They all look exactly the same in my slicer Simplyfi3D. Something about opening it as metric must be the issue. can you leave it in inches then scale it up? 2540% if mm or 254% if cm.
Does anyone know what the default Q is for shelf filters in Soundeasy? I need to make sure it matches miniDSP.
EDIT: SE also lists options of -6db and -12db for shelf filters (actual gain is adjustable so I don't know what this refers to), but miniDSP doesn't list any db value. Thoughts?
EDIT: SE also lists options of -6db and -12db for shelf filters (actual gain is adjustable so I don't know what this refers to), but miniDSP doesn't list any db value. Thoughts?
Last edited:
MiniDSP uses a slightly odd definition so there is a good chance it does not. A easy way to check is with a loopback measurement of a known EQ.
REW has the option to set it's EQ to MiniDSP then you can compare the traces and get verification.
Eyeballing the plot of SE vs miniDSP, it looks like miniDSP is the -6db version of SE's shelf filter, and Q=.5
I forgot I had went through all of this years ago, but couldn't remember the specifics
I forgot I had went through all of this years ago, but couldn't remember the specifics
First order shelf with Q=0.5 is what MiniDSP use as far as I remember which matches your plots.
@augerpro
In the above SE graph- What have you implemented?
SoundEasy's "+/-6dB/oct Low-pass shelving filter" ?
If so, then it is equivalent to miniDSP's "Low_Shelf" function with a Q=0.5.
For REW users, REW accurately models miniDSP's Low_Shelf and High_Shelf filters if one uses the "LS Q" filter function.
REW's "LS 6dB" and "LS 12dB" equalizers do NOT closely approximate it.
In the above SE graph- What have you implemented?
SoundEasy's "+/-6dB/oct Low-pass shelving filter" ?
If so, then it is equivalent to miniDSP's "Low_Shelf" function with a Q=0.5.
For REW users, REW accurately models miniDSP's Low_Shelf and High_Shelf filters if one uses the "LS Q" filter function.
REW's "LS 6dB" and "LS 12dB" equalizers do NOT closely approximate it.
First order filters don't have a Q factor by definition. That counts for any first order filter.
For shelving filters, there is a 2nd order shelving with a Q=0.5.
If memory serves well that's equivalent to a first order shelving filter.
Just a little practical tip, instead of eyeballing we can just trace both in VituixCAD and plot both 🙂😎
Exactly. Orange is traced from Brandon's graph... there's some imprecision because SE Y axis is 90dB... but they're essentially the same.
Yes, I meant setting Q=0.5 in MiniDSP makes a First order shelving filter which is their default.
well then I stand by what I commented on previously.
And those 2 images you posted in #2105 ARE equivalent.
@augerpro : I am considering a waveguide for my near field monitor (1-1.2m distance to listener). The tweeter is sb26cdc. Since the midrange is a BMR driver, TEBM46C20N with wide directivity, off-axis 45 degree go down -1 dB at 3kHz, -2 dB at 4kHz, -3 dB at 5kHz, it does not match directivity of sb26cdc either at stock form or your 4” elliptical waveguide. My speaker is for near field, where ctc should be smaller than for midfield listening, and I also want to match directivity of mid and tweeter, so I would like to cut the waveguide from 4” wide to around 2-3” wide, and scale down the height of waveguide also. I am new to waveguide, so I would like to ask if this approach is feasible or the new waveguide needs to be developed? Thanks in advance.
You don't need directivity matching, or lower crossover possibility or higher SPL with this tweeter and this midbass. Why would you then use waveguide and try to shrink it? I mean they are matched better without waveguide as you noticed...and you can cut plastic frontplate a little bit in order to bring ctc closer...
I need waveguide for directivity matching since SB26CDC in stock form at 45 degree off axis has 0db at 3 kHz, -1 dB at 4 kHz and -2 dB at 5 kHz, so there are mismatch with the BMR midrange. Small waveguide can correct that, it is just that 4” waveguide is not. Secondly, waveguide can reduce diffraction at HF and that is what I also want. Thirdly, even without waveguide, the ctc between mid and sb26cdc will not be smaller than wave length/4 to avoid lobing completely, so while I am trying to make it as small as possible, ctc is not a deal breaker.
I have to say now when tried both versions , phase shield version is a must. It's a night and day difference.
What you need is a shallow waveguide, rather than a smaller diameter. You could try scaling it, it's easy to print and test right?