Bit Ten Install and Tuning

arun1963

Moving the discussion to a new thread. Don't want a specific issue cluttering up the in house random thread:biggrin:.

pentoncm wrote: »

I don't know what their deal is because it really isn't that hard if you know all of the steps.

Ok, a list of things to check for.

1. Was the HU volume turned up to an appropriate level?
2. Was the output active coming from the HU, it wasn't faded in the wrong direction or anything?
3. Was the Bit Ten on, blue light?
4. Did someone try hooking up a pc to the Bit Ten and select the Aux input? And if it is the issue then they are morons. It will come out of the box in Master Output mode. I really suspect this is the issue.

As far as allocating drivers, I don't think it matters if it is registering some as rears. I'll look at the software later.

1. Yes 53/63

2. Yes output was active. Eq set set to flat and TA disabled. In active mode the p-88 by default, gives a 12db slope on the sub and a 6db on the tweet. The only driver you can set to full range is the mid, so those pre outs/rcas were used to plug into the aux in at the bit ten.

3. Yes blue light was on, the unit was powered up. But tweaking the gains knobs did nothing. No flashing lights.

4. Yes I hooked up my laptop to the bit ten via the usb connection and went thru the basic install menu like 5-6 times. Tried both the aux mode and the master mode. Nothing, no sign of life.

pentoncm

You hooked up the laptop and went through the configuration and selected the Aux channel and still the LED's never came on. And you were using Track 1 on the disc right?

Everything else sounds fine then, I wonder if you got a bad Bit Ten.

arun1963

pentoncm wrote: »

Arun when you set up the Bit Ten, when it says to set up the the Master inputs, it doesn't matter which one you select. What matters is the following screen where you have to select your speakers by clicking on them. Set it up so that you have rears. I just re did my configuration tonight for the Front only, you basically unclick everything but the sub, front tweeters and front woofers. And then if you uncheck crossover on the next screen it will set your tweeters and woofer on different channels. The downside to this is that it joins the channels together so now Left Tweeter and Left Woofer share an equalizer and right the same way.

Put the tweeters on the front and woofers on rear or vice versa. It will not affect anything and the delay settings are identical. It just determines what the title of the channel will be. As long as you have fronts and rears you will have independent EQ's which I find really useful. Example I do not cut 3.15 or 4kHz on my near woofers while I do cut these frequencies on my near tweeters.

This is how I would like to setup the bit ten. One channel per driver. I understand that you can even set it up as front+rears and use the front for tweets and the rears for mids. But having one output channel per driver seems like a cleaner way to do it. Truth be told, at this point I'd just be happy if it was installed and done with.

Did you run a seperate power line from the batt or did you hook it up from the distro block?

pentoncm

I just like having the ability to adjust the tweeter level from the DRC. That is why I have it set up fronts and rears. The only issue with having fronts and rears is that you can't see the response of an entire side, sub/woofer/tweeters together in the software program. It just affects the visual display.

I figured out the joint EQ between tweeters and woofers. Make sure you are in Expert mode. In expert mode if tweeters and woofers are set to front, EQ is independent between them. In standard mode they share the EQ for simplicity. Did not find this out from crappy manual, or the Electrommedia Rep that I called (he was not helpful about this). I decided to change over to Expert mode just to see what it does and bingo.

I don't know about the power wire because installer put it in so I have no idea. The Rep from Electrommedia was helpful about gains. He said that O scoping for the gains at the amp is a good starting point, but if you do a lot of cutting chances are that you will have to raise the gains. Especially because you O scope with a sine wave. He said to play it by ear with music when you get it done.

arun1963

I'm not an idiot :biggrin:. I got the bit ten installed. The installer at the shop wired and grounded it and I set it up. Only took me 4 attempts but I got it all done. The Audison rep never showed up.

Everything works perfectly. Connection is through speaker level. Couldn't get the pre out to Aux In, to work. No worries. It works fine this way. On the output side, the SR 6.5" are on channel 1&2 and the Scan tweets are on 3&4, sub is on channel 5. Control level is set to Expert. I can set TA on each driver independently. For a start I copied the TA settings from my p-80 and then fine tweaked them in increments of 0.01 ms. That's insane and the funny thing is you can hear that difference.

At the top of the bit ten menu you have the gain levels for all 5 channels. From here you can eith mute one or all channels, link/sum them so that the change you make on affects all equally, or you can select solo where each channel is independent. Selected solo here.

There has to be a thorn with this rose, right? Xover points and eq is summed for L&R. I should be able to eq each driver independently, but I can't. One eq for both mids and one for the tweets and third for the sub. I can't for the life of me figure this one out. I probably messed up somewhere while configuring it.

First Impressions:

1. 16 band to a 31 band eq is a big jump. Control over 15 additional frequencies can really clean up the sound. For the first time I am able to manipulate the roll off of drivers an octave above and below the xover points. Much better clarity, much better low end.

2. 15 extra bands is one thing, but the ability to adjust each frequency in +/- 0.3db and to hear that difference, just wow.

3. I already mentioned the fine tuning on the TA. I think eventually I'll spend a fair bit of time here.

4. Keep saving the results of each tweaking session to A/B later on.

All I need to figure out is how to get independent eq on each channel, but at the end of day 1 the jump in sq is palpable. I'll know for sure tomorrow morning. Seriously though this is going to be a long trip. I'll be getting an SPL meter later this week and thats when I'll seriously get to work. I'd like to kind of measure the progress. Here's what I'm thinking:

1, Start be measuring FR at speaker level, one speaker at a time. Then Measure the same speaker at ear level. Plot that chart for each driver so that we can identify frequencies where the environment is causing cancellations and those where its causing summing
.

2. This would also identify L/R response imbalance.

3. Once the L/R balance is done then go in with the RTA app and identify the peaks. You can actually do this with the spl meter readings as well. Once you have flattened the FR across the 10 octaves to within like +/- 2db

4. Then go in and set it by ear and see how different the FR graph is from 'flat'.

pentoncm

arun1963 wrote: »

There has to be a thorn with this rose, right? Xover points and eq is summed for L&R. I should be able to eq each driver independently, but I can't. One eq for both mids and one for the tweets and third for the sub. I can't for the life of me figure this one out. I probably messed up somewhere while configuring it.

Everything is independent, its not the configuration. Right below the fine delay is a button that says Xover L/R link. Unclick the button. Look at the EQ, directly to the left of 20HZ is a button, it says EQ L/R link, unclick. Now you have everything independent.

Also I think that the EQ is adjustable in .25db and that the program just rounds up on the display.

Here is a cool trick. Look at the bottom left in the software, where it lists the channels. Hold Control, click on Left woofer, Left tweeter, and subwoofer. Now the entire left side will be displayed showing all the crossovers and effects of the EQ. This will show you the effect of the different crossovers on the total response as well. Do the right the same way when you want to look at the right side.

The Bit Ten actually can do something the Bit One can't. The Bit Ten has assymetrical crossover slope capabilities on each channel. You would think the Bit One could do this but no. The only thing that it doesn't do that could be useful potentially is letting you have different crossover types on the same channel. Example, Linkwitz on Low Pass, Butterworth on High Pass. But that is really nitpicky. I don't even know of another processor that even has Linkwitz filters.

pentoncm

You actually can use the SPL meter as a mic for a PC RTA program. I'm considering TrueRTA as a program. Only problem is that I would need to buy a usb soundcard for my laptop in order for it to work.

I really need to get a tripod for my SPL measurements.

arun1963

pentoncm wrote: »

Everything is independent, its not the configuration. Right below the fine delay is a button that says Xover L/R link. Unclick the button. Look at the EQ, directly to the left of 20HZ is a button, it says EQ L/R link, unclick. Now you have everything independent.


Here is a cool trick. Look at the bottom left in the software, where it lists the channels. Hold Control, click on Left woofer, Left tweeter, and subwoofer. Now the entire left side will be displayed showing all the crossovers and effects of the EQ. This will show you the effect of the different crossovers on the total response as well. Do the right the same way when you want to look at the right side.

*facepalm*, thanks dear, worked like a charm. Now everything is independent.

pentoncm wrote: »

The Bit Ten actually can do something the Bit One can't. The Bit Ten has assymetrical crossover slope capabilities on each channel. You would think the Bit One could do this but no. The only thing that it doesn't do that could be useful potentially is letting you have different crossover types on the same channel. Example, Linkwitz on Low Pass, Butterworth on High Pass. But that is really nitpicky. I don't even know of another processor that even has Linkwitz filters.

My p-80 allowed this in active mode. I fooled around with it a bit it always caused integration issues. I could never get under lapping / overlapping xover points to work either. Just didn't sound right.

pentoncm wrote: »

You actually can use the SPL meter as a mic for a PC RTA program. I'm considering TrueRTA as a program. Only problem is that I would need to buy a usb soundcard for my laptop in order for it to work.

I really need to get a tripod for my SPL measurements.

I have the True RTA on my wife's droid and was fooling around with it last night while playing the test tones. I was a bit surprised at the accuracy of the mic on that phone. It picked the tones within like +/- 1-2hz. I'll check the accuracy of the spl readings from the phone, once I get the spl meter.

On a side note, how can I take a screen shot on my laptop of the bit ten menu? Not very tech savy

Mac, about the pink noise disk you made, I'm not sure I understand this bit:

Best thing to use is a cd with filtered pink noise, 1 for each track. If you know how to do it on your computer, make a cd with each track being one frequency but only 1 channel. I made one with 80 Hz left channel only then track #2 was 80 Hz right channel only and so on.

Here are the dumb questions:

1. How would having 80hz as two different tracks one for L and other for R be different from playing the 80hz with only your left mid playing and then only with the right?

2. How do you record a cd that way?

3. Test tones typically are not very accurate above 800hz due to reflections. Is there a site where you can download the pink noise tracks?

pentoncm

To take a copy a picture of the interface. (1) Hit control+Print Screen. (2) go into a program like word or powerpoint and paste in the picture. (3) Right click and hit save picture as. (4) Now you have an image you can easily attach.

arun1963

I'm still waiting to get my hands on the spl meter. The guy who has it is non commital on when I can have it. So Googled 'buy spl meter' and this was the first site that came up. Lots of meters on there that would be more expensive than my setup and a couple that are probably worth more than the car:biggrin:
I'm looking for a $ 40-50 unit from a site that will ship to India. Hopefully I'll close this tomorrow.

For a start I figured I'd plant my TA settings from the p-80 and the eq settings for the common frequencies and then play a bit with the new frequencies. One thing that I wanted to play with was the transition rage between the mids and tweets. An octave above and below the xover point. I look at this range as a baton exchange in a relay. One guy is slowing down while the other is breaking into a run, timing that right for a smooth transition.

Based on what I'm hearing, I don't have this right. Vocals are brittle and stretched. I think I need to bring up the 1-3khz range on my mids, maybe a touch on the tweets as well. I would like to open up the highs a bit as well, but I need to sort out the upper mids first. 300-800 seems ok for now. I like the lower end though. I ran the test tones and 30-60hz is a little left of centre (I sit on the right side). For some reason 80-125 is much hotter from my near side. Maybe this range is hotter depending on the side where the steering column is. IIRC, from a year or so ago when I used the spl meter, 125 was like 5dbs louder on the near side and 80 was 3db's louder. I have decent transition from sub to mid bass, but the mb could definately be snappier.

I'm still cutting my teeth on having way more processing power than I had earlier, but as of now, my active setup on the p-80 was better for both tonality and imaging. That is 100% due to my limitations.

A quirky thing about using the bit ten is that I set the laptop on the passenger seat and turn sideways to tweak. That put my ears out of phase vis a vis the driving position. So sounds right while facing sideways, sounds out when I face in front . So now it's tweak and then face forward to validate. I also need to get some serious seat time, I think I've spent all of two hours on it since Sunday. This is going to take a while. I really want to get the spl meter and get started measuring things.

arun1963

pentoncm wrote: »

To take a copy a picture of the interface. (1) Hit control+Print Screen. (2) go into a program like word or powerpoint and paste in the picture. (3) Right click and hit save picture as. (4) Now you have an image you can easily attach.

Tks!! I'm not computer savy

pentoncm

Did you try the linwitz filter yet? Seriously, Butterworth gives you a +3db peak at the crossover point causing unneeded Equalization all over the place. Linkwitz will give you a flat response. If you do try it, it will probably sound bad at first because of how you have things cut. You will probably have to bring up a lot of stuff around the crossover points. I'm not sure that you really have to cut so far away from the crossover point. That might be affecting the blending of your drivers.

pentoncm

For instance if you run a Linkwitz Riley and cross your mids at 3kHz, 3kHz will already be cut by -6db. 4kHz will already be cut by -12db and so forth. I would only cut here if when you run your analysis, you still have a peak when just running the mid for instance.

Also, try and run your mids and sub at 60Hz with a Linkwitz, just to see how it sounds. You might like it.

arun1963

I'm used to the BW on my p-80. At the end of the day its a slope that's causing a roll off based on a fixed progression, a pattern if you will. either ways you'd use an eq to tweak that roll off so that it sounds better. For now I'm working on getting the tonality and imaging up to speed. Thats why I'm sticking to the BW for now.

There's a poll over on diyma somewhere on how people use the eq. With possible responses ranging from, run it flat to just a bit, to extreme. Not sure what percentage understand extreme as setting each frequency a certain way relative to how everything else is set.

Look at Andy Wehmeyers graphs for speaker level response vs that at you ear, keeping everything else constant. You'll understand why you need to use your eq to cut in large swathes. That while maintaining balance. Balance for L/R and balance across the 10 octaves. There's a difference between flat and balanced. Although a certain part of the range you will want flattish.

pentoncm

arun1963 wrote: »

Look at Andy Wehmeyers graphs for speaker level response vs that at you ear, keeping everything else constant. You'll understand why you need to use your eq to cut in large swathes. That while maintaining balance. Balance for L/R and balance across the 10 octaves. There's a difference between flat and balanced. Although a certain part of the range you will want flattish.

I totally agree with you here, I just like trying to have things set up in order to minimize the amount of EQ. I just don't like BW because it causes more peaks. You may have to still tweak the rollof of L/R but it should be less.

pentoncm

Hey, a really useful tool are the selector buttons above each band, you can select multiple bands and adjust them simultaneously. I was messing around tonight with the EQ and I just couldn't get things right. Basically I couldn't get enough upper midrange out of my near mids, and I the far sound sounded tinny in comparison. What I did was select everything beneath 1.2kHz and I brought it down by 1.5db. Then on the far side everything above 1kHz I brought down an additional 2db. Now sounds much more balanced. Its cool to select a bunch of bands in an area if you know you have a general issue and take them down simultaneously and listen to it in real time.


New MM1240 sub going in this weekend, I'm excited to get rid of ported sound. Will only have 300W until xmas when I will be getting new sub amp.

arun1963

Selecting a bunch of bands and tweaking them together may not be the best solution, because everything will be cut or boosted by equal amts. Your FR at ear level is going to be audibly different for each frequency.

Select two speakers at a time. You can do the two mids together and then add the tweets. The only issue here is that based on your mid to tweet xo point the 'sound' you're aiming for will be different. Eg With a 5khz xover point, the mids would be playing upto ~7khz even on a steep slope. That's one type of sound. However lets say your xo point is 2.5khz. Now your mids are only playing to ~4khz. Totally different sound. If you're not very sure here then the easier way is to play the far mid and tweet only and then tune for best tonal accuracy. Now do the same for the near mids/tweets. Then play all 4 drivers and correct a bit for L/r and level match across the frequency ranges. It may not be 100% accurate but it will atleast start giving you an idea on which frequency has what effect on the sound.

How you set the eq will vary based on the cars environment and the install lcations of the speakers. But Generally speaking:

1. 50-100hz - This is the range where you mids are taking over from your sub, its the meat of your MB. 50-60hz is where the rumble on your mid bass is. 80 and 100 can make your MB bloated so watch that. Cut a bit in this range and balance for L?R.

2. 125-250hz: This is the upper half of your MB. Keep this range about 2db's lower than the previous range. There are some frequencies here that whacked out on L/R balance. Eg for me 160hz is about 6db louder on the right. 125 is about 4-5dbs louder on the right. 125 makes my door panel hum so watch for that. Again 250hz is much hotter on the right. These frequencies should be hotter for you on the left. It seems it's either related to how close you are to the driver or it has something to do with the steering column. 125hz is the low end of male vocals and 200-250hz is the bottom end for female vocals. If these frequencies are hot it will have a negative impact on vocals.

3. 315-800hz - You're into the midrange here. Cut 315 more on the left. This will help a bit with keeping the vocals centered to your right and will help a bit with clarity. 400-600 run this range flat you would cut here the least if any. check for L/R balance a bit at 400&630. 500 is fairly even for both sides.

Will continue in a couple of hours. Gotta drive home.

pentoncm

It actually sounds much better now that I made those changes last night. When I tune, I only listen to the mids, then I will add the tweeters later. For now I am mainly worrying about the mids. My near mid and far mid both sounded dramatically different to me. So I took a global type approach because it didn't seem like the differences were isolated to a few bands on either driver. It sounded to me like the upper mids needed cut quite a bit on the far mid, this helped to boost up the midrange and low mid punch. Also, the near driver seemed too thick below the upper mids. I'm only using this technique to get in the ballpark and adjust the balance of major areas (midbass, low mid, mid, upper mid) between the two speakers and focus on balancing each subsection later. I'll post what my EQs look like now when I get home tonight.

This all came about after running Droid-RTA last night. It confirmed what I thought I was hearing, not enough midrange/midbass on far side, too much upper mid on near side.

500Hz and 600Hz show significant peaks that need a bit of cutting on my near side, about 4-5db. Not much cut on far side at all.

Now that I look at your recent settings, mine are starting to look like yours more now.

arun1963

4. 1khz-3khz- There's a lot going on here. This would be your most attenuated range. Reflections will make this range much loder than everything else. Try running the tones from 80-500 and get an idea of the average intensity. Then run the 1-4khz range and hear the average intensity level. Plus you're dealing with the beaming of your mids in this range. A 6.5" mid will beam ~2.5-3khz. You're going to use the eq a lot here. An average cut of 4-6dbs is not out of place. You would cut the far side more. 1-1.6khz can make the sound hollow, brittle and vocals sound stretched. If this range is hot it will also take away a lot of your lower end presence.

Also this range or a large part of it, is being played by both your mids and tweets. What % of sound for each frequency you want to give to your mid or tweet so that the sum is right, is something you should play around with. Play 2khz with all drivers playing, then with only the mids and then with only tweets. Depending on your xover point do this for an octave above and below. Hence if you want to cut a frequency, do you want to cut the mid or tweet or a bit of both. When you do two drivers at a timeit will also help you correct for L/R balance.

5. 4kz and up- This is tweeter terriotory unless you're crossing high. 4 khz lets you add a bit of warmth to your upper mids. Use it carefully. I use 5khz to balance the tonality of the upper mids and highs and it also lets me move the stage away from or towards me, a bit. 6-8khz is about sibilance its best cut a fair bit. 10khz is the shimmer on hats 12khz is about air. But everything below has to be set with reasonable balanceto get the right sense of air. Your tweets would be beaming in the 12-13khz range. The more on axis one is going to be louder.

arun1963

pentoncm wrote: »

It actually sounds much better now that I made those changes last night. When I tune, I only listen to the mids, then I will add the tweeters later. For now I am mainly worrying about the mids. My near mid and far mid both sounded dramatically different to me. So I took a global type approach because it didn't seem like the differences were isolated to a few bands on either driver. It sounded to me like the upper mids needed cut quite a bit on the far mid, this helped to boost up the midrange and low mid punch. Also, the near driver seemed too thick below the upper mids. I'm only using this technique to get in the ballpark and adjust the balance of major areas (midbass, low mid, mid, upper mid) between the two speakers and focus on balancing each subsection later. I'll post what my EQs look like now when I get home tonight.

This all came about after running Droid-RTA last night. It confirmed what I thought I was hearing, not enough midrange/midbass on far side, too much upper mid on near side.

500Hz and 600Hz show significant peaks that need a bit of cutting on my near side, about 4-5db. Not much cut on far side at all.

Now that I look at your recent settings, mine are starting to look like yours more now.

If you're tuning the mids and then adding the tweets, when you add the tweets it should feel like a natural extension up from what you had with the mids only. Thats a good transition. This is what you're trying to achieve when tuning the range being played by different sets of drivers.

The near and far mid will sound different. They should sound right together, for the range they play. Will post up my current settings tmrw.

pentoncm

Here are my new settings, sounding much better now. Getting closer but not done yet.

I think the first thing I'll need to change is increasing hte 800Hz and 1kHz region a bit on the far mid.

arun1963

This is where I am now.

On the far side, I've cut 40-63 a bit to pull this range a bit towards centre. 1-4khz is cut further for better clarity on vocals and to get rid of most of the edginess. Since I'm cutting this range on the mids, I've brought up the tweets a bit in this range. Ive cut the far tweets further in the 6-8khz range. Much better now.

On the near side I've raised 40-60hz a bit, 160hz gets a much deeper cut and is now better balanced for L/R. 250hz is cut further and 500hz pulled up a bit. 1 khz is brought up a bit but 1.25khz is cut deeper, helps with mid range clarity a bit. 1.6khz is cut much deeper on both sides. Cutting 800, 1khz and 1.6khz seem to have a big effect on making the vocals less stretched. Next 2-4khz is cut a bit more on the near mid. Again this is balanced by raising the near tweets a bit for this range. 6.3 and 8khz get deeper cuts on the near tweets. These are two frequencies you definately want to cut a bit aggressively.

The sound is now much cleaner but I'm still not happy with the mid to tweet transition, I need to work here. This unit is going to keep my ocd tendencies well fed for a while.

pentoncm

My system definitely has reached a point in SQ that I did not think was previously attainable in a car. These MM components are absolutely amazing the more I tweak the system. Its far from perfect but you know its good when it doesn't sound like there are speakers up front, its sounds like a deep and wide stage. Definitely the mids and tweeters are blending very well now, it sounds like the tweeters just take over the sound from the mids.

Arun, how is your noise floor? I don't hear any static or noise on well recorded tracks. It helps that my deck is all the way up so the gain knob on the Bit Ten Input is turned just a bit past 1. Its a little weird that Audison doesn't put Voltage markings on gain, it just goes on a scale from 1-5.

arun1963

I don't have an issue with noise either. I'm connected via speaker level outputs and gains at the bit ten are set about 11ish. Gains were set with the engine running and volume at 52/63, which is the max I would run with the preouts. With speaker levels the voltage is obviously higher cause its much louder and now 44-46 is max volume. There is a digital silence track on the focal disk. I ran it at about 47 and nothing.

Get to a baseline setting that you are happy with. Now spend a few days just tweaking narrow bands say 60-100hz, then 125-300hz and so on. Boost 60 a bit and hear the difference it makes, then cut it a bit to hear that difference. Do this over several different cd's so that you get an idea first of what each frequency does to the overall sound. Next try it two frequencies at a time say 60 and 80 then 60 and 100. Cut one boost the other, see how the frequencies work together. When boosting or cutting stay within say +/- 0.5db. That will get you used to hearing subtle differences. You eventually want to get to a point where you're picking the frequencies instinctively while tuning, based on what you're hearing. Be patient, this will take a while.

Its also important to have a reference sound. If you have a good stereo store around you or if you have a home system, spend some time listening to music you're familiar with on a 2 ch set up then hear the same in your car.

arun1963

Going to pull the trigger on this spl meter. Getting it for about $ 70 online will land me here ~ $ 170. Any thoughts on the meter?

pentoncm

Did you find a method for generating pink noise in 1/3 increments?

pentoncm

Found it, http://ampguts.com/index.php?pageid=audio_downloads#

arun1963

Thanks man for sharing the link for the pink noise. Downloaded and ran them and yes they seem much more accurate than the sine tones. Another big difference is tha with the tones I didn't hear anything past 12khz. I figured age plus cranking music, finally got to my ears. However, with pink noise, no issues hearing the 16khz note. I'm still waiting for the spl meter, should be here sometime next week.

I have spent some time on tweaking and will post up the current screen shots and details on changes made this weekend. Just one issue that I'd like to highlight. I'm starting to get an issue of hiss from the right tweeter ocassionaly. I think its happened 4-5 times over the last ten days or so. The hiss is audible but goes once I turn the hu off and then turn it on. Not sure what it is. I wish I could get the pre outs to work i/o of the speaker level inputs.

arun1963

It's been a month now, with the bit ten. In a nutshell, it's been like going back to school. I had all these notions in my head, things I mentioned in the thread that I thought would help dial in the sound. They do, but it's nowhere near as easy as I thought it would be. I would get in the ballpark and then lose it. So I'm now learning stuff like knowing when to stop and saving settings when my ears tell me too. Doing things in baby steps (this is a biggie), knowing that a few frequencies out of whack by 0.5db is enough to throw the balance out of whack.

Getting the 600-4khz range right is critical. This range has both everything you want and that which you don't. Your ears are most sensitive in this range, it has all your imaging cues, clarity of vocals plus its the range where you can set the balance between vocals and instruments. Do you need to make the vocals stand out a bit or push them a bit deeper into the mix. It's the range where you have the early harmonics on practically all the instruments. To get a piano to sound at least 80% like a piano, you have to get both the fundamentals and the harmonics right. Harmonics are about decay while the fundamentals are about the rise of that note. A lot of the perceived impact comes from the fundamentals. But if the harmonics are overcooked, it can take away a huge chunk of this impact. You will lose a ton of the lower end if the 600-4khz range is overdone.

On the flip side this range has a ton of issues. On a top twenty list of negatives in this range 1-10 would be the same, REFLECTIONS!!! Frequencies starting from ~800hz up are highly prone to reflections, the environment we sit in doesn't help one bit either. Reflections limit the perceived space and add comb filtering and crosstalk to the mix. All wonderful things to kill your imaging and tonality. Add stuff like beaming from your mids, and the fact that this range is normally also the transition between different sets of drivers. You have a lot that can go wrong here.

Of the time spent tuning over the past thirty days, about 20% has been devoted to the 50-500hz range another 20% to the 5khz+ range. A full 60% of the time though has gone into the 600-4khz range. I still don't have it where I want it, but I have my first 'real' and validated baseline setting going.

Oh! I got my spl meter and that helped me tighten up the 50-400hz range much better. The base of the stage is now much better aligned with the layers on top. I had this range set by ear earlier and while nothing was seriously out of whack but it's much better with the meter. However for some reason the 1khz-6khz range seems to sound better done by ear than with the spl meter. I've tried this range 3-4 times with the meter and then gone back and tweaked a bit by ear to what sounds right. I'm not sure why this happens. Could be the impact of reflections or just the fact that our ears are more sensitive in this range compared to the mic. The differences are not that big, just enough to be noticeable. All in all it's a great tool to have.

Pentoncm, thanks to your insistence I tried and liked the LR filters . I'll post up my current settings tmrw. There's some things I have to work on a bit. :cheesygrin:

pentoncm

Arun, glad that you like the L-R filters. I would love to try out the 48db L-R filters on the Bit One if I had it. That would limit having to EQ past the xover point. I don't know why the new Alpine PXA H800 doesn't have L-R filters, they can't be that hard to incorporate. To me it is only more of a pain to be limited to crossovers that create a 3db boost in the crossover region.

I too have been learning a lot. I have been tweaking about 3 -4 times a week and have yet to get it close to perfect. It sounds really good, way better than any other system I have been able to hear, but not as good as I know it can sound.

One of the major things I am tackling now is using the EQ past the crossover point on my woofers and tweeters. My Xovers are at 4kHz as I mentioned before. I used to stop cutting at the crossover point and didn't think that I would have to go beyond. I was way wrong! For instance, my woofers need cutting at 5 and 6 kHz, I even have to cut 8khz on my far woofer. Also I had not cut the 3.15 -6kHz region enough on my tweeters, I have to work on getting that balanced so that it sounds like the tweeters don't add anything but just take over for the woofers.

Is my car weird? I have a null at 400Hz, then two huge peaks at 500 and 630Hz. I first detected this by ear listening to music. Then I confirmed it by using Droid RTA with pink noise and by using my SPL meter and sine tones. I have to cut 500 and 630 by 7.5db each on the near side. Far side is cut by 2.5db and 3.5db, respectively.

My sub did not sound as though it was hitting low enough so I added a .5 pound of polyfill to the box. Sounds just about right now. Still very tight response and getting great low end response. Also, I'm now applying EQ to my sub. The main spots are a 3db cut at 80Hz and 1db cut at 100Hz (these are preliminary but sound pretty good).

I totally agree that differences of .5db have a huge effect on the image between the left and right. I have stopped trying to get the left and right perfectly balanced with the spl meter. By only using the meter, certain frequencies pull to one side or the other. In the end I get the left and right in the ballpark and then finish by ear

I'm still disappointed at the Soundman site. The owner of the site is adamant that EQ has nothing to do with stage depth and that it only has to do with speaker placement. That is the most ridiculous thing. Maybe that would be the case if we weren't talking about audio in CARS with tons of reflections and peaks!

arun1963

There's a running thread on the other forum about stage width. It discusses all three dimensions. Speaker placement plays a big role in setting up the dimensions of your stage. No questions about it. Mac had mentioned this a while back. I remember him saying this, ofcourse I only understood the significance much later on. Anyway, there is a quote from Mark Eldridge in that thread

Speaker placement is always the first,and most important part of designing a great sounding system. If you want to (or have to because of time/financial requirements) you can use factory locations. But even if you only care about the driver's seat listening location, I'd bet you can make a significant improvement in the system SQ by working on the speaker placement.

Do you have any room for a midbass enclosure in the bottom of the dash assembly, above your feet?

Midbass is not very critical for stage width or height, but is essential for depth. Also, if the pathlength difference is too large, say over 6 inches for the midbass range from 100 to 400 Hz, the imaging will be shifted to the close side.

Mounting them under the dash will help with equalizing path lengths, and leave the kick panel area for the mids, which will have a strong influence on width.

In every case I have worked on in a serious SQ competition car, I have found that placing the primary tweeter as close to the mid as possible yields the best results. With them close together, it is much easier to achieve a coherent wave front from both channels, and minimize comb filtering. The staging and imaging will be much more stable, and easier to tune. If after tweaking the system with this configuration you find that height and width might be improved, you can add a secondary tweeter in the pillar, turned down, and crossed over somewhere above 8,000 Hz, only enough to bring up the corners of the stage, and add some width.

If you have a solid midrange and high frequency stage with good perceived stage height, locating the midbass lower will not affect it.

Comb filtering is a result of the same sound arriving at a point like your ear or a microphone tip, but at two different times. For example, a midrange and tweeter mounted on the dash top pointed at the windshield will result in your ears hearing the direct sound from the speaker, followed a fraction of a milisecond by the reflection of the sound off of the windshield. You get essentially the same sound twice at two different times. The phase differences in the arrival sounds will combine either destructively, or constructively, depending on the frequency and the additional distance/time the reflected sound incurs before arriving at your ear. The result can be some very wildly variating frequency response.

Try sitting in front of a single home seaker with no hard surfaces near you or the speaker. Listen to a song you are familiar with. Then have someone hold a board, large mirror, or some other hard surface so that it creates a reflection that arrives at your ear also. That is comb filtering.

Apart from the bit about mounting the tweet down low next to the mid, the rest is exactly like Mac had mentioned.
I'm not sure I'd mount the tweets at my feet for tonality and then use an extra pair to pull up the height. I'd rather mount the tweets up high and deal with the reflections via a dash mat and angling the tweets + baffles if required. Oh and BTW the foam at the corners of the windscreen works well for dash mounted tweets. Looks fugly, but it works:eek:.

Your midbass give you depth so get them as far deep as you can, that will also help with pathlength. Your width comes from your midrange. Ultimately your width is limited by the boundries as defined by the reflections and crosstalk. So get these drives as close to the physical boundry. Venting the midrange in your footwell will give you good width. It will also keep the drivers on better relative axis, your midrange frequencies are more sensitive to axis response. Get the tweets up high for the height. Then use TA to set arrival times.

There is dsp available that will help in giving width perception way beyond the physical boundries of the car. Abiophonics software like R.A.C.E. coupled with speakers placed close together give you the feel that you're listening in a room. You are then using the dsp to go beyond not only the physical boundries but even the recording itself. Is that accurate? Is a 20' wide 3' high and 4' deep stage believable? Is it accurate? That was the other debate. For me accuracy is linked to whats on the recording. If it's recorded in a matchbox size studio thats what I should feel. If its recorded in a 40'x40' studio then I should get a sense of some extra stage. In reality both the recording spaces are larger than the car, so I'm never going to get the exact feel, but I should feel and sense the difference.