Good Response Curve?

[timc31610]

If WinISD has calculated any parameters from previous parameters entered, then do not change the calculated ones as you will get that error.

Krakin, you were right. I let it do it's thing and it worked!

[timc31610]

Joe, how is that measured? Is there an easy way or should I shoot for the flattest response?

If you take your box specs from BBP and put them into WinISD I can tell you how to roughly approximate cabin gain.

Ok, thanks to Krakin I got my sub in Winisd, any help you can give me would be great!

[mothra]

It is one of those things where you do not want too much pv because it will cause a whistling or chuffing sound from the port, but if it is too little you will loose a lot of air flow and reduce spl I believe. Someone please correct me if I am wrong. I try to keep my port velocity at the least between 35-45 ft/s.

I'm a big believer in keeping port velocities low to eliminate port noise & compression, that's why I'm curious about why someone would want it to be higher.

I do not claim to be any kind of SPL expert, but I find it very hard to believe that higher port velocities do anything to improve SPL. I've someone has any data to prove otherwise I'd be very interested to see it.

When I saw your 35-45 number I was thinking in meters/sec and I was like "wow, that high" but then I saw it was in feet so actually your number is pretty modest.

Just over the years I have always had good results keeping port velocity in the 20s. Yes PV in the 30s can and will cause port noise but I find that having a PV below the 20 have issues with overall output at all frequencies. Too low of a PV usually leaves the listener wanting more or with a very boomy sound which I feel is worse than port noise. To increase velocity just decrease the amount of port area.

I understand how to increase and decrease it. By 20 you mean 20 m/s right?

I was actually just answering the other guy and yes I do mean 20 m/s. also BBP will change the color of the curve when it exceeds 32 m/s b/c that's usually when port noise occurs.

[CleanSierra]

20-25 m/s is the absolute highest acceptable port velocity. And as far as too much port,, you still need to satisfy the driver's need to have port given the Xmax, Sd and tuning frequency. Keeping the port velocity low should be the practice.

As far as TOO much port, you just want to avoid first port resonance by keeping the resonance at or above 120 Hz. Two reasons for that is one, to avoid chuffing, or audible port noise. The other reason is to avoid port compression, which is when there is too much impedance to airflow caused by the port which can result in non linear performance at higher SPL levels.

Another thing to remember is that typically those rules of thumb when modeling subwoofers and enclosures in Win, the end goal is usually a decent bandwidth and a bump around tuning for a normal bass reflex style box. Cabin gain changes it all very much, resulting in a much peakier response than what would occur in an anechoic environment.

We get greedy in the mobile environment and want a HUGE bandwidth and huge bump in output around tuning. The true nature of a normal 4th order(vented or bass reflex, or just normal ported box as most know it) is to have a good bump in output at and around tuning, followed by a fairly steep rolloff on the high and low sides of that tuning.

Edit- Sorry if any of this response is unclear, I just woke up and may not have put together as coherently as I could have I'm sure

[Joe X]

Joe, how is that measured? Is there an easy way or should I shoot for the flattest response?

With an RTA for example, the response wili be altered by sub location/orientation (etc) and is not possible to simulate in software by (almost) any means, the Linkwitz transform fix for WinISD may produce (or not) less ridiculous results, best bet for a "flat" response in a vehicle is sealed and I bet you don't want that like most of us, my take is build a box which you can change parameters a bit and experiment with it.

[CleanSierra]

I like ^^^^^ response honestly. Modeling MIGHT show you that your enclosure isn't crap, but it all changes so much per vehicle. Especially since orientation with sub, positioning of box, port direction/clearance, and cabin gain, the enclosure model and sub response changes a bunch.

[Triticum Agricolam]

Here is the link to how to roughly approximate cabin gain in WinISD. http://linearteam.proboards.com/thread/283

Like Joe X and CleanSierra have said, this method is only a rough estimate at best and what you actually get in your vehicle may be quite different and will be affected by things like placement, contents of your vehicle, etc.

While this method is far from perfect I do think its a whole let better than nothing though. We all know cabin gain has a big effect on how a system sounds and just because we can't predict it very accurately doesn't mean we shouldn't try to predict it and plan around it at all, IMHO.

[timc31610]

Thanks Joe, Cleanseirra and Triticum, I understand there are too many variables for any simulation to be completely accurate. I'm just trying to get an understanding of the basic process. I have built numerous ported boxes over the years and truthfully most of them were just wild guesses that luckily turned out to sound very good. I would like to remove at least a small portion of that luck from any future builds.