I have a friend with a 15" driver and he wants to use it for public address and light music, and to mount it near a bricked ceiling corner. I have measured 44Hz and Qts=1.3.
I could use some suggestions on what enclosure type to use. I am hoping not to need to resort to active EQ, though the corner boost with the high Qts concerns me.
I could use some suggestions on what enclosure type to use. I am hoping not to need to resort to active EQ, though the corner boost with the high Qts concerns me.
Hi,
Seems to me your only option is a sealed box and knocking back
the bass EQ knob somewhat. The high Q in box will then control
the effective bass extension. More common than you think, see :
http://www.partsexpress.com/projectshowcase/goldenboys/index.html
:0/sreten.
Seems to me your only option is a sealed box and knocking back
the bass EQ knob somewhat. The high Q in box will then control
the effective bass extension. More common than you think, see :
http://www.partsexpress.com/projectshowcase/goldenboys/index.html
:0/sreten.
I have use drivers with higher Qts and it is not possible to get lower system Q unless with active solutions. With passive solutions an IB would give Qs=Qts at best. Smaller boxes with produce Qs even higher than that.
Some options to consider
1) aperiodic vents for weak motor drivers (ie acceleration Bl/mms)
2) Open or semi-open back enclosures for stronger motor drivers (like the old guitar combo amps)
Some options to consider
1) aperiodic vents for weak motor drivers (ie acceleration Bl/mms)
2) Open or semi-open back enclosures for stronger motor drivers (like the old guitar combo amps)
I couldn't find anything on "alpha tl" but out of curiosity i modified the parameters of a 15" pro woofer to equal the limited specs that were given above and ran some simulations. The results were interesting. This hypothetical driver shows a huge bump of 6 to 7 dB in a rather large closed box (about 4dB IB) So OB is the obvious usage but not too practical for PA.
So I tried a few things in TL and was able to get a sim that showed broad ripple of about 2dB over the range from just under 30Hz to about 80Hz with sensitivity of around 105dB/W. But get this. The pipe cross sectional area was 8000 cm^3.
An interesting thing to play with but I wonder about the accuracy of simulations when such way out sizes are needed. It would be interesting to have the complete parameters of the driver for further play.
mike
So I tried a few things in TL and was able to get a sim that showed broad ripple of about 2dB over the range from just under 30Hz to about 80Hz with sensitivity of around 105dB/W. But get this. The pipe cross sectional area was 8000 cm^3.
An interesting thing to play with but I wonder about the accuracy of simulations when such way out sizes are needed. It would be interesting to have the complete parameters of the driver for further play.
mike
I like the idea of a lossy box of some sort, even if I'll need to resort to a little EQ.
I've loosened the driver up a little and fs has dropped to 37Hz. Haven't remeasured Qts but working with 1.2 till I do. Added mass Vas measured 122l (4.3'^3).
Edit: This driver has a fairly light paper cone for its size (in my experience), adding 147g of mass reduced fs from 37 to 26.5Hz. Suspension seems a little stiff (pleated doped cloth) and magnet is approximately 15cm*2cm.
I've loosened the driver up a little and fs has dropped to 37Hz. Haven't remeasured Qts but working with 1.2 till I do. Added mass Vas measured 122l (4.3'^3).
Edit: This driver has a fairly light paper cone for its size (in my experience), adding 147g of mass reduced fs from 37 to 26.5Hz. Suspension seems a little stiff (pleated doped cloth) and magnet is approximately 15cm*2cm.
lndm said:
Yes. An aperiodic vent is a resistive loss term in an otherwise closed box system. IMO it is a variable depending on pressure in the box. Also adding mass might help things here as well. Your cone mms is most likely around 50g, so I would not add more than 100% depending on suspension and possible reduction of Xmax and sensitivity. EQ as required for source and placement.
<edit>oops just noticed it was a 15" so mms would be a little higher than 50g, you can estimate by looking at similar drivers with known T/S.
OK, I was tossing up whether to go with a box much smaller than Vas and deal with the consequences, or make one closer to Vas. I began to sim the baffle step and luck kicked in. I chose a box as large as Vas and used the two largest dimensions for the front.
The red curve is from 'The Edge' baffle sim, the blue is the box response (it is plotted where it is for better visibility), and the green is the resultant.
The red curve is from 'The Edge' baffle sim, the blue is the box response (it is plotted where it is for better visibility), and the green is the resultant.
Attachments
Hmmm, I am fearing that.
On the other hand, if I make the baffle any smaller I'll just create a dip in the response, larger, a peak. I do still have the option of the aperiodic enclosure. I suspect that the overshoot may slightly compensate the apparent levels for adding a variovent after the fact. At least I hope so. Am I being reasonable?
On the other hand, if I make the baffle any smaller I'll just create a dip in the response, larger, a peak. I do still have the option of the aperiodic enclosure. I suspect that the overshoot may slightly compensate the apparent levels for adding a variovent after the fact. At least I hope so. Am I being reasonable?
forr said:
Come on Forr
Yes, maybe if you could stabilize it, But NOT practical! I would definitely NOT recomend it, except if you want to spend alot more time working on amplifier development. IMO More of science fair projects.
Best solution if going for an active solution, would be a Linkwitz Transform circuit. It's well documented as a working solution to Q and Fs problems besides.
Hi,
Corner mounting will bump up the response such that BSC is not needed.
Aperiodic vents aren't really appropriate for PA speakers.
E-sides suggestion is easily implemented by upping the input
capacitor value of the power amplifier - or can probably be
approximated by knocking back the bass tone control.
TBH quality wise getting reasonable bass balance is not that
diificult as long as you don't want / need real bass extension,
the highish q point simply defines the lower limit when Eq'd back.
Quality wise the real issue is getting an overall good balance
with decent dispersion and low distortion across the range.
/sreten.
Corner mounting will bump up the response such that BSC is not needed.
Aperiodic vents aren't really appropriate for PA speakers.
E-sides suggestion is easily implemented by upping the input
capacitor value of the power amplifier - or can probably be
approximated by knocking back the bass tone control.
TBH quality wise getting reasonable bass balance is not that
diificult as long as you don't want / need real bass extension,
the highish q point simply defines the lower limit when Eq'd back.
Quality wise the real issue is getting an overall good balance
with decent dispersion and low distortion across the range.
/sreten.
sreten said:
Aperiodic vents aren't really appropriate for PA speakers.
A Sealed system (closed box) from Sreten's earlier advise is definitely NOT advisable for higher Qts (> 0.7). Aperiodic (vario) vents will make a a bad situation tolerable at higher spl's. IMO They are very useful for weak motor drivers. Whether or not this driver fits that bill is not known. Otherwise use open back type cabs.
Or even better use an active solution via a LT circuit. Not too hard to implement. A PCB is available on a currently running group buy on this forum as we speak.
About negative impedance
Infinia
---Yes, maybe if you could stabilize it, But NOT practical! ---
You have to stabilize it if you want to nullify the voice-coil DC resistance a la Stahl. However dividing it only by 2 is very easy, you can do it with only three resistors around the power amp. Qt is then less than 0.7, a more manageable value than 1.3.
I have not made simulations of a Linkwitz transform with this precise speaker which has a Qt of 1.3, but previous attempts with high Q's failed.
Infinia
---Yes, maybe if you could stabilize it, But NOT practical! ---
You have to stabilize it if you want to nullify the voice-coil DC resistance a la Stahl. However dividing it only by 2 is very easy, you can do it with only three resistors around the power amp. Qt is then less than 0.7, a more manageable value than 1.3.
I have not made simulations of a Linkwitz transform with this precise speaker which has a Qt of 1.3, but previous attempts with high Q's failed.
Forr I Haven't read the AES papers in a long time. AFAIK pwr amp should be configured as a current source (AB w collector outputs). Small resistor samples the current at output return then applies neg feedback with trickey compensation network in feedback loop. A stability analysis would be required. IMO this technique needs a special dedicated pwr amp for woofer only.
I can simulate LT circuit if poster is interested in going active. This way he can use "off the shelf" voltage mode power amp.
I can simulate LT circuit if poster is interested in going active. This way he can use "off the shelf" voltage mode power amp.
The equipment I have to work with was purchased on the advice of 'the guy at the other end of the line' , and this is a not for profit organisation so what we have is all we have.
There is one 2ch power amp that will do well if we don't need to go active, and there is a mixing board which seems to have three fixed EQs per channel. Not sure whether the bass controls are centred or tone control shaped.
There is a reasonable chance I could apply passive EQ at the input to the power amps.
I am feeling that the power response will be more important here than the on axis response. Because of this, I feel I need to take the baffle step with a grain of salt, rear wall loaded or not.
I am concerned with the cancellation effects and placement difficulties I may encounter with an open baffle. I am concerned that a closed box would ensure overshoot and ringing in the bass. I am wondering whether EQ will make this tolerable.
, and this is a not for profit organisation so what we have is all we have.There is one 2ch power amp that will do well if we don't need to go active, and there is a mixing board which seems to have three fixed EQs per channel. Not sure whether the bass controls are centred or tone control shaped.
There is a reasonable chance I could apply passive EQ at the input to the power amps.
I am feeling that the power response will be more important here than the on axis response. Because of this, I feel I need to take the baffle step with a grain of salt, rear wall loaded or not.
I am concerned with the cancellation effects and placement difficulties I may encounter with an open baffle. I am concerned that a closed box would ensure overshoot and ringing in the bass. I am wondering whether EQ will make this tolerable.
See what design the other guy had in mind.lndm said:The equipment I have to work with was purchased on the advice of 'the guy at the other end of the line'
Hopefully you'll get all the credit when all your work pays off
With just 3 tone controls most prob not where you need them to be.lndm said:
Could work ok but may have to tweek the notch a bit.lndm said:
I think the LT circuit would be your cup-o-tea.
Most likely if you couldn't control the placementlndm said:
Yes they can cancel where you want them, but they add in phase at a central frequency as well. IMO The best EQ for this is the LT circuit. Rod Elliot has a project of the LT on his web site.lndm said:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.