I am currently designing a 6.5" ported subwoofer in WINISD. I am still playing around with the box dimensions just a bit, but inner dimensions will be about 23" x 7" x 5". Driver will be mounted on the 23x7 face, and the port must also exit on this face. I am looking to use PVC for port(s) because I already have it, it's easy to work with and elbows are readily available. I will need a 90 degree elbow so it can fire out the front but run the long way in the box to get the length I need.
Problem is... for this design a 2.5" port seems to be ideal with regards to keeping port velocity down. But the only PVC in 2.5" is conduit. Which is fine but the elbows for conduit have a huge sweeping bend and won't fit in my box. I modeled using two 1.5" ports and velocity seemed a bit high (26.5). Using two 2" ports the length is too big for the box (28.5")
So...I was wondering if I could use one 1.5" port and one 2" port?
Has anybody tried this?
How would I tune them?
Is this just asking for trouble?
Any other ideas for solutions?
Thanks
Problem is... for this design a 2.5" port seems to be ideal with regards to keeping port velocity down. But the only PVC in 2.5" is conduit. Which is fine but the elbows for conduit have a huge sweeping bend and won't fit in my box. I modeled using two 1.5" ports and velocity seemed a bit high (26.5). Using two 2" ports the length is too big for the box (28.5")
So...I was wondering if I could use one 1.5" port and one 2" port?
Has anybody tried this?
How would I tune them?
Is this just asking for trouble?
Any other ideas for solutions?
Thanks
I don't know the answer but I'm interested to discover what it is. I guess if you have two different pipe lengths then they can be tuned equally.
Where are you based? I'm in the UK and used this supplier for the port on the Heliums:
2 1/2 Inch 90 Degree PVC Elbow for Pressure Pipe Systems
Where are you based? I'm in the UK and used this supplier for the port on the Heliums:
2 1/2 Inch 90 Degree PVC Elbow for Pressure Pipe Systems
In my limited experience your quandary is irrelevant. You are trying to squeeze and extra 0.5dB out of a system, a task than can be achieved by closing the front door and putting away the washing. A single 32mm port 300mm in length will do the job.
I am located in the US
Surtsey - I really wasn't looking at it from a "squeezing out db" perspective. Just going with what WINISD is recommending. And a single port of 32mm (1.25") seems way too small for this subwoofer. I have limited experience, but from what several others have recommended we should keep port velocity down somewhere below 17. I have heard others say 25 is fine. But then there is the issue of port compression as well. I can easily fit a 2" port with an elbow...so if anything I would just use that instead of a 1.25"
Any other feedback out there?
Thanks
Surtsey - I really wasn't looking at it from a "squeezing out db" perspective. Just going with what WINISD is recommending. And a single port of 32mm (1.25") seems way too small for this subwoofer. I have limited experience, but from what several others have recommended we should keep port velocity down somewhere below 17. I have heard others say 25 is fine. But then there is the issue of port compression as well. I can easily fit a 2" port with an elbow...so if anything I would just use that instead of a 1.25"
Any other feedback out there?
Thanks
Don't mix them!
I have mentioned many times on the forum before.. Port compression and noise are not an issue if you design it properly up to 35 m/s.
A few basic rules:
Large flares at the end (flare diameter twice of port diameter minimum. For a 32mm ID pipe would be 16mm+ radius round over bit, 20mm good)
All bends to have generous inside radius (1.5D minimum)
All cross sections to be circular
NO restrictions to flow in duct, e.g. grills / vanes
NO geometry within 2D of either entry or exit of port
I have mentioned many times on the forum before.. Port compression and noise are not an issue if you design it properly up to 35 m/s.
A few basic rules:
Large flares at the end (flare diameter twice of port diameter minimum. For a 32mm ID pipe would be 16mm+ radius round over bit, 20mm good)
All bends to have generous inside radius (1.5D minimum)
All cross sections to be circular
NO restrictions to flow in duct, e.g. grills / vanes
NO geometry within 2D of either entry or exit of port
zerokelvin99 - I am not disputing your recommendation but curious the reason why they should not be mixed? Is there a specific reason of some kind...like iit will create unwanted distortion etc?
I have a 3/4" round over bit that I can use on both ends of a port...might not be the flares you suggest but much better than a completely flat ending.
As far as radius of the bend it will not be a stark 90 degree kink, it will have some smoothness to it. I am working with plumbing parts from the hardware store so essentially limited by what they carry and what will fit in my enclosure.
I guess I would be looking at either two 1.5" ports or a single 2" port. Still I am a bit cautious about a single 2" port because the velocity is about 30 m/s in that case
I have a 3/4" round over bit that I can use on both ends of a port...might not be the flares you suggest but much better than a completely flat ending.
As far as radius of the bend it will not be a stark 90 degree kink, it will have some smoothness to it. I am working with plumbing parts from the hardware store so essentially limited by what they carry and what will fit in my enclosure.
I guess I would be looking at either two 1.5" ports or a single 2" port. Still I am a bit cautious about a single 2" port because the velocity is about 30 m/s in that case
Hornresp can simulate it using ABC aperiodic bi-chamber topology, setting internal port to enormous section almost not lengh, and setting chamber 2 to 0,01cm²/LC2 to 0,01cm2. It works. Hornresp show higher velocity for shorter port, and different resonnance profile from both port.
But hornresp doesn't take account for loose dues to increased airflow wall resistance.
But hornresp doesn't take account for loose dues to increased airflow wall resistance.
Apologies, I seem to have misread your first message! For some reason I thought you were looking at changing port lengths, not diameters..
Re, your original question:
One of each (1.5"+2") should be technically fine. Maybe upsetting to look at, but that's about it. So long as you maintain the same open area as the sim, it should match up more or less.
Otherwise, I would say to go for a single 2" rather than two 1.5". The additional area increase for the 2x1.5 is minimal but wetted area is up significantly (a major factor in noise generation). 30m/s with sensible design is OK in my experience.
Re, your original question:
One of each (1.5"+2") should be technically fine. Maybe upsetting to look at, but that's about it. So long as you maintain the same open area as the sim, it should match up more or less.
Otherwise, I would say to go for a single 2" rather than two 1.5". The additional area increase for the 2x1.5 is minimal but wetted area is up significantly (a major factor in noise generation). 30m/s with sensible design is OK in my experience.
papasteak - you'll have to excuse my ignorance but I'm not quite sure what you are saying. I think you are trying to say that using different diameter ports will affect the airflow in each as well as resonant frequency. Not sure if that would have any audible negative consequences.
Zerokelvin99 - can you explain what "wetted area" is? I am not familiar with that term.
The additional area increase for the 2x1.5 is minimal but wetted area is up significantly (a major factor in noise generation).
I am thinking it is the perimeter of the port(s) vs the area. For instance, two 1.5" ports have a combined perimeter of about 9.42" and combined area of about 3.54sq.in. whereas a single 2" port has about 6.28" perimeter and 3.14" area. Ratio of perimeter to area is 2.66 vs 2.0, thus the 2" port provides more area per perimeter. Yes?
I have a commercially available sub that I bought used that has a 12" driver and a port that is no larger than 2.5". By no means is that likely to be ideal, but it sounds fine up to a certain point and then I can hear some port noise. I guess a 2" port with a 6.5" driver should at the very least do no worse than that 12" sub. I will probably just not complicate things and go with the 2" port and just make my life a lot easier for once
Zerokelvin99 - can you explain what "wetted area" is? I am not familiar with that term.
The additional area increase for the 2x1.5 is minimal but wetted area is up significantly (a major factor in noise generation).
I am thinking it is the perimeter of the port(s) vs the area. For instance, two 1.5" ports have a combined perimeter of about 9.42" and combined area of about 3.54sq.in. whereas a single 2" port has about 6.28" perimeter and 3.14" area. Ratio of perimeter to area is 2.66 vs 2.0, thus the 2" port provides more area per perimeter. Yes?
I have a commercially available sub that I bought used that has a 12" driver and a port that is no larger than 2.5". By no means is that likely to be ideal, but it sounds fine up to a certain point and then I can hear some port noise. I guess a 2" port with a 6.5" driver should at the very least do no worse than that 12" sub. I will probably just not complicate things and go with the 2" port and just make my life a lot easier for once
If one port has higher airflow, it will chuff more than the other, compress more. Different port in a same volume share the same tuning frequency determinned by the full system. I talk about higher frequency first resonnance of port due to port lenght. It's a nasty spike hard to avoid, often in the 100-1000hz range. Not all simulation tool are enough advanced to show this. Easy to see measuring response at port output. Contrary to tuning frequency, those resonnance only depend of port lenght. So with multiple port they should be spread over different frequency. The other way to deal with it, is using a stub to damp at least first resonnance by tuning stub frequency. I got a 1khz resonnance in my actual design. It will ne easy to add a stub to damp it. The higher frequency, the power volume is needed to damp resonnance. Playing with slidders in hornresp is so convenient to see those effects in real time.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.