Greg Oshiro

Amazing deal. Welcome to the forum. About the amp... The LaScala sensitivity spec is 105 dB-SPL, 1 Watt input at a distance of 1 Meter. Let's say you place the speakers 5 meters (16.4 feet) apart and the speakers and the listening position form the default (for me, anyway) equilateral triangle in plan view. So the speaker-to-listener distance is also 5 meters. The attenuation due to distance will be 20*log10(5/1)=14 dB, so 1 Watt into one speaker will be 105-14=91 dB at the listeining position. Since there are two speakers with different signals driving each speaker, we have power addition of 2 equal level sources: Add 10*log10(2) = 3 dB. Now we have 91 + 3 =94 dB at the listening position for 1 Watt into each speaker. Live music has peaks that are about 20 dB higher than the average level. Average level is what we measure with a typical Sound Level meter. This corresponds to a power ratio of 10^(20/10)=100. So in order to get all the peaks, we need 100 Watts of *peak* amplifier power. Power amps are rated with a sine wave, which has peaks 3 dB higher than average level. This is a power ratio of 10^(3/10)=2, so an amplifier rated at 100/2 = 50 Watts is what we need for super-clean audio at a level of 94 dB at the listening position. Now... If we turn things up so that we are asking the amplifier to deliver more power than it can cleanly deliver, the positive and negative peaks of the waveform will get chopped off, or "clipped". It turns out that human hearing can tolerate some short duration clipping, so we can make it louder and still be happy with the sound quality. How much louder is open to debate. YMMV. In the commercial audio world, the rule of thumb is: Power amp rating = 10*(power required for desired average SPL). 10 Watts for the current setup. I occasionally listen at 100 dB. At these levels, I'm concerned with emotional impact more than cleanliness. I expect I'm clipping my amps. This level would require a 200 Watt amplifier in the current setup, assuming we don't want to clip the amps. Most of the time I listen at about 90 dB, which means a 20 Watt amp would be adequate. So... How loud and how clean does your sound need to be? I would recommend an external audio interface for the computer. Internal sound cards tend to have lots of computer interference noise *which you will hear* with speakers as sensitive as the LaScalas.

If imaging is the primary concern, I suggest: The difference between left and right needs to be addressed. If the two channels are not identical within a few dB, imaging will suffer. Try spacing the speakers farther apart so the speakers and the listening position form an eqilateral triangle in plan view. Experiment with toe-in angle. The optimum angle may not be pointed directly at the listening position. Do the best you can to make the room symmetric about the center line. We're eager to learn what you discover.

deleted erroneous double post

You might discover that more attenuation is required. I recommend allowing space and connection points for series resistors followed by shunt resistors in front of the pots. As usual, "It depends ..." I think the input impedance only needs to be high enough so the DAC output does not distort. I'm not familiar with the HRT DAC. Maybe sending this query to their tech support folks is in order. If there is significant capacitance loading the output, as in long cables between the output of the volume control and the amps, lower output impedance will cause less high-frequency rolloff.

There are "standard" FIR filter types, but I think they have questionable value for speaker equalization and crossovers. The cool thing about FIR filters is that you can take a measurement of a speaker, do some math on the data and have a set of FIR coefficients that will flatten the *measurement*. Note that the set of coefficients is only good for the microphone position at which the data were taken. Another thing FIR filters can do is independently define the magnitude and phase characteristics of a filter (at the expense of signal delay). With analog and IIR digital filters, the magnitude and phase characteristics are closely interdependent (all-pass filters can change the phase and not the magnitude, but there are still limits). So anything useful for our purposes is significantly more complex than specifiying frequency, gain and Q for a bunch of PEQ sections, delays, LPF/HPF settings and gains as we do for analog and IIR digital speaker processing. Long story short, FIR filters take a lot more work. Break out Oppenheim & Shafer and prepare to generate a string of at least 512 seemingly meaningless numbers that define the FIR filter. If a DSP box is sophisticated enough to let us split an input signal and run it through an "A" FIR and a "B" FIR, each with a mute switch, and sum the filter outputs, we can do an A/B comparison in real time. Voicing systems would be a slow process. It seems more suited to evaluation by measurement than by ear, where real-time twiddling is the order of the day. Real-time twiddling of the frequency response of an FIR EQ filter probably exists in some software/hardware system, but I'm not aware of it. The IRIS-NET software will let you adjust some FIR filter parameters by dragging a point on a graph (on the FIR filter for the EV N8000, I haven't looked at the DX46). I'm kind of old school when it comes to EQ and crossovers, so my knowledge is limited here and I haven't had the motivation to research it. Maybe this thread will change that. Have you looked at MiniDSP? Low cost, FIR capable, other forum members have experience with it.

In my very brief dabbling in FIR filters I have discovered that the major difficulty with FIR filters is creating the FIR filter coefficients from the desired frequency (magnitude & phase) response. The filter coefficients are a string of numbers that represent the sampled impulse response of the desired frequency response. The string is usually some power of 2 (256, 512, 1024, ...) long. If you have a way of determining the set of coefficients that you want in an ASCII file, there may be DSP boxes out there that you could program with the ASCII file. I'm not aware of any software that can create the coefficent file other than the MATLAB scripts I've written, but I'm sure there's some software out there that's more user-friendly than MATLAB, but I haven't looked. How were you planning to define the FIR filters for your system?

I'll try calling EV tech support after the weekend. IRIS-NET does allow building a very limited set of filters. In the FIR dialog, un-click the "BYPASS" button and click on the (edit: (I think) "Define filter" "GENERATE FIR") button. You can make some brick-wall approximation low-pass, band-pass and high-pass filters, but no detailed EQ. I have only recently dabbled in FIR filters. I think building an FIR equalization file takes some serious math. What is it you hope to accomplish with FIR filters?

I just did some work for a speaker manufacturer client with the same concern. They wanted to import an FIR filter into an IRIS-NET controlled device. IRIS-NET needs a *.gkf file that appears to be a proprietary format. Hopefully EV will provide some kind of *.gkf creation utility. Unless, of course, the suits are trying to force the end-loser to buy EV speakers so the FIR is useful..... Who me? Cynical?

It sounds like a failed tweeter. Disconnect the tweeters from the crossovers and measure DC resistance with an ohmmeter. They should both measure the same within a few tenths of an ohm. This is not a definitive test as there are failure modes that do not change the resistance. Try swapping tweeters. Try swapping networks. Try driving the muffled speaker from the amp output currently driving the non-muffled speaker, etc. Swap components/channels from the known-good channel into the muffled channel *one at a time* until you hear a change. If the problem moves with the networks, re-capping *might* be in order. My AA networks are ~32 years old and the capacitor values (capacitance and ESR) measure OK. I'm not a fan of replacing film or paper capacitors just because of age. If a cap measures poorly enough to indicate replacement, it might be prudent to replace the other caps in the same network or stereo pair of networks.

I'm with tube fanatic and JJKIZAK on this one. I think the physical location of the wiring ("lead dress") is at the root of the problem. Does the hum occur if nothing is connected to the amp inputs? How about if RCA shorting plugs are connected to the inputs? Once it is certain that the amp (not the interconnection of the amp with other equipment) is humming, try this: Connect amp to speakers. Connect shorting plugs to amp inputs. With amp upside down, move wires around with a non-conductive tool (plastic ballpoint pen barrel?) and see if you can change the loudness of the hum. The lead dress as it stands now does not appear to be optimized for minimum hum, so it could be coming from any number of sources and pickups. The power switch wiring and the filament wiring, as mentioned by others are likely culprits.

The amp is stable *with the amp turned on*. When the amp is turned off, the power supply capacitors discharge causing the power supply voltages to decay toward zero. The operating characteristics of the transistors/IC's change because of the lower supply voltages, causing the amp to be unstable. My rig has Crown D-75 amps and I hear faint squeals/pops/thumps upon shutdown. I don't worry about it because the level is low enough that I don't think any damage will occur. The load seems to be part of the problem since the center does *not* squeal. If someone can post a schematic of the RF-7 and the center channel speaker (RC-7?), I'll take a stab at a fix. There are amplifier circuit tricks that might remedy the problem (I've never seen a schematic for an Outlaw 7125), but it would probably be a major hack. Many amplifiers have output relays that disconnect the amplifier circuit from the speaker terminals when the amp is shut down. This avoids the squeal/thump/whatever problem, but then there are relay contacts in the signal path... the amp designer has to pick your poison. It's amazing what we can hear with high-sensitivity speakers!

re: issue 1 Are the amps "about 8 feet away" ffrom the AVR and EQ? Is the "rack" where the amps are located? Is all the equipment powered from the same power strip? Is this the case with or without the subwoofer connected? re: issue 2 I'm a 2-channel guy, so this may reflect my ignorance. I would expect that the AVR does all the "correct" summing/crossover/delay processing and sends the result out the subwoofer output. What am I missing?

I don't understand the question. Is your concern the power or signal connection(s)? One thing about hum/gounding problems is that *everything* matters: physical proximity, AC power connections, whether or not the chassis are touching (electrically), physical position of interconnect cables etc. If you can hook everything up except for the (powered?) subwoofer without objectionable hum, then things have been narrowed down considerably. Once you determine which equipment connection causes hum, please provide manufacturer and model information so I can find the manual online. > Never install a subwoofer cable with the sub on. Especially with RCA connectors. The tip usually connects before the sleeve, so there is no ground connection until the sleeve connects.

Hum is gone, after removing audio cables from the back of the Onkyo. If your RCA >> XLR cables have 2 conductors plus a shield and are wired like this: RCA tip >> conductor A >> XLR pin 2 RCA sleeve >> conductor B >> XLR pin 3 RCA sleeve >> shield >> XLR pin 1 Disconnect the shields at the XLR ends and see if things improve. There is a possibility that a wire from the Onkyo ground screw to a chassis screw on the dBX will help when the shield is disconnected. It might also make things worse. If the hum is not sufficiently reduced, the sure-fire hum killer is a pair of transformers. I recommend Jensen Isomax. IIRC they have a version with RCA in and XLR out. Expensive but worth it.

I don't think you can get SPDIF or AES3 from an HDMI source. Something about copyright.... So you will have to go the route of multiple D/A and A/D conversions. All the more reason to get a 24-bit/96 KHz DSP speaker processor if you eventually get a digital processor.

The EBtech box has transformers in it. It might add some distortion. I would see if the hum can be reduced by other means first. Chasing hum can be an arduous task, but let's get started anyway: With all controls set to the "normal" listening settings, turn off the power amps and let them discharge for a bit. Disconnect the audio cables between the Onkyo and the dBX. Turn the power amps back on. Is the hum still there?

Mark-- I use the stepped attenuators so that the overall gain after the DSP is repeatable. With stereo tri-amp that's 6 gains that have to track with fractional-dB repeatability. There's no way the level controls on the amps can do this. I built the attenuators myself. SMT resistors (painful), a double-sided circuit board and an Electro-Switch 2-pole 8-position switch. Balanced ladder configuration at 1210 ohms. 1210 ohms was chosen because it is a standard 1% resistor value, it's higher than the 600 ohms that the processor can drive, but low enough that a reasonable length of cable on the output won't produce significant high-frequency roll-off at 20 KHz. I don't use an external DAC in front of the Yamaha processor. The SPDIF from the CD player or DVD player gets converted to AES3. The AES3 signal goes into the Yamaha processor. All processing is in the digital domain until the DACs at the Yamaha outputs. This way there is only one A/D or D/A conversion in the signal path. If we use a DSP processor with analog inputs the signal path is: CD (dig) >> D/A >> processor input >> A/D >> DSP processing >> D/A >> power amps A total of 3 A/D or D/A conversions. This is why I recommended the OP get a processor with a digital input *if his sources are digital*.

If your sources are digital, I would recommend a digital speaker processor *with a digital input* (AES or SPDIF). This will allow you to minimize the number of A/D and D/A conversions in the playback chain. In my experience, this makes a huge difference in sound quality. I also recommend a DSP running at 96 KHz. I've measured a number of DSP boxes, and the 96 KHz boxes generally behave better in the top octave. I also recommend a digital speaker processor so you can "time-align" the various ranges. IIRC member Cask05 has a bi/tri-amping FAQ somewhere on the forum. Another thing that helps, especially with "professional" speaker processors is a set of passive fixed attenuators between the processor and the power amps. This allows running the processor at the highest possible level, therefore using the most number of bits in the processor. The attenuators reduce the signal in the analog domain to a level that won't blast you against the wall and/or destroy your speakers. This also reduces any noise upstream of the attenuators. I have a Yamaha SP2060 speaker processor, Crown D-75 (not A) amps tri-amping Klipschorns. SPDIF sources are converted to AES3 with a HOSA CDL-313. The SP2060 has an AES3 input. 0 dBFS in the SP2060 is +24 dBu (about 12 Volts RMS). The amps have 26 dB of gain and 0.812 Volts sensitivity. So running the SP2060 wide open into the amps (with the amp level controls wide open) would overdrive the amps by about 27 dB. The stepped attenuators inbetween the SP2060 and the amps are usually set for 36 dB of loss. This allows setting the amp level controls wide open so the gain is always calibrated. The relative levels of woofer, squawker and tweeter are set in the SP2060. As far as determining crossover settings goes, I recommend measuring the transfer function of the speaker and adjusting for flat, then roll off the high end to taste. Transfer function can be measured with Room EQ Wizard, but room reflections may limit what you can measure. I should also warn you that I have a strong measurement bias. I'm looking forward to reading about your adventures!

Gary-- Standing waves produce peaks *and* nulls. Anywhere the graph amplitude is zero is a null. Anywhere the graph amplitude is maximum is a peak. So, the software predicts that there will be a 23 Hz null in the center of the 12-1/2 foot dimension *for a standing wave in the direction of that dimension*. Since there can be standing wave in 3 dimensions (as well as oblique and tangential), in order for a frequency to "disappear", your head would have to be in a location where all the possible standing waves have a null at one frequency. Toole's book http://www.alibris.com/booksearch?qwork=10813619&matches=22&cm_sp=works*listing*title, Chapter 13 explains this much better than I can. I can't recommend this book enough.

What are the results using your approach? Do you have any measurements? I do have measurements, see attached image. I don't have anechoic data to compare, however.

Ebay has a bunch. Variac is the name trademarked by General Radio, which is no longer in business. Other generic terms are "Variable Autotransformer", "Autoformer", "Variable Autoformer", "Autotransformer".... Manufacturers include Staco and Superior Electric. Make sure you get one rated for 50-60 Hz and enough power for the intended load. Some are rated only for 400 Hz operation and will not work at 50-60 Hz. I'd recommend getting one with built-in fusing and oulet(s). It'll save you the hassle of building an enclosure and wiring. Some things to watch out for if buying used: They have a brush that slides along a transformer winding. Sometimes the brush needs to be replaced, so buying a unit that is currently manufactured is recommended, so you can get repair parts. Sometimes, if the variac was overloaded, the winding will get damaged at the spot where the brush was at the time it was overloaded, quite often at the 120V in/120V out position. They are pretty heavy, so beware shipping costs.

John-- Hey! Another measurement-type guy! What do you think of this LF measurement idea? http://community.klipsch.com/forums/t/160262.aspx --Greg

Weighting is the frequency response of a filter between the microphone and the detector/display: http://en.wikipedia.org/wiki/File:Acoustic_weighting_curves_(1).svg If you are balancing subs with the rest of the system, The best weighting is *no* weighting.

Thanks Greg, I will surely be taking you up on your offer. I'm having my first kid some time in the next couple weeks, so lets plan to synch up some time after April/May. Maybe we can even turn it into a little clinic if other DFW'ers want to come out and learn some things about measuring and tweaking. > I'm having my first kid... Congrats! > some time after April/May 2012? 2013? 2029? You never know with kids.... Whenever you're up for it, send email. Also, I have a triamped K-horn rig that was tuned by measurements and I'd like to get some other ears involved. Again, send email and we'll schedule something. > Maybe we can even turn it into a little clinic... I'd be interested in doing a clinic *after* I get some small room measurement chops. All my experience is in larger spaces or outdoors. Another thing that would make more sense would be my learning REW so that all the demonstrations could be done with a system that everyone can get their hands on.

I replied in my usual long-winded pedantic way, so I started a new thread http://community.klipsch.com/forums/p/160262/1696183.aspx#1696183 I added a Crayola bit at the end. This build thread rocks, and my offer to help thaddeussmith with measurement still stands, as we are both in the DFW area.