• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Reasonably priced good quality 6 or 8 channel USB DAC?

I use JRiver for this, it works very well. It installs a Windows sound driver, you route the sound to that, and then you use the DSP facility in JRiver. It has crossovers built in already, although I use FIR ones from rePhase.

I'm currently using an Asus u7 sound card, and use 6 of the 8 available channels. I also have an M-Audio 10 channel interface, and am working out whether it's worth bothering to use the Lynx AES16 and getting five stereo DACs for the horn system.

I've used other systems to host the DSP, such as Reaper (I think that foobar might be able to do it, not tried), but JRiver is dead easy to set up, fiddle with and it's very stable.
Cool. Yes I plan on doing something similar (though not with Jriver, there are very lightweight DAWs around as well meant only for hosting VST's etc) but it depends on what I'll choose in the end for crossover work.

As for getting 5 stereo DACs, this is technically not the ideal solution I believe as those 5 stereo DACs will all be running from their own clock (unless they can sync perfectly to a master clock comming from the Lynx, some studio DACs are able to do this but almost no consumer DACs I've seen). Better to get a multichannel DAC like the MOTU UltraLite mk4 discussed here I think.
edit: I mean, if one of those DACs is off sync by just one sample at 44.1kHz, and this is between for instance the DAC for the mid-driver and the DAC for the tweeter, then the resulting phase shift between those drivers is audible already at certain off-axis locations (around the -3dB point off-axis for LR for instance where it is very sensitive, or at the flat on-axis for a Butterworth crossover).
 
Last edited:
As a side note: the Anaview amps are very picky with the source: you need a balanced source with low impedance.
That also mean you will not be able to use passive attenuators in fornt of the amp.
26dB of gain is a lot, especially with a unit that outputs 20+dBu like the Motu.
 
As a side note: the Anaview amps are very picky with the source: you need a balanced source with low impedance.
That also mean you will not be able to use passive attenuators in fornt of the amp.
26dB of gain is a lot, especially with a unit that outputs 20+dBu like the Motu.
Ahaa. Very good info, thanks!
This is something I was wondering already. The effect of passive attenuators on impedance (I had not looked into this yet).
Ok.. I was indeed thinking of setting it up with the MOTU and it's trimmers and then see in practice how much passive attenuation I need and then in the end place those.. but so my little plan has a problem.. Hmm.

I do like the Anaview AMS amps otherwise. I have an AMS0100 right now (2x50W @4ohm) and I've loved it from the very first moment right up untill now several years later. Not much equipment does that for me. I am aware of the gain and input impedance and preference for balanced signals (it is possible however to hook it up unbalanced and add a resistor to the open cable to match the resistance of the unbalanced cable and still get the same result as balanced I read).
But the AMS0100 sounds just fantastic and it actually measures a little bit better around the 1W output and lower (where most of the music is when in actual use) than the smaller NCore amps from Hypex (though the NC400 and NC500 still beat the AMS0100 and AMS1000). So I was giving it preference over the Hypex NC252MP and NC122MP etc (which are used in the Fusion plate amps).

So time to rethink things.. No passive attenuators :( And I'm really really not looking forward to making active attenuators..
 
As a side note: the Anaview amps are very picky with the source: you need a balanced source with low impedance.
That also mean you will not be able to use passive attenuators in fornt of the amp.
26dB of gain is a lot, especially with a unit that outputs 20+dBu like the Motu.
My hypex amps have 26dB gain and work fine with my 8A and ultralite. No additional attenuation required and software volume control.

Just, what's the sensitivity of your drivers and the input impedance of the anaview?
 
Last edited:
Btw, since I am very lucky to be amongst knowledgeable people here.. :)
How does one actually calculate the Volt RMS input level needed at an amp to reach for instance 1W at 4ohm output of the amp?
Is it as simple as calculating the voltage level input at the amp (lets say 1Vrms) times the amp gain (20.6 for AMS0100 and 27.5 for AMS1000) to get the output voltage? Which then can be converted to Watt at a specific impedance/ohm of the driver at that freq?
So 1Vrms input * 20.6 = 20.6 Vrms at the amps output?
And then I know only that 2.83 V (don't know if this is V peak or rms actually) is 1Watt at 4ohm and 2W at 8ohm. I assumed it is Vrms but maybe I'm making an error here please correct me if I do. So I'm assuming that 20.6Vrms = (20.6 / 2.83) Watt = 7.28 Watt @4ohm? Hmm.. I don't trust my calculation one bit! haha.
 
From hypex 252 data sheet

Input sensitivity
Below a formula is given to calculate the balanced input signal level for a desired output level. An example is
given using PRATED 250W, Load 4Ω, Gain 25.6
√ ∗ Ω
(

)
= ∗ √ = ∗ (

. ) =

√ ∗
(
.

)
= . . ∗ √ = . ∗ (
.
. ) = .
 
My hypex amps have 26dB gain and work fine with my 8A and ultralite. No additional attenuation required and software volume control.

Just, what's the sensitivity of your drivers?
I was planning to use 1 AMS0100 with 20.6dB gain for the mid drivers which will be the sb17nbac35-4 which have a sensitivity of 90dB at 4ohm (and I'll actually use them in this range so no drop in sensitivity or only very minor from minor EQ corrections unless one also counts baffle step correction then sensitivity will drop for some of the range).
And 1 AMS0100 for the tweeters BlieSMa T34B-4 which have a sensitivity of 97dB at 4ohm, but I'll need to EQ them a bit to get them to 1400-1500Hz so in reality they'll have a 94dB sensitivity or so though at 1400Hz or so their impedance is more like 7-8ohm. (edit: this is not true as 1400Hz is -6dB for the crossover so the sensitivity will truly be about 96-97dB not counting for baffle step correction)
I haven't decided on the woofer yet, will know in about a week I hope, but planned on using the AMS1000 for them which has 27.5dB of gain. Sensitivity will likely be 80dB or so in reality as I will probably do some EQ / Linkwitz transform on the low end of the woofer.
 
Last edited:
Screenshot_2018-10-27-22-00-34-88.png
 
Oh and the imput empedance of the Anaview AMS1000 is: (where I think 2.5K is min and 14.5K is max, numbers for the AMS0100 are similar)
Input impedance single ended (*1) Non symmetrical on positive and negative inputs 2.5 14.5 kΩ
Input impedance balanced (*1) Non symmetrical on positive and negative inputs 1.37 14.5 kΩ
(*1) The input impedance on IN+ and IN- is not identical and also different between channels. See application notes below for more information.
APPLICATION NOTES Optimizing input stage CMRR This is simplified drawing of the input of AMS1000. It is a typical circuit used where the source impedance is well known and does not vary too much. Input current are calculated when a balanced signal is applied. As can be seen the input impedance is not the same on both inputs and depending on which type of signal is applied (single ended or balanced) the input impedance changes. This is however not a problem as long as a few precautions are made. Common mode rejection CMRR will be significantly improved by having the same source resistance on both the inputs. Below is shown a setup with an impedance balanced single ended source. This requires a balanced cable.
It is quite common to have a series resistance of 50ohm or more on the signal output so if the same resistance is placed in the opposite side of the signal of either sending or receiving side of the cable the CMRR rejection is intact. If a balanced signal source is used the following setup applies.
If long cables are used the cable impedance itself can contribute a lot to the series impedance and since that impedance is not very well defined (symmetrically) it can be an advantage to increase both the diff mode and common mode input impedance. In such a case an additional circuit as below can be added before the AMS module.
(some images missing from the copy paste above)
edit: full pdf manual for all the Anaview amps can be downloaded here: http://www.lautsprechershop.de/inde...prechershop.de/hifi/anaview_uebersicht_en.htm
 
Last edited:
I think a 24dB/oct Linkwitz-Riley crossover is a beatiful perfect and natural thing including its phase shift :)
Yeah me too, I always end up with 4th order when IIR. I'm brick wall FIR right now while playing with crossover points - I use horns, and the flexibility to cut them off quickly is handy, they can get quite funky if you stretch them wide. Another benefit of FIR is being able to unroll the phase of the reflex port, which does tighten up the bass nicely.

But anyway, do post when you've decided on software to host the DSP, I'm always interested to hear about alternative solutions.
 
No...

Many useful calculators here: http://www.sengpielaudio.com/Calculations03.htm

For this problem:

http://www.sengpielaudio.com/calculator-gainloss.htm

1V in, 20.6dB gain yields 10.7 Vout

View attachment 16927
Aah thanks! I suspected I was messing up :)
So voltage ratio gain, does this translate to the following:
20dBU = ~7.75 Vrms
7.75 Vrms with 20.6dB gain = 7.75Vrms * 10.72 = 83 Vrms?
Now if I can translate 83Vrms to a specific Watt at 4ohm then I can see how badly the amp is clipping at this input level.
 
20dBU = ~7.75 Vrms

http://www.sengpielaudio.com/calculator-db-volt.htm

1540681094954.png


7.75 Vrms with 20.6dB gain = 7.75Vrms * 10.72 = 83 Vrms?

That calculation appears to be correct.

1540681209833.png


Now if I can translate 83Vrms to a specific Watt at 4ohm then I can see how badly the amp is clipping at this input level.

http://www.sengpielaudio.com/calculator-ohm.htm

1540681302921.png


That's a lotta watts.

What are we talking about here?

1540681513607.png


Work backwards, find the (unspecified) sensitivity - the input voltage for maximum (1% distortion) output:

50W at 4 Ohm = 14.1Vrms out

1540681631687.png


So, 14.1Vout reduced by 20.6dB gain = 0.093 Vin

oops.

So, 14.1Vout reduced by -20.6dB gain = 0.093 ratio of Out to In = 0.093 x 14.14Vout = 1.31Vin for rated output.


1540681730159.png


1540683656439.png


Interesting.

More reasonable.
 
Last edited:
(see correction above)
 
http://www.sengpielaudio.com/calculator-db-volt.htm

View attachment 16929



That calculation appears to be correct.

View attachment 16930



http://www.sengpielaudio.com/calculator-ohm.htm

View attachment 16931

That's a lotta watts.

What are we talking about here?

View attachment 16932

Work backwards, find the (unspecified) sensitivity - the input voltage for maximum (1% distortion) output:

50W at 4 Ohm = 14.1Vrms out

View attachment 16933

So, 14.1Vout reduced by 20.6dB gain = 0.093 Vin

oops.

So, 14.1Vout reduced by -20.6dB gain = 0.093 ratio of Out to In = 0.093 x 14.14Vout = 1.31Vin


View attachment 16934

Interesting.

More reasonable.

:D
Thank you again!
You even calculated the input voltage for 1% THD which is exactly what I also wanted to see.
So the AMS0100 seems perfectly reasonable for a 2V output DAC. Just not so much for the MOTU with 7.75V..
Since passive attenuators are not an option and I don't want to build an active circuit (since I know absoutely 0 about making such a thing and it probably involves an op-amp and sound quality reduction) all that is left is the trim controls of the MOTU which likely operate the internal digital volume of the DAC (32-bit, no digital resolution loss, but no lowering of the noisefloor either so dynamic range reduction). So if we start with a dynamic range of 123dB at 7.75V we get a dynamic range of about 107dB at 1.31V. Which isn't that bad still actually.. Little bit of a waste though and the AMS0100 has a dynamic range of 115dB. Am I correct in thinking that the THD is not affected by the trimming down of the level, only the relative noise / dynamic range?
 
Since passive attenuators are not an option

Why not?

Am I correct in thinking that the THD is not affected by the trimming down of the level, only the relative noise / dynamic range?

Someone with more experience/measurement capability can reply on this.

I don't know that I have any measurable (with what I have as tools) electrical THD to investigate (too much noise in the rigging, using PC on-board ADC across the room from the analog outputs).

For speakers, though, I'd say I've noticed the THD rises faster than the signal as level is increased. None, none (or buried in noise floor), then up it goes.

My recent little experimentation with excessive digital attenuation - preamp at maximum and attenuation of digits by 48dB - didn't raise any notable nasties here (to my deaf ears).
 
The amp power for the mids seems reasonable.
http://www.sbacoustics.com/index.ph...7nbac35-4-br-font-color-c70039-shipping-font/
60W rated power, but I don't want to push them to their limit anyhow.
And at 1W for 90dB/1m this means 32W for 105dB totally clean from the amp's point of view for one driver, so 111dB/1m for 2 speakers. And I will actually be listening at 1.2m to 1.4m distance so rough guess 108dB at my actual listening position for the mids seems loud enough :)
Tweeters need a bit more trim (and loss of dynamic range) at the DAC for this level.
 

Since the Anaview AMS amps have a strange input stage which doesn't take high impedance outputs well. As I understand it adding a passive attenuator will raise the impedance the Anaview sees at its input.
But I really don't know my things when it comes to electrical. It was something that was in the back of my mind as a worry to look into and then Pos posted this a few posts up: "
As a side note: the Anaview amps are very picky with the source: you need a balanced source with low impedance.
That also mean you will not be able to use passive attenuators in fornt of the amp.
26dB of gain is a lot, especially with a unit that outputs 20+dBu like the Motu. "
And that made my fears come through.. It makes sense the MOTU has too high a level as soon as you see +20dBU but there are no worries if passive attenuation is an option. When it's not, as is apparently the case now, I am worried.
 
Since passive attenuators are not an option and I don't want to build an active circuit (since I know absoutely 0 about making such a thing and it probably involves an op-amp and sound quality reduction)

There are tons of good articles here for that sort of DIY. An opamp buffer shouldn't make too much of a difference from a measurement point of view.

A few good articles to start with:

http://sound.whsites.net/dwopa.htm
http://sound.whsites.net/articles/balanced-2.htm
 
Back
Top Bottom