B&C provide plans for a coaxial wedge using the 12HCX76.

http://bcspeakers.com/media/W1siZiIsIjIwMTkvMDUvMDkvMTJfNDVfMzhfODYxX01DMTIyQU4ucGRmIl1d/MC122AN.pdf" rel="nofollow - https://bcspeakers.com/media/W1siZiIsIjIwMTkvMDUvMDkvMTJfNDVfMzhfODYxX01DMTIyQU4ucGRmIl1d/MC122AN.pdf

http://bcspeakers.com/media/W1siZiIsIjIwMTkvMDUvMDkvMTJfNDVfMzhfODYxX01DMTIyQU4ucGRmIl1d/MC122AN.pdf" rel="nofollow -

Is this tuned slightly higher than might be ideal for a dance music monitor? Or is it giving everything the 12" has low down? I was thinking of putting a 15" sub underneath it anyway but just wondering if this wedge plan is a good starting point or whether to look at the design from scratch.

Second question (and forgive the potential ignorance here). In terms of the tuning, do I understand correctly that the main component is the port size in relation to the cabinet volume? So if the volume remained the same and the port size remained the same but the cabinet shape was changed the tuning should in theory be the same?

The only other thing that struck me was the ports are just cut outs in the front baffle of the cabinet, there's no extension to them. From memory believe the baffle on our old Res 1's was similar but just wondered if this was simply how the ports work at this frequency range or something else that I'm missing (I've only built subs before).

This will be driven 2 way active.

A port that is simply a cut out on the baffle still has depth, the thickness of the material. Provided that air velocity remains acceptable then this is perfectly valid, reduces construction costs and complexity etc. If air velocity is too high, port cross section can be increased by also increasing depth whilst tuning remains constant.

Putting the driver and cabinet into WinISD though, I get port lengths that don't connect with the B&C reference design - WinISD is suggesting a vent length of 15cm whereas in the plans it is the depth of the baffle - 15mm

Am I missing something?

I've found this thread which mentions triangular ports being reduced by ~10% from what is calculated to account for the position of the port, but not a 10x difference

https://forum.speakerplans.com/reflex-design_topic92067.html" rel="nofollow - https://forum.speakerplans.com/reflex-design_topic92067.html

Have you tried simming it with the same volume but a port of 15mm to see what it does to the tuning?

Accounting for a 10% difference in port length due to triangular ports brings the tuning down to 105Hz

Still not a full range box by far with 60hz at -12dBs. You will need some subs to get a proper dj monitor. But subs are also excellent stands!

Look also at 18sound monitor enclosure and at the "smithers wedge" plans online.

Link should work.

While a traditional wedge was more aimed at piercing tru the mix in the midrange for voices and instruments to be heard by performers, a dj monitor needs to be more fullrange, you don't need to feel the 30hz wobbles but you need the boom and the tiz for mixing. 

Classic wedges rely also on being on floor for 2pi loading, this rounds the bass, if you plan to use this at ear height this has to be compensated.

So if you need to use it standalone for dj monitors  you need to tune lower than regular wedges, and loose some efficency. 

Using separate subs will allow to compromise less on the midtops and put less strain on them, so they will sound better at same level.

Crossover will play a crucial role, you need a specific one, or dsp and amplifiers for every channel. Proper designed filters, either dsp or passive, will result in better feedback control, this is a monitor after all.

Either build longer and cut or start short and add pieces until you are happy. 

Any guide point on where to start with the tuning frequency? Adding is probably going to be easier - especially with triangular corner ports - can just take the driver out and glue, pocket screw and clamp more sections on

Once you're in the business of tuning the port length you may as well modify the design to whatever you need. I would use tube ports that are easier to adjust the length

That said, I feel I'm a little out of my depth on a few counts:

1. Tuning frequency, Fb, what should this be for a full range box? Taking this any lower than 50Hz seems to produce excessive port lengths, or when increasing the port CSA to decrease the lengths, port velocities in excess of 20m/s. In most scenarios this will be run with a sub but it would be good to use it standalone as well, does this have any impact on the choice of Fb?

2. Box Volume, is 50l a good starting point? Most designs in this area appear to be around 35-50l for a 12" mid/top box. Reducing the volume increases the port length and decreases the LF response.

3. Does WinISD take driver + port volume into account, or do you need to add those onto the cab volume when building

4. Do I need to be worried about the apparent exceeding XMax at rated power (350W)

3) nope, got to add driver, ports and handles before calculating box dimensions. 

4)yes, but add your high pass filter in and it should be fine.  Tickling the xmax line is ok, sailing way past isnt. 

5) check your port speed, there's a tab for this,  it should be <30m/s to avoid port noise

 

PS: In your plot, the HPF in place is definately to low, set it to 45 Hz and max excursion should be far less.

Re your point 1, this depends a lot on who will be using the box and their expectations. In general, as long as they can hear some bass it should be ok, but you can easily do some experiments for yourself.

Playback some typical program material that you think the box will be used for, and add in a high pass filter at various frequencies to see just how high you can cut it off and still hear when the kick is happening. You can do this easily on any digital sound desk these days or in software - Audacity is free and can do it no probs.  (Think of this stage as "virtual prototyping" if you want )

As you're finding out, trying to get the LF flat down low in a smaller box leads to a cascade of other compromises, so I'd recommend accepting the highest cutoff (& thus tuning frequency) you can get away with.

Remember a monitor (in a live gig setting) isn't first and foremost about pristine hifi sound quality; rather it's about making sure the performer can hear what the need in order to deliver a great performance. Hearing when the kick is happening makes perfect sense for a DJ monitor, feeling the full impact of it, not so much.

My hunch is that somewhere in the 60-70Hz range as suggested by smitske96 will turn out to be better than trying to get down into the 50's.

HTH,

David.

Couple of things 

1/ will the baffle will need bracing or is the cab sufficiently small and rigid to not need it? 

2/ do I need a gasket on the mounting flange?

Pics of dry fit attached - need to domino the top and rout in the handles and speakon plate before I can glue it up and line it.

It's driven actively via a DCX2496 and then both HF and LF are running off two channels of a NX4-6000

LF 0dB 44Hz 24dB Butterworth - 1.21kHz 24 dB Butterworth

HF -8dB 1.21kHz 24dB Butterworth - 20kHz 24 dB Butterworth

Expected tuning at this port length was 60Hz

Any thoughts?

Messed around with the various port lengths as well 64mm, 53mm and finally removing the extensions and just using the 18mm baffle thickness. Also ran a Genelec 8030A I had handy through the same test as a comparison (appreciate running a studio monitor vs. a PA speaker is not a like for like test).

Noticing quite a bump at the x-over point - is this down to filter type selection or does it need moving or EQ'ing out? Haven't yet run the drivers independently with the same test.

Tou can move to linkwitz riley filters wich are -6 at crossover and sum flat to try remove the hump

Measure the raw response of each driver on axis

Measure the raw response off axis to check if there is an area where the attenuation off axis matches

Measure the response of a filter on the DCX - this will be your acoustic target response (HF first your case as filter will be near lower limit of HF driver) Pay attention too phase versus target response.

Use low 'Q' filters to flatten response of HF driver

Overlay target response of DCX filter (eg LR24)

Attempt to use EQ and filters to match the HF response to the target.

If not possible to match to given target then adjust target to make it achievable by iteration.

Follow same process with LF to match ideal HF filter response.

Pay attention to the phase response of the drivers compared to the DCX target response.

Use delay in DCX to get phase response of LF and HF to overlap neatly through crossover region

Remember that it is the combined acoustic response of filters and drivers that matters - not the filters alone.

Your measurements look a bit wiggly, I presume your not using time domain gating.  Try getting the speaker up on a tall pole and away from surrounding objects along with gating to get quasi ancoic results:

https://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements" rel="nofollow - https://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements

https://community.klipsch.com/index.php?/topic/182892-using-rew-to-determine-time-delays-between-drivers/" rel="nofollow - https://community.klipsch.com/index.php?/topic/182892-using-rew-to-determine-time-delays-between-drivers/

This won't cover bass measurements unless you have a very big pole.  The usual way to get those is to perform them on a ground plane or near field and then splice with the gated response.

The best way to see the box tuning frequency is to measure the box impedance minimum frequency which can be done using a power amp, a resistor and a multi-meter. 

New measurement attached with that enabled and LR24 filters on the crossover.

Will take it a fairly dead studio in a few weeks and get it on a pole.

If you could explain/link me to the method of measuring minimum impedance it would be appreciated

http://www.mh-audio.nl/calculators/impedancecurve.html" rel="nofollow - http://www.mh-audio.nl/calculators/impedancecurve.html

if you look at the impedance curve in the above link your aiming to find the minimum between two peaks.

1) connect speaker in series with resistor (8 ohm for example)

2) connect multi meter in AC volts mode across resistor

2) connect to power amplifier output

3) drive amplifier with sine tone generator (EG from pc software)

4) adjust the frequency until you find the first peak in impedance (local maximum volts reading)

5) adjust the frequency until you find the 2nd peak

6) between those two points find the minimum impedance (minimum volts) this is your tuning frequency.

You can confirm this by observing the cone when driven to visible excursion it should stop moving at the tuning frequency.  Be careful not to run at significant power for long in this condition as the coil is not moving so poorly cooled.

Listening test wise no noticeable HF issues, it isn't quite as bright as the Genelec, but then they do have their characteristic sound. The B&C has an impressive amount of clarity and detail for a PA speaker and the coax arrangement works brilliantly for a monitor.

While the genelecs should sound good with no processing, some lifting of the highs is usually done to compression drivers on most horns.

Consider applying a shelving eq, might get some brightness back, just test also at higher volumes, bright at low volumes could result harsh at high levels, as hearing sensitivity to highs increases with level.