This is a review and detailed measurements of the Grace Design m900 DAC and headphone amplifier. It is on kind loan from a member and was drop shipped to me. It costs USD $575 but I see it on sale for USD $545.
The unit has a substantial feel to it despite being small. Its design highlight is the metal rotary control for the volume control:
The rotary control feels a bit light to me but otherwise, it is one of the best controls in the business. Unlike similar looking ones from JDS Labs, this one has a metal knob giving it more of a luxurious feel. A couple of bright white LEDs show the current volume level as you see in the picture.
The unit comes with an optional external USB power supply which is beefier than the typical mobile phone one. The manual says it will work without it but it will sense the lack of sufficient current from the USB port and throttle the output. For my testing, I left the power supply plugged in.
Typical of modern DACs, the device is UAC2 compliant so no need for drivers or anything. Plug it in and you are in business.
There are three inputs in the back: USB, S/PDIF and Toslink. I only tested the USB port.
Massdrop sells a version of this called m9xx. A local member has offered that for testing so I will be able to verify how different they may be.
Strangely, the web page for m900 is missing from Grace Design's own website.
Let's get into measurements and see how she does.
DAC Measurements
As usual, I test the DAC portion of these combo units before testing the headphone output. Here, the RCA out is variable and goes to multiples of what we need (2 volts) indicating RCA outs are just tapped from the headphone amplifier. In these cases I adjust the level to 2 volts to match DACs normally tested and run the dashboard:
Wow, these are excellent numbers! Second harmonic distortion is down to -115 dB or so. Combined and added to noise level, the Grace Design m900 lands solidly in our tier 1 performance chart:
Likewise dynamic range is quite good with 2 volt output:
Story gets even better with the m900 nailing the jitter test with very low noise level and almost total absence of spurious tones.
The good news doesn't end there. Here is linearity:
Intermodulation distortion and noise doesn't quite match match our reference Topping DX3 Pro in noise level (sloping down part of the curve) but has almost no level related distortion, beating the DX3 Pro that way at max volume (of again, 2 volts):
Can this keep going when we run the 32-tone test?
Oh, what happened here? We have those intermodulation between our spikes going down to the lowest frequency and accumulating heavily as we go up in frequency. Why didn't this show up in previous tests? I am not 100% but this test is run at 192 kHz sampling rate so perhaps there is a problem there?
To make sure the setup was not the cause, I ran the identical test on Topping D50 and got the expected clean results:
As you see, we used the same power supply as Grace Design.
Let's see if we can get a clue as to what may be wrong by running the THD+Noise versus frequency:
Look at that massive rise in THD+N starting at 5 kHz. Note that this test uses a wide spectrum of 90 kHz versus the dashboard which runs at 22.4 kHz. Let's look at the wideband spectrum at different frequencies and see if we can find the source of this problem:
On the left is 1 kHz and all is well more or less. But notice what happens when I switch to 8 kHz on top right. We have that large spike at 40 kHz. The more I increase the frequency, the taller it gets. To wit, the inset is at 16 kHz and by then the 40 kHz tone is at very high levels.
The frequency of the 40 kHz tone is not changing with source frequency. It is its amplitude that increases. I wonder if the output stage is oscillating at that frequency?
EDIT: The above issue was due to selection of filter F4 which provides no filtering in transition band. Selecting F1 filter remedied that: https://www.audiosciencereview.com/...of-grace-design-m900-dac-amp.6470/post-144171
Headphone Amplifier Measurements
Let's switch to the headphone out jack and run our usual power versus distortion+noise at 300 ohm:
The m900 matches the power output of Topping DX3 Pro but can't keep up with it with respect to distortion. Distortion sets in earlier and climbs higher. On noise level, due to lack of gain settings, it loses a lot to the DX3 Pro in low gain mode.
Changing the load to 33 ohm gets us this:
We get nearly 1 watt of power but distortion sets in pretty early (around 2 milliwatts). Clearly the amplifier in m900 is less at home with a low impedance load than high.
Output impedance is a low and comfortable 1.2 ohm:
This shouldn't give you trouble with any headphone.
Channel balance was excellent as is typical of these DAC and Amp combos where they can control the level digitally through the DAC chip:
Listening Tests
By accident, I started my listening tests with the m600 running purely on USB power (i.e. no external power supply). With my Sennheiser HD-650s, the m900 barely got loud enough and fidelity was just OK. Switching to much lower impedance Hifiman HE400i caused additional trouble. Getting close to anywhere close to max volume would cause distortion. In addition, the unit would shut down with "OC" indicator which I assume to be over current.
Scratching my head that the above should not be happening, I realized my power strip was shut off so the power supply was off as well. Turned that on and there was a larger transformation. Bass became thundering with both headphones and I could not detect any distortion even at the limit. As such, I highly recommend that you run the m900 using its external power supply with power hungry headphones such as the ones I used.
Conclusions
The Grace Design m900 starts the show well with excellent measured performance. It was hard to find a fault until I tested the unit with wider bandwidth than 22.4 kHz I normally use. There, there are high frequency oscillations which sharply reduce performance. Mind you, the problem is ultrasonic and not audible but from engineering point of view, they should not be there. It is amazing how close to winning the m900 the measurement crown for dac+amp but didn't quite get there at the end.
Subjectively, performance is excellent. There is plenty of volume there to drive just about any headphone. No, it is not distortionless like some of the other products we have measured but subjectively, it is fine.
Price is high of course but with it you get excellent build, super nice volume control and attractive and unique packaging.
For those of you wanting to find a western design version of Topping DX3 Pro, the Grace Design m900 is it. On that basis, I am going to recommend the product even though I wished there were no ultrasonic distortions visible in measurements.
------------
As always, questions, comments, recommendations, etc. are welcome.
They say you should save money for a rainy day. It rains for about 6 months here so I need plenty of money!!! Please consider donating funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
The unit has a substantial feel to it despite being small. Its design highlight is the metal rotary control for the volume control:
The rotary control feels a bit light to me but otherwise, it is one of the best controls in the business. Unlike similar looking ones from JDS Labs, this one has a metal knob giving it more of a luxurious feel. A couple of bright white LEDs show the current volume level as you see in the picture.
The unit comes with an optional external USB power supply which is beefier than the typical mobile phone one. The manual says it will work without it but it will sense the lack of sufficient current from the USB port and throttle the output. For my testing, I left the power supply plugged in.
Typical of modern DACs, the device is UAC2 compliant so no need for drivers or anything. Plug it in and you are in business.
There are three inputs in the back: USB, S/PDIF and Toslink. I only tested the USB port.
Massdrop sells a version of this called m9xx. A local member has offered that for testing so I will be able to verify how different they may be.
Strangely, the web page for m900 is missing from Grace Design's own website.
Let's get into measurements and see how she does.
DAC Measurements
As usual, I test the DAC portion of these combo units before testing the headphone output. Here, the RCA out is variable and goes to multiples of what we need (2 volts) indicating RCA outs are just tapped from the headphone amplifier. In these cases I adjust the level to 2 volts to match DACs normally tested and run the dashboard:
Wow, these are excellent numbers! Second harmonic distortion is down to -115 dB or so. Combined and added to noise level, the Grace Design m900 lands solidly in our tier 1 performance chart:
Likewise dynamic range is quite good with 2 volt output:
Story gets even better with the m900 nailing the jitter test with very low noise level and almost total absence of spurious tones.
The good news doesn't end there. Here is linearity:
Intermodulation distortion and noise doesn't quite match match our reference Topping DX3 Pro in noise level (sloping down part of the curve) but has almost no level related distortion, beating the DX3 Pro that way at max volume (of again, 2 volts):
Can this keep going when we run the 32-tone test?
Oh, what happened here? We have those intermodulation between our spikes going down to the lowest frequency and accumulating heavily as we go up in frequency. Why didn't this show up in previous tests? I am not 100% but this test is run at 192 kHz sampling rate so perhaps there is a problem there?
To make sure the setup was not the cause, I ran the identical test on Topping D50 and got the expected clean results:
As you see, we used the same power supply as Grace Design.
Let's see if we can get a clue as to what may be wrong by running the THD+Noise versus frequency:
Look at that massive rise in THD+N starting at 5 kHz. Note that this test uses a wide spectrum of 90 kHz versus the dashboard which runs at 22.4 kHz. Let's look at the wideband spectrum at different frequencies and see if we can find the source of this problem:
On the left is 1 kHz and all is well more or less. But notice what happens when I switch to 8 kHz on top right. We have that large spike at 40 kHz. The more I increase the frequency, the taller it gets. To wit, the inset is at 16 kHz and by then the 40 kHz tone is at very high levels.
The frequency of the 40 kHz tone is not changing with source frequency. It is its amplitude that increases. I wonder if the output stage is oscillating at that frequency?
EDIT: The above issue was due to selection of filter F4 which provides no filtering in transition band. Selecting F1 filter remedied that: https://www.audiosciencereview.com/...of-grace-design-m900-dac-amp.6470/post-144171
Headphone Amplifier Measurements
Let's switch to the headphone out jack and run our usual power versus distortion+noise at 300 ohm:
The m900 matches the power output of Topping DX3 Pro but can't keep up with it with respect to distortion. Distortion sets in earlier and climbs higher. On noise level, due to lack of gain settings, it loses a lot to the DX3 Pro in low gain mode.
Changing the load to 33 ohm gets us this:
We get nearly 1 watt of power but distortion sets in pretty early (around 2 milliwatts). Clearly the amplifier in m900 is less at home with a low impedance load than high.
Output impedance is a low and comfortable 1.2 ohm:
This shouldn't give you trouble with any headphone.
Channel balance was excellent as is typical of these DAC and Amp combos where they can control the level digitally through the DAC chip:
Listening Tests
By accident, I started my listening tests with the m600 running purely on USB power (i.e. no external power supply). With my Sennheiser HD-650s, the m900 barely got loud enough and fidelity was just OK. Switching to much lower impedance Hifiman HE400i caused additional trouble. Getting close to anywhere close to max volume would cause distortion. In addition, the unit would shut down with "OC" indicator which I assume to be over current.
Scratching my head that the above should not be happening, I realized my power strip was shut off so the power supply was off as well. Turned that on and there was a larger transformation. Bass became thundering with both headphones and I could not detect any distortion even at the limit. As such, I highly recommend that you run the m900 using its external power supply with power hungry headphones such as the ones I used.
Conclusions
The Grace Design m900 starts the show well with excellent measured performance. It was hard to find a fault until I tested the unit with wider bandwidth than 22.4 kHz I normally use. There, there are high frequency oscillations which sharply reduce performance. Mind you, the problem is ultrasonic and not audible but from engineering point of view, they should not be there. It is amazing how close to winning the m900 the measurement crown for dac+amp but didn't quite get there at the end.
Subjectively, performance is excellent. There is plenty of volume there to drive just about any headphone. No, it is not distortionless like some of the other products we have measured but subjectively, it is fine.
Price is high of course but with it you get excellent build, super nice volume control and attractive and unique packaging.
For those of you wanting to find a western design version of Topping DX3 Pro, the Grace Design m900 is it. On that basis, I am going to recommend the product even though I wished there were no ultrasonic distortions visible in measurements.
------------
As always, questions, comments, recommendations, etc. are welcome.
They say you should save money for a rainy day. It rains for about 6 months here so I need plenty of money!!! Please consider donating funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
Last edited: