|
|
|
| Re: EAW UX8800s [message #242224 is a reply to message #216485 ] |
Sun, 02 September 2007 01:06   |
 |
 Langston Holland
Messages: 486
Registered: August 2004
Location: Pensacola FL |
Has No Life |
|
|
Geri O threatened to stop sending Dave hog meat for Christmas this year if EAW didn't get me one of those processors - so now I have one and I'm thrilled with it. To quote part of a note I sent to the UX8800 beta test list:
...Had an outdoor concert (the perfect subjective test environment) last night with a couple of extremely good vocalists with two acoustic guitars (Shane & Shane - contemporary/worship Christian), about 1,000 folks, 4 KF730 and 2 SB730's per side flown at a pick point of 21', used the "rev 2" KF730 load (it would be interesting to find out the diff. b/w rev 1 and rev 2). (Edit: Rev 1 had a limiter ratio of 1:1 for some reason. Rev 1 is no longer available with EAWPilot versions 1.0.43 and later)
Generally I expected what I heard, but nowhere near that extent. I didn't think a line array could sound that real. I normally do a little playback of some known tracks after setup to verify that everything sounds right - but I was so thrilled after hearing the first couple of seconds that I spent 3 hours out there just playing everything I could find. I heard stuff that I didn't know was in the recordings, low level backing vocal parts, low level percussion embellishment, early reflections that give one a sense of room size, reverb tails... The live music was the best I've ever heard - no EQ on vocals or guitars - just high pass filters and riding the faders and playing with the comps a bit. On the comps (tube based Summit TLA-100A's), when I was matching gains b/w having them inserted vs. bypassed on the console (Crest X-VCA), I heard a character change in the guy's voice (no compression applied) who unknowingly was helping me setup the comp - the high frequency was different - some kind of emphasis that Summit built into to the unit was there that I never had heard before and it wasn't subtle - it make me think about where that sound would be most appropriate and when it might not be. Stunning. Basically it was a similar experience to the first time I started playing with my NT29's and the wonderful Danley SH50's, except the KF730 is world's louder, much easier to deploy, etc...
BTW, the high frequencies of line arrays get squirrelly sounding pretty easily with a bit of wind whereas an arrayable/properly arrayed point source system can tolerate more wind in my experience. Of course the line array has other advantages the point source doesn't have. My point is that the event I've detailed above was a perfect evening with very little wind. Indoor events provide their own forms of interference with reflections and can hide much of the excellent detail available with this processor and loudspeaker system or any well designed time coherent loudspeaker system.
Thus, I don't always get that "hope the band is late so I can listen to my recordings a bit longer" feeling - but overall I think that anyone that owns an EAW loudspeaker for which the Gunness filters are available is crazy not to buy this processor. It is not hype.
It also is not something you can use on the KF730's without a bit of tweaking, specifically in the HF. It is too bright without a bit of downward shelving and you have to measure even EAW subs to get the proper alignment delay. I'm sure this will change with time, and at this point I'm so busy that I've only had time to align the KF730's lows with the Danley TH115's that I use most often. In the next couple of weeks I'll have to do the same with the SB730's due to gig that will use them and I'll post the results at that time.
A very cool thing this processor can do that seems to sound right is a Distance/Temperature/Humidity adjustment for the HF's. If you set the distance to 1 meter, it ignores the Temp and Humidity fields. This distance field is intended to be the distance to the closest listener that the bandpass will cover. You'll notice that that there is a reasonable amount of boost for these varying conditions to the point where the boost becomes a cut after a certain point because it's impossible to overcome the atmospheric attenuation of the HF's. Thus, EAW chose to boost the lower treble region at that point to keep the tone balanced for long throws and/or cool and/or dry environments.
The following screenshot is a comparison of the older MX8750 recommended Mid/High processing curve in white, the UX8800 curve in cyan, and my modified curve in light green. The MX curve resulted in a proper HF balance in my opinion and it will be interesting to find out why the stock UX curve has so much boost. Other UX/KF730 users should try the settings at the bottom of the screenshot:
The following screenshot removes the stock UX curve and adds an acoustic measurement at 50' from a single KF730 flown 21' above the ground in an open field with the microphone in the ground plane. This loudspeaker was the center of a 5 box array to add a typical amount of radiation resistance to the measurement. This acoustic measurement was made with the MX processor settings (white curve) and includes the low bandpass, whereas the processor curves shown are relevant only to the Mid/High bandpass. You can see the fine detail in the corrective EQ with the UX curve that will result in a much smoother acoustic output from the KF730. I do not yet have the time to make new measurements, but if you look carefully, you can see that every reasonably sized bump is dealt with. You'll also notice they added a warmth boost around 600Hz and a much higher cutoff that should allow this loudspeaker system to go louder cleanly. Phase is radically different and I'm looking forward to studying this when the insane crunch season settles down.
If you happen to use Danley's TH115 with your KF730's, add .74 msec delay to the KF730 and leave the TH115 delay at zero and you'll be set. This assumes you use an 81Hz 4th order LR highpass on the KF730 and 88Hz 4th order LR lowpass on the subs as EAW recommends for the SB730. This is a pretty high lowpass on the subs, but the KF730's need the help in that region. It's important to achieve a wide phase curve overlap when having this much magnitude overlap. Screenshot of the result:
Now for something truly interesting. I really don't like v6 of Smaart yet, but it does allow you to make pretty pictures of lots of curves all at once:
[Updated on: Sun, 02 September 2007 09:16]
God bless you and your precious family - Langston
Soundscapes <><
|
|
|
| UX8800 Limiters [message #242330 is a reply to message #242224 ] |
Sun, 02 September 2007 14:58 |
|
Langston Holland
Messages: 486
Registered: August 2004
Location: Pensacola FL |
Has No Life |
|
|
UX8800 limiter measurements using continuous sine waves. Bursts yielded higher peak voltages for the first couple of cycles as viewed on the scope but I didn't bother to try to interpret their RMS voltages.
To keep the amplifier specs from biasing the limiter's calculations downward, I lied to the EAWPilot software and told it the amplifiers were capable of 110v max with a 32dB gain into the specified loads. This translates to an 11dBu output from the processor, which as you can see below was not reached, though probably would have been with single cycle bursts.
All voltages (dBu) were measured at the output of the processor.
SB730 Greybox:
1. Rated at 1,200w RMS at 4 ohms with 6dB peaks.
2. 40Hz limiting begins approx. 5dBu (750w RMS).
3. 40Hz peak approx. 9dBu (1,900w RMS).
KF730 Rev. 2 Greybox:
1. Low rated 700w RMS at 16 ohms with 6dB peaks.
2. 150Hz limiting begins approx. 5dBu (190w RMS).
3. 150Hz peak approx. 9dBu (470w RMS).
1. Mid/High rated 350w RMS at 16 ohms with 6dB peaks.
2. 2kHz and 16kHz limiting begins approx. 4.5dBu (170w RMS).
3. 2kHz and 16kHz peak approx. 7.5dBu (335w RMS).
Based on the soft knee and conservative thresholds, EAW is trying to achieve a hi-fi sound at high SPL's by employing a bit of compression prior to driver protection limiting. I've never done this personally, opting for fairly hard limiting at RMS +3dB, but have read convincing arguments in favor of it. The best argument I've heard is the fact that even the best loudspeakers starting sounding ugly when driven near max RMS, much less their peak mechanical limits. Thus limiting the output of each bandpass before it becomes audibly distorted is a reasonable goal. It will certainly give the loudspeaker a longer life span, though it would be important not to rob the system of apparent SPL. The UX's max RMS output of 9dBu (470 watts) into the 700 watt RMS rated low bandpass did surprise me a little, but I can't argue with the sonic result.
I'm extremely impressed with the virtually inaudible limiting in the NT29 boxes and assume the UX8800 uses the same logic given the same guy designed it. After about 10 shows so far with the UX and KF730's, I'm very pleased with the system.
[Updated on: Sun, 02 September 2007 15:07]
God bless you and your precious family - Langston
Soundscapes <><
|
|
|
|
Welcome to the ProSoundWeb sound reinforcement forums, home of more than 50,000 audio professionals. 
Forum Home
Login
Forum Search
Register
FAQ
