DBX DriveRack 4800

I am currently using a Behringer DCX for my ScreenArray speakers in my home theater, and it does a very poor job in my opinion. Much hiss in my HF sections and not very refined sounding.

I have found a DBX DriveRack 4800 under half the price of the MSRP and in perfect condition, so I am very tempted to buy it. I have read somewhere on this forum though, that DBX products are not very good and what goes in is NOT what comes out.. Is this true for the 4800 as well, or is it only for the cheaper models?

Would the BSS London units be a better choice? (I know these are used in the Synthesis systems).

Hi Jonas

Your hiss problem is common with many other PA digital crossovers. While not state-of-the-art, the DCX remains a very good device. The hiss is related to the input/output gain and it can easily be solved. Here's a quick read:

http://www.diyaudio.com/forums/digit...-over-280.html

Quote:

---

Originally Posted by Lee in Montreal

Hi Jonas

Your hiss problem is common with many other PA digital crossovers. While not state-of-the-art, the DCX remains a very good device. The hiss is related to the input/output gain and it can easily be solved. Here's a quick read:

http://www.diyaudio.com/forums/digit...-over-280.html

---

Thanks for the response. Would you think the hiss would be the same with the DBX 4800? No matter what, I want to replace the DCX. I think its DA/AD convertes are way to harsh. I can clearly hear the change in sound when I put it into the chain even though all settings are set neutral.

I'm very interested in what for example the BSS London Soundweb's can do, that the DBX 4800 cannot. As far as specs goes, they seem very similar. (I dont need all the input/outputs of the soundwebs though)

Most household systems have a 1v sensitivity, while PA/pro systems need something near 3v. In both cases, the input at the crossover is the same. Therefore it exlains the much higher gain of a "professional" crossover. You may want to reduce the input level at each of the amps (at least by half), and turn that knob a bit more on your preamp. That should solve your problem.

It would be great if manufacturers could include a switch on the back to select your output. 1v vs 3v.

No experience with the DBX unit.

BTW The DCX is not exempt of faults. My first unit had the dreaded frying egg floor noise and was replaced within 3 months. I suspect I was sold a refurbished unit. The second one I got works perfect. That problem is also very well documented and can easily be fixed.

Quote:

---

Originally Posted by Lee in Montreal

Most household systems have a 1v sensitivity, while PA/pro systems need something near 3v. In both cases, the input at the crossover is the same. Therefore it exlains the much higher gain of a "professional" crossover. You may want to reduce the input level at each of the amps (at least by half), and turn that knob a bit more on your preamp. That should solve your problem.

It would be great if manufacturers could include a switch on the back to select your output. 1v vs 3v.

No experience with the DBX unit.

BTW The DCX is not exempt of faults. My first unit had the dreaded frying egg floor noise and was replaced within 3 months. I suspect I was sold a refurbished unit. The second one I got works perfect. That problem is also very well documented and can easily be fixed.

---

I'm using a surround processor with XLR out.. they would think they shouuld provide enough input.

I wouldnt be surprised if my DCX has a fault.. It is mostly one of the inputs (or one of the outputs) which has most hiss.

Hi
Don't hesitate, if you can get it for half the prize, just buy it.

The dbx4800 is better than the one you have. You have settings for the Cinema screen arrays and that goes for the dbx260 as well.

I have myself the dbx260 and PA+ and have no hiss problem whatsoever. Very satisfied with both of them but dbx4800 is much better.:)

Thanks

Quote:

---

Originally Posted by Mctwins

Hi
Don't hesitate, if you can get it for half the prize, just buy it.

The dbx4800 is better than the one you have. You have settings for the Cinema screen arrays and that goes for the dbx260 as well.

I have myself the dbx260 and PA+ and have no hiss problem whatsoever. Very satisfied with both of them but dbx4800 is much better.:)

Thanks

---

I have already contacted the seller, and am awaiting reply. I think the dbx260 would be more than enought for my needs, but I would need two of them as I am using it for three front speakers in a home theater, and then the used dbx4800 is actually cheaper than two dbx260's. With the dbx4800, I can process all three LCR's and my two subs. Ideal situation for me.

Hi
Perfect!!

So, I just picked up the DBX Driverack 4800... It had been used in a hifi-setup in a living room, so it was in absolutely perfect condition.

I dont have the time to try it out at the moment, but in a weeks time, I will try it. I hope the improvement is noticeable:)

Quote:

---

Originally Posted by Jonas_h

So, I just picked up the DBX Driverack 4800... It had been used in a hifi-setup in a living room, so it was in absolutely perfect condition.

I dont have the time to try it out at the moment, but in a weeks time, I will try it. I hope the improvement is noticeable:)

---

Pictures please of current set-up and new set-up!
:useless:


I don’t get hiss now unless I turn the amp level up for the LCR HF and as long as I can hear pitches around near 16KHz on the rare occasion I’m not fussed. I think the cost is unbeatable for the DCX2496 its how you use it over high priced models that would cost 5 or more DCX2496!

Quote:

---

Originally Posted by JBL 4645

I don’t get hiss now unless I turn the amp level up for the LCR HF and as long as I can hear pitches around near 16KHz on the rare occasion I’m not fussed. I think the cost is unbeatable for the DCX2496 its how you use it over high priced models that would cost 5 or more DCX2496!

---

I know I can turn the gain down on the amp, but I dont like those "hack"-solutions... I'm not satisfied with the sound quality either. Have tried listening with and without the Behringer in the chain, and everything is much more transparent without it - more smooth and refined. I'm hoping the DBX will be unnoticable in the chain.

EDIT:
A picture of the system without the screen in front of it:
http://dl.dropbox.com/u/10759510/For...nemaV1/6_1.jpg

The Behringer amps are all gonna be replaced with Crown CTs1200 for the LCR's and a Crown CTs2000 for the subs.. They are on their way.

Quote:

---

Originally Posted by Jonas_h

I know I can turn the gain down on the amp, but I dont like those "hack"-solutions... I'm not satisfied with the sound quality either. Have tried listening with and without the Behringer in the chain, and everything is much more transparent without it - more smooth and refined. I'm hoping the DBX will be unnoticable in the chain.

EDIT:
A picture of the system without the screen in front of it:
http://dl.dropbox.com/u/10759510/For...nemaV1/6_1.jpg

The Behringer amps are all gonna be replaced with Crown CTs1200 for the LCR's and a Crown CTs2000 for the subs.. They are on their way.

---

It’s no wonder you’re hearing hiss! I’ve seen that model before. I can’t see it on the JBL pro site unless its discontinued? I think the HF horns would have sensitivity of around 110db or less so its no wonder you’re hearing hiss. Turn the levels down! Those horns would blitz you’re ears in small room. My tiny little JBL control 5 tweeters can be rather bright hush if set at higher level in small room.

Quote:

---

Originally Posted by JBL 4645

It’s no wonder you’re hearing hiss! I’ve seen that model before. I can’t see it on the JBL pro site unless its discontinued? I think the HF horns would have sensitivity of around 110db or less so its no wonder you’re hearing hiss. Turn the levels down! Those horns would blitz you’re ears in small room. My tiny little JBL control 5 tweeters can be rather bright hush if set at higher level in small room.

---

Its a custom model.. Its the JBL 5641 enclosure with the 2226H driver (the LF section of the 3731) and the 4722-HF horn. They are 115db sensitivity :) I see no point of getting the 4722 with dual woofers in a home environment, therefore the "custom" model.

My room is acoustically treated and the horns sound very smooth and pleasant. The amp gains are set the same for LF/HF but the HF has been lowered 18db on the DCX. (Well, lowered 15db and the LF raised 3db:)). So they are level matched (the LF is 97db sens.) The system sounds absolutely amazing to my ears! As you know, it is a lot like the gear in the Leicester - i visited that cinema about a year ago, and was very impressed with the sound.

Quote:

---

Originally Posted by Jonas_h

I know I can turn the gain down on the amp, but I dont like those "hack"-solutions... I'm not satisfied with the sound quality either. Have tried listening with and without the Behringer in the chain, and everything is much more transparent without it - more smooth and refined. I'm hoping the DBX will be unnoticable in the chain.

---

To put the level down is not a hack, but instead a proper way to setup gain structure.
99% of the hiss problems when using DSPs are related to an improper setup.

There are many threads discussing "gain structure" here and on many sites.

I suspect your DBX 4800, even if it's a very good unit and better than the Behringer, will also introduce hiss in your HF if the gain of your amplifier is all the way up.

Quote:

---

Originally Posted by Jonas_h

Its a custom model.. Its the JBL 5641 enclosure with the 2226H driver (the LF section of the 3731) and the 4722-HF horn. They are 115db sensitivity :) I see no point of getting the 4722 with dual woofers in a home environment, therefore the "custom" model.

My room is acoustically treated and the horns sound very smooth and pleasant. The amp gains are set the same for LF/HF but the HF has been lowered 18db on the DCX. (Well, lowered 15db and the LF raised 3db:)). The system sounds absolutely amazing to my ears! As you know, it is a lot like the gear in the Leicester - i visited that cinema about a year ago, and was very impressed with the sound.

---

I’ve been going to Empire for 21 years since it had the original JBL 13KW Lucasfilm THX sound system that was damn hard edge hitting for Dolby Stereo 70mm!

Well which screen at Empire Leicester square because they have I think its 8 or 9 screens now, screen 1 is the big THX 56KW JBL screen! All the other screens use JBL I think the horns are the ones used for JBL 3678 I think some screens have duel 15” while ones not big enough to swing a cat around have single 15”.

I think you’ve set the HF far too high for small room. Try setting the level over LCR HF at 75db while using wideband pink noise. Use an RTA and SPL db metre to make sure each have the same pressure and that each tonally sound the same some might have more output in different parts of the frequency response only an RTA can see that where the SPL db metre can’t see it unless its plugged into an RTA! :D

Mute the HF LCR and look at the level on LF LCR and set the level at 75db. The subs should be within a few db they should merge smoothly.

They look kinder funky I could fit a model like that in my living room but not enough space for 5 fronts. I like the HF horn that I could easily fit in the room. How you managed it the LCR are far too close to each other where in small digital screen at Empire the spacing between LCR would be a lot greater to insure widest possible stereo.

Quote:

---

Originally Posted by Eaulive

To put the level down is not a hack, but instead a proper way to setup gain structure.
99% of the hiss problems when using DSPs are related to an improper setup.

There are many threads discussing "gain structure" here and on many sites.

I suspect your DBX 4800, even if it's a very good unit and better than the Behringer, will also introduce hiss in your HF if the gain of your amplifier is all the way up.

---

They are not all they way up.. I think they are about 20% max. But how would you use the JBL presets for example? They assume, that the amp gains are set the same for both LF and HF. The same goes for the Synthesis systems which use SAM1HF/SAM2LF. The DBX should perform equally to the BSS London units (at least according to specs). But I'll try the DBX in a couple of weeks and see how bad the hiss is.

Quote:

---

Originally Posted by JBL 4645

I’ve been going to Empire for 21 years since it had the original JBL 13KW Lucasfilm THX sound system that was damn hard edge hitting for Dolby Stereo 70mm!

Well which screen at Empire Leicester square because they have I think its 8 or 9 screens now, screen 1 is the big THX 56KW JBL screen! All the other screens use JBL I think the horns are the ones used for JBL 3678 I think some screens have duel 15” while ones not big enough to swing a cat around have single 15”.

I think you’ve set the HF far too high for small room. Try setting the level over LCR HF at 75db while using wideband pink noise. Use an RTA and SPL db metre to make sure each have the same pressure and that each tonally sound the same some might have more output in different parts of the frequency response only an RTA can see that where the SPL db metre can’t see it unless its plugged into an RTA! :D

Mute the HF LCR and look at the level on LF LCR and set the level at 75db. The subs should be within a few db they should merge smoothly.

They look kinder funky I could fit a model like that in my living room but not enough space for 5 fronts. I like the HF horn that I could easily fit in the room. How you managed it the LCR are far too close to each other where in small digital screen at Empire the spacing between LCR would be a lot greater to insure widest possible stereo.

---

My horns are not the ones that are used on the 3678 if that is what you mean?

As I said, I have level matched the HF and LF sections by setting them according to their sensitivity. I have verified this with REW, so no reason to turn down the HF :)

Dont underestimate the size of the horns...:
http://dl.dropbox.com/u/10759510/For...horn_vs_tv.jpg
Almost same size as a 32" TV.
I know that the space between them is not optimal, but my room is not big enough for more space. My next place, I will have a dedicated room for the home theater (currently its both living room and home theater), and hopefully it will be big enough to space out the speakers a little more. I'm getting good seperation of the sounds though. I can easily pinpoint the sound to the screen when the sounds travels from one speaker to the other. (The Pixar intro is a very good example!).

Quote:

---

Originally Posted by Jonas_h

My horns are not the ones that are used on the 3678 if that is what you mean?

As I said, I have level matched the HF and LF sections by setting them according to their sensitivity. I have verified this with REW, so no reason to turn down the HF :)

Dont underestimate the size of the horns...:
http://dl.dropbox.com/u/10759510/For...horn_vs_tv.jpg
Almost same size as a 32" TV.
I know that the space between them is not optimal, but my room is not big enough for more space. My next place, I will have a dedicated room for the home theater (currently its both living room and home theater), and hopefully it will be big enough to space out the speakers a little more. I'm getting good seperation of the sounds though. I can easily pinpoint the sound to the screen when the sounds travels from one speaker to the other. (The Pixar intro is a very good example!).

---

Why are you covering yourself up? Why not place the HF at head height at least then you’ll have some headroom. :D

I find pink noise to be the best test signal than any film soundtrack! I use wideband first to set up then run the AVR pink noise that is narrow band cycling over LCR and I’m listening for the same tone on the LF or HF one at time then as whole! If the tone is out it could be only a few bands that need adjusting its no big deal. Its not conspiracy you know. :D

I’d use a smaller 15” HF horn then if a set of them ever turned up the ebay for cheap price with compression driver and plastic horn.

I’d use a smaller 15” HF horn then if a set of them ever turned up the ebay for cheap price with compression driver and plastic horn. But I’d still have the damn things turned down! I know what they sound like in cinema and often they sound too damn loud and that’s not good for the ears!

LOL its no wonder you have hiss! :D I think you need DOLBY NR! One you have it up far too high! Two you’re far too close to them, where in cinema the front row would be at least 10 feet or more from the front of the screen and its only a few inches behind the screen where the JBL are! If you put you’re ear near to the screen before the films start between a music break I bet you’ll hear some type of hiss.

Upon a visit a few years ago for Toy Story 3D in Empire #1 I took a few pictures of the small digital JBL screens. So a little peak behind one of the JBL screens.

Quote:

---

Originally Posted by JBL 4645

LOL its no wonder you have hiss! :D I think you need DOLBY NR! One you have it up far too high! Two you’re far too close to them, where in cinema the front row would be at least 10 feet or more from the front of the screen and its only a few inches behind the screen where the JBL are! If you put you’re ear near to the screen before the films start between a music break I bet you’ll hear some type of hiss.

Upon a visit a few years ago for Toy Story 3D in Empire #1 I took a few pictures of the small digital JBL screens. So a little peak behind one of the JBL screens.

---

What does hiss have to do with how high they are placed? Dont see the connection. I only have the hiss when the DCX is in the chain... With the amps alone, the speakers are dead silent. So I'm still pretty confident, that the DBX will remove the hiss - or at least bring it down to an acceptable level. I will update the thread when I have tried it. I'm not putting in another component to do Dolby NR.

THe speakers in your picture are the 3678 as you mention. They have a tiny horn compared to the 4722, but the LF is pretty much the same. Just a little smaller, but same driver.

EDIT: The hiss with the Behringer DCX is not as bad as you make it sound... I'm just annoyed that its noticeable when I am listening for it! But when watching a movie, I dont notice it.

Quote:

---

Originally Posted by Jonas_h

What does hiss have to do with how high they are placed? Dont see the connection. I only have the hiss when the DCX is in the chain... With the amps alone, the speakers are dead silent. So I'm still pretty confident, that the DBX will remove the hiss - or at least bring it down to an acceptable level. I will update the thread when I have tried it. I'm not putting in another component to do Dolby NR.

THe speakers in your picture are the 3678 as you mention. They have a tiny horn compared to the 4722, but the LF is pretty much the same. Just a little smaller, but same driver.

EDIT: The hiss with the Behringer DCX is not as bad as you make it sound... I'm just annoyed that its noticeable when I am listening for it! But when watching a movie, I dont notice it.

---

That was joke about the Dolby NR.:p

Easy the level down or place a few equal HP filters over LCR HF and cut the level back. I trust the LF gives planet of bass mid and often deep lows around the 40Hz range. :D

The soundtrack level will mask the noise of hiss because its running many db higher over the hiss!

The soundtrack level will mask the noise of hiss because its running many db higher over the hiss!

Currently I have the levels on the amps set at like this a few nights ago the levels on the Alesis RA300 was at full but the levels on the MDX4600 was set differently. The levels for the HF on the Marantz 1050 for LCR HF are set way, way low I need to put my ear up to tweeter horns to hear any kind of hiss.

Set the levels for a computable level if the soundtrack level is higher than the hiss how high is the HF on its own over LCR with LF muted and all other channels as well is the HF within a peak of 85db (90db or over) will be far too bright I'll be reaching for the volume for mine.

I’m going to do a simple test! I’ll be back in jiffy!

This thread is too noisy, I quit.

I guess you hasn't got DOLBY SR type :D

Quote:

---

Originally Posted by Eaulive

This thread is too noisy, I quit.

---

Well said, sadly incredible isn't it.

Hi
What is this fuzz about hiss thing:confused:

I am running my M/T with McIntosh MC252 and for my subs two Crown MAi5000 and have no problem with hiss no matter where my volume knob is. My system is dead silence even if I use dbx260 in the signal chain.

There is alot of reason that can create hiss like power supply or bad interconnect or ground loop and so on....

Jonas.. with your new Crown amps and dbx4800 you will have a brand new system and with no hiss.

Thanks

Jonas_h

Why don't you make a video so we can hear the hiss first hand. I made a quick video because I had one idea I haven't yet tried.

http://www.youtube.com/watch?v=P38Le...layer_embedded

Just curious about the DBX 4800. What is the current street price?

Quote:

---

Originally Posted by Lee in Montreal

Just curious about the DBX 4800. What is the current street price?

---

In Denmark where I live, its about $6000.. but Denmark is very expensive with electronics! I got it for $1800, so it was a no-brainer for me when I was looking for a new speaker processing component anyway. I dont know the price in The US.. Probably $4000-$4500?

yes, appears to be ~$4000USD

Quote:

---

Originally Posted by Jonas_h

In Denmark where I live, its about $6000.. but Denmark is very expensive with electronics! I got it for $1800, so it was a no-brainer for me when I was looking for a new speaker processing component anyway. I dont know the price in The US.. Probably $4000-$4500?

---

$1800 is a gooood price considering it sells for more than three times this amount in your area. You got yourself a great deal. I like the fact that the 4800 is an 8 channel crossover. I wished my DCX had an extra channel for summed subs. But hey, for $340.00cad I can't complain. ;)

Quote:

---

Originally Posted by grumpy

yes, appears to be ~$4000USD

---

Denmark can be a pain when it comes to prices.. Even though one dollar is 5,5 Danish crowns, a rule of thumb says, that we can multiply the US price with 10 to get the Danish price. I actually think, that the one I got was bought in the US... Its a universal power supply, so it fits perfectly to our 230V power grid.

Quote:

---

Originally Posted by Jonas_h

Denmark can be a pain when it comes to prices.. Even though one dollar is 5,5 Danish crowns, a rule of thumb says, that we can multiply the US price with 10 to get the Danish price.

---

Must be a very high import tax. Something's rotten in the kingdom of Denmark :D

Quote:

---

Originally Posted by Lee in Montreal

Must be a very high import tax. Something's rotten in the kingdom of Denmark :D

---

Haha, yeah taxes are very high! But we get a lot of stuff for free then. But when you do not need all those things at the moment, you tend to think the money cuold have been spent better elsewhere (like new gear for the home theater :)).

But I am very happy for my purchase of the DBX. A week ago, I got a similar offer on 4 Crown CTs amps (just under half the price), so it has been a good couple of weeks... Bad for the bank account, but in the long run, I have saved a lot. :)

You mention you have the DCX? Do you have any problems with the SN ratio?

Quote:

---

Originally Posted by Jonas_h

You mention you have the DCX? Do you have any problems with the SN ratio?

---

I do have some floor noise, caused by the in/out digitsl amplification. I just reduce the input of my amps and increase the output of my preamp. BTW The noise is only audible when no music plays.

Strangely, my analog Behringer CX3400 is exempt of such noise.

Quote:

---

Originally Posted by Lee in Montreal

I do have some floor noise, caused by the in/out digitsl amplification. I just reduce the input of my amps and increase the output of my preamp. BTW The noise is only audible when no music plays.

Strangely, my analog Behringer CX3400 is exempt of such noise.

---

An analog crossover has physical attenuators at the output stage when a DSP has not.
The output of the DAC is fed directly to the output driver and then to the output connector.
Attenuation is done digitally, when you attenuate the output signal you're really not doing anything in the analog domain. You only have less headroom, that's why it's so important to set the gain structure properly with a digital unit, even more than with an analog one.
You have to use all the headroom possible, meaning you have to feed the unit with the maximum signal. Most units will accept in excess of +20dBu., many times more than any consumer preamp is able to supply.

Quote:

---

Originally Posted by Jonas_h

My horns are not the ones that are used on the 3678 if that is what you mean?

As I said, I have level matched the HF and LF sections by setting them according to their sensitivity. I have verified this with REW, so no reason to turn down the HF :)

Dont underestimate the size of the horns...:
http://dl.dropbox.com/u/10759510/For...horn_vs_tv.jpg
Almost same size as a 32" TV.
I know that the space between them is not optimal, but my room is not big enough for more space. My next place, I will have a dedicated room for the home theater (currently its both living room and home theater), and hopefully it will be big enough to space out the speakers a little more. I'm getting good seperation of the sounds though. I can easily pinpoint the sound to the screen when the sounds travels from one speaker to the other. (The Pixar intro is a very good example!).

---

Ah yeah, I like those!:) One day............................................... ....:D

A little update... Yesterday I hooked up the new amps and the DBX 4800... The gain on both amps are about 30% up for both the HF and LF driver, and there is absolutely no hiss. Im really satisfied with the DBX. S-sounds are now very good and not fatiging at all and it just sounds like, a thick fabric has been removed from the speakers.

Excellent. :applaud:
Thanks for the update ;)

In regards to gain structure... How is the best approach to ensure that this is done correctly? I can set the input/output sensitivity on the DBX. If I set the output sensitivity to 14dBu, the hiss is completely gone, but when I listen to movies on reference levels and there is a lot of action in the movie, the DBX almost hits the clipping point. This does not happen if I set the output sensitivity to 20dBu, but a small hiss is introduced though.

Is the best approach to adjust the gains on the amplifiers? I dont like the idea of not being able to level match the HF/LF sections on the DBX only by using the sensitivities of the individual drivers. If the amp gains are not equal, then I cannot be 100% sure that they are level matches as they should be.

Hi Jonas;

When you are at maximum signal level input, almost clipping is where you want to be, not clipping, but almost. That is where the device is capable of delivering maximum signal to noise.

What is currently in you signal path?

Quote:

---

Originally Posted by 1audiohack

Hi Jonas;

When you are at maximum signal level input, almost clipping is where you want to be, not clipping, but almost. That is where the device is capable of delivering maximum signal to noise.

What is currently in you signal path?

---

My path is Onkyo PR-SC5508 (surround processor, balanced outputs) -> DBX Driverack 4800 -> Crown CTs amps.

The gains on the amps are not set very high - I have read that Crown recommends setting them at 3/4 for best performance, so maybe they should be set higher for better performance?

Would it be a good idea to send pink noise from my processor set at 0dbFS and make sure, that the DBX is not clipping (but at "almost" clipping)? Of course with the amps turned off - otherwise I would go deaf :)

EDIT: The Crown gain comment edited.

The DBX makes me setup the sensitivities like this:

http://srforums.prosoundweb.com/index.php/fa/11236/0/

I can't see what you tried to post.

The short of it is since everything in the chain makes its own noise and adds it to the signal as it goes, you want to get all the gain you need in as early as possible as not to amplify the self noise further down the chain.

With the Onkyo wide open, you want the input stage of the 4800 almost clipping. I would just adjust the input of the 4800 by your listening trends, if you see it clipping when your running it hard, just pull some input gain out, it wont take long to get that set and if you clip it one in a while its not the end of the world.

If you can then set the output levels of the 4800 to unity, zero dB gain or cut (found in the last page of the crossover parameters I believe) and still have adequate voltage (and it should) to drive the amps to satisfactory level (loudness) with out the amp input attenuaters wide open that will likely be as good as it gets.

Keep me up on what and how you do with it. If I did a bad job at explaining let me know and I will try it again.☺

That was a good explanation! I will try it as soon as possible. When you say, unity gain on the output, does this mean setting it to the same value as the input?

I was linking to ths in the post before: http://www.martinloganowners.com/for...1&d=1211144106

This is the options I have for setting the input/output sensitivity.

In the crossover section as you page through I think it's the last parameter, all it says is gain with a little arrow pointing to a number, 0.0 is unity meaning if one volt comes in, one volt goes out, so now your not amplifying anything done after the input stage, just passing whats there along to the amp.

You can run that number from -60dB to +20dB if I remember right. If you can leave that at 0.0dB you will acheive the best signal to noise ratio while keeping the most dynamic range.

This individual channel output gain stage is independant form the input gain stage.

Okay, thanks. Now I know what you mean. (This is already at 0 though).

But the image I linked too. How would you set this for the output? The input should be set, so full scale on the Onkyo results in "almost clipping". Do I connect the amps to the DBX and do the same clipping test?

Quote:

---

Originally Posted by Jonas_h

Okay, thanks. Now I know what you mean. (This is already at 0 though) Do I connect the amps to the DBX and do the same clipping test?

---

You could do that if you needed maximum output from the amps, but chances are, and hopes are that you don't. If for instance the HF amp input attenuator knob needs only to be at say 10 o'clock for the HF part of the system to play as loud as you would ever play it, then having it there gives you all the volume you need, leaving all the remaining headroom available for dynamics with the minimum amplification of noise. If it is turned up any higher, it only raises the noise floor.

The LF amps may very well require the input attenuator be much higher, like say 3 o'clock or more but the same principal applies, advance the attenuator only as far as needed to achieve the desired maximum loudness. In the event that you don't get enough output with the amp attenuators wide open, (check the amp for switchable input sensitivity) you would then have to increase the 4800's output level to those channels that are not loud enough. At least woofers don't hiss.

One test would be to set all atenuators on the amps to zero and the volume control on the preamp at max. Then raise the amps' attenuators until the maximum sound level to be sustained at home is reached. This is most likely where the system will produce the least hiss...

Quote:

---

Originally Posted by Lee in Montreal

One test would be to set all atenuators on the amps to zero and the volume control on the preamp at max. Then raise the amps' attenuators until the maximum sound level to be sustained at home is reached. This is most likely where the system will produce the least hiss...

---

But the problem, and as I understand, the concern of Jonas, is to know exactly where to set the levels on the amplifiers for the different drivers.

Example, for my 4520 setup, I have to have the MF set 7dB lower than the LF, and the HF 11dB lower than LF.
It's easy to do with the driverack, but if you want to preserve headroom it's better to do it directly on the amps.
Now the problem is to know where to set the amplifier attenuator to have exactly this value.

What I do in this case is that I disconnect the speakers, drive the system right into the onset of clipping with the correct attenuation set on the driverack, then I reset the driverack to 0dB and lower the gain on the amp to return back to the onset of clipping.

Quote:

---

Originally Posted by Eaulive

But the problem, and as I understand, the concern of Jonas, is to know exactly where to set the levels on the amplifiers for the different drivers.

---

Yes, exactly. I found this in the Crown manual though: http://dl.dropbox.com/u/10759510/ForumPics/ctsLevel.gif

If the listings are precise, I can se exactly how much I'm lowering the gain on each channel.

In regards to the "analog gain structure" function on the DBX4800 I linked to earlier, can anyone explain this? Not much info in the manual, and I am a little confused if lower or higher values are preferable. With lower output-clip values, I am able to raise the gain on the amp and be closer to the 3/4 gain setting recommended by Crown.

I'll see what I can find.

Quote:

---

Originally Posted by 1audiohack

I'll see what I can find.

---

I would really appreciate that - thank you :)

Is that analog gain structure screen shown on your DBX 4800 screen? Or on a PC with the GUI open? I don't have a 4800 here to check. Are you working to cal an RTA mic to the 4800 or is this something else?

Sorry I am losing focus on what we are trying to do, but I am easily confused.☺

Do you have a mic for the DBX RTA? After you get the gain settings sorted out in the DBX you could certainly use the RTA to get the amp levels set as long as you stay out of the region where the room stores energy, RTA is time blind and will lie to you in that region.

Yes, the analog screen is shown on the PC. I have not used the psysical buttons yet, as System Architect is much easier. When I change the settings, a higher output clip value results in louder volume. But I dont know what the goal should be.. High amp gains and low output clip or the other way around. I will be home this weekend (have been travelling because of my work) and I will play with the settings while playing full scale pink noise and find the settings where "almost" clipping appears and set them amp gains for that.

Someone told me once though, that the Crown's would not come alive if its gain was set very low.

I have no plans on connecting a mic to the DBX. I have done some measurements in REW, but I would rather use the drivers sensitivity for level mathing rather than measurements. As you say, room acoustics play a big role in the measurements, and I will risk that a room mode gives me incorrect level matching between the HF/LF.

Hi Jonas;

The region where the RTA is incapable of giving useful data is easily calculated. RTA is a usefull if somewhat limited tool and you have it so, at the very least you could use it to get the three haystacks balanced ie subs, bass and HF. If you don't mind sharing, what are your room dimensions?


This whole gain structure concept is just like a compound mechanical linkage system with movable levers and fulcrums. I don't know about anyone else but once I got my head wrapped around it in a way that made sense to me, it seems fairly intuitive.

If my understanding of the concept and the attempted explanation thereof is only confusing, please let me know.

I figure it's only knobs and it is easy to return them to where you were before if you're unhappy with the results.

Sorry I have been away from the thread. I have played a lot with the gain structure "issue" and read about it, and I think I have it dialed in. I set the sensitivity so it was almost clipping when getting full input from my processor. I then set the amp gain so no measurements should be done at the processor to reach reference level when set to 0dbFS.

I measured the response with the amp gains set equal where the LF/HF were level matched on the DBX. Then I set the HF to unity on the DBX and lowered the amp gain to the HF until the measurements where identical. (Adjusted 10 "click" down, and lowered 2db on the DBX.). This should eliminate the influence of room gain on the measurements, as I used the "correct" level matching as reference.

I am very satisfied with the results now. It sounds amazing and absolutely NO hiss. Next upgrade will be a larger room - but this requires a new home, so its far out in the future :)

Thanks for all your help so far!

Quote:

---

Originally Posted by Jonas_h

I am very satisfied with the results now. It sounds amazing and absolutely NO hiss

---

After all is said and done, this is the most important thing!

I'm glad you enjoy your setup :applaud:

You're right! :)

But I still have to get everything dialed in, in regards to EQ etc. I have used Audyssey Pro for the calibration so far, and it does an ok job.. But I feel that it "removes" punch and other things in the sound to hide possible room artifacts. The acoustic in my room is very good, so I think that simple PEQ filters to the <1000hz area will yield in a better result.

I only did a quick read on Audyssey and can not argue it's strengths or limitations. I doubt it's execution is bullet proof however.

That said, if we are considering a reasonably standard acoustical space that does not have dimensions with severe aspect ratios or major openings into other rooms, the region where a room is mode dominated can be calculated easily as follows;

Multiply the speed of sound by three, divide the answer by the rooms smallest dimension, eg 344 X 3 = 1032 / 2.44m = 422.9Hz

Divide the speed of sound by two, divide the answer by the rooms largest dimension, eg 344 / 2 = 172 / 5.8m = 29.6Hz

In the above example the region between 30Hz through 423Hz is where modes dominate and cannot be controlled by EQ. Lowering certain offending bands of energy put into the room can only help by not aggravating the beast, but you then have limited the dynamics of the system.

RTA cannot give you good information in this region and EQ really can't help. Do the math and see if this is the region that you don't like what Audyssey is doing.

Quote:

---

Originally Posted by 1audiohack

I only did a quick read on Audyssey and can not argue it's strengths or limitations. I doubt it's execution is bullet proof however.

That said, if we are considering a reasonably standard acoustical space that does not have dimensions with severe aspect ratios or major openings into other rooms, the region where a room is mode dominated can be calculated easily as follows;

Multiply the speed of sound by three, divide the answer by the rooms smallest dimension, eg 344 X 3 = 1032 / 2.44m = 422.9Hz

Divide the speed of sound by two, divide the answer by the rooms largest dimension, eg 344 / 2 = 172 / 5.8m = 29.6Hz

In the above example the region between 30Hz through 423Hz is where modes dominate and cannot be controlled by EQ. Lowering certain offending bands of energy put into the room can only help by not aggravating the beast, but you then have limited the dynamics of the system.

RTA cannot give you good information in this region and EQ really can't help. Do the math and see if this is the region that you don't like what Audyssey is doing.

---

Why are you multiply the speed of sound by three?

It's a quick method to calculate axial modes out to the sixth. In a "normal" rectangular room this will get all modes out to where they are statistically spread out with insignificant energy even in a poorly dimensioned closed acoustical space.

Quote:

---

Originally Posted by 1audiohack

Multiply the speed of sound by three, divide the answer by the rooms smallest dimension, eg 344 X 3 = 1032 / 2.44m = 422.9Hz

Divide the speed of sound by two, divide the answer by the rooms largest dimension, eg 344 / 2 = 172 / 5.8m = 29.6Hz

---

That's an interesting formula. I suppose that the smallest measure is the ceiling's height, and that circa 400Hz is the usual upper range affected by the room, unless one has cathedral ceilings. :D

So, to lower that frequency value, I suppose that room treatment shall begin on the ceiling. Either raise it, or dampen it...

Yup.

The region where the acoustical performance of the space is mode dominated is bounded by the first mode of the space where f=.5C/RLD up to highest where f=3C/RSD. This is where RTA will lie to you, EQ can't help you, absorption is really your only weapon and your ears are your best tool.

Above and below this region RTA is an effective and efficient tool for individual driver and channel level setting and more. Knowing it's limitations is the key to getting good results, same with EQ, especially an auto EQ system.

Hi
Where did you get your formulas from?

This is what is explained in "Master Handbook of Acoustics"

A 10x16x23,3 ft room has a volume of 3,728 cu ft and the reverberation time is 0,5 sec.

F1=565/23,3=24,2 Hz
F2=11,250 x sqrt(0,5/3,728)=130 Hz
F3=4 x 130=520Hz

Between 24,2 and 130 Hz the wave acoustical approach of modal resonances is essantial.

Between 130 and 520 Hz is the transition region. It is a diffucult region dominated by wave lenghts often too long for rays acoustics and too short for wave acoustics.

Above about 520 Hz the modal density is very high, statistical conditions genarally prevail, and the simpler geometrical acoustics can be used.

This is the problems in small rooms witch we all have.

Thanks

Oh right,, make me go back to the books! ;)

I found in my fourth edition of the Master Handbook of Acoustics the equation you correctly quoted in the Modal Resonances in Enclosed Spaces chapter.

One thing to look out for is when formula for statistical spaces finds it's way into small room acoustics. Few of the intelectual tools of the trade for large rooms can be applied to small rooms. I am not arguing that the math is not correct, just the validity of some of it as it applies to small spaces. The 0.5 second reverb time is an indicator that this has happened.

If you look back a couple of paragraphs it is stated;

Quote:

---

Region B is that region we have studied in detail in which the dimensions of the room are comparable to the wavelength of the sound being considered. It is bounded on the low frequency end by the lowest axial mode, 565/L. The upper boundry is not definite but an approximation is given by what has been called the cutoff or crossover frequency given by the equation;

F2 = 11.250 (RT60/V) 1/2

Where F2 = cutoff or crossover frequencyin Hz
Where RT60 = reverberation time of the room in seconds
Where V = volume of the room in cubic feet

---

What is often overlooked in the attempted measurement of RT60 in small rooms is that the definition of RT60 has two parts, the first of which is often overlooked or ignored;

1 RT60 is the measurement of the decay time of a well mixed reverberant sound field well beyond the Dc', or the critical distance.

2 RT60 is the time in seconds for the reverberant sound field to decay 60 dB after the sound source is shut off.

Since in small rooms there is no Dc', no well mixed sound field, hence no reverberation but mearly a series of early reflected energy, the measurement of RT60 becomes meaningless in these environments.

What becomes meaningfull is the control of the early reflections because there is no reverberation to mask them.

Finally, the formula I used comes from the Handbook for Sound Engineers, 4th edition, Small Room Acoustics chapter by Doug Jones. I have also had the good fortune to have received some formal training from him and could use a great deal more!

All the best,
Barry.

Hi

Thats why I really don't get to mutch anal about theoretichal aspect when talking about small rooms. You are absolutely right when talking about RT60, thats why RT30 is more usefull.

It is more easier to do measurement over a particular room. You see more data how your room is rather then trying to calculate it by hand.

That's why I have abandonded the use of full range absorption in small rooms and use broadband diffusion integrated in helmholtz principle. The decrease in reverberation is as good as in full absorption and will not over dampend the room at the same time.

The stuff we read in varoius books is mainly for big auditorium and churches and so on and not for small rooms.

I can read therory just to get me more understanding about acoustics but it dosen't say how to approach it in a practical solutions, construction and so on.

If you have a good acoustical room as I have, the auto EQ will be easier to perform, but in my case I don't need it.:)

Thanks

Hi;

I remember the magnitude and decay charts you showed for your room and when you say it sounds good in there, I beieve it!

You obviously read and understand the acoustics stuff, and since you are interested in it, for accuracy I would just like to share a fundamental point;

Modal decay rates are not reverberation. Reverberation is the "time in seconds that it takes a diffuse sound field, well beyond a real critical distance, to lower in level by 60 dB when the sound source is turned off."

Modal decay rates are dB-per second (dB/s) rate of decay for a specific frequency.

By definition, reverberation does not exist in small rooms. RT nothing.

Not to be picky, really, it's just a very common mistake and some of my instructors are / were quite militant about it. I guess it rubbed off. OK last on that from me.

All the best and enjoy!
Barry.

Jonas_h

What video source components are you running in your home cinema?

Betmax
VHS
S/VHS
Laserdisc
DVD
HD-DVD
Bluray

Quote:

---

Originally Posted by JBL 4645

Jonas_h

What video source components are you running in your home cinema?

Betmax
VHS
S/VHS
Laserdisc
DVD
HD-DVD
Bluray

---

A computer with a bluray drive. All movies are stored on a NAS.

Quote:

---

Originally Posted by Jonas_h

A computer with a bluray drive. All movies are stored on a NAS.

---

Interesting! Let’s see the current new pictures of the set-up and around the room.

Amp Gain & Volume Control
Professional amplifiers do not have gain or volume knobs. They have attenuators. The amp will deliver full output no matter where the attenuators are set. Back in the analog days, broadcast level was @+8dB, pro was +4dB and consumer was -10dB. In the 70s & 80s doing rock concerts and discos, the amps were typically set all the way up, as the biggest amps were in the 200- 250 watt range. They weighed around 50 lbs a piece and were expensive. So you ended up being under- amped and trying to squeeze every last watt out it. Of course, then, you could walk by the pa before the show and tell if it's on- lots of hiss.

When you have the input attenuators at full on, the amp is amplifying the noise floor. Most amps do about 36dB of gain, which means at idle, your noise floor is 36dB louder.

In my world, we deal with mostly pro (+4) level. I set the amps around 11 o'clock and might have to knock them down to 10 o'clock. That's because our gain structure along the way is very good. Any kind of hiss is unacceptable to our clients. The other side of this coin is, of course, headroom. We want gobs and gobs of headroom. I want to see -8dB, -10dB on the processor input at a maximum. Any kind of dynamic range will kick the system way above that and we don't want to clip the amps. Clipping the processor gets real ugly, real quick.

I've seen some equipment manuals instructing the user to set levels at 0db. That is so wrong. Leave some headroom at every step along the way. Most cd players don't have VU or dB meters (Tascams do). It's interesting to watch the meters on different cd's to see the dynamic range.

How about DBX 480 VS 4800? I can get a DBX 480 for a reasonable price. F.N i have beringer ultradrive and ultracurve.

Edit: On what I have found it has only smaller changes and mostly cosmetical. Mabye I`ll have to buy it:)

Regards Karl