EAW VR21 and VRS12 DSP Settings

Michael John

Michael John

Junior
Jun 25, 2011
352
9
18
Sydney, Australia
eclipseaudio.com
For anyone who's interested, I thought I'd share the DSP settings I've created for the EAW VR series speakers that I use. These cabinets were the best I could afford a few years back and when I got access to Lake processing (for reasons I won't go into here), I figured I'd bi-amp the VR21's to squeeze a little more power and quality out of them. (I know, for many people, these cabinets are on the cheap end of the scale. Please, no flaming. :)~:)~:smile:)

I'd welcome any comments or feedback on this. I've kind of made this up as I go.

Here are screen shots of the module.

Delay's
- 0.44ms to align the VR21 12" with the horn
- additional 5ms on the VRS12 to match the delay of the linear phase crossover used in the VR21
delays.jpg

Gains
gains.jpg

VRS12 2 x 12" sub.
- EAW recommended HPF and parametric.
- Linkwitz-Riley LPF to cross with the VR21
sub.jpg

Linear phase brick wall crossover for the VR21
- LPBW chosen to reduce lobing around the crossover region
- 5ms delay
- 1.2kHz somewhat arbitrary
xover.jpg

EQ for the VR21
- Created using pink noise and SysTune at about 1m.
- Note the notch at 3.6 kHz. (I could hear some weird resonance.)
- The highs are adjusted to taste in each venue using another overlay
eq.jpg

Best,
Michael
 
Re: EAW VR21 and VRS12 DSP Settings

Thanks for posting this Michael.
We have 4 x VR21's at my church but I've never been happy with the sound. When you talk about aligning the horn with the 12" does that mean you're bi-amping the cabinet? If so, how does that work? I don't think the cabinet has bi-amp sockets. I find the horn doesn't have much "sizzle". I just noticed it's a 1.4" driver which seems kind of unusual for this type of cabinet. I suspect the larger horns have a bit deeper sound.
Cheers, Lee.
 
Re: EAW VR21 and VRS12 DSP Settings

Lee Wright said:
.. When you talk about aligning the horn with the 12" does that mean you're bi-amping the cabinet? If so, how does that work? I don't think the cabinet has bi-amp sockets. I find the horn doesn't have much "sizzle". I just noticed it's a 1.4" driver which seems kind of unusual for this type of cabinet...
Click to expand...
Hi Lee,

Yes, I added Speakons as well as switches to bypass the crossover board completely. I can run them passive (1+,1-) or biamped (Low - 1+,1- and High 2+,2-). To me the sound is better and more consistent, especially vertically with the linear phase brick wall crossover. I agree the HF isn't very exciting but I don't have many comparison points. My JBL MRX512M's are a little flatter and higher, but I really like the BMS compression driver in the 12CN680 coax driver I made into a wedge. I'm considering changing the VR21 compression drivers for BMS's. They do a model 4554 1.4" ferrite.

What I didn't show was the limiter settings. I emailed EAW support but they wouldn't give me any information so I made some educated guesses. With this module and my limiter settings, I've run some fairly loud indoor events and some moderate outdoor events throwing about 30m.

Best,
Michael
 
Last edited:
Re: EAW VR21 and VRS12 DSP Settings

Having learned a lot more on alignment from Timo's blog and other sources, I checked VRS12/VR21 sub crossover and found I was way off...

The VR21 is a bass-reflex cabinet and, at least from theory, has a natural 24 dB/oct (4th order) phase rotation in the low end. Adding a 24 dB/octave Linkwitz/Riley high-pass at 100 Hz results in roughly 48d/oct (8th order) equivalent phase rotation in the low end.

The VRS21 on the other hand is flatter in it's phase at around 100 Hz and the 24 dB/oct Linkwitz/Riley low-pass at 100 Hz doesn't complement the VR21. From measurement, I found neither the phase nor the phase slope matched at all. (Thankfully it wasn't sounding too bad.)

From measurement and experimentation, I found that the following settings give almost identical phase, roughly between about 70 and 170 Hz. And, most importantly, it sounds better than before.

VR21
- Already has 5.44ms of delay from linear phase biamping. (See previous posts.)
- High-pass: 24 dB/oct Linkwitz/Riley @ 100 Hz.
VRS12
- Low-pass: 36 dB/oct Butterworth @ 90 Hz.
- Delay: 1.5ms
- Polarity: normal (i.e. not inverted)
 
Re: EAW VR21 and VRS12 DSP Settings

Hey Michael,

I know that linear phase for the masses is the Lake processor's thing, but is it possible to use one of those topologies without it being brick wall? I just hate to throw away all that output.

Another thought, have you done sound quality comparisons of your sub/top crossover if you remove two orders of crossover? 2nd order HPF is enough to equalize excursion, and would let you get your LPF on your sub (which is where I think the penalties are really paid, as all its effects lie within your bandpass) down to 4th order.
 
Re: EAW VR21 and VRS12 DSP Settings

Just a question Bennett, is there any penalty to only going 12db/oct on tops hpf? You mentioned it is enough to equalise excursion, but does it pay to be a little more cautious with limiter settings? With less of a rolloff, surely it's going to "excurt" further down lower, yes?
 
Re: EAW VR21 and VRS12 DSP Settings

Steve Anderson said:
Just a question Bennett, is there any penalty to only going 12db/oct on tops hpf? You mentioned it is enough to equalise excursion, but does it pay to be a little more cautious with limiter settings? With less of a rolloff, surely it's going to "excurt" further down lower, yes?
Click to expand...

Steve,

Well, there would be more excursion than you might have with a higher order crossover, for sure. But with a 2nd order HPF at or just above the F3 of the box, excursion an octave (or two) below F3 should have the same maximum as at F3. I like to use a slightly under-damped filter response here if possible, to add a little boost and compensate for the rolloff of the filter right around that critical frequency range. Considering how much easier it makes alignment with 2+ orders of filter removed, it's worthwhile to me. I find this especially true in subwoofer HPF applications, where the likelihood of large signals octaves below the HPF frequency is very small, and the shallower rolloff makes the box sound like it goes lower than it does. Perhaps I have just been lucky to tune boxes that mostly had excellent excursion control, but I haven't seen an overexcursed woofer in some time.
 
Re: EAW VR21 and VRS12 DSP Settings

Bennett Prescott said:
Hey Michael,

I know that linear phase for the masses is the Lake processor's thing, but is it possible to use one of those topologies without it being brick wall? I just hate to throw away all that output.

Another thought, have you done sound quality comparisons of your sub/top crossover if you remove two orders of crossover? 2nd order HPF is enough to equalize excursion, and would let you get your LPF on your sub (which is where I think the penalties are really paid, as all its effects lie within your bandpass) down to 4th order.
Click to expand...

Hi Bennett,

Re linear phase, the Lake boxes have Linkwitz/Riley magnitude equivalent linear phase FIR filters with various slope options. The beauty of these is that it's easier to tune a system (in terms of measuring relative delay and level). Once happy with the system, they can be left in place, or switched for the corresponding regular IIR Linkwitz/Riley filter. Depending on the filter slope/order, this may mean flipping polarity when switching.

Re using 2nd order on the top and 4th on the sub, good idea. I haven't tried it, nor done a lot of listening on any of my tunings. EAW recommend a 4th order HPF on the VR21 and it became my starting point. I have heard and read comments that >4th order filters can become audible and I guess I should listen more to the 36 db/oct. :)

Thanks very much,
 
Last edited:
Rob Timmerman said:
This due to the phase response of the IIR filter/delay of the FIR filter? Ringing of a particular filter implementation? Or something else?
Click to expand...

I find I like lower group delay more, but in this case I am speaking specifically about ringing. Even FIR filters ring, and the sharper ones ring more just like IIR filters.
 
Re: EAW VR21 and VRS12 DSP Settings

Here's a low-pass filter example at fo=200 Hz. The plots show the magnitude, phase and group delay for 4th order Butterworth, Bessel and Linkwitz/Riley filters.
Butterworth gives the flattest magnitude response in the passband.
Bessel gives the flattest group delay in the passband. (a. la. linear phase.)
Linkwitz/Riley is a cascade of two N/2 order Butterworths. (Hence the -6dB @ fo rather than -3dB for the N order Butterworth.

4th_order_filters.jpg
 
Top Bottom