This is a review and detailed measurements of the JDS Labs Synapse and Neutron Hifi V1 USB isolators. They were both sends to me by the respected companies. Synapse costs US $69, and Neutron US $118.
As you see, the styles are very different. Synapse comes in a playful red color. Neutron in a more serious case, resembling a USB power supply.
Some background: USB bus carries data, power and ground. Data is used by all downstream devices naturally (e.g. DACs). Power may or may not be used. The ground however, is always used as the reference for both data and power. Being normally connected to an active and high power computing device, USB ground and power can both be quite polluted. As can the data lines. If you are using balanced output in your downstream device (again, such as DAC), there is little import to this. However, if you are using unbalanced RCA connection, the noise on the USB ground can travel through the DAC and manifest itself as noise and ground loop. How much this happens, and how audible it is, is complex and not predictable. Suffice it to say, once it becomes audible, you are kind of stuck, having to possibly change your computing device, DAC or other games to try to hopefully eliminate such noise. It can be quite frustrating.
A USB isolator such as the ones being reviewed here, have two USB connections: an input and and output. They are electrically isolated from each other so have the potential to solve the noise problem above. Some expensive DACs have an isolator inside but I would say 99% do not.
Both of these isolators also regenerate the USB power. In the case of Synapse, you are only provided 100 milliamps whereas Neutron V1 passes through 400 milliamps. Further, it accepts power over micro-USB which can augment what it can produce. To the extent you have USB powered devices, the Neutron may work better depending on power consumption of the device.
Testing such devices is difficult. I could create an artificial ground loop but that would not be representative of real life situations. Fortunately, I just reviewed the JDS Element IV DAC and HP Amp which due to inclusion of its own AC power adapter, objectively shows the noise transmitted over USB. So I am going to use that for analysis of how well these isolators work.
USB Isolator Measurements
Let's first look at the dashboard of Element IV with direct USB connection:
I can always tell leakage of computer noise in the dense spectrum it creates in the FFT spectrum, top right. That noise while unlikely to be audible, adds up to the same level as distortion specs so we don't want that. Please note that I am seeing this with my Audio Analyzer which has floating input, i.e. its "RCA" connection is not grounded. When faced with this situation, I usually resort to different ways of grounding the analyzer to device being tested. Unless the noise is internal to the device itself, I am usually able to reduce it fair bit as is the case here:
Notice how a lot of that hash is gone in high frequencies. But we still have a tilted up noise floor in lower frequencies. SINAD is improved by 2 or so dB.
Now let's connect JDS Sypanse inline with the USB connection (now without grounding):
We see complete clean up of the noise floor with SINAD elevating 2 to 3 dB yet again.
I replaced Synapse with Neutron V1 and performance improvement was the same (there are tiny run to run variations):
We can zoom into the noise floor without the signal present:
Please ignore the differences in mains noise. That is again, nature of the floating analyzer input. If I use both Synapse and grounding, that gets almost eliminated:
Conclusions
Of all the devices I test, I say around 5 to 10% of time I run into noise issues. Per my explanation above, I am usually able to reduce their impact to be negligible. But in a few cases, I have failed to do so. You might say that is significant but my analyzer is not 100% stand in for whatever amplifier you use. You may have bigger or smaller problem or none. When hit with that issue, you can use Toslink optical but that may not exist in your source. And at any rate, Toslink is usually limited to 96 kHz sampling. With either one of these isolators, you can get maximum performance using USB bus and eliminate any dependence on the source.
I am happy to recommend both JDS Synapse and Neutron V1 USB isolators. Both function very well and are reasonably priced.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
As you see, the styles are very different. Synapse comes in a playful red color. Neutron in a more serious case, resembling a USB power supply.
Some background: USB bus carries data, power and ground. Data is used by all downstream devices naturally (e.g. DACs). Power may or may not be used. The ground however, is always used as the reference for both data and power. Being normally connected to an active and high power computing device, USB ground and power can both be quite polluted. As can the data lines. If you are using balanced output in your downstream device (again, such as DAC), there is little import to this. However, if you are using unbalanced RCA connection, the noise on the USB ground can travel through the DAC and manifest itself as noise and ground loop. How much this happens, and how audible it is, is complex and not predictable. Suffice it to say, once it becomes audible, you are kind of stuck, having to possibly change your computing device, DAC or other games to try to hopefully eliminate such noise. It can be quite frustrating.
A USB isolator such as the ones being reviewed here, have two USB connections: an input and and output. They are electrically isolated from each other so have the potential to solve the noise problem above. Some expensive DACs have an isolator inside but I would say 99% do not.
Both of these isolators also regenerate the USB power. In the case of Synapse, you are only provided 100 milliamps whereas Neutron V1 passes through 400 milliamps. Further, it accepts power over micro-USB which can augment what it can produce. To the extent you have USB powered devices, the Neutron may work better depending on power consumption of the device.
Testing such devices is difficult. I could create an artificial ground loop but that would not be representative of real life situations. Fortunately, I just reviewed the JDS Element IV DAC and HP Amp which due to inclusion of its own AC power adapter, objectively shows the noise transmitted over USB. So I am going to use that for analysis of how well these isolators work.
USB Isolator Measurements
Let's first look at the dashboard of Element IV with direct USB connection:
I can always tell leakage of computer noise in the dense spectrum it creates in the FFT spectrum, top right. That noise while unlikely to be audible, adds up to the same level as distortion specs so we don't want that. Please note that I am seeing this with my Audio Analyzer which has floating input, i.e. its "RCA" connection is not grounded. When faced with this situation, I usually resort to different ways of grounding the analyzer to device being tested. Unless the noise is internal to the device itself, I am usually able to reduce it fair bit as is the case here:
Notice how a lot of that hash is gone in high frequencies. But we still have a tilted up noise floor in lower frequencies. SINAD is improved by 2 or so dB.
Now let's connect JDS Sypanse inline with the USB connection (now without grounding):
We see complete clean up of the noise floor with SINAD elevating 2 to 3 dB yet again.
I replaced Synapse with Neutron V1 and performance improvement was the same (there are tiny run to run variations):
We can zoom into the noise floor without the signal present:
Please ignore the differences in mains noise. That is again, nature of the floating analyzer input. If I use both Synapse and grounding, that gets almost eliminated:
Conclusions
Of all the devices I test, I say around 5 to 10% of time I run into noise issues. Per my explanation above, I am usually able to reduce their impact to be negligible. But in a few cases, I have failed to do so. You might say that is significant but my analyzer is not 100% stand in for whatever amplifier you use. You may have bigger or smaller problem or none. When hit with that issue, you can use Toslink optical but that may not exist in your source. And at any rate, Toslink is usually limited to 96 kHz sampling. With either one of these isolators, you can get maximum performance using USB bus and eliminate any dependence on the source.
I am happy to recommend both JDS Synapse and Neutron V1 USB isolators. Both function very well and are reasonably priced.
------------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited:
