[toobluvr]

Don't mean to sidetrack this thread, but I just click the insert image button then browse to the file on my hard drive and upload.
No gallery involved or anything.  Was the gallery required to accomplish this previously, and is no longer needed on the new site/software?

Steve

Yes...previously, only images that existed in your Gallery (or elsewhere online) could be inserted into a post.

With the new version of AC, they can now also be pulled in from your hard drive.

[dmatt]

Here's an update to all those who have helped with my problem.

I swapped in my Logitec Transporter for the Channel Island VDA2 DAC/Duet Receiver Combo.  Using balanced inputs to the Melody I2A3, there was no distortion on bass notes.  So I put the CI DAC back in the system and I tried both another Duet Receiver and the Transporter as digital inputs to the CI DAC into the amp using balanced -- distortion once again.

Summary (at the risk of TMI):
Using Balanced inputs:
Transporter to I2A3 amp (OK)
Transporter to CI DAC to I2A3 amp (bass distortion)
Duet Receiver to CI DAC to I2A3 amp (bass distortion)
Duet Receiver to CI DAC to I880 amp (OK)

Using single ended inputs:
Duet Receiver to CI DAC to I2A3 amp (OK)
Duet Receiver to CI DAC to I880 amp (OK)

The mfg. specs say the CI DAC outputs a balanced signal at a higher value than the Transporter (4.5vRMS vs. 3.0vRMS).  So I set-up the digital out from the Duet Receiver to the CI DAC balanced out to the I2A3 and lowered the digital output on the Duet receiver (from 100 to 94) until I could no longer hear the distortion (I turned up the gain on the amp to keep the same level).  With only -6(dB?) of attenuation I don't think I am throwing bits away.  I still don't know why the Melody I880 handled the balanced signal from the CI DAC without distorting, but at least I can fully enjoy the I2A3 balanced from the CI DAC now. 

Ah, the mysteries of electrons and iron.

Thanks everyone.

David

[JoshK]

You problem is simple.  You are overdriving the input stage of your I2a3 amp. Since the CI dac has higher voltage output, it is overdriving your amp's input stage. 

In guitar amps they do this on purpose; overdrive a stage to get that clipping/distortion.  4.5vRMS is a LOT!  I bet your I2A3 amp is looking for 0.7vrms.  Do you have the input sensitivity of your amp?

Essentially is works like this.
Take the input stage of your amp.  It has its operating point set based on some assumptions your manufacturer made.  They probably never expected 4.5vrms.   

So they set the input stage for some amount of gain, let's guess it is 50x (12ax7 for instance).  The B+ is somewhere, let's say 300V, with resistive plate load that leaves about ~150V across the 12ax7.  The 12ax7 then can't swing more than 150v peak to peak, and realistically a lot less.  Therefore with 3V peak to peak and 50x gain (150V peak to peak output from this stage) you are clipping the stage as it has run out of voltage. 

3V peak to peak is 1.05vrms.  You are providing more than 4 times this amount.  So unless your input stage has a lot less gain than 50x or you have a lot more B+ available you are going to overdrive the input stage. 

The thing about component matching and synergy is, IMHO, mostly about impedance matching and appropriate gain structure.  Its not as black artsy as some make it sound.

What tubes does your amp use?  How many gain stages does it have?  You might get away with loosing a gain stage.

[JoshK]

So I saw the schematic on the first page.   Haven't worked out the operating point but just to give a rough guess, you have 265v B+ for the input stage and a 6SN7 (mu of 20) which is probably in the 15-17x gain range.  Let's say 15x as it is more conservative. 

You have to know the rp of the tube at operating point to work out how much voltage is across the tube but I am going to take a guess its probably around 150v.  So that means if you provide more than 10V peak to peak you are going to overdrive this stage into gross distortion.  In reality, because the tube can't conduct all the way down to 0V, actually very far from it, you probably can only give it 6v p-p.  That means something like 2.1vrms (6vpp * .35 = 2.1vrms).  Drive it with more than 2.1vrms and you are in distortion land.

Now these numbers were just wild ass guesses, but its in that ballpark.  The transformer for the XLR sums the phases to give you the full Vrms that the balanced signal delivers, in this case 4.5Vrms.

[dmatt]

Thanks for the analysis Josh, you confirmed what I guessed was happening.

Can I trouble you to explain (now type slowly, I'm just learning here) what is the relationship between the digital input of the DAC and the output?  I'm assuming that some absolute reference volume (0dB?) corresponds to full balanced output.

In other words, since I can digitally reduce the max output of the Duet Receiver, is it just a simple matter of "turning down" the digital volume, reduce the max balanced output from the DAC, and not overdrive the amp?

One more quick question.  I always thought that balanced outputs had a standard voltage -- isn't the 3vRMS out of the Transporter fairly common for balanced?  If that's the case, most balanced inputs would be more than 2.1vRMS and overdrive the amp.  Or does the balanced output voltage spec vary just as much as all the rest of the specs in a design?

Thanks,

David

[JoshK]

When you turn down the volume in the digital domain it is turning down the voltage output of the DAC.  In your case, your best bet is to use the SE output of the CI DAC.  I don't think that in your case you are going to get any benefits from using the balanced out and reducing the volume in the digital domain.