Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Wayne Smith2]

http://www.rane.com/note109.html

http://www.mackie.com/pdf/onyx1620_om.pdf  (pages 26 – 28.

I need some advice on how to make a summed output from the main XLRs on a Mackie Onyx1620.
The sum of he stereo fader would be my main PA feed, but I still want to allow a stereo recording mix on the two parallel TRS outputs.
 
I found Rane's method, but there might be a few points here beyond my experience.
It mentions requiring a 600 ohm line-drive capability; Mackie lists the balanced mains as 150 ohms. (Is my guess correct that lower ohm source fits that requirement?
Second is the chance this arrangement could adversely effect the panning or separation at the TRS outputs? I found post 12 here..
http://forums.prosoundweb.com/index.php/topic,126260.msg1185803.html#msg1185803

Thank you in advance
Wayne Smith

[Brian Houchin]

If you're going to use this setup a lot, buy a Radial J48 and a couple of xlr-trs cables.

If it's a one-off, rent a Radial J48 and a couple of xlr-trs cables.

[Wayne Smith2]

I don't follow, how would the direct box apply?

[TJ (Tom) Cornish]

I don't follow, how would the direct box apply?

Some of the Radial boxes have summing resistors built in, including the J48.  For this particular application I disagree with that recommendation - you want something that doesn't require phantom power.  I would recommend the Radial Pro AV series:
http://www.radialeng.com/pdfs/smartsheet-radial-proav1.pdf

[Geoff Doane]

 
I found Rane's method, but there might be a few points here beyond my experience.
It mentions requiring a 600 ohm line-drive capability; Mackie lists the balanced mains as 150 ohms. (Is my guess correct that lower ohm source fits that requirement?
Second is the chance this arrangement could adversely effect the panning or separation at the TRS outputs? I found post 12 here..
http://forums.prosoundweb.com/index.php/topic,126260.msg1185803.html#msg1185803

You are to be commended for your research on this before you asked the question. 

The fact that the XLR and TRS outputs have different output impedances would seem to indicate they already have some isolation.  The flow diagram near the end of the manual seems to bear this out.  The XLR output (which you intend to sum) seems to be buffered separately from the TRS outputs. 

I would build the summing circuit with 200Ω resistors, plug it in, and see if you can see any detrimental effects on the TRS outputs.  I doubt you will, and this will be a pretty cheap test (certainly less than a J48 DI, that wouldn't have worked anyway  ).

GTD

[Tim Perry]

here is an alternate suggestion:  feed left and right into a dual band equalizer (or anyother dual channel equipment like a compressor.  Preferable these have both xlr and 1/4: inputs.

Hardwire the outputs together. Most modern gear has resistors in the output which will prevent damage.

Pick the the stereo feed off the unused input. 

[Geoff Doane]

Hardwire the outputs together. Most modern gear has resistors in the output which will prevent damage.

The [output] resistors may prevent damage, but they won't prevent distortion, particularly if you get close to the maximum output.

We've covered this before.

The external resistors are simple, cheap, and more or less foolproof.  The only tradeoff is a slight loss of signal and a slightly raised output impedance.  Neither of those factors is a problem if you're feeding another device like an equalizer or DSP nearby in the signal chain.

GTD

[Wayne Smith2]

...I would build the summing circuit with 200Ω resistors, plug it in, and see if you can see any detrimental effects on the TRS outputs.  I doubt you will, and this will be a pretty cheap test ...(

Excellent, thank you very much.
I've sort of stepped through the logic here, between my 150 ohm source, your 200 ohm start point and what I've gathered from Rane and Ivan Beaver's post so far. I was curious why you chose 200 ohms, but It looks like you've given me a good start point that might need some trial and error experimentation.
Presumably then this combination would degrade (from Rane's examples) the separation at the XLRs, but might be fine between the upstream TRS pair? ..Also I don't need perfect 100% panning capability.

Rane mentions 'scaling all the resistors' if changing to balance separation with final impedance. From their 475' to your 200 ohms, should I apply similar ratios to their 20K ohm resistor as well?

[Geoff Doane]

Rane mentions 'scaling all the resistors' if changing to balance separation with final impedance. From their 475' to your 200 ohms, should I apply similar ratios to their 20K ohm resistor as well?

As long as you don't want to use the XLR outputs for anything else, you don't care if there is crosstalk.  I grabbed 200Ω out of thin air, figuring that the resultant, worst case load of 400Ω, shouldn't be a problem for a 150Ω output.  You could easily go higher without any problem, as long as you don't have a long run (100s of feet) to the EQ or DSP.  I'm not not sure why Rane bothered with the 20KΩ resistor.  It's just in parallel with the input of the next stage, which will typically be about 20KΩ.  Leaving it out won't cause a problem.

When you check the summing and crosstalk, send a tone in two channels of the mixer with a wye cord.  Pan one left and the other right, and mute them one at a time.  Using a single channel and just panning L-C-R won't work because the centre is typically 4.5 dB down from hard left or right.  Adjust each channel for nominal output on the console's meters.  The level at the sum output should be exactly 6 dB down when either channel is muted, compared to both channels on.  The level at the stereo recorder inputs should disappear completely when the corresponding channel is muted.

GTD

[Nick Hickman]

Hi Geoff & Wayne,

I grabbed 200Ω out of thin air, figuring that the resultant, worst case load of 400Ω, shouldn't be a problem for a 150Ω output.  You could easily go higher without any problem, as long as you don't have a long run (100s of feet) to the EQ or DSP.  I'm not not sure why Rane bothered with the 20KΩ resistor.  It's just in parallel with the input of the next stage, which will typically be about 20KΩ.  Leaving it out won't cause a problem.

I'm not sure either.  I don't think a shunt resistor should be necessary when combining two outputs where the series resistors are selected so that the sum of the output Z itself and the series R is equal to the minimum impedance the output is happy driving.  A shunt may be desirable if you combine more than two outputs as it allows smaller series R to be used.  As this is for line outputs and there'll be no great line capacitance hanging on the output, I'd personally scale up the resistors so the sources are driving around 2Kohm.

(Now, TMI!  With two identical outputs combined, the worst case for the current demand on the outputs is where one is at +Vpk and the other is at -Vpk.  If the series resistor plus build-out is R, a current of Vpk/R must be okay.  In the worst case, with a high-Z load, there's 2*Vpk between the outputs across 2*R, so the current is still Vpk/R.  If more outputs are combined, the worst case gets worse.  If three outputs are combined and one is at +Vpk and the others at -Vpk, for example, each sees a load of 1.5*R and one of them sees a current of 1.33*Vpk/R, i.e. more than tolerable.  To fix this, you need to increase the series resistors, but you can increase them less if you also add a shunt.)

When you check the summing and crosstalk, send a tone in two channels of the mixer with a wye cord.  Pan one left and the other right, and mute them one at a time.

Try driving each output to the point of clip and inverting the polarity on one side.  Depending on relative levels, the combined signal should get very quiet but stay clean.

Nick

[ProSoundWeb Community]