• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Digital Crossover/EQ DSP Stuff

Ziinda

Active Member
Since there doesn't seem to be a place to post anything on this topic I figured it mostly pertains to multiple amplified speakers and who knows more about this than people who have actually done it and found out very quickly that anyone who is buying manufactured speakers is truly getting burned! Even with mediocre drivers and reusing old cabinets I can make speakers that will simply amaze most people and ruin it for yourself once you've achieved that level of quality sound, making it hard to listen to anything but multiple amplifiers that you are barely tapping their power because it's loud enough.

Everything analog but the source is what I'm doing, I already know about records and hated them 50 years ago then hated them 40 years ago and gave mine away. The clutter of records was obnoxious and keeping them in good condition was an impossible task for me. All the needle cartridge, tone arm combinations just to drive you crazy and hear crunching sounds as it spun then hits a warp and watch your speakers move to max excursion. .......... Blah, blah.. enough of that.

So here I am now fully understanding how speakers work and trying to understand why the audio world never seems to learn from their mistakes and constantly scrapes the edges of perfection only to abandon it fir the next or lastest craze or crazy idea that never seems to last long enough to even consider plopping down the cash to own it let alone how quickly it disappears and is just more worthless junk. I never bought into DVD, SACD, owned a bunch of players and sold them but never bought a disc. I did like MP3s and once lossless was available I really liked that idea. USB drives were not it for me but SD cards struck me as the answer to audio and still can't see how it's not the obvious go to medium.

But all of this is just to give you my insight and how I have been thinking, figure in 45 years if car audio, never had kids, divorced, lived life without restraints and sometimes a bit out of control. Now I'm old and car audio is still there but I hardly ever drive very far, so my blaring system is more fir show. Stacks of amps working and not, a few rooms full of speaker drivers, boxes of useless crossovers, ports and other parts. Every once in a while there's some parts I'll use but not often enough to justify keeping it much longer.

After dealing with refinishing and making some existing speakers better than they were new by doing a good job of sealing and bracing only to find little to no difference except they looked much better but cost me a fortune in supplies and hundreds of hours but still sounded awful as far as I was concerned. I couldn't go d myself agreeing with anyone on most speakers. Tried a few times after reading through this site and figured I could trust the majority and bought a few different sets as I saw them pop up fir sale locally and every time I was underwhelmed and was wondering how people came to these conclusions? Where my ears defective or are so many others simply not knowing what sound reproduction is? I always thought it was supposed to sound as natural as possible but what I heard my entire live was blasting loud garbage.

My sister was seriously into audio stuff when I was a kid she always had the best she could afford and played AC/DC and Rush when they were unheard of. I was listening to Kiss on a llyods stereo. So I knew there was something that was better. What I remember (since she is 4 years older and I had no clue) was a Harman Kardon receiver with ohm speakers. Very simple looking system but it had volume. Many years later I stop by here place and she has a stack of electronics and giant Klipsch corner horns. She said they cost $10k and looked like concert speakers. Everything was over the top. It sounded ok but really I couldn't get passed the fact that she had $10k in 2 speakers?!? And they honestly weren't what I thought a pair of $10k speakers should sound like at all. I always imagined something more along the sound of an electrostat, a very smooth and lifelike sound that filled the room with whispers and sonic fullness with subtitles and nuance of things you never heard from those songs before. Not 150db of screeching sqwaukers piercing my ears and lacking in bass.

Fast forward another 40 years and i decided to ditch the surround sound receiver that I never used since I don't watch rediculously childish movies that are so popular now and such a waste of time I couldn't stay awake even with the crashing and booming constantly happening, just a headache. I rounded up tons of 2 channel stuff and speakers upon speakers in a quest to find something that was decent. I did find some that had potential using electrostatic panels but had terribly mismatched low drivers.

Now working without a crossover frees up how I design my lower end. I started with 2 way then went with 3 way. All analog pre, EQ and active crossovers. I tried 4 different crossovers and found them all very easy to set up and quick to get good sound.

But the buzz lately has been thess DSP things, now I already knew about them from 20 years ago in car audio and found them to be quite useless and impossible to work with in a car. Millions of button pushes to get to the setting then each click was too much or not enough and very hard to see even while stopped and not possible to change while driving. I figured 20 years later (and made for home audio) they should have things sorted out with better UI and more steps for fine adjustments, etc....

Well, I was wrong on all accounts, if anything the UI is worse now I guess they figure more people are used to this kind of awful way to use stuff now they have no need to improve it? So I started on my quest to find the digital replacement for my dual 31 band EQ, my dual 15 band EQ with an RTA that has pink noise and tones built in along with a microphone input so you can make accurate adjustments. The idea being send the signal out and make adjustments on the 31 band then remove the 15 band (or keep it in if you run Bose 901s you'll need the extra 15db cut and gain on top of the 31 bands 15dbs to get them flattened out). The entire process takes less than 30 minutes to run the RTA and have a decent sounding tri-amped system.

Enter the digital fixer, after trying the miniDSP (destroyer of sonic perfection), there was no way of getting it set up to sound close to anything I had before it. 3 days of dicking around, I finally set everything flat and bypassed everything and it still was changing the sound dramatically for the worse. I sent it back and moved to the next, it's Brazilian made expert 2.0 DSP made for both car and home has a phone interface as well as a screen on board so no need to involve a computer, which I thought was a terrible idea with the mini, nothing can be adjusted without it, pain in the ass.

The Expert was connected and looked promising but output number 2 or 3 made a loud buzzing noise. I tried new cables, switched amps to be certain it wasn't anything other than the Digital Scratch Provider. It was coming from the expert, 20 minutes later I had it packed up. Then I spotted 2 other speaker management systems which is the same thing only made with a much higher quality design with more features but most are useless and unneeded. One made by behringer the other DBX, both priced around $500 but I also found another piece by Behringer that's similar but is more an EQ and autotuner system with RTA and mic input. I picked one up at a reduced price and gave it a go. I was very interested in the autotuner, I found it needed to have volumes set to concert levels to get the auto tune to even start the process. In the time it was testing I had 1 of my electrostats electtonics burn up. A coil burned the core and wrapper off while pushing 400W through it to achieve the needed sound level. I made a new coil then studied the circuit and realized I could bypass it and improved sound dramatically allowing me to run the stat panels down below 500 if I wanted too. They will reproduce sound even lower but I think that they sound best around 650 mine are at 647 now. I sent the DEQ2496 back and waited for a DCX2496 which is the crossover version 3 in 6 out configured any way you want with fully adjustable delays and single degree alignment (0-180) with other drivers plus an auto mode that sets phase and delays using a measurimg mic which has been sold out for months.

The reason I'm here writing this is after a few days working with it, doing all the suggested ways that are needed to properly set up this piece, using a db meter, tone generator, special microphone, 3 devices with a slew of apps installed on each one to perform all the tasks. I'm finally done, pop in the SD card and it sounds like ****, I mean just f'in awful. I turn off all settings go and turn all gains back to max on my amps, turn on my BBE maximizer all set at zero and it sounds like heaven.

So I'm really wondering how so many have such great things to say about digital equipment but I find it to be the least user friendly, near impossible to get close to analog quality sound and the time and money invested seems to be a complete waste. It's kinda like the speaker recommendations that are posted.
One thing I did notice was my old NAD pre amp was robbing quite a bit of sparkle from the overall sound. Not sure if it's made like that or it's showing its age but I'm in the market for a new pre. I picked up a cheap one so I had one to work with. I discovered this after I tested a proaudio pre made by pyramid that was part of a package deal along with a rack, audio control EQ and a digital amp. The pyramid actually sounded good but I'm not using it it's part of another sale that is geared for a dj set up.

So what are your thoughts on Digital Equipment? I can't see paying any more than $500 for an EQ/crossover when I can get an analog crossover with delays and EQ with RTA set up for half the cost and never struggle to get good results.

Here's the simple setup I'm running with stellar results, my amps are hidden but they are old fanless QSCs. Those are 2 different sets of DIY electrostats, the wood grain have a 12" sub on the side facing to the opposite speaker.

I'm going to be looking for ideas on making new cabinets for both of them. I have a table saw that's all set up and cuts perfectly straight and square, modified to cut longer lengths so pretty much anything is possible. I'll make a new post with a few sketches of ideas I have and I'm sure there's a bunch of ideas that I've never considered since making my own cabinets that are more than just a perfectly square cube or rectangle will be my next step. I know that making a simple cube is beyond many peoples skills but won't know exactly how difficult it can be until they actually try it on their own saw.IMG_20211010_213715.jpg IMG_20211010_213739.jpg
 
Register to hide this ad
i have a 3 way active system. My mantra has always been to take something that is basically good (i.e. of sound design) and to then improve it by understanding the design and then through careful research, coming up with an upgrade scheme.

Over a period of many years I tried all sorts of passive and hybrid semi active systems and eventually grew a fully active system. The only downside of course is complexity and the difficulty in simply swapping itey ms in and out for camparison or upgrade, as you can do with a passive system. One reason why so few manufacturers offer active domestic hi-fi I guess. Also you can destroy tweeters though misconnection or mis-programming. How many people out there would even know how to set up an active system anyway?

My present system comprises a modified NAD C320 amplifier used purely as a preamp - The power amps are completely disconnected and the preamp class A amplifier modules have been replaced with Burr Brown 2134 OPAmp circuits. This also allowed replacement of the in-line capacitors (there to mitigate DC offset from the amp modules) with low value film caps. The sound difference was astonishing.
The preamp is followed by a pair of Behringer CX2130 crossovers set up as 3 channel mono. These have high quality linear power supplied fitted and some capacitor minor circuit changes changes.
The crossovers feed a Rotel RB976Mk2 60W x 6 channel power amp with upgraded power supplies and modifications to the voltage amp stages in the amplifiers.
The speakers are of my design and build and consist of an 8" bass unit in a separate transmission line cabinet. These have an Fs of 28.8Hz and with the average 6dB/octave rolloff of the cabinet means that they have a surprisingly deep and clear bass response. The upper frequencies are handled by a 61/2" midrange and a 1" horn loaded ferrofluid tweeter in a separate cabinet on top. That cabinet is set up as an aperiodic vented cabinet as bass loading is not required.
I also have 7 channel and 15 channel analogue EQs available in switched circuits which are used occasionally (mostly for testing, but normally switched off)

The crossovers are set at 240Hz and 1.8Khz. I can do this as the crossover ramps are steep enough that the tweeter doesn't get overloaded. The frequencies have been arrived at such that the woofer and midrange crossover at very flat response regions and the mid/tweeter frequency is chosen such that both drivers exhibit similar dispersion values at that point. Baffle step effects are dealt with by crossover level balance and relative vertical driver positioning obviates the need for artificial response delays. These loudspeakers are well balanced and tend to 'disappear' with no single driver attracting attention to itself.

I too tried DSP and didn't like it. I played with it for over a year. Although it was very flexible I just couldn't get the system to sound natural.
 
Hi,
In the last few years I had the chance to get my hands on this one after a few, many years agoI started with analogs like Orban674 which I really liked but wanted to go further.


DSP is not a solution It's only a part of it. Even if I understand because I passed this stage where I felt overwhelmed and one day I found my way, I must say that I'm not a Pro and I learned by myself and with a lot of reading and then of tests and errors i ended up figuring out how to play with that. It is not the easiest way but it is precisely all the possibilities that these tools offer that make me appreciate them and I always continue to learn. You can choose different paths and get great results, but when you understand what to do and how to do it, they are very powerful tools for certain multi-channel system configurations. It takes time and in my case it was years but now I really like what they can do. These tools depend on who uses them and therefore they are not for everyone many of my friends have thrown in the towel. Keep going and one day you will be happy with what you can get, otherwise change your plans and go to other paths. (Thx google translate:)
 

Attachments

  • IMG_1191.jpg
    IMG_1191.jpg
    107.9 KB · Views: 25
i have a 3 way active system. My mantra has always been to take something that is basically good (i.e. of sound design) and to then improve it by understanding the design and then through careful research, coming up with an upgrade scheme.

Over a period of many years I tried all sorts of passive and hybrid semi active systems and eventually grew a fully active system. The only downside of course is complexity and the difficulty in simply swapping itey ms in and out for camparison or upgrade, as you can do with a passive system. One reason why so few manufacturers offer active domestic hi-fi I guess. Also you can destroy tweeters though misconnection or mis-programming. How many people out there would even know how to set up an active system anyway?

My present system comprises a modified NAD C320 amplifier used purely as a preamp - The power amps are completely disconnected and the preamp class A amplifier modules have been replaced with Burr Brown 2134 OPAmp circuits. This also allowed replacement of the in-line capacitors (there to mitigate DC offset from the amp modules) with low value film caps. The sound difference was astonishing.
The preamp is followed by a pair of Behringer CX2130 crossovers set up as 3 channel mono. These have high quality linear power supplied fitted and some capacitor minor circuit changes changes.
The crossovers feed a Rotel RB976Mk2 60W x 6 channel power amp with upgraded power supplies and modifications to the voltage amp stages in the amplifiers.
The speakers are of my design and build and consist of an 8" bass unit in a separate transmission line cabinet. These have an Fs of 28.8Hz and with the average 6dB/octave rolloff of the cabinet means that they have a surprisingly deep and clear bass response. The upper frequencies are handled by a 61/2" midrange and a 1" horn loaded ferrofluid tweeter in a separate cabinet on top. That cabinet is set up as an aperiodic vented cabinet as bass loading is not required.
I also have 7 channel and 15 channel analogue EQs available in switched circuits which are used occasionally (mostly for testing, but normally switched off)

The crossovers are set at 240Hz and 1.8Khz. I can do this as the crossover ramps are steep enough that the tweeter doesn't get overloaded. The frequencies have been arrived at such that the woofer and midrange crossover at very flat response regions and the mid/tweeter frequency is chosen such that both drivers exhibit similar dispersion values at that point. Baffle step effects are dealt with by crossover level balance and relative vertical driver positioning obviates the need for artificial response delays. These loudspeakers are well balanced and tend to 'disappear' with no single driver attracting attention to itself.

I too tried DSP and didn't like it. I played with it for over a year. Although it was very flexible I just couldn't get the system to sound natural.

It seems we have a very similar way of thinking as far as speakers go. I tried running dual 8s for woofers and electrostats crossed at 600Hz but I was kinda missing that real low umph that is a big part of much of the music I listen to. I made 2 sets of test speakers both with dual 8s one uses seismic audio Richter 8s and the other is a bit different with poly cones and rubber surrounds, knowing I was going to add a dedicated sub to that set I chose with as high of a power rating I could find but still would be able to play low mid that handed off to a sub without a large timber difference and I ran into some interesting and very cheap drivers by a company called MTM, they are new and offer a full line of drivers for just about any use. They had 4 or 5 different 8s all with reasonable prices that made taking a chance with them to not really be much of a loss. The seismic were almost 3 times as much and they have a build quality that looks similar to what you'd think a car sub would have but with paper cones and surrounds. I was not looking for large excursion and specs have them good to 47Hz with 175W. I originally planned on making a isobaric clamshell design with them, borrowing a bit from Allison speakers. I've always done thus with my car audio sub's and i have been running cheap sub's at high volumes with results that most don't believe is possible without running high quality drivers. Plus in 20 I have her to blow any sub's while I see others going through them every month and they aren't close to my output.

Back on track, the set with the MTM, I originally bought a set of 15s as well to go with the 8s. They sit in the box new yet, I haven't decided. I ran into a pair of monitor audio Bronze 10 inch sub's locally that were part of a box full of loose drivers for $25! They were exactly what I needed. The only thing that really cost any money was the Infinity electrostats. Found a set of A2000s locally for $220. They looked mint but sounded off. Different woofers. Identical but 1 year apart close examination revealed they weren't exact at all. Plus the design is simply awful as was the JansZen set of Z400s I found locally. They had a bad overall idea and the woofers were just not matched for the task. I had no problem tearing both speakers apart to start my own.

I agree on the digital pieces, I was so frustrated I put all my analog stuff back together since I have a really nice looking curve when using the 1 phone app that actually works, I picked up a calibrated mic and found my new cheap phone is actually better than the mic and they cost the same. But when I do a sweep using spectroid its written by a Russian sound engineer who made the app including exactly what you need to do final sweeps. Its free and it works very good. I only have a spot around 200Hz that needs any correction and I was able to correct it with my old AZDEN 15 band which has an RTA, tone generator, pink noise and a mic input with adjustable gain for the RTA to keep it on screen.

To solve that simple problem using digital stuff seems like it sound be a no brainer. I mean I have control of everything in 1 shot? I can set the amount of eq I need plus or minus and the width of that point? I'll be dammed if I didn't end up using every available (9) adjustment setting to get it close to flattened! The problem is the settings are not fins enough to really gain control over the exact amount of each settings Q. There's no choice in the rolloff, its simply running out to the ends of the spectrum which is not at all realistic and they narrow settings are too narrow and do nothing since another close setting can overide it easily creating a fight between the 2. Its nearly impossible to get rid of any large deviations, like a spike followed on the down side of a 88 degree angled edge that turns into a large dip.

For example say a peak starts at 167 to 196 and its 7db over ideal slope line shown but that exiting line is going sharply down passed the line 8db below. The app uses percent from a zero line at the bottom so you don't have a db representation from the curve. Its my estimate to explain. When you try to remove the lower Hz spike with a narrow Q it won't remove it unless you widen it to cover that band but as it widens its tapering off so much that it effects the dip as well. Narrowing it reduces it effect on the spike unless you set the reduction to max or close to it but its not wide enough to do anything but create 2 smaller spikes on either side. So you pick the leading edge and then move it and see how it reacts with the dip which is much wider that the spike. The dip runs from 200 to 450 but it trails off smoother as it moves toward the 450 mark. Setting any eq on the dip is going to end up far to wide and the center is not where the most gain needs to be. It needs to have a sharp cut off on 1 side and a long roll off on the other or it will just create a peak in the middle of that dip with w smaller dips on either side. 1 will be a new peak on the outer edge and if you move it, it interacts with the peak which us much sharper due to the q width. The dip Q being wider requires the Q to be wide but as it widens the edges loose their sharp slope and flatten out. So the area you are moving can't cover that entire dip, if it did then the trailing edge us going to be way lower than the peak which you already needed to max out just to get it partially under control! So it affects that peak back up but makes the 2 smaller peaks larger and more slopped!

Until they make one with a dual adjustable Q (my idea already have it written on paper and mailed to myself) where you can set the slope angle of the Q similar to what's offered on a high pass or low pass setting but offer it on the typical band pass setting not only as a crossover setting but offer it on maybe 4 eq settings as well without needing to enter into dynamic eq settings or overlapping eq centers multiple times to exaggerate the settings. When that's done it just sounds awful. It might look OK on a graph but the sound is destroyed in the process.

I have put it back in and started from scratch and once again I ended up using 8 of the 9 available eq settings to flatten it out. Now I know it should not take more than 4 settings to resolve that issue but every time I try a different approach it turns back into the same convoluted mess! There may very well be better pieces that have considered this huge problem and they would be very easy to correct anything as long as the Q slopes can be set differently on either side or even the same on both sides could dig you out of a sharp back to back rise and fall problem. Done in steps that gets the shallow sloping Q fare enough away from the steep issues would be helpful but having a lock on each setting might be even easier.

As you set each center and width (even if its using an uncontrolled slope) if you could lock if then make that setting be unaffected by any other settings that might overlap it! You only need to lock 1 side of a steep rise/fall since it will enable you to put a wider higher setting right next to it without changing what's inside the locked column. Of course super wide Qs could run past the locked ones but I'm sure common sense will be able to figure out a happy center and width that has minimal outward affect after the locked area. 1 more thing, the locked area can now be set to cover the entire peak without worrying about where it ends and will never need to consider the second adjustments affect as was needed to be considered and precompensated for by over exaggerating the amount of correction knowing it will be affected. This is where so much guessing is involved. Back and forth constantly , move one wider lessen the the other then increase height on one and decrease the other then test just to find out you need to move them closer or further apart and it continues in a huge circle!

There is no representation of the Q slope for each number shown since the graph is way too low resolution to be any use at all. 2 clicks and it spreads out from 3 wide to 10 wide but can't be seen since its overlapping existing settings! The picture is showing 8 or 9 settings but looks like its only 3 different spots, there's 4 or more overlapped around 250Hz and 2 more set in the output EQs to make the final corrections.IMG_20211024_180638.jpg IMG_20211024_182421.jpg IMG_20211024_182630.jpg
This is where I left off I'm not quite done yet, it took me 4 hours to get it from the no eq shot to the first one I have tamed the bass spike but didn't save the shot. Its just a but shorter and moved a bit lower on frequency since I was able to increase the amp and I still have about 1/3rd power left to bring up the lower end more after I'm able to trim down the peak a but more tonight.
 
Hi,
In the last few years I had the chance to get my hands on this one after a few, many years agoI started with analogs like Orban674 which I really liked but wanted to go further.


DSP is not a solution It's only a part of it. Even if I understand because I passed this stage where I felt overwhelmed and one day I found my way, I must say that I'm not a Pro and I learned by myself and with a lot of reading and then of tests and errors i ended up figuring out how to play with that. It is not the easiest way but it is precisely all the possibilities that these tools offer that make me appreciate them and I always continue to learn. You can choose different paths and get great results, but when you understand what to do and how to do it, they are very powerful tools for certain multi-channel system configurations. It takes time and in my case it was years but now I really like what they can do. These tools depend on who uses them and therefore they are not for everyone many of my friends have thrown in the towel. Keep going and one day you will be happy with what you can get, otherwise change your plans and go to other paths. (Thx google translate:)

You're the only other XTA user I've seen here. I also run a DP448 and love it.

There's a steep learning curve involved for sure, about as far from plug and play as you can get. But properly implemented, it's killer - and I love the XTA user interface, it is pretty sweet.
 
Hi,
In the last few years I had the chance to get my hands on this one after a few, many years agoI started with analogs like Orban674 which I really liked but wanted to go further.


DSP is not a solution It's only a part of it. Even if I understand because I passed this stage where I felt overwhelmed and one day I found my way, I must say that I'm not a Pro and I learned by myself and with a lot of reading and then of tests and errors i ended up figuring out how to play with that. It is not the easiest way but it is precisely all the possibilities that these tools offer that make me appreciate them and I always continue to learn. You can choose different paths and get great results, but when you understand what to do and how to do it, they are very powerful tools for certain multi-channel system configurations. It takes time and in my case it was years but now I really like what they can do. These tools depend on who uses them and therefore they are not for everyone many of my friends have thrown in the towel. Keep going and one day you will be happy with what you can get, otherwise change your plans and go to other paths. (Thx google translate:)


I have been working with it for the last few weeks and I have swapped speakers to try a fresh start. I also finally got my ECM8000 mic. I've been waiting for it to be in stock I ordered it before I had the D
 
I have been working with it for the last few weeks and I have swapped speakers to try a fresh start. I also finally got my ECM8000 mic. I've been waiting for it to be in stock I ordered it before I had the D


This is my 4th try at replying, the pages keep disappearing on me! This was the original attempt and it's the only one that remained as a partially posted reply even though I never posted anything.

The other 2 were much more complete and I was about 1 sentence away from completing it!

I'm going to explain the mainpoints, the way DSP have been set up for use has not changed since they were first made available from what I saw 25 years ago. Back then DSP may have been pricey so they weren't as used as now but I saw the main issue with DSP has always been the interface. Since nobody seems to think it needs to be overhauled to make what it can do become a far more useful tool, saving time and allowing it to be a much more predictable outcome. I believe that this has never been addressed because people want to keep it difficult and on the verge of useless to others that don't want to spend weeks adjusting and running sweeps over and over just to see exactly what the DSP does when changes are made. The results are far from predicable!

My initial sweep shows a very decent curve without any corrections and has a very nice overall sound when any recording of music is played. So I have a peak/ dip back to back to back. This seems to be an impossible correction for any DSP simply due to their selection of tools available. I am implementing 2 EQs on the input and 3 EQs on the output (counting the crossover) since I can move the XO settings anywhere and not worry of any damages to my driver's I can use the XO to avoid bad areas or overlap or void if I need plus there's 7 or 8 slopes on each point for each driver. That alone makes a vast amount of possible settings before ever setting any EQs. So after testing each driver individually with all flat settings and full range I can find each drivers best crossover point. Sound logical? If I see sharp dips and peaks I can try to cut that out and use the next driver to cover it as long as it's within the same db range. I can also boost it to match then cut the peaks at each end of the XO range by using a dynamic shelf low pass or high if I have a flat area in the middle, I won't always have a flat middle but I will always have the end control on each driver.

After days of testing and also testing the testing apps to make sure they are consistent in results (most are junk is what I found out 1 called SPECTRUM I reported the wrong one earlier) I can run 4 sweeps and get 4 graphs all exactly the same so I know it's correctly reporting changes. Now I'm ready to smooth out 3 peaks and 3 dips. Should be simple. It you would think a DSP can take care of this in a few clicks and off we go never to think about it again.

Days later and many thousands of button clicks over 6 screens using graphic and dynamic EQs for 2 inputs (that's 4 EQs) then each output has 2 EQs total of 12 EQs) plus the crossovers (6 points with 6 or 7 slopes basically 4 points for each left or right if no setting is applied to highest and lowest plus possible overlap, underlap!) If I'm using all of these EQs 3 dips/peaks should take seconds to adjust? That would be so if what I'm setting is exactly what is changing after each setting is made. Example: I make a change to 384Hz - moved down 2db - Q set at 10 changed to 8.9. this should show a very predictable change that would make a single dip centered at 384Hz and tapering off affecting a range from 250 to 600Hz or less. But I will see changes made up at 2800 and/or 8000Hz with a hump going out a the same width as height from those centers. But not always at those centers and sometimes might not affect anything and barely change the 384Hz at all.

Once I use Dynamic and raise it lower entire ends from each crossover setting to match the db where I'm cutting out big dips or peaks that would otherwise be showing if I used crossover points that were ideal for each driver but might include big dip/peaks if I chose them. Since I know those points are impossible to smooth, I have to go with the driver that's flatest at those points to remove them even if the driver is rolling off at that point. Dynamic leveling will cure it on the output EQ of each stretched range as needed. Then set slopes to get smoothest transitions at XO points

Am I correct so far?

After days of trying layers of EQ settings back and forth, eventually settling on a plot that's realy not much off from my original one? Test it and it sounds like the recording has lost all natural sound. It lacks depth, stage and realistic tone.

So I run the delay and phase auto correction. It plots as such from left to right: 1-Left Low 2- left mid 3-left high 4-right low 5-right mid 6-right High
I get:
1-normal 52.05"
2-normal 68.43"
3-inverted 69.71"
4-inverted 0 "
5-nornal 67.87"
6-inverted 69.90"

My speakers are set 115" from front of cab to mic. Drivers are mounted from floor to center of driver; low-7', mid-28' high-41"

I was under the impression the delay numbers should have been less than 12"?

I have to say that after running the correction (after getting close to flat sweep) and it made some improvement to EQed sound but still not much different than flat, but when I ran the sweep I was back to worse than I had before I started!?!
Tried to flatten curve and reran delay/phase and was back to the same corrupted curve again.

My other set of speakers have a 12" sub on the side of the cab I'm going to start over and see how it plots those delays.
This is my final sweep after corrected delays!IMG_20211030_081529.jpg
 
Back
Top Bottom