@Tyett71... i believe Wilson's original approach is to have the Puppy's low passed at around 120 hz, so your opinion seems sound.
i've modeled LOTS of drivers in WinISD now - some 50+ - to get an idea of how they might work in this application. below is a short list of some options suggested by you all as well as others that seem to offer a good balance of Qtc, Extension, and Output. What is interesting is the frequency at which the Xmax is reached on some of these - notably the smaller drivers. Thank you guys for calling this important parameter to my attention, as it clearly eliminates some otherwise nice looking options from consideration.
The drivers are sorted by the expected Qtc at 22L, which made comparisons easier for me.
i've modeled LOTS of drivers in WinISD now - some 50+ - to get an idea of how they might work in this application. below is a short list of some options suggested by you all as well as others that seem to offer a good balance of Qtc, Extension, and Output. What is interesting is the frequency at which the Xmax is reached on some of these - notably the smaller drivers. Thank you guys for calling this important parameter to my attention, as it clearly eliminates some otherwise nice looking options from consideration.
The drivers are sorted by the expected Qtc at 22L, which made comparisons easier for me.
An externally hosted image should be here but it was not working when we last tested it.
Something wrong with your image linking here...
I would appreciate to see one or more design proposals with simulations from you - that way there is something specific to discuss.
In this context I would strongly recommend taking sreten's (bloody good) observations and recommendations regarding the Wilson design into account.
odd thing on the image linking... i see the file in the thread replies, as well as in your reply above.
i'll try to submit it a different way here. it's basically a table of drivers sorted by 22L Qtc showing a few variables calculated through WinISD in an attempt to weigh the merits of each.
regarding specific designs, dual, opposed 8" drivers is intriguing, and the data in WinISD looks promising. does front / back firing vs side / side firing make a difference when crossing over as high as 100 Hz? i'd like to explore this option if the budget allows. otherwise, a single 10" down firing driver will likely what i try. there are some impressive smaller drivers, but they reach their Xmax pretty quickly.
Attachments
I would recommend you to conclude on a few driver's that you find interesting with regards to output capacity, size and money. Then you could do simulations with these and present curves for frequency response, stroke, impedance and with info on applied power. It tells so much more than just figures in a table.
If you have the opportunity in the software you use, take a look at how group delay and impulse and/or step response varies with the different configurations.
Hi,
Sidefiring will make no difference at 100Hz and the driver opposition
will remove the headache of dealing with single subwoofer vibration.
Also bass is mono at low frequencies, otherwise the two channels cancel.
Checking 8" subs they all seem to be 4 ohms, so I'd suggest a single 4
ohm amplifier, with a built in bass boost option (can't hurt to have the
option) in one cabinet driving both cabinets, budget friendly too.
There is an unusual option you could try for crossing them over.
Reduce the input capacitance of your valve amplifier to roll off the bass.
Take the drive for the sub from the amplifiers outputs via an inverse 6dB
per octave filter, easily done passively as you have loads of gain to reduce.
At this point you could try a passive 6dB/octave roll-off for the sub, setting
the subs actual filter somewhat higher to keep out midrange grunge. This
mainly passive set-up (with the passive BSC correction) would be very
difficult to match transparency wise without very expensive active filters.
Even if the sub amp does not have built in bass boost it would be
relatively easy to passively modify / extend its roll-off somewhat.
The 6dB/octave c/o points don't have to match (at -3dB) and likely for
the best sounding option they won't. Here we are trying to maintain
good transient response between the subs and sats, rather than SPL.
Tina-Ti is a free circuit emulator and could be used to design the filters.
Search on PLLXO's if you are confused, passive line level crossovers.
Playing around with some drivers there are 12mm Xmax Tangbands
that will end up around 0.5 Q for a pair in a 22L sealed box, and
some that will be more like Q=0.7 in a well stuffed sealed box, 8mm.
All things being equal the latter will require less power, but the former
should be more flexible if you have the extra power and also EQ a bit.
Also note that if you have enough power and the driver can take it,
the only thing that matters is clean Xmax, it determines max SPL,
which rolls off at 12dB/octave as you go down at excursion limits.
FWIW two 8"ers are very near the equivalent of an 11" driver.
rgds, sreten.
FWIW though, two of these per side with a form of
passive LT* and lots of power IMO is the best bet :
Tang Band W8Q-1071F 8"x12" Subwoofer 264-838
It would the highest combination of cone area and Xmax at a reasonable price.
* Box Q would be around 0.9, so you'd need some mild cut in the 100Hz
area and then electrically boost the 25Hz to 50Hz octave, doable passively.
Last edited:
I hadn't heard about it before & it looks similar to Multisim which i have, but FREE - Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.