Hello everyone! I'm trying to design a compact home subwoofer using a pair of Hertz HS300 car subwoofers. I'm using them in a push-pull, 4th order bandpass configuration. They model pretty well on WinISD and BassBox in a 15lt(sealed chamber)+5lt(vented) bandpass enclosure tuned at 60Hz. I get a nice flat curve with low group delay with, even if I lose 9dB of efficiency it stays manageable (should be of around 82dB/W after being put in this bandpass box).
But doing so I encounter the usual issue with small displacement bass reflex subwoofers: the port is too long to fit in the box, even if I fold it. Reducing the diameter by a lot seems to help (I still need to fold the port) but because of the high Xmax and diameter of the subwoofers the vent air velocity is too high. This will lead to power compression issues and potentially port chuffing.
This issue is usually solved by replacing the traditional cylindrical or slotted port in favour of a more manageable passive radiator. However, I have no clue about how to deal with this particular situazion in which the bandpass enclosure and a PR are used in conjunction. How do I model the enclosure, and how do I find the correct mass to add to a PR of my choice?
I've been able to find some info on the internet about this type of use for the PR but no design tips.
Has anyone ever dealt with this problem?
Thanks
Piergiorgio
But doing so I encounter the usual issue with small displacement bass reflex subwoofers: the port is too long to fit in the box, even if I fold it. Reducing the diameter by a lot seems to help (I still need to fold the port) but because of the high Xmax and diameter of the subwoofers the vent air velocity is too high. This will lead to power compression issues and potentially port chuffing.
This issue is usually solved by replacing the traditional cylindrical or slotted port in favour of a more manageable passive radiator. However, I have no clue about how to deal with this particular situazion in which the bandpass enclosure and a PR are used in conjunction. How do I model the enclosure, and how do I find the correct mass to add to a PR of my choice?
I've been able to find some info on the internet about this type of use for the PR but no design tips.
Has anyone ever dealt with this problem?
Thanks
Piergiorgio
Some programs give one a PR option, otherwise it was/can be done by designing a vent using the PR's (Sd) = vent area (Av) and once the length is determined, calculate its air mass weight to find the PR's total weight.
60 hz is not very low tuning so you should be able to fit a port in that box.
If your looking at port velocity using max SPL then I bet port velocity is extremely high.
Look at the cone excursion screen and only apply enough power till it hits xmax @ 30 Hz
that is a more realistic power level. port velocity would be much lower.
60 Hz tuning seems really high, power handling would not be very good since it would unload quickly.
WinIsd has a passive radiator option.
usually for Subs you need double the passive radiator area so it does not hit distortion.
So you need 2x passive radiators.
be a lot easier, more SPL and less power to build normal reflex boxes.
If your looking at port velocity using max SPL then I bet port velocity is extremely high.
Look at the cone excursion screen and only apply enough power till it hits xmax @ 30 Hz
that is a more realistic power level. port velocity would be much lower.
60 Hz tuning seems really high, power handling would not be very good since it would unload quickly.
WinIsd has a passive radiator option.
usually for Subs you need double the passive radiator area so it does not hit distortion.
So you need 2x passive radiators.
be a lot easier, more SPL and less power to build normal reflex boxes.
You need to understand this is a bandpass design, and therefore 45 to 60 Hz is actually a NORMAL tuning range. Too low limits upper bandwidth, too high limits lower bandwidth, in combination with the sealed box highpass and the tuning of the vented/PR lowpass.
The other points i need to make are modeling the PR chamber needs to be very small, and PRs need to be larger than active drivers. Since most of them have been designed to tune low in a standard PR/Vented alignment, the mass tends to be quite high initially. Decreasing the volume of the vented chamber raises the tuning frequency to ranges that are then usable for a bandpass.
For example, I used Unibox to model this in stages as it can't do it as a single system.
1- model the bandpass with large port so you can see what volumes and tunings work. You may have to reduce the front volume later and retry.
2- Model the PR volume to see how/what the tuning is in that volume and how the PR and its Xmax are handling your output requirements, and model your mass added to get it there. You cant take away mass, so you may have to go back to step one and alter it and retry step 2.
3- Make sure everything will physically fit in the volumes or such. If they will not, return to step one and alter the cabinet dimensions.
How do i know? Check out my Overdrive10 design. 10" bandpass with 12" PR. 18 and 8 ltr volumes to make it work, and it's a shoehorn.
One last bit of advice, you may need boost at 25-30Hz if;
1- driver can handle it with xmax
2- you cant tune low enough to get lower end extension
3- you try to stay at roughly the initial sensitivity of the driver.
Later,
Wolf
The other points i need to make are modeling the PR chamber needs to be very small, and PRs need to be larger than active drivers. Since most of them have been designed to tune low in a standard PR/Vented alignment, the mass tends to be quite high initially. Decreasing the volume of the vented chamber raises the tuning frequency to ranges that are then usable for a bandpass.
For example, I used Unibox to model this in stages as it can't do it as a single system.
1- model the bandpass with large port so you can see what volumes and tunings work. You may have to reduce the front volume later and retry.
2- Model the PR volume to see how/what the tuning is in that volume and how the PR and its Xmax are handling your output requirements, and model your mass added to get it there. You cant take away mass, so you may have to go back to step one and alter it and retry step 2.
3- Make sure everything will physically fit in the volumes or such. If they will not, return to step one and alter the cabinet dimensions.
How do i know? Check out my Overdrive10 design. 10" bandpass with 12" PR. 18 and 8 ltr volumes to make it work, and it's a shoehorn.
One last bit of advice, you may need boost at 25-30Hz if;
1- driver can handle it with xmax
2- you cant tune low enough to get lower end extension
3- you try to stay at roughly the initial sensitivity of the driver.
Later,
Wolf
Love it, another BP fan! A 60hz BP4 is ideal for "musical" car audio. That tune should get you a response from 30-90hz.
The only problem I have with a PR BP4 is the driver's vented pole piece cannot get fresh air to the voicecoil for cooling. I like designing my BP4's with the driver's basket in the vented section of the enclosure.
The only problem I have with a PR BP4 is the driver's vented pole piece cannot get fresh air to the voicecoil for cooling. I like designing my BP4's with the driver's basket in the vented section of the enclosure.
My Overdrive10 was 30 to 65 without the boost at 25 that yields about 22Hz F3 before room gain. In a car, i would not use or even need the boost.
Wolf
Wolf
Greetings GM,
I tried calculating the PR weight once by comparison to air mass in a vent as you mention, but found it wasn't very close for a small 4.5" full range driver. Maybe the method works better for larger woofers or subwoofers.
I believe that the PR suspension (spring constant/compliance) needs to be included, and almost as important, some damping in a model. Some manufacturers provide a graph or table as a guide for tuning their PRs (Accuton, e.g.)
My ability to work with differential equation modeling has dulled over the past few decades 🤔 , so something like Unibox or Winisd, etc. is probably the best option.
Ya know, I never had luck with the mass = air equivalency either.
I guess it isn't just me then,
Wolf
I guess it isn't just me then,
Wolf
Greets!
Didn't take time to elaborate since it should be on-line 'somewhere', but as noted it takes a minimum 1.75x Sd for high excursion drivers and the basic need for the vent to be at least Sd = Av large/long if vented, so simmed will be considerably larger, hence much longer, as in TL huge, which BTW was how I use to quickly design various type TLs.
As for the math, mine from a heat/AC textbook is long gone, though got 'close enough' without going over using the last version of BoxPlot? for the last couple I did with only some fine tuning required and yes, all were 12-15" low excursion with < ~27-32 Hz tuning, but can't think of a good technical reason ATM why the design routine would 'fall apart' with smaller, higher tuned systems unless its effective Qts (Qts') was > ~ 0.383, i.e. too weak to drive such a heavy mass and/or the PR chosen had too stiff a suspension.
You're welcome! Considering Qts' would be even higher, kind of surprised it worked well!
PR model used?
PR model used?
VituixCAD can do 4th order PR bandpass systems (or used to).
A bandpass is basically just a acoustic lowpass filter, so the benefits in an active system are pretty minimal.
A bandpass is basically just a acoustic lowpass filter, so the benefits in an active system are pretty minimal.
Vifa NE123W-04 (4" full range) with Dayton DSA175-PR (6.5" aluminum cone) in a 3.8 liter box (Qtc ~ 0.61).
PR used as is, with no additional weights. Box tuning with PR was close to 60 Hz, IIRC.
The SPL response of the NE123 was a bit ragged, so a contour circuit was added to smooth it.
End result was really quite good.
Dave R
Hello! Thanks for all the replies.
I've got some questions still. But I'm happy with the fact that at least it seems like there's a way out.
I'm using WinISD and I think I'm following the same procedure you described. I'm modeling a 17lt+5lt 4th order bandpass, tuned this time at 55Hz as I noticed it gives me some more bass extension losing only 1dB of efficency. I've chosen the passive radiator I want to use (Dayton Audio DSA315-PR) which is a high Xmax (13mm, way more than the woofers which have 9mm) 12" passive radiator.
Then I created another project on WinISD, loaded the subwoofers' and PR TSPs, with the same volume as the vented chamber (5 liters) and started adding mass to the PR until the box's resonant frequency was 55Hz. I got pretty close with a 56Hz Fh and 400g of added mass to the passive radiator.
I've got some questions still. But I'm happy with the fact that at least it seems like there's a way out.
I see how this could work, but at that point, how would the PR's other parameters affect the end result?
Why do you think that is a very high tuning? The sxcursion is ok thanks to the very stiff suspension, According to WinISD the bandwidth of this enclosure would start at 30Hz and end at 130Hz (both are measured at -3db from the mean level). I can make the tuning lower (around 55Hz), and the volume slightly higher in the sealed chamber, but I'd end up losing even more efficiency which is already quite low. Power would be plenty either way (around 600W RMS for each sub, and I'm planning on building a stereo pair), but I wouldn't want it to dip below 80dB.
I've seen your project and it's indeed like what I have in mind for this, the only difference being that mine is also in a push-pull configuration. You used UniBox for modeling the speaker enclosure, which lacks the possibility to use a passive radiator in conjunction with a bandpass enclosure.
I'm using WinISD and I think I'm following the same procedure you described. I'm modeling a 17lt+5lt 4th order bandpass, tuned this time at 55Hz as I noticed it gives me some more bass extension losing only 1dB of efficency. I've chosen the passive radiator I want to use (Dayton Audio DSA315-PR) which is a high Xmax (13mm, way more than the woofers which have 9mm) 12" passive radiator.
Then I created another project on WinISD, loaded the subwoofers' and PR TSPs, with the same volume as the vented chamber (5 liters) and started adding mass to the PR until the box's resonant frequency was 55Hz. I got pretty close with a 56Hz Fh and 400g of added mass to the passive radiator.
Can you fit both drivers and PR into a 5 ltr box? Or are you using one driver/PR per sim? A single liter is about a 4" cube, so i do not see this working physically.
Wait a minute...
One more question, i wonder if we are not seeing one detail. Push-pull means 2 drivers mounted to said box, and one reversed polarity as well as mount. Ie; not mounted isobarically with a chamber between the drivers' cones. If you have them to either side of a tunnel mounted back to back or face to face, then you must say this is also an isobaric system. Push Pull is not equal to Isobaric Push Pull. They are different.
Additionally, if you are adding 400g to the PR, you are not being as efficient as possible. You should not have to add that much mass to get back down to 60Hz. -9dB is a huge loss.
Later,
Wolf
Wait a minute...
One more question, i wonder if we are not seeing one detail. Push-pull means 2 drivers mounted to said box, and one reversed polarity as well as mount. Ie; not mounted isobarically with a chamber between the drivers' cones. If you have them to either side of a tunnel mounted back to back or face to face, then you must say this is also an isobaric system. Push Pull is not equal to Isobaric Push Pull. They are different.
Additionally, if you are adding 400g to the PR, you are not being as efficient as possible. You should not have to add that much mass to get back down to 60Hz. -9dB is a huge loss.
Later,
Wolf
Well, can't recall ever seeing any definitive answer to ensure best overall performance, so can only use mine derived from a theoretically ideal reflex and IB tuning alignment = ~ > 4-10x driver Vas (i.e. IB alignment spec wise) with an Fs ~1.56x driver Fs per: https://archive.org/details/HowToBuildSpeakerEnclosuresByAlexixBadmaieffDonDavis/page/n75/mode/2up
Yes, my fault, it's an isobaric box. I thought that isobaric was just another word for push pull!
How do I lower the PR box tuning frequency without adding that much mass? I don't mind the -9db sensitivity loss as long as I can get some extension from those subwoofers (which are meant for car audio), because even if I was -9dB under the driver's sensitivity it'd still be 83db/W which is ok for a subwoofer with that much power handling. Consider that I'm making a stereo pair also.
Last edited:
Increase the front volume. That or the mass addition.
I still don't think you can get the driver backside and the PR in 5 ltrs, unless you mount the PR backside-out.
You may have to make it larger just to fit the stuff in there without stuff colliding.
Wolf
I still don't think you can get the driver backside and the PR in 5 ltrs, unless you mount the PR backside-out.
You may have to make it larger just to fit the stuff in there without stuff colliding.
Wolf
Hello. Yes, you're right. I checked the space requirements and I need at least a chamber that is 32cm wide and 15cm deep, this is without considering the PR. Even if I wanted to mount the PR backside-out, I still need to make the front chamber bigger because the woofer itself won't fit as the existing 5 liter (net volume) chamber is 11cm deep. I tried increasing the volume until I could fit both the speaker and the passive radiator (mounted backwards), but I ended with a very narrow response although manageable if I can cross the speakers at 50Hz (15l+13l). I decreased the tuning to 40Hz.