That is exactly what I want to avoid so thanks Jan for making it more clear. In my case the "SE output" is to speakers/headphones so these are truly floating and don't need/want a ground connection.
I was wondering if this "ungrounded SE signal" could simply be connected to the analyzers hot and cold poles of an XLR input leaving ground/shield floating😕
It's probably not ideal for headphones given that the outputs have 50 ohms in series, nevermind speakers. It's a line output.
Well it's still two connections. In your case you might be better off connecting the HP to the BAL output.
Jan
@NicMac,
From what I understood, your amplifier has a balanced input.
But what is the "floating (SE) output" configured? Is it also balanced or what do you mean by "floating"?
From what I understood, your amplifier has a balanced input.
But what is the "floating (SE) output" configured? Is it also balanced or what do you mean by "floating"?
Yes - normal balanced (+/GND/-) input. By floating I mean that the output is without reference to input GND (i.e. not +/GND or -/GND). The output is only referenced to floating (+/-) rails.
That should be no problem as long as the voltages are in the common mode range and there isn't a large HF common mode component. The supplies should reference to the input ground. With no ground reference back to the RTX there can be a lot of common mode which could swamp the input. The RTX has a nice ground Jack in the front that references to an optimal internal ground and is good for shunting ac leakage around the input circuits.
Sent from my LG-H811 using Tapatalk
Sent from my LG-H811 using Tapatalk
I would be interested in seeing how the noise will change between the two methods.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
I remember, but can't find it back, a post by JensH where he stated that the SE output is formed from one of the balanced output phases and Gnd, and that therefor the SE output was half of the BAL output.
It triggered me at the time but then I forgot about it until something triggerred me again.
My point is that this method degrades the SE output S/N ratio. A better way is to form the SE output from taking the difference between the two BAL output phases as the SE output, and not the difference between only one phase and Gnd. This also has the advantage that the BAL and SE outputs are equal. See attached for a possible way.
Of course, at the end he always need to get Gnd involved in a SE output but this is a bit cleaner; a fully diff amplifier would be ideal.
And the Gnd connection of the SE output connector should be directly connected to the Gnd of that input resistor (which in turn should be directly connected to whatever the signal Gnd is).
Jan
It triggered me at the time but then I forgot about it until something triggerred me again.
My point is that this method degrades the SE output S/N ratio. A better way is to form the SE output from taking the difference between the two BAL output phases as the SE output, and not the difference between only one phase and Gnd. This also has the advantage that the BAL and SE outputs are equal. See attached for a possible way.
Of course, at the end he always need to get Gnd involved in a SE output but this is a bit cleaner; a fully diff amplifier would be ideal.
And the Gnd connection of the SE output connector should be directly connected to the Gnd of that input resistor (which in turn should be directly connected to whatever the signal Gnd is).
Jan
Attachments
Last edited:
The circuit you have shown is very similar to the output that I use, except that I use an LME49990 with an LME49600 buffer instead of the NE5534. And the output resistor is 50 ohm in my case.
The balanced input to the circuit is taken from the balanced output of the post-DAC filter.
This is how the SE output is made.
And the balanced output is just two of these circuits with the inputs swapped on one of them, to get the opposite polarity.
In terms of S/N ratio I don't think that you loose anything. The noise is dominated by the DAC/post-DAC filter, not the output amplifier. On the balanced output you get twice the output signal, but also twice the noise.
The balanced input to the circuit is taken from the balanced output of the post-DAC filter.
This is how the SE output is made.
And the balanced output is just two of these circuits with the inputs swapped on one of them, to get the opposite polarity.
In terms of S/N ratio I don't think that you loose anything. The noise is dominated by the DAC/post-DAC filter, not the output amplifier. On the balanced output you get twice the output signal, but also twice the noise.
OK, thanks for the clarification. From the remembered post I understood it differently but that may have been me.
Jan
I thought the subject was a balanced out but taking in single ended from the amp...
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
Hi
HPW & JENSH, I was wondering couple of weeks ago if there was a perspective to get the hpw software bundled (at a super duper price) with the rtx.
that would be verygreat, and I m sure we would pay for regular upgrade and with about 70 people at a start their migth be a good business case for all of us
cheers
It would be wrong to offer a deal that works only with a specific operating system, while the Jensalyser does not at all depend on a specific OS. In addition I would feel cheated if the deal was Windows only, because I don't have and don't want Windows.
Gesendet von meinem D5803 mit Tapatalk
okay, I m also having a iMac on my bench which I much prefer, but I feel REW not enough powerful going forward with RTX, so I m afraid I ll have to put a PC on the bench too. also audio tester which is ok, is on wintel...
I'm flexible using either OS if it works with Windows bootcamped on a Mac.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
I suggest we (the "RTX Community") should use MATAA and develop some more functions which are more tailored to electrical measurements since the current SW is more focused on acoustic environments at the moment.
The Input/Output-Range-Control could also be implemented in MATAA, if the API is available.
@JensH: Any update on the programming API? Thanks!
The Input/Output-Range-Control could also be implemented in MATAA, if the API is available.
@JensH: Any update on the programming API? Thanks!
Personally I'm not interested in any software bundles with the RTX. I have software that I'm familiar with already and I don't plan to switch.
No objection if there is some sort of optional bundle offers, but mandatory bundles that raise the overall price is a big no-no for me.
No objection if there is some sort of optional bundle offers, but mandatory bundles that raise the overall price is a big no-no for me.
Hi cwtim01,
-Chris
I agree with you completely on that. I'm up in the air whether to run this on Windows or Linux. I feel more comfortable on Linux, but it depends on all the other applications the bench computer has to run. I'm afraid that I might be stuck with windows because I only want 2 PCs on the bench (one dedicated to the GPS timing application).
-Chris
I don't want bundled software either although it would be great if one or more of the existing software makers offered a version tailored to the RTX..