RME ADI-2/4 Pro SE - RIAA mode measurements

[sarumbear]

Maybe you could start by explaining why the RMI comments are incorrect i.e. why 6 dB is not enough headroom. Hopefully I could follow along. Thanks

I have tried. Maybe you can search for the meaning of headroom?

I have seen cartridge output waveform using lab grade amplifiers. Many people have seen too, back in the day when vinyl was the only source. Those findings refute RMI. You may search and find those old tests.

 

[badboygolf16v]

According to the RME manual the analogue headroom is 41.3dB so unless I'm missing something, this discussion about lack of headroom is moot.

 

[sarumbear]

According to the RME manual the analogue headroom is 41.3dB so unless I'm missing something, this discussion about lack of headroom is moot.

Why do you say 6dB then?

 

[badboygolf16v]

You've confused me with someone else.

 

[levimax]

I have tried. Maybe you can search for the meaning of headroom?

I have seen cartridge output waveform using lab grade amplifiers. Many people have seen too, back in the day when vinyl was the only source. Those findings refute RMI. You may search and find those old tests.

I have seen all of those "spikes" on oscilloscopes as well but I have never seen a rigorous explanation of what makes "click and pops sound bad or less bad" all the way through the chain. Lots of anecdotes about how much headroom and how much "soft clipping" and how much "graceful recovery" and the like but no "evidence" I have ever seen. With the RME product things get a little harder to follow when talking about digital headroom vs analog headroom but I think at the end of the day what they are saying is 6 dB of headroom above the music playing capability of the cart is enough. As there is some confusion see below for the RME explanation again.

 

[sarumbear]

I have seen all of those "spikes" on oscilloscopes as well but I have never seen a rigorous explanation of what makes "click and pops sound bad or less bad" all the way through the chain. Lots of anecdotes about how much headroom and how much "soft clipping" and how much "graceful recovery" and the like but no "evidence" I have ever seen. With the RME product things get a little harder to follow when talking about digital headroom vs analog headroom but I think at the end of the day what they are saying is 6 dB of headroom above the music playing capability of the cart is enough. As there is some confusion see below for the RME explanation again.

1) oscilloscope is not a tool to see them correctly
2) RMI is wrong

However, @badboygolf16v gives a much larger value for headroom. I will leave you two to agree on the actual headroom level.

 

[sarumbear]

When I'm curious, I measure.

Here I created a test:
I've generated a multitone with a 5mV 997Hz sine wave + a 330mV 10kHz sine wave, simulating a "click" 37dB above the signal level.
I've recorded the FFT at 192kHz, from RME ADI-2/4 Pro SE in RIAA mode (+14dB)

Here is the result.

The 10kHz at 330mV "click" does NOT saturate the input.

View attachment 249812

Your simulation is wrong. Do please find a relatively much LP and test the cartridge output.

 

[sarumbear]

You've confused me with someone else.

I sure did. I’m sorry.

 

[LTig]

1) oscilloscope is not a tool to see them correctly

Not? What about a DSO?

Or an RME ADI2 PRO fed directly by an MC [1] pickup? I did this not long ago and was astonished that despite a serious mismatch between output level of the pickup (0.65 mV) and the RME's input level (+4 dBu), after doing RIAA EQ and amplification both digitally in Audacity noise was almost inaudible after the music started.

[1] The RME's input impedance of 9 kOhm is too low for MM.

 

[sarumbear]

Not? What about a DSO?

An oscilloscope is a tool to analyse the waveform. A Digital version may have integrated extra tools in it but that’s not the job of an oscilloscope. I was simply commenting against the definition of the tool, not on a specific model or type.

 

[levimax]

[1] The RME's input impedance of 9 kOhm is too low for MM.

I used this simple "Balanced Transmitter", the PCB version, https://sound-au.com/project87.htm and changed R102 to 47K. Loading for the cart is then right, you can have the transmitter close to the TT so easy to keep capacitance down, and the signal is transmitted to the ADC balanced. I found this to work very well ... ..especially no noise.

 

[LTig]

An oscilloscope is a tool to analyse the waveform. A Digital version may have integrated extra tools in it but that’s not the job of an oscilloscope. I was simply commenting against the definition of the tool, not on a specific model or type.

Isn't this exactly what we need to do here? To find out how much higher spikes are compared to music?

 

[sarumbear]

Isn't this exactly what we need to do here? To find out how much higher spikes are compared to music?

No. We need to measure the values against the average level. You need a device that stores those peaks as they are very fast and averages your normal level.

 

[LTig]

I used this simple "Balanced Transmitter", the PCB version, https://sound-au.com/project87.htm and changed R102 to 47K. Loading for the cart is then right, you can have the transmitter close to the TT so easy to keep capacitance down, and the signal is transmitted to the ADC balanced. I found this to work very well ... ..especially no noise.

This a balanced receiver. Since the RME is already balanced two simple impedance converters would be sufficient. However I prefer not to worsen the great SINAD of the RME by putting something in front.

@MC_RME himself wrote to me here long ago that I could replace the two 9 kOhm resistors inside the ADI2 with 100 kOhm each which would be the perfect solution for me. I just did not find the guts to open the ADI2 and do it. Although I've built my own analog preamp and digital patchbay I've only had bad experience soldering SMD parts so far. Well warranty has run out so maybe ...

 

[LTig]

No. We need to measure the values against the average level. You need a device that stores those peaks as they are very fast and averages your normal level.

This is what a DSO can do.

 

[sarumbear]

This is what a DSO can do.

I understand that, however a DSO by name does not necessarily define the set of tools you get. I was commenting on an oscilloscope, not a composite tool. In an audio scope you know what to expect.

 

[Rja4000]

Your simulation is wrong. Do please find a relatively much LP and test the cartridge output.

I don't know if my simulation is wrong.
What I'm sure of is that you don't know either.

I suggest YOU now do what you say I should do, and come back with evidences of your statements.

I deleted my post (before noticing you answered) because I don't want to fuel this discussion further, while we still have no characterisation of what the clicks look like, in amplitude and frequencies.

From what I understood so far, the only agreement is that they are random high frequency noises, of unknown (or disputed) amplitude.

 

[sarumbear]

I don't know if my simulation is wrong.
What I'm sure of is that you don't know either.

I suggest YOU now do what you say I should do, and come back with evidences of your statements.

Bye. Enjoy your life.

 

[LTig]

I understand that, however a DSO by name does not necessarily define the set of tools you get. I was commenting on an oscilloscope, not a composite tool. In an audio scope you know what to expect.

Nowadays almost all oscilloscopes are DSOs. And almost all of them have the tools on board to show true RMS and peak value. Mine has, and I payed less then € 400 for it (new, 2 channels, 200 MHz, EDIT: incredible performance for this price).

 

[sarumbear]

Nowadays almost all oscilliscopes are DSOs. And almost all of them have the tools on board to show true RMS and peak value. Mine has, and I payed less then € 400 for it (new, 2 channels, 200 MHz).

Good to know. Thank you. What is the input sensitivity they normally have in that price range?