Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jerry Turnbow]

Hey folks -

I got called in to help resolve some issues for a small church that has had their own DIY sound guy for quite a few years, who's put together a system using off-the-shelf MI grade gear - certainly not the stuff I was used to 16+ years ago when I worked for a system contractor.

The good news is that most of the "fixes" are fairly straightforward - changing the way his monitors are set up, a little acoustic treatment, splitting up some circuits, etc.

Their biggest problem is no GBF for the pastor's wireless mic.  After going through several purchases prior to calling me, they've actually got a pretty decent mic and wireless (EV, formerly known as "Telex" with RE97TX mic), but for monitors they're using the dreaded Galaxy Hot Spots (three of them) pointed at the dais where the pastor and several singers are.

The singers all use Shure wireless mics with the tried and true SM58 capsules, and are not a problem, but the boom mic on the pastor, being omni, combined with the peakiness of the response of the Hot Spots, is unusuable.

As I said, their main problem is simple - the Hot Spots right now are on the same amplifier and EQ as the mains, so they don't even have the option of pulling the Pastor down, or even out of the monitors.  I'll be correcting that with a separate amp, driven by a monitor buss, etc. for the monitors, and am wavering between putting a decent graph in or going with something like a Driverack or Sabine Navigator to try to go in and "tweak" the system to try to flatten out the response of the Hot Spots to the maximum extent possible.

Galaxy doesn't publish any frequency or polar plots for these things (and before you suggest it, replacing them is not an option at this time), so all the "tweaking" and measuring will need to be done on site, using Smaart or perhaps a simple RTA.  

Has anyone had any experience with taming these units down, and or done any response curve plots that I could use as a baseline to save time (and the client's money) in the field?

[Tom Young]

I have rather strong opinions about HotSpots, the class of mini spot monitors that employ more than one driver covering the same frequency range.

From a loudspeaker engineering standpoint this configuration is just plain "wrong". The manner in which the two drivers interact with one another is the textbook definition of destructive interference. Perhaps the greatest flaw, or operational roadblock, is that the response changes considerably as you move about the coverage of this loudspeaker. So even if you equalize one of these to have smooth response at one position, a few degrees in any direction from that measurement point will exhibit completely different response. In your case (with a "moving target" mic'd pastor) ....... all bets are off.

Here's a graph showing my measurements of the frequency response of a brand new Galaxy HotSpot, on axis and at 10-degree increments. Despite the poor picture quality you should see that the blue trace is the on axis measurement.



Note the 8kHz, 1/2-octave wide peak. This is probably what your pastor's mic is interacting with, along with its own presence peak.

The reason 2 drivers are deemed necessary is to increase acoustic output and especially at lower frequencies. In many 2-way fullrange loudspeakers with dual cones plus a HF device, no effort is made to compensate for the dual driver interaction. But in this type of 2-way design the destructive interference occurs over a narrow band of mid-frequencies because the HF driver takes over at some point.

In some very well-engineered loudspeakers of this type, they actually roll-off one of the two drivers (above the frequencies where they sum fairly well and add bass) so that the mid-frequency combfiltering is almost completely eliminated. Cool.

Not to completely rain on your parade: it will be worthwhile to equalize these as best you can. But when possible, torch them.

[Brad Weber]

If the pastor has the mic properly positioned then the increased gain allowed by it being close to his mouth should offset the cardioid pattern of the SM58's, especially if they are not well versed in keeping the null directed at the monitors.  I think that getting the monitors onto their own mix will definitely help, as will having their own processing.

The comments that they have three Hot Spots to cover the pastor and several singers and that these are fed the house signal indicates that they might be misapplied.  The Galaxy Hot Spot is intended to be a personal monitor located near the performer, typically on a mic stand.   As Tom noted, the Hot Spots have an interesting polar response, here are polars for the current model - http://www.galaxyaudio.com/pdf/HotSpotPolar.pdf.  The -6dB horizontal pattern is apparently around 30 degrees at 250Hz, 90 degrees at 500Hz, 60 degrees at 1,000Hz, 40 degrees at 2,000Hz and 20 degrees at 5,000Hz, so the pattern varies pretty significantly and they have have a very narrow horizontal coverage within which the response will not differ greatly even if you did EQ them on-axis.  The Hot Spots are also intended to be primarily a vocal monitor and if they are being fed full range instruments or music then the low end below 150-200Hz might need to be EQ'd out.

So while you might get things much better by breaking out the monitor send and being able to EQ the signal, the Hot Spots are not intended to be full range monitors or to serve several people, in fact the Hot Spot manual actually suggests there should be one per performer located within arm's length of the performer they serve.  It sounds like this is perhaps a misapplication of the product and that this would not change, which would likely limit the potential results.

[Jerry Turnbow]

Thanks, Brad!

I appreciate the polar plot on the Hot Spots.

I'm aware that they're intended for use as vocal monitors, which was the reason for considering the use of a Drive-rack or Sabine Navigator unit instead of a garden-variety 31-band graph.  That would allow me to tightly bandpass just the vocal range where the things want to operate, with true shelving filters, etc. and some parametrics to smooth out any peaks in the boxes, instead of trying to "fix" the problem by just pulling down a bunch of 1/3 octave filters, and maybe adjusting a LPF on the graph. And getting them on a true monitor buss would allow me to route only the vocals to them.

Do you know if there's a frequency response plot on that site as well?  I poked around, but got an error message when simply trying to access Galaxy's .pdf page.

The units are mounted on mic stands, and while perhaps not as close as recommended, they are pointing toward the null points for the handheld wireless mics.  They work surprisingly well, considering they're getting the same signal as the house, so I'm fairly confident that the situation can be made tolerable by splitting them up on their own buss/processor/amplifier chain.

Interestingly enough, the feedback that they get with the omni RE97TX on the pastor is extremely high-pitched.  I'm guessing 3-5K, which I wouldn't have expected.  Given the comments from other posts regarding these boxes, I'd have expected problems in the 800 Hz-1.2KHz range.  That being said, if they decide they can't afford to put a full-fledged processor on the monitors, I might be able to get by with a 31-band graph, and then split the pastor's mic at the direct out and into a spare channel, where I could use a different EQ for his monitor.

Thanks, everyone for your feedback (okay, pun intended there.)

Tom - I'm still interested in your frequency plot if you have it, unless you're referring to the polar plot that Brad sent.

[Ivan Beaver]

I would say the whole situation is not worth any discussion untill you get the monitors on their own eq and aux.

Any attempt to fix it or guess at the problem is just wrong.

Kinda like saying you have different size tires on your car and it doesn't steer or handle right. Don't discuss front end alignment, type of tires etc.  Get the same tire and start from there and you may not even have a problem anymore.

Same thing with your situation by running mains and monitors off the same eq and amp.

[Don Boomer]

If you're gonna use a Navigator why not just let the FBX filters set up in auto mode.  Then you can take a look at where the filters set ... maybe use the PEQs, maybe not.

You didn't say how you were using the Navigator in the system.  Can you dedicate one Navigator channel to the Pastor's wireless mic (you'd also get comp/limiting in the deal).  If money is a big deal maybe just use a single FBX in line with the wireless.

[Jerry Turnbow]

Don Boomer wrote on Fri, 11 April 2008 21:46

If you're gonna use a Navigator why not just let the FBX filters set up in auto mode.  Then you can take a look at where the filters set ... maybe use the PEQs, maybe not. You didn't say how you were using the Navigator in the system.  Can you dedicate one Navigator channel to the Pastor's wireless mic (you'd also get comp/limiting in the deal).  If money is a big deal maybe just use a single FBX in line with the wireless.

Thanks for the idea, Don!  I didn't know that Navigator would tell you what the filters "found" and locked on to.  I currently used the Driverack 260 on my rig, but have looked into getting the NAV 480 for my own monitor setup, so I don't have any hands-on experience with it at this point.

The idea behind using the Navigator would be to use the model without front panel controls, put it in the signal path between the monitor buss and the monitor amp, then dial in the curve to band-limit and shape the response for optimal use with the (ugh) Hot Spots, then take away the computer so the user can't putz with the settings.  I think there's a two-input version, where I could use the other channel on the house, or perhaps to shape the bandwidth and limit the feed to the rest of the church (cry room, vestibule, rectory, etc.)

I do own and have used the FBX solo units in the past - particularly to solve problems with acoustic guitars, but in my experience, they need to be "trained" each time for reliable use, and I'm not sure my client wants to go through that every Sunday, but they are great little problem solvers when I need them.

The other thing is, I'm not sure this client will spring the $$ for any system processor.  Their mode of operation in the past has been to go to GC and buy something inexpensive off the rack, so I may be limited to using an MI-grade graphic EQ to get the monitors separated, but even that will be a major improvement over the current setup.  I'd just like to propose to them a couple of options (good, better, best sort of thing.)

Thanks again for your input!

[Jerry Turnbow]

Tom Young wrote on Fri, 11 April 2008 01:38

Note the 8kHz, 1/2-octave wide peak. This is probably what your pastor's mic is interacting with, along with its own presence peak.

Tom - Thanks again!

I think you're correct in that I was hearing the results of that 8K peak being picked up by the pastor's mic.  My ear's still not the greatest at picking out frequencies, although as a violinist (and later, telephone systems engineer) for many years, for some reason I can usually sing and/or identify A440 pretty reliably.     (440Hz is the dominant of the two tones that comprise North American standard dialtone).

Looking forward to getting my new laptop and installing Smaart so I can retire my ancient Rane RTA and do some real measurments in the field.

Cheers!

[Ivan Beaver]

Jerry Turnbow wrote on Fri, 11 April 2008 22:58

The other thing is, I'm not sure this client will spring the $$ for any system processor.  Their mode of operation in the past has been to go to GC and buy something inexpensive off the rack,

Well it seems to be working well so far  and they do not have any problems with their current setup  do they?

Maybe they need to take some real advise, not what GC has sold them.

[Philip Roberts]

Here's a idea for you.

Install a high current relay that disconnects the hot spots during the when needed.

P&B now part of Tyco is a major manufacture, you'd be looking for a DPST relay a minimum.

Philip