It’s freeware now I noticed. Competition from REW? I also have a license from way back. I will see if I remember the basics. Do you have any advices, beginner mistakes to avoid, settings to make sure?
Start with loopback tests.
You will be operating the RTX manually. I think you want to keep the input level -10dBFS at the highest for the best performance. Jens can confirm.
In general, just be aware of what tests you are running and the frequencies. Make sure your DUT isn't clipping. All the basics. Leave the RTX inputs AC coupled unless you have good reason not to.
I try to send signals out at the industry accepted levels for the device for "0 Vu", reducing when required. I use 19 KHz and 20 KHz for IMD tests. 1 KHz and 400 Hz for THD as required. It is also used as a general signal generator for doing other tests. Sometimes modulating an RF generator.
You will be operating the RTX manually. I think you want to keep the input level -10dBFS at the highest for the best performance. Jens can confirm.
In general, just be aware of what tests you are running and the frequencies. Make sure your DUT isn't clipping. All the basics. Leave the RTX inputs AC coupled unless you have good reason not to.
I try to send signals out at the industry accepted levels for the device for "0 Vu", reducing when required. I use 19 KHz and 20 KHz for IMD tests. 1 KHz and 400 Hz for THD as required. It is also used as a general signal generator for doing other tests. Sometimes modulating an RF generator.
Cabling. What cables do you use for the 6001?
Loopback: Common XLR female to male?
Output: XLR Female to whatever is on the DUT for balanced and BNC to whatever is on the DUT for SE?
Input: XLR male from whatever DUT provides for balanced and XLR male with pin 2 to plus and pin 3 unconnected from SE DUT or do you use some kind of line transformer or SE to BAL converter?
Loopback: Common XLR female to male?
Output: XLR Female to whatever is on the DUT for balanced and BNC to whatever is on the DUT for SE?
Input: XLR male from whatever DUT provides for balanced and XLR male with pin 2 to plus and pin 3 unconnected from SE DUT or do you use some kind of line transformer or SE to BAL converter?
Anyway, I did a some first loopback tests.
Here is outputs set to 0 dBV and inputs to 10 dBV.
And here is outputs to 0 dBV and inputs to 0 dBV.
Shall I dare go to -10 dBV on inputs or does it overload?
Here is outputs set to 0 dBV and inputs to 10 dBV.
And here is outputs to 0 dBV and inputs to 0 dBV.
Shall I dare go to -10 dBV on inputs or does it overload?
Last edited:
It seems like your unit has firmware V1.10, so you should upgrade to V1.19.
Run the rtxusbaudioDfu.exe, select the RTX6001_i2o2_V0119.bin and let the program finish the update.
For single ended input signals I typically use something like this:
With R=100ohm and C=10nF.
The RC circuit can also be built into the XLR housing with e.g. a BNC connector on the input. Not quite as ideal, but more flexible. In this case I have used a 500ohm PTC resistor (makes it robust against abuse). Don't use a standard BNC to XLR adaptor where the shield is connected directly to both pin 1 and pin 2 of the XLR connector.
Regarding your question: Shall I dare go to -10 dBV on inputs or does it overload?
If your input signal is -3dBV, yes it will overload if you select the -10dBV range. It doesn't destroy the analyzer, but you will get a lot of distortion.
Run the rtxusbaudioDfu.exe, select the RTX6001_i2o2_V0119.bin and let the program finish the update.
For single ended input signals I typically use something like this:
With R=100ohm and C=10nF.
The RC circuit can also be built into the XLR housing with e.g. a BNC connector on the input. Not quite as ideal, but more flexible. In this case I have used a 500ohm PTC resistor (makes it robust against abuse). Don't use a standard BNC to XLR adaptor where the shield is connected directly to both pin 1 and pin 2 of the XLR connector.
Regarding your question: Shall I dare go to -10 dBV on inputs or does it overload?
If your input signal is -3dBV, yes it will overload if you select the -10dBV range. It doesn't destroy the analyzer, but you will get a lot of distortion.
Thanks alot Jens! I managed to upgrade to V1.19 yesterday thank you. Will start making adapters according to above!
I just use XLR to XLR for loopback testing.
Jens, thank you for the RCA to XLR wiring. I'll make a set of cables, it will probably reduce mains leakage in the readings. I bought audio isolation transformers, but even the expensive ones add distortion.
Jens, thank you for the RCA to XLR wiring. I'll make a set of cables, it will probably reduce mains leakage in the readings. I bought audio isolation transformers, but even the expensive ones add distortion.
Ok, gave that a try (yes, it's been a long while). The distortion spectrum is entirely dominated by the 1:100 probe (Micsig DP10007). I don't think that's going to work for low-distortion tests with high-voltage outputs from estat amps.
I found in the calibration lab that many HV probes were not linear. These were for DC volts (or AC I guess), and they varied the drop as you increased the applied voltage. So, their resistance did change with voltage.
You may have much better luck stringing together some HV resistors with a low tempco and voltage coefficient. Many in series will probably improve your results. Your capacitance compensation also has to be linear in the same way, several in series.
You may have much better luck stringing together some HV resistors with a low tempco and voltage coefficient. Many in series will probably improve your results. Your capacitance compensation also has to be linear in the same way, several in series.
The voltage coefficient will be the silent killer of THD and it's rarely specified, even for fancy resistors. I've had good luck with Susumu RG as long as the voltage across them doesn't go much beyond 20-or-so volt. The exact limit depends on what kind of THD you're trying to measure. I often operate below -130 dBc (1 kHz) and it doesn't take many gnat farts to muck that up.
Tom
Tom
More than that for AP. But yeah. I use the multiple-in-series trick in the Modulus-286 to get the performance.
Tom
Tom
Ok, so I tried a resistive voltage divider to attenuate the AC signal:
- DUT output without attenuator:
30 Vrms sine into RTX@100V input gave 0.02% THD - DUT output with 1:100 attenuator:
30 Vrms sine into RTX@1V input gave 0.04% THD
IIRC the sweet spot of the RTX is about 1V.
Try lower attenuation, you just have two data points now which doesn't say a lot.
Jan
Try lower attenuation, you just have two data points now which doesn't say a lot.
Jan
I did play with other attenuation ratios (1:10 and also something like 1:500 or so). The results were the same.
I agree with Tom, several resistors in series to keep the voltage drop across each low. One pair is obviously much better and there is the law of diminishing returns.
It is difficult to figure out distortion when you have uncertainties with the DUT and the divider network.
It is difficult to figure out distortion when you have uncertainties with the DUT and the divider network.
Multiple resistors in series also lowers the power dissipated in each resistor which then helps you with the TCR. Especially at low frequency the thermal cycling of the resistors can cause distortion due to the TCR of the resistors.
Bruce Hofer of AP fame has a good presentation on this topic. You can find it through AES and also on YouTube. He presented it at Oregon State U.
Tom
Bruce Hofer of AP fame has a good presentation on this topic. You can find it through AES and also on YouTube. He presented it at Oregon State U.
Tom