Help Needed: Level matching

kapvin

Well-Known Member
Joined
Jul 23, 2009
Messages
1,141
Points
113
Location
Gurgaon / NCR
da6r.jpg


The picture above explains mostly everything.

my trusty two decade old radioshack meter is reading 98dB. Average, not peak. that's really quite loud.

but if you look at the crossover level behind, only the -40dB level is reading on the woofer channels, and the signal going to the mids and tweeters are much below that.

so I am losing dynamic range and musical information.

the culprit is level matching between pro-audio and consumer audio gear. my crossover outputs at +22dBu (0dBFS). that really loud, given the high input sensitivity of the emotivas, it would easily clip them at about 10 to 20db below that.

how do you suggest I get the levels to match?

some options

1. get power amps with input sensitivity / level controls or use an an integrated for each channel pair.. woofer, mids, tweeters

2. insert a level control inbetween. need 6 channels of level control.

3. create a fixed attenuation between the crossover and power amp? any ideas on how to go about this.

any other solutions that comes to mind. Please let me know

I am not keen to spend spectacular amounts to achieve this. any help or suggestions will be extremely welcome

best wishes
 
source ->(RCA)->emotiva umc-1(volume control)->(RCA->XLR)->DEQ->(AES-XLR)->DCX->(XLR-RCA)->Power amps.(emotiva UPA)
 
Last edited:
My mind is slow, and I'm not sure that I understand, but if the problem is something like the different levels on pro balanced and domestic unbalanced, then maybe something like this might be useful...

Balanced/Unbalanced Line Level Converter


Balanced/Unbalanced is a topic that I made a mental note to investigate a day or two ago. Thanks for the nudge :)
 
Hi, one suggestion is using S-convert or its alternative.

Don't know about the availability of S-convert though.

Samson S-convert

UMC-1-> S-convert -> DEQ -> DCX-> S-convert -> UPA


You are also using the INUKE ?
 
The picture above explains mostly everything.

my trusty two decade old radioshack meter is reading 98dB. Average, not peak. that's really quite loud.

but if you look at the crossover level behind, only the -40dB level is reading on the woofer channels, and the signal going to the mids and tweeters are much below that.

so I am losing dynamic range and musical information.

the culprit is level matching between pro-audio and consumer audio gear. my crossover outputs at +22dBu (0dBFS). that really loud, given the high input sensitivity of the emotivas, it would easily clip them at about 10 to 20db below that.

how do you suggest I get the levels to match?

some options

1. get power amps with input sensitivity / level controls or use an an integrated for each channel pair.. woofer, mids, tweeters

2. insert a level control inbetween. need 6 channels of level control.

3. create a fixed attenuation between the crossover and power amp? any ideas on how to go about this.

any other solutions that comes to mind. Please let me know

I am not keen to spend spectacular amounts to achieve this. any help or suggestions will be extremely welcome

best wishes


Kapvin,

Having a proaudio gear setup and yet you are finding it difficult to setup the level. I thought you were into studio setups.

There is a golden rule, set every equipment -3dB below the clipping point and then check for every overloading in subsequent stages in chain.


BTW what do you mean by XO reading -48dB and your meter showing 98dB, don't you think you are mixing many things here? :o
 
Last edited:
thank you both!

i was also reading up about a highly recommended similar product

ART CLEANBoxPro Stereo Balanced/Unbalanced Converter 245-868

the samson has slightly better specs than this but I have not heard about the samson before. but I was looking more at a passive solution (there are already too many active links in the chain). I would need about 15-20dB of gain reduction at the least.

one alternative is to look at a pro-amp solution and the Behringer a-500 presents itself in the silent category and the crown xls-1000/1500 in the fan category.

any experience of either?
 
Kapvin,

Having a proaudio gear setup and yet you are finding it difficult to setup the level. I thought you were into studio setups.

There is a golden rule, set every equipment -3dB below the clipping point and then check for every overloading in subsequent stages in chain.


BTW what do you mean by XO reading -48dB and your meter showing 98dB, don't you think you are mixing many things here? :o

Kanwar,

1. I am not into studio setups. I work in CPG industry, so as far away as possible for electronics and studios. I visit studios as a client of the people who book the studios, when they want to show me the finished product. (if you get what I mean)

2. this is for my home -> my DIY active system. and the pro part was introduced into my system about 1-2 weeks ago

3. the problem is the mixture of pro and consumer gear (and the fact that my power amps have no gain setting and have a very high input sensitivity)

Now to the point I was making

1. 98dB "C" wtd is the average SPL in the room when the photo was clicked. (As measured with the radio shack SPL meter) at this high SPL, my amps would be putting out 10 watts total (give or take). which is about -16 dB of full rms power.

2. Now look at the DCX outputs level on channel 1 and 3 (which feeds the midbass). the output level for -40dB is just lighting up occasionally (there is a peak hold function) . therefore the output from the crossover is between -40dBfs and -30dBfs. (ie) there is atleast between 30 and 40dB of headroom still left in the crossover. (let us for arguement sake take 35dB as the head room)

3. but there is only 16dB of headroom left on the amplifier - which means that when the amp clips - there will be still some 20dB of headroom to left unused on the crossover.

I know about the golden rule. point is. I cannot change gains on my power amp.

3 options that come to mind:

1. passive attenuation (fixed or variable ).
2. active attentuation - using a cleanboxpro or something similar
3. get an amp with gain controls. (need 4 channels)-- so are you making those lower power amps quickly or what?

hope you have a clearer picture of my problem.

Cybercat --> the inuke is for the sub, which is right now not part of the equation. but that has never been an issue, because 1. gain controls. 2. UMC outputs balanced for for the sub.
 
My mind is slow, and I'm not sure that I understand, but if the problem is something like the different levels on pro balanced and domestic unbalanced, then maybe something like this might be useful...

Balanced/Unbalanced Line Level Converter


Balanced/Unbalanced is a topic that I made a mental note to investigate a day or two ago. Thanks for the nudge :)

It is absolutely that. Your mind is anything but slow! :-)

the problem is accentuated by the fact the the crossovers output at the highest possible pro level and the amp is very sensitive consumer gear. (opposite ends of the spectrum, so to speak)
 
the culprit is level matching between pro-audio and consumer audio gear. my crossover outputs at +22dBu (0dBFS). that really loud, given the high input sensitivity of the emotivas, it would easily clip them at about 10 to 20db below that.

how do you suggest I get the levels to match?

some options

1. get power amps with input sensitivity / level controls or use an an integrated for each channel pair.. woofer, mids, tweeters

2. insert a level control inbetween. need 6 channels of level control.

3. create a fixed attenuation between the crossover and power amp? any ideas on how to go about this.

any other solutions that comes to mind. Please let me know

I am not keen to spend spectacular amounts to achieve this. any help or suggestions will be extremely welcome

best wishes

Please have a look at this relationship between different audio scales: World Audio Levels and Scales Document |

At +22 dBu your signal's voltage is 9.752 Volts. In fact 0 dBFS is 24 dBu = 12.277 V. Since your crossover is pro device, its nominal line level voltage is 1.228V which equates to 0 on the VU scale. And 4 dBu on the absolute scale. 1.228 V = nominal line level voltage for home audio devices, is the nominal voltage level expected by your non-pro devices. Even assuming overload margin of 6 dB for your non-pro devices, there is very good chance of overloading the non-pro devices when fed with +22 dBu signal (22-4 = 18 dB above nominal). At the mildest, it could lead to distortion or some oscillation/instability. At its worst, it could blow up something.

Now, where does the crossover get so much signal level from? Does it have some inbuilt master volume control?

PS: one person who has extensive experience of the Behringer DEQs is Sridhar of ARNS Bangalore as they are used in equalising the Emerald Physics speakers (for which he is the dealer).
 
Last edited:
It is absolutely that. Your mind is anything but slow! :-)

I picked on a couple of key words to identify the issue without understanding the. It's the sort of thing bad managers do to look clever in meetings :lol:. Or how good auditors instinctively home in on the dodgy transaction :ohyeah:

But I am interested in this balanced/unbalanced thing. It is not actually a problem for me now, but it is going to be shortly! I do know that simple conversion can be done in cabling (what you are doing now?) but that doesn't deal with levels. As an indication of level differences: my balanced->balanced interface/powered-speaker connection sees the volume knob at 9 o'clock, unbalanced sees it at 3 o'clock.
 
Kanwar,

1. I am not into studio setups. I work in CPG industry, so as far away as possible for electronics and studios. I visit studios as a client of the people who book the studios, when they want to show me the finished product. (if you get what I mean)

2. this is for my home -> my DIY active system. and the pro part was introduced into my system about 1-2 weeks ago

3. the problem is the mixture of pro and consumer gear (and the fact that my power amps have no gain setting and have a very high input sensitivity)

Now to the point I was making

1. 98dB "C" wtd is the average SPL in the room when the photo was clicked. (As measured with the radio shack SPL meter) at this high SPL, my amps would be putting out 10 watts total (give or take). which is about -16 dB of full rms power.

2. Now look at the DCX outputs level on channel 1 and 3 (which feeds the midbass). the output level for -40dB is just lighting up occasionally (there is a peak hold function) . therefore the output from the crossover is between -40dBfs and -30dBfs. (ie) there is atleast between 30 and 40dB of headroom still left in the crossover. (let us for arguement sake take 35dB as the head room)

3. but there is only 16dB of headroom left on the amplifier - which means that when the amp clips - there will be still some 20dB of headroom to left unused on the crossover.

I know about the golden rule. point is. I cannot change gains on my power amp.

3 options that come to mind:

1. passive attenuation (fixed or variable ).
2. active attentuation - using a cleanboxpro or something similar
3. get an amp with gain controls. (need 4 channels)-- so are you making those lower power amps quickly or what?

hope you have a clearer picture of my problem.



Thanxz for letting me know about yourself.:D


Best option is to go for passive attenuation, its nice and easy way to do. It will help you to match the high sensitivity consumer level gear with proaudio ;)
 
Please have a look at this relationship between different audio scales: World Audio Levels and Scales Document |

At +22 dBu your signal's voltage is 9.752 Volts. In fact 0 dBFS is 24 dBu = 12.277 V. Since your crossover is pro device, its nominal line level voltage is 1.228V which equates to 0 on the VU scale. And 4 dBu on the absolute scale. 1.228 V = nominal line level voltage for home audio devices, is the nominal voltage level expected by your non-pro devices. Even assuming overload margin of 6 dB for your non-pro devices, there is very good chance of overloading the non-pro devices when fed with +22 dBu signal (22-4 = 18 dB above nominal). At the mildest, it could lead to distortion or some oscillation/instability. At its worst, it could blow up something.

Now, where does the crossover get so much signal level from? Does it have some inbuilt master volume control?

PS: one person who has extensive experience of the Behringer DEQs is Sridhar of ARNS Bangalore as they are used in equalising the Emerald Physics speakers (for which he is the dealer).

the pro levels come from the DEQ which converts the analogue signal into digital and sends it to the dcx over an aes/ebu connection. also a digital optical signal can be sent in the same way.

i am not too worried at all about about instability / oscillation or blow ups. the amps will clip well before that, which I will hear, i will find the sound unbearably loud well before that as well.

thanks for the heads up on sridhar. do you have any contact details?
 
Thanxz for letting me know about yourself.:D


Best option is to go for passive attenuation, its nice and easy way to do. It will help you to match the high sensitivity consumer level gear with proaudio ;)

I was hoping you would say that. :)

(or give me a top secret/ trial model of a lower power zypher labs amp to test against future sale):D

could you guide me through this? there 2 changes required - balanced to unbalanced and then attentuation. cost dictates that we do the attentuation on the unbalanced leg (for variable) but i am thinking fixed attenuation.

any guidance on how to calculate the resistor values and the topology. (is it as simple as a 10k pot each between hot and ground and cold and ground?

if so then the attenuation would be by using 2 resistances X and 10k-X between hot and ground and the output hot to be wired to the junction of the 2 resistors? (and similarly for cold)?

all help would be really welcome!
 
Of the various product links, I like the look of the Art box. Great selection of Inputs and outputs. I wonder, though, if these kind of boxes are audio transparent?
 
any guidance on how to calculate the resistor values and the topology. (is it as simple as a 10k pot each between hot and ground and cold and ground?

if so then the attenuation would be by using 2 resistances X and 10k-X between hot and ground and the output hot to be wired to the junction of the 2 resistors? (and similarly for cold)?

all help would be really welcome!

I am assuming that your consumer amp is unbalanced.

Get a 10K pot [easily available], Use only HOT and GND, leave COLD from the Balanced source. Wire it just like a potentiometer for volume. I hope you know this
and take the output from the middle pin of the pot for the amplifier. Also earth the metal body of POT with GND. Set the level as desired, simple.
 
Last edited:
Get the Award Winning Diamond 12.3 Floorstanding Speakers on Special Offer
Back
Top