Not according to the block diagram, this shows the controls behind the adc and in front of the dac - but your conclusion is still correct.
If a bit loss of 5 is audible? Maybe, maybe a matter of taste.
You also write that the maximum attenuation is 30dB, but it is actually only 15db referenced to the digital signal input level - so there must be actually amplification to boost the signal to 15dB above the input - or adding of bits?. That can be seen when feeding the dcx digitally from any digital source - at 0 setting the signal level indicator reaches quite comfortable into the red zone.
Using the src which allows immediate switching between sampling modes one can perceive a difference especially between 16 bit and 24 bit, more so than between 48kHz and 92kHz sampling.
The difference is perceptible in the upper region - even with my ears that only can hear discrete sine up to 14kHz. The sound up there seems more extended - brighter - but I do not perceive a "harshness" there at lower bitrates. This harshness is actually more noticable at the midrange when comparing 24/96 to16/44.1, but not by any means earsplitting - one has to listen fairly concentrated.
That is what I have done as well, using six gang pot's from thel in germany, as I in this way use the full signal level of digital inputs into the processing chain.
But I understood his question that he just wanted to use the gain/attenuation controls for level matching of the speakers, not for the volume control, for which they are quite unsusable and cumbersome, considering the menu structure of the DCX.
Hi !
Another solution is to implement a dividing network on each of the outputs.
I use 20db attenuators made of two resistors soldered into the XLR connectors : 10Kohms in series with signal pin and 1Kohm between signal pin and ground pin.
This is called also a T-pad : Attenuator calculator or even Another calculator
It is very efficient, cost is close to zero (a bunch of resistors can be bought for a few cents) and it works a treat !
Thierry
Another solution is to implement a dividing network on each of the outputs.
I use 20db attenuators made of two resistors soldered into the XLR connectors : 10Kohms in series with signal pin and 1Kohm between signal pin and ground pin.
This is called also a T-pad : Attenuator calculator or even Another calculator
It is very efficient, cost is close to zero (a bunch of resistors can be bought for a few cents) and it works a treat !
Thierry
This is the unit I am using as I am only using 1 input and 5 outputs on the DCX to my subs, this unit allows me to boost the input signal and attenuate the output. It can be used purely as a 6 channel attenuator.
Behringer MX-882
Behringer MX-882
An externally hosted image should be here but it was not working when we last tested it.
Spearmint said:
Yes, but can it change the volume with one knob for 6 individual channels? It changes the volume from ONE input to 6 individual outputs or from 6 inputs to ONE output, right? I checked that one earlier but figured it couldn't be used as a volume control for a tri-amped setup.
Thmartin said:Behringer MX-882
Behringer has an answer for every situation haven't they
Is the MX882 "transparent" enough ? I mean, does it bring noise or distortion or whatever to the sound ?
Thanks
Thierry
I haven't noticed any colouring or noise from the MX882, but then again I am only using it with my subs.
UrSv said:
You are quite correct, and for my situation of using it only for my 5x subs it works well for me.
As has been posted previously here are a couple of cheap kits one as an individual & and the other as master level control.
An externally hosted image should be here but it was not working when we last tested it.
6 Channel individual
An externally hosted image should be here but it was not working when we last tested it.
6 Channel remote
BUT wait!
Preamble:
Some (20+) years ago I built a clone of a threshold (I think) solid state moving coil preamp.
consisted of 1 gain stage of 7 transistors in parallel and ran off 1 (-22v) supply rail.
To test, I connected it to a lab variable power supply, put signal generator on input, and oscilloscope on output
Could easily observe wave form etc and unit just sounded great!.
BUT
I varied power supply voltage and gain also varied..(can you see the question coming?)...
I can't design electronics hence the following question:..
Is it possible to design a simple amplifier stage of sufficient quality to give no gain but allow volume adjustement based purely on power supply voltage????
If so one could "easily" gang as many of these units as you liked and have them all syncronised in their attentuation
The intention is to be able to control outputs on DCX2496 dig XO
this has also been posted under the solid state forum
Any comments appreciated
cheers
Preamble:
Some (20+) years ago I built a clone of a threshold (I think) solid state moving coil preamp.
consisted of 1 gain stage of 7 transistors in parallel and ran off 1 (-22v) supply rail.
To test, I connected it to a lab variable power supply, put signal generator on input, and oscilloscope on output
Could easily observe wave form etc and unit just sounded great!.
BUT
I varied power supply voltage and gain also varied..(can you see the question coming?)...
I can't design electronics hence the following question:..
Is it possible to design a simple amplifier stage of sufficient quality to give no gain but allow volume adjustement based purely on power supply voltage????
If so one could "easily" gang as many of these units as you liked and have them all syncronised in their attentuation
The intention is to be able to control outputs on DCX2496 dig XO
this has also been posted under the solid state forum
Any comments appreciated
cheers
George, what you are describing is a voltage controlled amplifier. As it happens, Analog Devices builds a nice VCA chip for that application. Here's a project Dave R. built to control the volume of his Behringer. One pot controls all 6 VCAs.
http://home.comcast.net/~dreite/preamp/volume_control.htm
http://home.comcast.net/~dreite/preamp/volume_control.htm
The "Class A or Class B resistor" ist just to change the output of the circuit to match either a Class A or Class B amp. If you should built it to whatever amp jou use to get the best result (why is there a difference anyway?).
This is kind of what I would be using if I had the DCX
I just read the whole thread, and it look very interesting. I'm very interested in this peace of fine engeneering
But I wonder: Were is the cometition in this price segment? What other nice 3-way stereo digital x-overs are there out ther, with kind of the same fearures, for arround the same price (let's say upto 800 euro's)? Surely ther must be something out there?
This is kind of what I would be using if I had the DCX
I just read the whole thread, and it look very interesting. I'm very interested in this peace of fine engeneering
But I wonder: Were is the cometition in this price segment? What other nice 3-way stereo digital x-overs are there out ther, with kind of the same fearures, for arround the same price (let's say upto 800 euro's)? Surely ther must be something out there?
@Ouroboros: Thanks, nice to know
Do you mean the ZonePro™ 1260 & 1261 Digital Zone Processor? Thay are way more expensive!
If I were to send that kind of money, I would go for this: http://www.deqx.com/dsp-product/pdc26.html
Any other thoughts on nicely priced competition?
rick57 said:
Do you mean the ZonePro™ 1260 & 1261 Digital Zone Processor? Thay are way more expensive!
If I were to send that kind of money, I would go for this: http://www.deqx.com/dsp-product/pdc26.html
Any other thoughts on nicely priced competition?
Hi,
Still an interesting thread, started following it again after a year orso
But about the six-way VCA attenuator:
I would not recommend this.
I have experience with the that 4301 rms/vca combi, and the modular that/dbx vca's (2151/2180/2181 etc).
Also with the SSM2018, but not a lot. (Used it as a direct gain-controllable bal input stage once, didn't they stop the production of this chip? not sure....)
VCA's: If used unity gain:
all fine.
If used as an attenuator:
possible but even with trimming, (mostly 2nd) harmonics are quite present....
Pro audio consoles (Midas, SSL) with VCA faders have good but not simple solutions. Proper PCB layout also has a lot influence on the VCA performance.
I should avoid this if not really nescessary.
A passive volume control is way more easy.
This can also be made with a handful of reed relays with resistor (sil arrays) networks, and a rotary binary encoder.... 8 relais per channel gives 64 log (or other self definable) steps. must be fine?
cheers
Still an interesting thread, started following it again after a year orso
But about the six-way VCA attenuator:
I would not recommend this.
I have experience with the that 4301 rms/vca combi, and the modular that/dbx vca's (2151/2180/2181 etc).
Also with the SSM2018, but not a lot. (Used it as a direct gain-controllable bal input stage once, didn't they stop the production of this chip? not sure....)
VCA's: If used unity gain:
all fine.
If used as an attenuator:
possible but even with trimming, (mostly 2nd) harmonics are quite present....
Pro audio consoles (Midas, SSL) with VCA faders have good but not simple solutions. Proper PCB layout also has a lot influence on the VCA performance.
I should avoid this if not really nescessary.
A passive volume control is way more easy.
This can also be made with a handful of reed relays with resistor (sil arrays) networks, and a rotary binary encoder.... 8 relais per channel gives 64 log (or other self definable) steps. must be fine?
cheers
You should read the app. note from AD. The SSM2018T has been optimized for class AB (lowest distortion), so that's where I'd recommend you run it. That's what I"m doing. The 6 channels should track very well indeed, but one thing I've been thinking of is thermally linking the 6 chips together. THATCorp mentions mounting their VCA chips touching end to end to aid in thermal tracking.
Just to throw another gain control method into the pot.
I've been experimenting with 'chopper' gain control circuits. Have a look at the attached schematic (From my SPICE sim) to see what I mean. You can have as many channels as you want controlled by the variable mark-space clock. If you've got simulation on your machine it's worth having a look at how it performs.
You will only get about 30dB max range though, but this may well be adequate.
I've been experimenting with 'chopper' gain control circuits. Have a look at the attached schematic (From my SPICE sim) to see what I mean. You can have as many channels as you want controlled by the variable mark-space clock. If you've got simulation on your machine it's worth having a look at how it performs.
You will only get about 30dB max range though, but this may well be adequate.
Attachments
hmm
The distortion gets more at the lower controlled levels.
It is all possible to make it very low-dist, audiophile and elegant, and I'm sure the 6way 2018 of transducer is so...
But there is nothing better to it compared with the passive thing, exept maybe for the tracking.... within 1% accuracy will do for me....
IMO: the chopper will give a lot of residual side effects to cope with?
The distortion gets more at the lower controlled levels.
It is all possible to make it very low-dist, audiophile and elegant, and I'm sure the 6way 2018 of transducer is so...
But there is nothing better to it compared with the passive thing, exept maybe for the tracking.... within 1% accuracy will do for me....
IMO: the chopper will give a lot of residual side effects to cope with?
All the speculation in this thread is......amusing.
Transducer,
I wouldn't worry about thermally linking the six chips. I investigated this awhile back and I couldn't identify any appreciable mistracking with normal temperature variations. Since most likely a unit like this would be left on all the time and the ambient conditions will be unvarying I don't see it as a problem.
Regarding distortion performance of the SSM2018T: It's very low. Class AB biasing of the gain-core and providing an input signal that is nominally near the 1 volt RMS level is the optimum setup. Depending upon the output levels of the DSP crossover used a bit of optional padding might be required, but it's unlikely.
I think you'll find that multiple SSM2018T VCA's will track very closely with each other. The worst-case matching I've found between any of six channels at VCA settings between 0db and -30db is approximately 0.08db. That's close enough for me.
You know where to find me if you need a hand. How's the dipole project coming?
Cheers,
Davey.
Transducer,
I wouldn't worry about thermally linking the six chips. I investigated this awhile back and I couldn't identify any appreciable mistracking with normal temperature variations. Since most likely a unit like this would be left on all the time and the ambient conditions will be unvarying I don't see it as a problem.
Regarding distortion performance of the SSM2018T: It's very low. Class AB biasing of the gain-core and providing an input signal that is nominally near the 1 volt RMS level is the optimum setup. Depending upon the output levels of the DSP crossover used a bit of optional padding might be required, but it's unlikely.
I think you'll find that multiple SSM2018T VCA's will track very closely with each other. The worst-case matching I've found between any of six channels at VCA settings between 0db and -30db is approximately 0.08db. That's close enough for me.
You know where to find me if you need a hand.
How's the dipole project coming?Cheers,
Davey.
Hey Davey,
The dipole project is stalled, while I await the arrival of the DCX2496 at my local pro audio shop. Which reminds me, I should call them today and bug them about it.
Trying to find the time to put the other bits together, including the VCA volume thing, an LM1875 amp for the tweeters, and cabling.
Once the DCX arrives, it's time to do more measurements and put the system together. How much padding did you find was necessary using the DCX into the SSM2018T?
Cheers,
Ron
The dipole project is stalled, while I await the arrival of the DCX2496 at my local pro audio shop. Which reminds me, I should call them today and bug them about it.
Trying to find the time to put the other bits together, including the VCA volume thing, an LM1875 amp for the tweeters, and cabling.
Once the DCX arrives, it's time to do more measurements and put the system together. How much padding did you find was necessary using the DCX into the SSM2018T?
Cheers,
Ron