how to calculate resonance in dipole H-frame?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'd like to minimize the size of my dipole bass arrays by using an H-frame but do not want to sacrifice clarity for increased distortion (ie a noticeable increase). I'd love to mess with designs, but have no idea how to calculate the resonance for the particular woofers (I'm using 2 towers of 4 RSS390HFs per side in a line array). Crossover point for these (crossing to a rather large ESL) would be in the 150-350hz range with 48dB slopes.

Im crossing these over to a diy ESL (~25" diaphragm + frame) and may experiment by using it as an extended modular baffle of sorts for the woofers.
 
Hi,

I´d suggest not to use a H-Frame but just an U- or an A-frame.The difference between those two is, that the A-frame slightly reduces the Fs of the driver, which might be useful in some cases and depending on the driver´s parameters.
Still though it´ll be very hard to have the resonance far enough above the working freq-range with 15" drivers. I used an array of 8x 6.5" drivers in A-frames of ~8"width and 8" depth and the resonance is just above 300Hz! I doubt that You might reach as high a resonance as You need in Your application without compromising other parameters too much.

In case You can afford it money- and placewise, You might be better off with a concept like the ML Statement. Their design is very straightforward, adding an array of 6.5" drivers in kind of a U-frame to fill the range between 250Hz and 60Hz and a Sub-tower with CBs below that.

It´s nearly impossible to combine such big drivers as Yours with a Panel without sacrificing seriosly on dynamics. The restriction lies within imo any(!) ESL not beeing too capable below ~200-300Hz and dipole basses with big drivers not beeing capable to play well up to this freq-range. If You´re restricted to a just 2-way-system a dipolar array-tower with drivers of max. 8" in U- or A-frame is capable to supply for high dynamics down to ~30Hz and up to max. 250Hz. That is how I did it and got excellent results.

jauu
Calvin
 
I recently built two H-frames for my 15" woofers A&D 15308.
An externally hosted image should be here but it was not working when we last tested it.


Inner width of the frame is 40x40 cm, cavity depth to each opening side is 19 cm. In this configuration I get a resonance "peak" (more like an end of a ramp) at 230 Hz (near field measurement black curve). Since it is not very pronounced the usable range of the H frame gets up to that point - unlike U- and A-frames, which usually "feature" much more pronounced resonance peaks. :D
An externally hosted image should be here but it was not working when we last tested it.


I compensate the dipole loss with a simple 6 dB LP and get the red curve measurement at the opening plane of the frame.

1 m into the room and with an additional 300 Hz/24 dB LP I get the purple curve (ungated and unsmoothed measurement already including some room response).
All dB values are only relative (no calibrated measurement) and for didactical reasons the curves have been moved to peak all at 230 Hz.

If you limit the depth of the H-frame to 10 cm (instead of 19 cm), you will loose 6 dB efficiency, but the resonance peak will probably move to 350 Hz.

I´m in no way arguing with Calvins recommendations. They look very reasonable. I just want to point out that H-frames are by no means as unsuited as Calvin appears to believe.;)

Rudolf
 
The ESL is extremely large (25" wide by 7' high) and should be able to handle a 150/200hz crossover point, but that severely lowers its power handling. The RSS390HF from reviews Ive seen are perfectly capable of extending extremely high in frequency with low distortion, so I thought why not take advantage of it? I am not deadset on crossing 300+ although it would be fun to mess with.

I am however pretty set on using the RSS390HF as
#1 I already have 2 of them
#2 I would have a driver covering 1 maybe 2 octaves to go between the ESL and RSS390HF and that seems like a waste of $$$
#3 2-way requires much less amplification
#4 the RSS390HF are cheap and have excellent distortion performance

The low distortion in the bass and midbass is what im striving for in this design, and will sacrifice hf extension if thats necessary. Is there anyway I can personally calculate the resonance, so that I can play with various designs and see what works best for me.

What exactly is an A frame?
 
Member
Joined 2018
Paid Member
Any idea what units of measure are to be used for the circled parts?

There is a mix of imperial and metric in the sheet.
 

Attachments

  • mathcad frame units.jpg
    mathcad frame units.jpg
    31.6 KB · Views: 361
I'd like to minimize the size of my dipole bass arrays by using an H-frame but do not want to sacrifice clarity for increased distortion (ie a noticeable increase). I'd love to mess with designs, but have no idea how to calculate the resonance for the particular woofers (I'm using 2 towers of 4 RSS390HFs per side in a line array). Crossover point for these (crossing to a rather large ESL) would be in the 150-350hz range with 48dB slopes.

Im crossing these over to a diy ESL (~25" diaphragm + frame) and may experiment by using it as an extended modular baffle of sorts for the woofers.

The "line" resonance of each side of an H-frame is a 1/4 wave resonance. So you measure the depth from opening to the middle baffle where the driver is located, multiply that by FOUR, and then calculate the frequency that corresponds to that wavelength, e.g. f=c/l where c is the speed of sound and l the line length, that will be where your resonance (peak) will appear. The shape (Q) of the peak depend on the aspect ratio of the line.

In your case where you have multiple drivers stacked one above the other within the H-frame there will likely also be an internal resonance along the longer up-and-down direction. I tried this once, and I did not like the multitude of resonances that were created. A single driver in an H-frame is easier to work with IMHO.

Although others have suggested using a U- or other "frame" I prefer the H-frame because front and rear line resonances are the same, and you can therefore treat both using EQ. This is not the case with a U-frame. The downside is that for the same LF output the H-frame must be twice as "deep" as the U-frame.
 
Last edited:
The "line" resonance of each side of an H-frame is a 1/4 wave resonance. So you measure the depth from opening to the middle baffle where the driver is located, multiply that by FOUR, and then calculate the frequency that corresponds to that wavelength, e.g. f=c/l where c is the speed of sound and l the line length, that will be where your resonance (peak) will appear. The shape (Q) of the peak depend on the aspect ratio of the line.

Doesn't the cross-sectional area of the cavity also have to be taken into account?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.