It is known that the DM1 can reduce the noise floor, is this method feasible for calculating the actual EIN?

[Richarrd]

This is my first time measuring such a device, and I hope the following process is correct. If there are any issues, please point them out, and I will correct them. Thank you!

I have a Topping E2X2, with an officially specified EIN of -130.5 dBu(A). I fed a known 2 mV signal into the INPUT (with the INPUT knob turned to maximum), calibrated both the generator and RTA in REW, then disconnected the cables and replaced them with a 150Ω load, obtaining a consistent -130.5 dBu(A) reading.

Next, I connected the DM1 (with the INPUT knob still at maximum) and fed it a calibrated voltage (since the DM1 has higher gain, the signal had to be reduced to 0.0x mV to avoid clipping). I then calibrated the RTA the same way as before, disconnected all cables, left the DM1 in place, and connected a 150Ω load at its output. The measured EIN was -129.3 dBu(A), indicating that it does not further improve the EIN of an already excellent audio interface and instead adds about 1 dBu of noise floor.

Later, I switched to an audio interface with a poorer EIN specification and followed the same method. The original EIN was -118.5 dBu(A).

After inserting the DM1 and recalibrating, and plug in load,the new EIN was -129.4 dBu(A), showing that it improved the noise floor of this interface and will increased the signal-to-noise ratio by nearly 11 dBu when actual recording.

Are there any issues with this process?

 

[Richarrd]

Additionally, I only have Topping E70 XLR output to feed the DM1 for distortion testing. Could this affect the measurement results?

 

[DVDdoug]

I didn't really "study" your procedure...

-130dBu seems hard to beat... Room/acoustic noise will probably dominate...

(since the DM1 has higher gain, the signal had to be reduced to 0.0x mV to avoid clipping). I then calibrated the RTA the same way as before,

For an apples-to-apples measurement the end-to-end gain should be the same. The booster has a gain of 28dB so the gain of the interface should be reduced by 28dB. With the same signal into the interface's ADC and the same digital signal level, the signal-to-noise ratios can be compared.

Of the end-to-end gain doesn't need to match if the gains are known. In that case you can measure the output noise and calculate EIN.

The main purpose of those microphone boosters is that many interfaces don't have enough gain for a dynamic or ribbon mic. But if the preamp in the interface is too noisy, turning down the gain may reduce the preamp noise and substituting the quieter gain from a booster may improve the performance.

 

[Richarrd]

Yes, but in this case, sine wave input is necessary for comparable data, right? If the audio interface's input gain knob is adjusted, calculating pure EIN without a sine wave might introduce additional variables?

From what I understand about EIN characteristics, it doesn’t increase as the gain knob is turned up**. In fact, EIN is typically worst at minimum gain and improves as gain increases because the signal-to-noise ratio widens.

The DM1 provides +28dB of gain, which seems equivalent to stacking additional gain on top of the interface’s maximum input gain (e.g., if the interface’s max gain is 50dB, inserting the DM1 would make it 50 + 28 = 78dB).

So, if we:
1. Keep the interface’s input gain at maximum,
2. Insert the DM1,
3. Recalibrate the input voltage,
then the resulting EIN measurement should still be meaningful—right?

For an apples-to-apples measurement the end-to-end gain should be the same. The booster has a gain of 28dB so the gain of the interface should be reduced by 28dB. With the same signal into the interface's ADC and the same digital signal level, the signal-to-noise ratios can be compared.

Of the end-to-end gain doesn't need to match if the gains are known. In that case you can measure the output noise and calculate EIN.

The main purpose of those microphone boosters is that many interfaces don't have enough gain for a dynamic or ribbon mic. But if the preamp in the interface is too noisy, turning down the gain may reduce the preamp noise and substituting the quieter gain from a booster may improve the performance.

 

[Blumlein 88]

Here is a thread about measuring EIN of audio interfaces. Gets into the weeds a bit at times.

How to measure EIN (Equivalent Input Noise) of an interface's Mic preamp

For audio interfaces with Mic preamp, low level / high gain noise is a key performance parameter. I therefore always measure EIN when reviewing an interface. But what is EIN, and how do we measure it ? I thought it was worth adding a proposed methodology that everybody could use. Of course...

www.audiosciencereview.com

 

[Richarrd]

Yes, but in this case, sine wave input is necessary for comparable data, right? If the audio interface's input gain knob is adjusted, calculating pure EIN without a sine wave might introduce additional variables?

From what I understand about EIN characteristics, it doesn’t increase as the gain knob is turned up**. In fact, EIN is typically worst at minimum gain and improves as gain increases because the signal-to-noise ratio widens.

The DM1 provides +28dB of gain, which seems equivalent to stacking additional gain on top of the interface’s maximum input gain (e.g., if the interface’s max gain is 50dB, inserting the DM1 would make it 50 + 28 = 78dB).

So, if we:
1. Keep the interface’s input gain at maximum,
2. Insert the DM1,
3. Recalibrate the input voltage,
then the resulting EIN measurement should still be meaningful—right?

Apologies—I realize describing it as "gain stacking" might not be accurate. What I meant was:

**If the DM1's self-noise is negligible**, could its +28dB gain effectively act as "headroom extension" for a lower-tier audio interface's EIN, thereby improving the EIN specification?

In this case, demonstrating the EIN improvement (e.g., from a poor original value to -128.4 dBuA at maximum gain) — **would this be a meaningful way to present the data?

 

[Blumlein 88]

Apologies—I realize describing it as "gain stacking" might not be accurate. What I meant was:

**If the DM1's self-noise is negligible**, could its +28dB gain effectively act as "headroom extension" for a lower-tier audio interface's EIN, thereby improving the EIN specification?

In this case, demonstrating the EIN improvement (e.g., from a poor original value to -128.4 dBuA at maximum gain) — **would this be a meaningful way to present the data?

When you put together an extra gain element in the mic preamp chain, it is the EIN of the first device which sets the overall EIN level. So yes a booster with lower EIN can improve upon the EIN of the device it is feeding.

 

[Richarrd]

Here is a thread about measuring EIN of audio interfaces. Gets into the weeds a bit at times.

How to measure EIN (Equivalent Input Noise) of an interface's Mic preamp

For audio interfaces with Mic preamp, low level / high gain noise is a key performance parameter. I therefore always measure EIN when reviewing an interface. But what is EIN, and how do we measure it ? I thought it was worth adding a proposed methodology that everybody could use. Of course...

www.audiosciencereview.com

I believe I can correctly measure EIN under normal circumstances. The main issue I'm encountering now is whether the EIN parameter measured in the same way still holds significance after adding such an additional device.

 

[Blumlein 88]

I believe I can correctly measure EIN under normal circumstances. The main issue I'm encountering now is whether the EIN parameter measured in the same way still holds significance after adding such an additional device.

Yes. I don't see any problems with how your measured or the results. As you saw, the basic EIN of the Dynamite DM-1 came thru with both devices. One of which was not an improvement, and one of which was. This measurement has the usual significance in that it shows what noise will enter your interface if you are using the DM-1 booster. No problems that I see with it or how you measured it.

 

[Richarrd]

Yes. I don't see any problems with how your measured or the results. As you saw, the basic EIN of the Dynamite DM-1 came thru with both devices. One of which was not an improvement, and one of which was. This measurement has the usual significance in that it shows what noise will enter your interface if you are using the DM-1 booster. No problems that I see with it or how you measured it.

thank you!