Behringer DEQ2496

Apparently TacT will be releasing thier own phono stage, maybe it may help....

 

The Tact pre-amp benefits GREATLY from a decent off board psu, as its just a glorified PC, with D<>A & A<>D cards inside, all the research money went on the software, and just the basics were appiled.
Though I did get a decent look inside the newer power/intergrated amps, and some proper thought and planning/layout has gone into the construction of these, a neat effort

 

Tone,
I'll let you know

 

Hi,

I tried to post this as normal post, but it refused.
I tried to attached an HTML Version of the essay, renamed to .doc, it says "too big". I give up.

If you want a copy of the essay please join my technical discussion group on Yahoo:

http://groups.yahoo.com/group/Thunderstone_technical/

Then, once you are member you can access the document here:

http://groups.yahoo.com/group/Thunderstone_technical/files/TL Circuits/Ultracurve/PADEQUSE.HTML

Ciao T

 

Thorsten,

i carefully read your review.

few questions:
1. i understood that 8024 in analog mode (ADC+DAC) did alter sound quality - your review provides one more evidence for that.

isn't deq2496 is better in this regard?
i don't know - other dacs, higher sampling rate, etc................

2. is there expirience of moding deq2496?

3. why don't you use it anymore?

4. isn't this thing PC controlled? if NO - what are all those computer screen screenshots at behringer's site?

5. i didn't undestand whether deq2496 also provides "Auto-Q" function?

6. as i intend to buy it online from US with no local waranty - it's important to know are there known reliability issues with behringers?

thanks in advance,
Gennady

 

Hi,

I find that a pair of connectors and a piece of wire alter sound quality....

As said, the stock DEQ2496 is marginally better, subjectively speaking than the 8024.

Nothing specific, however I can make a few "generic" suggestions, as I did below in the thread....

1) Change PSU Regulators for the Ap-Amp stage to 12V Types (I must check the detailed schematic, it may be more apropriate to change the negative reg to 10V and leave the positive at 15V if the analog circuitry matches that of the DCX2496 Digital EQ).

2) Change the Op-Amp's (SMD) to AD8066 (if on a budget) or AD8620 (if not on a budget).

3) Change all local digital circuit decoupling electrolytic capacitors to Sanyo Os-Con, all Audio Coupling Capacitors either to Os-Con as well or to Elna Silmic or to Black Gate NX-HiQ.

I think the above would go a long way, possibly change the clock system, but that is not required (or beneficial IMHO) if the DEQ is used in analogue I/O mode....

My system was changed to include speakers that have a more suitable room interaction, to the key job the EQ was for is no longer needed. I will eventually put the EQ back, but right now I have other priorities and needed to shrink the system somewhat.

It can be controlled via Midi from a PC, but I am not sure that there is any special software.

What screen shots? I cannot find any on the DEQ2496 page.

With the most recent software updates the functionality of the DEQ2496 should match the 8024, including Auto Q or whatever they call it now. I still have to upgrade the software on my (very early) unit.

I am not aware of any, but I suspect like any other mass manufacturerd article out of china each batch will contain a few duds. So I'd make sure of warranty, at leats of the kind that protects against DOA and early failure.

Ciao T

 

sorry, screenshots that i saw are from 8024:



what are they?
from MIDI interface?

regarding sound degradation - i guess it's about balance of sound altering against improvement of RC.

what people say - how is it compared in this regard (sound degradation) with tact and deqx?

 

Hi,

PC Software which communicates with the UC via Midi.

Yes.

I never had the TACT system at home and it may be marginally better, but if the levels are right (that is the key caveat) the 8024 and 2496 are pretty transparent and I doubt that any drastic improvement will come from switching to TACT or others, as that would require an improvement over a direct wire bypass....

However, I MUST repeat that you are required to get the levels right and that you must use good quality cabling. Fail that and the sound gets progressively worse.

Ciao T

 

what do you mean by "levels"?

i once did a test at home at the time that i had big standing waives problems:
- i took some old yamaha 10 band graphic equaliser and connected it between pre and power
- i measured each speaker's response suing redioshack SPL (yea, i know, it's not accurate at all, especially below 35 Hz) and testones disc with 5 Hz step
- made necessary correction to each speaker

the result was quite stuning - although it introduced "dirt" in the sound, the RC effect was very significant.

 

Hi,

Excactly that. Levels, as in Signal levels.

The DEQ2496 offers a selection of 3.1V or 9.8V for digital full scale on the input/input. The DEQ2496 offer a 113db dynamic range compared to these voltages.

Now if you feed the DEQ a level of 0.31V and have set it to use +22dbu full scale level your dynamic range is down to 83db, well below CD standard. The result will be audible distortion and general poor sound.

Now change your gainstrutcure so that your Amplifiers overload with exactly 3.1V RMS on their inputs, select the +12db full scale level and crack up your preamp and you have a dynamic range of 113db velow clipping, well past CD standard, so in fact you can attenuate the system overall by around 20db without loosing resolution from CD Sources.

As you can see, "getting the levels right" is absolutely crucial and the difference between great results and poor ones.

Ciao T

 

Right, my bits from farnell arrived...

So far, for the money this is a killer product. The performance as a straight DAC is pretty good, although recommendation as a budget DAC is difficult because of the AES/XLR connectivity, which does not fit in with a budget player or DVD player used as transport, or with cheaper amplifiers.

Compared to my modded superdac...

Not sure yet. The superdac has more "air" and separation, but the DEQ has the advantage of equalisation (which I quickly did a cursory experiment with by using the RTA to identify what frequencyish was causing the floor to vibrate, and dialling it down a little, and it worked). A proper room EQ with a mic will no doubt reap greater rewards.

In terms of character, the sound of the two is pretty similar, which you would expect given that they use the same DAC chip. I can't currently do an A/B comparison on a "fair" basis, using the same cables because one of the balanced outputs on my superdac isn't working properly (the -ve output is 0).

Cables currently are a pair of home-made XLR leads made from StarQuad mic cable with heavy-duty Neutrik XLRs, and a Deltron AES/EBU lead made from Deltron XLRs with Gotham GAC-2 AES cable (a whopping £13). I've also plugged the DAC's eupen into the DEQ, although I didn't perform critical before/after listening for that change so I won't vouch for anything on that front. It should at least attenuate some of the noise from the microprocessors and switch mode heading back into the mains.

 

have to ssay that i lost you.

till today never got to understand all this dbfs/dbu stuff

 

Hi,

Okay. In principle db is just a measure which is logarythmic and indicates a relative value compared to a given reference.

Now:

1db = 1.12
2db = 1.25
3db = 1.41
6db = 2
10db = 3.16
12db = 4
18db = 8
20db = 10

For digital audio the standard reference is "digital full scale" usually abbreviated to fs (full scale). This means once you get there you have turned every bit on and there is nowhere to go after this. So increasing the input level will not turn on any more bits because they are already all on. In analog terms we call that clipping.

So, 0dbfs is the point where the ADC/Bitstream clips. In digital all measures are commonly referenced to that 0dbfs even though commonly peaks should remain a db or two below that imaginary line.

So a dynamic range of 113db means the loest signal to still change the output is -113dbfs, which BTW is equivalent to 19 Bits. So even though the ADC and DAC are claimed to be "24 Bit" they and the analogue circuitry produce so much noise that only 19 Bits of this resolution is genuinely available, but that is still 3Bit or 18db more than with CD.

Now for dbu, dbu is a pro audio measure which is defined as 0.775V and comes from the old dbm where the reference was 1mW (the m) into 600 Ohm. Thus 0dbu = 0dbm if the load is 600 ohm and in voltage terms it is 0.775V.

Knowing that the Behringer DEQ2496 inputs are rated as requiring 22dbu Input level for 0dbfs or clipping of the bitstream we can look at our table above and find that 22db is 1.25*10.

Now we know that 0dbu is 0.775V so +22dbu is 0.775 * 1.25 * 10 or 9.68V. So we need to apply 9.68V to the input to use the full dynamic range of the ADC and that which is available from the DEQ2496. Equally, if the DAC in the DEQ2496 outputs 0dbfs (or clipping) we know the Output Voltage is 9.68V.

Lets drop the EQ between Amplifier and Preamplifier in a normal HiFi System like yours.

Your Amplifier is a Citation II, which if stock has a gain of around 27db for an output power of 60 Watt on the 8 Ohm Tap. I'll spare you the math to work out the input sensitivity and tell you that it comes to around 1V for full power. BTW, on the SF Cremona's you have this would in an "average room" translate into around around 100db for the pair of speakers at a listening position 3m from each speaker.

Now if our Amplifier will overload with 1V input but the Digital equaliser will output 10V when it overloads we have a clear level mismatch, which for the volume totally tuned up to the point where the speakers overload amounts to 20db. These 20db result in only 93db dynamic range for the system.

Okay, lets do something else.

We switch the DEQ2496 to +12dbu full scale. Now we know that the full scale output will be around 3V. This is still 10db higher in level than what we need to overdrivedrive the Amplifier, so if we then add around 6db attenuation (say a 3k3 series resistor and a 3k3 shunt resistor) on the amplifier input will will have matched the levels pretty well as now the level on the DEQ's output will be 2V to overload the Amplifier. In other words, instead of throwing away 20db dynamic range in the EQ we now only loose 4db, which is a sensible safety margin.

The 16db we have gained BTW are pretty much 3 Bit worth of resulution or in other woirds the difference between DVD-A and CD!!!!

Any clearer?

Ciao T

 

Ok, a quick comparison has put the superDAC on single-ended outputs ahead. The DEQ sounds congested and thick by comparison. I will investigate the superDAC's balanced out problem at some point, although it isn't particular fair for the superDAC to use those either, as I didn't replace the balanced output op-amps, only the single-ended.

It could, of course, be partially to do with the home-made XLR leads... more investigation will tell...

 

In fact, a quick check suggests that the problem on the superDAC output has gone away... not sure why that should be...

Anyhow, it sounds pretty good through the XLR outs, although it's even "hotter" than the DEQ on the XLR outs. I might be forced to lower the gain in the output stage when I get around to replacing the balanced output op-amps.

The thickening was slightly in evidence from the superDAC connected this way, so it could partially be down to the cables and the differential input stage of the amplifier (the unbalanced inputs go through a resistor and a relay before the volume pot. The balanced input goes through an AD711 based differential amplifier circuit before the relay and volume pot. However, the superDAC is definitely ahead of the DEQ at this time.

 

This is comparing the un-modded DEQ2496 to a modded SuperDAC right? So it's not really fair anyway is it.

 

Indeed. But it's an interesting comparison nonetheless. I would have been surprised had the Behringer been better... I think it's possible to get it closer, but it needs some work. I suspect that some work on the clocks might be necessary (as my superDAC has the pll filter mod, which combined with the DOB in the CD transport gets things pretty accurate..)

 

clearer indeed, although i need to re-read it few more times

really appreciate your effort.

 

i wouldn't beleive that - i've got contacted by local dealer sujjesting coming to my home for half a day with new deq2496, mike and all the adaptors.

WOW
of course i jumped on the oportunity.

T, i will start with 12dbu setting as you suggested and will go from there.
unfortunately the dealer don't know how to operate it so i will have to study the manual - which i will.

can you guys give any any usefull suggestions/tips from you expirience?

 

well,
i've heard deq2496 in my system.

the biggest WOW is that i couldn't hear any negative effect it made on the sound.

although i don't have big bass problems, RC worked quite nice.

it was quite easy to use.
i don't think that TACT is easier to use.