I think that in most cases a suitable cable will be sufficient.
With an electronic converter you would lose the advantages of the built-in attenuator/gain stage and protection.
With an electronic converter you would lose the advantages of the built-in attenuator/gain stage and protection.
Hi Jens,
-Chris
In your tests, did you find that open top shields still performed well, or would the top have to be a mesh?
-Chris
Thanks - I had ordered some cables and let me see the results. However if the signal levels are within 10V RMS max will electronic converter helpfull?
Unless you really need to run very long cables, I don't think so. Most likely you will just add noise and distortion.
One of the exceptions could be if you want to measure very low signal levels, like noise from a very low noise power supply, with e.g. 1 uV noise in the audio band. In that case a low noise preamplifier could be useful, since it is possible to design a preamplifier with a lower input noise than the one in the RTX6001, if you are willing to sacrifice the protection and the high input voltage handling capability.
One of the exceptions could be if you want to measure very low signal levels, like noise from a very low noise power supply, with e.g. 1 uV noise in the audio band. In that case a low noise preamplifier could be useful, since it is possible to design a preamplifier with a lower input noise than the one in the RTX6001, if you are willing to sacrifice the protection and the high input voltage handling capability.
Hi Jensh
no such need right now for such a low noise measurement. However the noise values from softwares that i tried for signal source ( HP 239A) are all over the place than what the spectrum looks ( which is very clean indeed with cable) Only Wavespectra and SpectralPlus -SC software seems reasonably close. Ihave a nice working AP Sys 2322 to compare. Most of the software seems to agree on THD ! Probably we will learn more from such a decent set of users about real life use of RTX
no such need right now for such a low noise measurement. However the noise values from softwares that i tried for signal source ( HP 239A) are all over the place than what the spectrum looks ( which is very clean indeed with cable) Only Wavespectra and SpectralPlus -SC software seems reasonably close. Ihave a nice working AP Sys 2322 to compare. Most of the software seems to agree on THD ! Probably we will learn more from such a decent set of users about real life use of RTX
Yes with 'ins' and 'outs' were left open and I noted the improvement of a simple vertical screen (using the clips) between the right out and left in. Improvement was such that a very noticeable cross coupled component was visually eliminated. Screen was cut from an aluminium larger can (cider and beer cans can also be used!). Its a simple vertical divide so would not expect any cooling implications. As Jens noted the cross talk would be reduced in reality by the source impedance presented to the 'in' being low. Confirmed by shorting the input. Haven't determined what sort of source impedance would start to cause material effects but given a very simple screen provides an almost something for nothing improvement then seems worth fitting.
@JensH
Wouldn't it be possible to run the in- and output of the RTX6001 at different sampling frequencies (at least with frequencies of the same family)? That way one could use the DAC up to its 768 kHz.
One thought why is that, at least in theory, it should be better to provide the DAC the real samples, instead of letting them be interpolated by the digital filter.
Wouldn't it be possible to run the in- and output of the RTX6001 at different sampling frequencies (at least with frequencies of the same family)? That way one could use the DAC up to its 768 kHz.
One thought why is that, at least in theory, it should be better to provide the DAC the real samples, instead of letting them be interpolated by the digital filter.
Output impedance?
The RTX6001 manual specifies the output impedance of the analog generator as 100 Ohm (XLR). Is this true for all output level settings, or just the 10 V setting?
I am asking because I observed that if I connect the analog out to an amp with a high input impedance (500k in this case), the noise level observed at the amp output is higher if the RTX output is set to 100mV or 1V than at the 10V setting. I suspect the input picks up more noise at the 100mV or 1V settings because the RTX output impedance is higher at these settings.
The RTX6001 manual specifies the output impedance of the analog generator as 100 Ohm (XLR). Is this true for all output level settings, or just the 10 V setting?
I am asking because I observed that if I connect the analog out to an amp with a high input impedance (500k in this case), the noise level observed at the amp output is higher if the RTX output is set to 100mV or 1V than at the 10V setting. I suspect the input picks up more noise at the 100mV or 1V settings because the RTX output impedance is higher at these settings.
The output impedance is the same for all output level settings.
The wideband (white) noise level is reduced when you go to 1 V or 100 mV compared to the 10 V setting. But there is some mains related hum, which does not get attenuated with the same amount as the white noise.
The wideband (white) noise level is reduced when you go to 1 V or 100 mV compared to the 10 V setting. But there is some mains related hum, which does not get attenuated with the same amount as the white noise.
Hmm. Why do I see more noise with the 100 mV or 1 V setting than with the 10 V setting?
How do you measure it? You mentioned that you connected it to an amp. Do you change the gain of the amp to compensate the reduced level from the generator?
I use the spectrum analyser function of REW to continuously show the noise spectrum at the speaker out. The generator (analog out) is muted. If I change the output level of the (muted) generator from the 10 V setting to the 1 V or 100 mV setting, the noise level goes up in the spectrum analyser. No other changes than the output level switch involved.
Well, actually, no, this is not correct. I just checked again. The above (silent output) shows the behaviour your described (slightly lower noise level with with the lower voltage settings). The "inverse" change in the noise level with the different output levels I described (higher noise level with lower voltage settings) happens only if a test signal is playing at the output (I used a 1 kHz sine signal). Could this be a software / data processing thing?
Hi Jens,
Would you make the XMOS USB to I2S board available for purchase?
There are hardly any good multichannel USB interfaces available.
thanks
Would you make the XMOS USB to I2S board available for purchase?
There are hardly any good multichannel USB interfaces available.
thanks
Last edited:
Tutorial
So now that we're mostly done with showing off everyones loopback tests, I think it would be useful to have some kind of a document with examples of how (not) to use the RTX analyser in real world tests. A tutorial showing how to connect things in typical measurements, and how to avoid noise/hum or similar issues.
Examples: How to set up the wiring for a typical amplifier test? How/at which point do you convert the RTX's balanced input and output if the amp does not have a balanced input or output? How to test speakers using a microphone (maybe directly connected to the RTX input without an extra microphone amplifier)? What are common errors or no-gos when setting up a test? What works, what doesn't? Etc.
I know that these things may be boring and obvious for some, but since most of us are just hobbyists (sometimes missing proper training), such a tutorial would be very useful for many others.
A good format for such a tutorial might be a WIKI page, which is much easier to update and maintain by the community than a static file that can only be edited or expanded by a single person.
What do you guys think?
So now that we're mostly done with showing off everyones loopback tests, I think it would be useful to have some kind of a document with examples of how (not) to use the RTX analyser in real world tests. A tutorial showing how to connect things in typical measurements, and how to avoid noise/hum or similar issues.
Examples: How to set up the wiring for a typical amplifier test? How/at which point do you convert the RTX's balanced input and output if the amp does not have a balanced input or output? How to test speakers using a microphone (maybe directly connected to the RTX input without an extra microphone amplifier)? What are common errors or no-gos when setting up a test? What works, what doesn't? Etc.
I know that these things may be boring and obvious for some, but since most of us are just hobbyists (sometimes missing proper training), such a tutorial would be very useful for many others.
A good format for such a tutorial might be a WIKI page, which is much easier to update and maintain by the community than a static file that can only be edited or expanded by a single person.
What do you guys think?