of 59 Hz, , with 0 dB offset at 100Hz or so, and only an 11.56 litre box, tuned to 64.7 Hz. That's a 3.23" diameter X10" long port (or a rectangular one about 9.42" L X 4.04" X 1.85")
In this case the driver is so far from appropriate for any closed box , and certainly for one 54 litres. Hope that helps.
stew
gives an Efficiency Bandwidth Product of 142.9, which indicates a 6th order bandbass would be best, (maybe I'm wrong , but I just added 4.7 Ω to the Re of 6.7 Ω , becoming 11.4 Ω ). This results in a Qtc of .475 with a 54 litre cabinet. Also your stuffing is making the box seem even larger, so the effected range will be greater than 135Hz-1000Hz
An EBP of 50 or less usually indicates a closed box, while one above 80 suggets a vented enclosure
F3 is at 135 Hz, but not smoothing out until 1000Hz. So any possible bass is too shallow, and everything up to 1000Hz is pulled back or "cupped"
now if making a BR or tuned enclosure, the results are much better, F3 of 59 Hz, , with 0 dB offset at 100Hz or so, and only an 11.56 litre box, tuned to 64.7 Hz. That's a 3.23" diameter X10" long port (or a rectangular one about 9.42" L X 4.04" X 1.85")
In this case the driver is so far from appropriate for any closed box , and certainly for one 54 litres. Hope that helps.
stew
The factory Qts of 0.71 was not very real, as I measured it at ~0.42 IIRC, using proven methods. This is why I added series resistance as to get Qts back up so Qtc doesnt get too low.
Bass is not the only problem, the mids are quite unlistenable, sound like they were routed through a telephone and EQ just cannot fix that, seems to be a built-in sound of the driver.
Win ISD modeled the bass response with f3 of ~72Hz, which would have been good. I did not measure, but it sounds more like a higher Qtc, at a higher Fb, boomy and not smooth. I tought of such a big enclosure to only contain the back wave and have minimal spring action from the trapped air.
I guess I could use T/S measurement methods with the driver in the box to see what kind of Fb and Qtc it's developping in there.
I'll do the above and try to play with stuffing, but I have a feeling that the drive units are just plain cheap ones and not much can be done in the way of hifi with them. Last resort I could also chop the tall boxes to smaller ones and add a port to try a BR alignement.
I'm bummed out, but I won't stop there, I have other build ideas, I'll just remember to spring for better drivers.
Thanks for the feedback!
Bass is not the only problem, the mids are quite unlistenable, sound like they were routed through a telephone and EQ just cannot fix that, seems to be a built-in sound of the driver.
Win ISD modeled the bass response with f3 of ~72Hz, which would have been good. I did not measure, but it sounds more like a higher Qtc, at a higher Fb, boomy and not smooth. I tought of such a big enclosure to only contain the back wave and have minimal spring action from the trapped air.
I guess I could use T/S measurement methods with the driver in the box to see what kind of Fb and Qtc it's developping in there.
I'll do the above and try to play with stuffing, but I have a feeling that the drive units are just plain cheap ones and not much can be done in the way of hifi with them. Last resort I could also chop the tall boxes to smaller ones and add a port to try a BR alignement.
I'm bummed out, but I won't stop there, I have other build ideas, I'll just remember to spring for better drivers.
Thanks for the feedback!
MJK's worksheets
Bob Brines is perhaps one of the few that has actually supported Mr. King financially (please correct me if I am wrong) by purchasing his worksheets. There are a few others as well (unfortunately can't name any others off hand). Martin does deserve support, as he has done a huge amount of work for the DIY community (and some commercial concerns) regarding loudspeaker modelling. These commercial concerns using his software have generally not paid any fees, but use it freely.
I guess I didn't even think about the MathCad worksheets (requires MathCad viewer or software).
My hope is that if enough interest could be developed that Mr. King would consider licensing for personal use, with occasional updates (maybe once a year?) as needed. Maybe that way Mr. King could also continue with his well documented speaker projects for those of us that are perhaps less technically inclined.
stew
Bob Brines is perhaps one of the few that has actually supported Mr. King financially (please correct me if I am wrong) by purchasing his worksheets. There are a few others as well (unfortunately can't name any others off hand). Martin does deserve support, as he has done a huge amount of work for the DIY community (and some commercial concerns) regarding loudspeaker modelling. These commercial concerns using his software have generally not paid any fees, but use it freely.
I guess I didn't even think about the MathCad worksheets (requires MathCad viewer or software).
My hope is that if enough interest could be developed that Mr. King would consider licensing for personal use, with occasional updates (maybe once a year?) as needed. Maybe that way Mr. King could also continue with his well documented speaker projects for those of us that are perhaps less technically inclined.
stew
Greets!
To completely remove a ~143 Hz null will require a diaphramatic absorber AFAIK, so we're talking a very dense pad centered in the pipe. Not having ever done it though I have no idea how it will affect system response, but I'm thinking not in a good way overall since both sides will have the same Vb (tuning).
Stuffing density will increase apparent cab Vb until it reaches ~1.5 lbs/ft^3 referenced to acoustic fiberglass insulation, so with Polyfil it would have to be considerably higher before it morphs from an acoustic 'sponge' to a reflection boundary.
'Muffled' bass makes sense due to driver, stuffing locations along with its density, creating a bass wave 'two step', but 'boomy' OTOH implies an under-damped response which in turn implies your Qts measurements are suspect, so how does it sound without any series R?
If you're satisfied that you've accurately measured the driver specs, then you can measure the speaker's Fb, Qt specs which will 'tell the tale' about what the cab is doing.
WRT mids/HF, you don't mention any lining of walls, top of the driver 'chamber', so these reflections will comb filter with the driver's output and to a lesser extent, ditto off the inner basket legs and any motor overhang.
Last, but not least, how does the mids/HF sound on a OB of the same baffle size? It shouldn't be any different in a cab except for the lower mids sounding more tonally balanced ('full') and of course the lack of mid-bass.
Bottom line, for half WL pipe designs, the driver needs to be located based on the pipe length:CSA ratio with minimal stuffing density along its entire length. FWIW, I calc a 17.04" offset in case you want to experiment and 0.2 lbs/ft^3 (~6.16 oz) should be plenty (see below).
GM
To completely remove a ~143 Hz null will require a diaphramatic absorber AFAIK, so we're talking a very dense pad centered in the pipe. Not having ever done it though I have no idea how it will affect system response, but I'm thinking not in a good way overall since both sides will have the same Vb (tuning).
Stuffing density will increase apparent cab Vb until it reaches ~1.5 lbs/ft^3 referenced to acoustic fiberglass insulation, so with Polyfil it would have to be considerably higher before it morphs from an acoustic 'sponge' to a reflection boundary.
'Muffled' bass makes sense due to driver, stuffing locations along with its density, creating a bass wave 'two step', but 'boomy' OTOH implies an under-damped response which in turn implies your Qts measurements are suspect, so how does it sound without any series R?
If you're satisfied that you've accurately measured the driver specs, then you can measure the speaker's Fb, Qt specs which will 'tell the tale' about what the cab is doing.
WRT mids/HF, you don't mention any lining of walls, top of the driver 'chamber', so these reflections will comb filter with the driver's output and to a lesser extent, ditto off the inner basket legs and any motor overhang.
Last, but not least, how does the mids/HF sound on a OB of the same baffle size? It shouldn't be any different in a cab except for the lower mids sounding more tonally balanced ('full') and of course the lack of mid-bass.
Bottom line, for half WL pipe designs, the driver needs to be located based on the pipe length:CSA ratio with minimal stuffing density along its entire length. FWIW, I calc a 17.04" offset in case you want to experiment and 0.2 lbs/ft^3 (~6.16 oz) should be plenty (see below).
GM
Attachments
Re: Re: What I built, impressions and solutions?
About inside walls, I lined them with some foamy undercarpet material I cannot name more precisely.
I did first try the drivers in OB's, but of different size, ~24" x ~40" IIRC. It had the same telephony sound, but I dismissed it as the result of a very "comby" response due to having 3 driver-edge offsets the same dimension.
I am confident in my measuring method as it is the one you gave me a link to in another thread some weeks back. I'll try it again on the whole system, driver+enclosure. Measuring both with and without the series resistance. I guess the formuals will require me to add the 4.7ohm to the Re value.
I'll try playing with stuffing and my driver offset is already 16", not to far from your estimate.
About inside walls, I lined them with some foamy undercarpet material I cannot name more precisely.
I did first try the drivers in OB's, but of different size, ~24" x ~40" IIRC. It had the same telephony sound, but I dismissed it as the result of a very "comby" response due to having 3 driver-edge offsets the same dimension.
I am confident in my measuring method as it is the one you gave me a link to in another thread some weeks back. I'll try it again on the whole system, driver+enclosure. Measuring both with and without the series resistance. I guess the formuals will require me to add the 4.7ohm to the Re value.
I'll try playing with stuffing and my driver offset is already 16", not to far from your estimate.
If you're really disenchanted with the labor put into those cabinets, the dimensions could lend themselves to converting to a TQWTL. Remove the bottom, fit and piece to form the taper, move the driver to the top. Wouldn't be perfect, but it might be an interesting exercise? I get most of my inspiration from old speaker builder books by D Weems. My first was using the old Radio Shack 5-1/4 with whizzer cone. Then I put Lineum tweeters on top, and replaced the drivers with lower frequency response. They have been entertaining us for years. Just 2c from the peanut gallery.
Re: Re: Re: What I built, impressions and solutions?
If it's the cheap patchwork of scraps, it's called 'rebond'. Regardless, all foam carpet padding is closed cell, ergo reflective, so open cell is required for lining speakers and is probably contributing to how they sound. Try 3/8" thick felt or 1" acoustic fiberglass insulation instead or just the minimal stuffing I suggested.
Hmm, if the mids/HF don't sound good on a baffle this big, then you may have to make do with adding a tweeter.
OK, the dims you listed implied only an 11.5" offset which yielded a hard to cure dip, but at 16" it's close enough for the negligible dip to be inaubile with the minimal stuffing I suggested.
GM
IG81 said:
If it's the cheap patchwork of scraps, it's called 'rebond'. Regardless, all foam carpet padding is closed cell, ergo reflective, so open cell is required for lining speakers and is probably contributing to how they sound. Try 3/8" thick felt or 1" acoustic fiberglass insulation instead or just the minimal stuffing I suggested.
Hmm, if the mids/HF don't sound good on a baffle this big, then you may have to make do with adding a tweeter.
OK, the dims you listed implied only an 11.5" offset which yielded a hard to cure dip, but at 16" it's close enough for the negligible dip to be inaubile with the minimal stuffing I suggested.
GM
Re: Re: Re: Re: What I built, impressions and solutions?
I'm not that bummed out. Total investment so far is sub 100$ and ~20 hours of work. I'm learning by doing all of this, so I'm glad.
I thought about TL conversion. I'll see, maybe I'll chop it up in two and re-use the other part for BR for small Vifa woofers I have.
I guess it makes sense what you say about the "rebond". I'll leave it for now as if it's not helping, it's not worse than bare wood for sure.
What I first did is remove the series 4.7 ohm. You know what, it sound better without it. Bass is not as boomy. Mids improved slightly, but are still thin sounding, especially on snare drum hits, this is what strikes me most.
I measured T/S parameters of the whole system without the 4.7ohm, and my box is big enough not to affect anything. Fb is at 70Hz (driver Fs 70Hz) and Qtc, if I'm right that in-box measured Qts is Qtc, is 0.49 (driver Qts is 0.49). Maybe I should be concerned about the too-close Qts and Qtc... Sound like that could be caused by a leaky box, I'll try to see if that's what's happening. Would that make sense?
I did not have time to measure with the series resistance to see just how much it did raise system Q, I'll do it this weekend. But I would not think it should raise it as much as to give me the annoyingly boomy bass it did. I'll see.
Thanks for the input guys!
Sparky OR said:
I'm not that bummed out. Total investment so far is sub 100$ and ~20 hours of work. I'm learning by doing all of this, so I'm glad.
I thought about TL conversion. I'll see, maybe I'll chop it up in two and re-use the other part for BR for small Vifa woofers I have.
GM said:
I guess it makes sense what you say about the "rebond". I'll leave it for now as if it's not helping, it's not worse than bare wood for sure.
What I first did is remove the series 4.7 ohm. You know what, it sound better without it. Bass is not as boomy. Mids improved slightly, but are still thin sounding, especially on snare drum hits, this is what strikes me most.
I measured T/S parameters of the whole system without the 4.7ohm, and my box is big enough not to affect anything. Fb is at 70Hz (driver Fs 70Hz) and Qtc, if I'm right that in-box measured Qts is Qtc, is 0.49 (driver Qts is 0.49). Maybe I should be concerned about the too-close Qts and Qtc... Sound like that could be caused by a leaky box, I'll try to see if that's what's happening. Would that make sense?
I did not have time to measure with the series resistance to see just how much it did raise system Q, I'll do it this weekend. But I would not think it should raise it as much as to give me the annoyingly boomy bass it did. I'll see.
Thanks for the input guys!
Greets!
You're welcome!
Why not take it out and reduce/'tease' out the existing Polyfil?
Hmm, adding series R to flatten the response typically doesn't cause what you're hearing until ~2-3x Re is inserted. That, or a really crappy out-of-spec resistor is used and why I prefer to use a cheap pot (RS 271-265 or similar) to dial in series R, then replace it with a high quality non-inductive power resistor.
FWIW, trying to guesstimate your mixed stuffing densities, the lowest Fb simmed is ~73 Hz and ~72 Hz with my suggested stuffing. If there was a measurable leak, then where the leak was would affect tuning to some extent, but mainly a leak is ~aperiodic, so typically in a near IB alignment it over-damps the system, i.e. lowers Fb, Qtc to at/below Fs, Qts, resulting in a slightly higher F3. I doubt it would be audible in this case unless it's leaking around the driver, but then it shouldn't sound 'boomy'............
There's some contradictions though WRT its bass response. I mean without the series R it should be ~critically damped with no ringing, i.e. very 'tight', 'dry' sounding and tonally balanced down low with it, so what's your signal chain, i.e. amp, speaker wire?
WRT converting them to a MLTL, adding a 3" i.d. diameter x 2" long vent 31.5" down from the top sims usable output to 40 Hz and if it sounds 'boomy', damp the vent as required, then dial in some series R to flatten it.
GM
You're welcome!
Why not take it out and reduce/'tease' out the existing Polyfil?
Hmm, adding series R to flatten the response typically doesn't cause what you're hearing until ~2-3x Re is inserted. That, or a really crappy out-of-spec resistor is used and why I prefer to use a cheap pot (RS 271-265 or similar) to dial in series R, then replace it with a high quality non-inductive power resistor.
FWIW, trying to guesstimate your mixed stuffing densities, the lowest Fb simmed is ~73 Hz and ~72 Hz with my suggested stuffing. If there was a measurable leak, then where the leak was would affect tuning to some extent, but mainly a leak is ~aperiodic, so typically in a near IB alignment it over-damps the system, i.e. lowers Fb, Qtc to at/below Fs, Qts, resulting in a slightly higher F3. I doubt it would be audible in this case unless it's leaking around the driver, but then it shouldn't sound 'boomy'............
There's some contradictions though WRT its bass response. I mean without the series R it should be ~critically damped with no ringing, i.e. very 'tight', 'dry' sounding and tonally balanced down low with it, so what's your signal chain, i.e. amp, speaker wire?
WRT converting them to a MLTL, adding a 3" i.d. diameter x 2" long vent 31.5" down from the top sims usable output to 40 Hz and if it sounds 'boomy', damp the vent as required, then dial in some series R to flatten it.
GM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.