Passive Radiator vs Ported

Posted 11 Oct 2000 02:58:58 PST to the BassList

OUTDOOR GROUND PLANE MEASUREMENTS SHOW NO ADVANTAGE FOR PRs over PORTS or vis versa WHEN TUNED TO THE SAME FREQUENCY!!!!

This is not to say that PRs do not have certain advantages such as no pipe resonances and the ability to operate in boxes too small to fit the needed ports.

Read on if interested in how these tests came about and the details.

Early this year, because of the various comments on the list about the possible better performance of PRs vs PORTED, I started a thread to try to find out why. It had been my understanding that there was no significant difference if both were tuned to the same freq.

Well, as every one is aware the opinions were split! Since I owned a copy of BassBox6 and BB6 showed considerably better performance for PRs, I started an extensive correspondence with David Harris without conclusion. David could not suggest anything that could be added to the acoustic ckt that made any sense that would give the gain that BB6 was showing. He was not willing to give me the details of his model. David then said that he would do some measurements this (past) summer. (As of this moment, David is in the process of completing his measurements.)

Since those results did not seem to be forthcoming, Mike Thompson and myself decided to do the measurements ourselves. We first did some preliminary measurements on a box that Mike put together - a 12" Lambda driver and a 15" Lambda PR in a 3cuft box tuned to 20hz. The next day I made some measurements on a ported box I had floating around with a 12" Dayton 295-120 tuned to 26.7hz. The simulated results from BB6 for the Dayton followed fairly closely to the measured results.

The simulated results for the Lambda w PR, however, showed that it performed poorer than the simulated ported box tuned to the same freq and MUCH poorer than BB6's prediction for a PR. Of course my expectation was that it should perform to the ported simulation as what tunes the box, IMO, PR or PORT, should not matter. I played around with losses and the various T/S parameters to try and match the predicted PORTED to the measured to no avail.

Although these preliminary tests would not clearly answer the question as to whether a PR would outperform a PORTED because they were in different boxes with different drivers - the tests did give a very strong indication that there would NOT be any significant advantage to a PR over the PORTED. The preliminary tests did however simplify the necesary test to prove one way or another the PR vs PORTED question. Originally, I planned on using an 8" driver and going all the way to dual 15" PRs to get a very large differential in driver area to PR area which now appeared totally unnecessary.

Mike did the dirty work and built a 2.8 cu ft box that was capable of handling a 12" driver and 1 or 2 PRs or PORTS. The ports and the driver were mounted outside of the box so that no volume was taken from the box. No stuffing was used and both the LAMBDA and Dayton drivers were used. As indicated above, Mike and I did outdoor ground plane measurements at 3 meters to guarantee full summation of outputs. I happen to be the only house on a cul-de-sac in So Ca. so I can do outdoor measurements all year round.

For PRs we used nominally 2-15",700gm and 2-15",1000gm also loaded to 1400gms. One 700gm PR or 2-1400gm PRs and 2-2" ports tuned the box to 20hz to within a few tenths of a hz. All the systems tuned to 20hz, PR or PORTED, had their responses fall esentially on top of one another within a few tenths of a dB down to 20hz. Although there were small variations in box volume due to the way the PR holes were blocked and the ports were added, this was not considered significant in the scheme of what we were trying to determine. In fact, reviewing the simulation shows that if we had maintained exactly the same volume, the measured results would have matched even closer!! We have not covered every aspect in precision detail, but the detail is clearly sufficient to say, response wise, there is no advantage to a PR over a PORTED system.

Mike has put the results together in a couple of graphs and we would like to post them on a website if someone is willing to offer the use of their site. I had hoped that David Harris would have been interested in releasing this info - he was given the first chance but didn't seem to interested as he felt that his measurements showed some benefit below 20hz. Our data shows good tracking below 15hz. Below that, noise becomes a problem and data is unreliable even though I am using 30dB bass preemphasis/deemphasis and integration of 3. We could gain 10dB by measuring at 1m instead of 3m, but I would not be as comfortable with driver/PR/port "integration".

Hope this is helpful.

John Koval/Mike Thompson

And a follow up from John on 12 Oct 2000 00:52:42 PST

I will try to respond to the comments (surprisingly mostly positive) to our PR vs PORTED study.

I would like to reiterate that a PR does have benefits - just not in response compared to ported. Again these benefits include less midrange "noise", no pipe resonance, no port "huffing", easy to tune, better in very small boxes, and of course as Phill Abbate points out - higher cost!!

************* 1) What about in room?

Considering the wavelengths involved, I would expect similar results in room. However, I am not an expert in how things affect radiation impedance. Mike left the box with me and I will at some point do a hopefully valuable but obviously limited study on PR vs PORTED in room. Mike has already stated that he did not see any gain in 2 PRs over 1 PR in room.

************ 2) Did you try to examine possible non-linear behavior at different SPLs, or compare different sized ports?

No. For what we were trying to prove, port size would not be important. PRs MAY have less audible distortion to reasonably designed (low mach #) ports, but I've not had an occasion to evaluate that. Most of my experience has been with ported systems and can't say that I've noticed any problem on music - on video explosions I don't know that I could tell without a direct comparison.

************ 3) Did you try any EBS alignments, tuning significantly below the maximally flat alignment?

The answer is yes as the system was tuned noticably below what would be optimum for a 4th order Butterworth response. That tuning freq would be in the vicinity of 35hz and also provides an F3 improvement of ~25hz!!! but with about 7dB less output at 20hz and about 9dB less at 10hz. This is according to BB6 for the ported system.

************ 4) John E. Janowitz said: "This also leaves me questions as to the Contrabass, where when modeled as a ported system will show to have 6dB less output at 16Hz than was measured in the PR system. Nick McKinney modeled the Contrabass with PR's in BB6 some time ago, and the results showed to correlate almost exactly what Tom Danley measured as the response."

I can't comment on the Contra. However, I know there is one in So Ca as it was at the last Bass List get-together. I would be very happy to have an opportunity to measure it if it were volunteered. It would take about an hour. I live near Santa Ana in Orange county.

************ 5)John E. Janowitz also quoted Tom Danley as:

> "Speaking of measuring, I do not splice port curves to driver
> curves etc. to get the response, in fact I generally do the
> measurements on big stuff somewhat differently than usual. The
> problem is that a large enclosure occupies some of the space
> within the 1 meter radius distance your measuring "from the
> source". Also while the contra is not a big box (abt 7 cu/ft
> internal), it does have a 15" driven and 18" passive on two
> opposing sides making the actual center of the acoustic source
> "non obvious" initially.
> What I do generally is measure out 2 or 4 meters from the
> geometric center........"

We did exactly the same for that reason. We measured at 3+ meters (exactly 10dB over 1m, 3.16m)

********** 6) Bearlabs said: >snip< "Personally, I have never been able to get a port to tune properly 1/2octave below where it 'ought' to be set - and have any decent output. PRs do this with absolutely no problem. Ports seem to "want" to tune to that "natural" frequency, or pretty close to it.

Also your results of the tests at least so far as the PR system is concerned will vary quite a bit depending upon how well you physically sealed the PR system, ie. the QL value..............."

First, the box was well gasketed and sealed.

I don't know what this "natural" freq is but the only limit to tuning with a port that I've found is what I can fit in a box including using elbows.

John

Bear sent me this email some time later:

Date: Mon, 25 Feb 2002 13:34:09 -0500
From: bear 
To: drees@greenhydrant.com
Subject: More input...

Vis-a-vis my post that is quoted on your webpage:

By "natural frequency" I meant to imply the tuning frequency that is close to
or at the "optimal" frequency found by standard simulation packages - usually
something close to a bessel or butterworth. This means in specific that the
F3 point is set to be at the point where it yields a fairly flat response, or
*above* that frequency, where it yields a  + bump in the response typically.

Whereas, you can tune a PR to a full octave or more below the Fs of the driver,
regardless of box volume, and get *it* to pump - of course the response will
not be flat, but that's not what I was getting at.

(why someone would *want* to do this is another question to ponder)

Tuning a port this far below the Fs of the driver/box system (the natural
frequency, as I've referred to it here) seems to be much more problematic.
My experience trying to get this to work worth a damn has been very frustrating,

since the port doesn't want to move in frequency very much in this case. I
suppose
that it might fly if one added many feet of port - but would that then still be
a port or a TL/Labrynth??

Anyhow, it is trivial to tune the PR to any frequency you fancy, without regard
to the driver and box, which is (I guess the main point I was trying to make).

As far as sealing, I don't want to reveal all my secrets, but there is sealing
and
then there is SEALING. The Lambda - with the pole peice - can NOT be fully
sealed. This *does* make a difference in the operation of the Passive. Losses
are still losses, and I have found substantially better results when some really

small holes and leaks are sealed. I wrote an article in the 80' for SB mage that

covered some of these points. The cover has an article on Klipsh, FYI.

Hope this clears up some of what I was getting at...

            Regards,

            _-_-bear

It sure does! -Dave

drees@greenhydrant.com	Home
Last Updated: Mon Feb 25 11:11:29 PDT 2002	www.greenhydrant.com

Passive Radiator vs Ported - The Real Story