Lockdown Experiments

doober wrote: I wouldn't say it sounds harsh. It sounds fine at a distance, but close to the top box is out of balance- too much upper mid which sounds like it could easily be tamed with some gentle EQ. I should have filmed a bit longer and moved the camera down lower before stopping, that would have shown the difference better. The lower box sounds much smoother at the same distance. The upper box output will be going over the heads of the first 2/3 of the audience anyway, how it sounds at a distance is most important. I can't work out why the upper mid would be more prominent close up, logic would suggest otherwise, that as the listener gets closer they hear less of the higher end due to the curvature of the exits and the smaller wavelengths. The analysis of my video is interesting, I never would have thought to do that. It prompted me to do some analysing. This is a transfer function of the audio from the youtube video used as measurement signal, and the audio of the same video straight from my phone as reference. Averaging is set to 5 seconds. One might expect it to be fairly flat... Next is single channel amplitude of the same music, from an mp3 player, averaged over 10 seconds I'm not sure what to make of any of this really.

The first thing that comes to mind is the dispersion differential between your High Mids & Low Mids. The High Mids may offer a narrower dispersion hence, the overwhelming amount of SPL in the High Mids as you move closer to the cabinets.

Since the bottom box sees two boundaries (the ground and the box sitting above it), it is not usual to have a box positioned between two boundaries to sound smoother.

Bear in mind you are using horns not direct radiators. The majority of Horns always leave their sonic imprint on the sound. So it is going to be very difficult, to achieve a very smooth response, in the High Mids, all the way to the Treble region without, a good dose of equalisation.

Best Regards,

Edited by Elliot Thompson - 12 April 2021 at 12:09am

Elliot Thompson wrote: The first thing that comes to mind is the dispersion differential between your High Mids & Low Mids. The High Mids may offer a narrower dispersion hence, the overwhelming amount of SPL in the High Mids as you move closer to the cabinets. Since the bottom box sees two boundaries (the ground and the box sitting above it), it is not usual to have a box positioned between two boundaries to sound smoother. Bear in mind you are using horns not direct radiators. The majority of Horns always leave their sonic imprint on the sound. So it is going to be very difficult, to achieve a very smooth response, in the High Mids, all the way to the Treble region without, a good dose of equalisation.

There is a huge dispersion differential, top box total is 5 degrees, bottom box is nearly 40 degrees. Both boxes are intended to be asymmetric with a gradual increase in dispersion from top to bottom. The original plan was to build a single box, this is how I designed the curve of the mid and high horns, with a gradual increase in curve. After considering the practicalities of a single box with 24 drivers in it, probably a 4 person lift and a very inconvenient size for van loading and storage, I decided to split it into 2 boxes.

I really don't mid using EQ within reason. The processors I use have 6 bands available per output, and a good selection of filter types. My aim is always to get good results with a maximum of 2 parametric and a shelf per output. I have found with other boxes that trying to use a lot of filters to flatten the response can end up sounding worse than an uneven response.

If you want to put the finishing touches on an excellent project, I would recommend putting away SMAART, putting away this thread and some of the baffling suggestions in it, and spending a few afternoons with cups of tea, your measurement mic and interface, the obscenely good REW manual and papers from Charlie Hughes’ Excelsior Audio site and/or the DIYRM Facebook page.

Start with learning what coherence is, how to window a measurement, how to spot room modes and reflections, and how to derive the minimum phase components of a system - those are the only bits you can fix with IIR EQ.

You can probably pick up a few things on measuring polar or directional data while you’re at it.

Your front room or garden in ground plane are fine for this. Once you’re done reading, you’ll be on the path to learning the most useful thing for box tuning - how to ask the analyser the right questions, and how to know what data that you get back is useful or not.

2 Years into the project....and.....they're still not finished! They have been out entertaining the masses though. They just need some grilles and a few little tweaks.







I'm very happy with them, and band engineers have been very complimentary. They seem to attract attention with their slightly unusual design on show due to the lack of grilles.

The first outing had a fair amount of cross wind. The mid did drift around a bit, as expected, but there was very little phaseyness compared to my previous setups. The coverage of the crowd was good but I do need to take more measurements outside to work out the best tilt angles for crowd coverage vs complaints from local residents.