Review and Measurements of u-turn Pluto and ART DJPRE II Phono Preamps

[amirm]

This is a review, comparison and detailed measurements of Review and Measurements of u-turn Pluto and ART DJPRE II Phono Preamplifiers. Both fall in the category of budget products with the Pluto retailing for USD $99 and the DJPRE II for just $49 (both include Prime shipping). Of the two, the DJPRE II has more bells and whistles:

As you see, the DJPRE II has "low cut" filter switch together with input capacitance selector (100 or 200 pf). It also has a clipping indicator and variable gain switch.

The u-turn pluto on the other hand, is fixed function. No switches, indicators, or anything else other than the power LED.

Between the two, I like the look of the Pluto better. The DJPRE II screams hobby project box. And that bright blue LED is deadly. Needs some black tape to cover it for sure. It is a sturdy box though so it is just a visual preference.

Both units have external transformers. Pluto has its own branded unit that has 9 volt AC at 670 milliwatts. The DJPRE II's transformer also outputs AC at 9 volts and 1000 milliwatts. Interestingly enough both are identical in size so not sure about the ratings being correct. The fact that these are AC adapters means that rectification to DC is inside each unit.

There is not much else to them. So let's get into measurements and see how they do.

Measurements
We are breaking new ground here with our first test of phono pre-amps so expect future measurements to vary based on feedback to this thread. In addition, I am using my new Audio Precision APx555 so bear with me as I refine these graphs and settings over time.

To make the testing easier, I only tested one channel of each unit. That allowed me to test them simultaneously. For input voltage, I picked 5 millivolt. Both of these phono preamps are for moving coil and the research I did showed range of cartridge outputs of 3 to 6 millivolts. I don't see an industry standard for measurement. Let me know if you know of one.

Here is a "dashboard" view of how the two perform using 1 kHz tone at 5 millivolts, with A-weighting filtering of the output (making it more perceptually relevant):

I know, lots of data . Let's take a bit at a time.

For starters, I adjusted the variable output of the ART DJPRE II to match that of u-turn Pluto. This setting was just past the "zero" dial (around +1). This is good as it means the DJPRE II has good range of input sensitivity.

As you see, the output is 0.306 millivolts which corresponds to a gain in db of 35.6 or so. This matches the specification of Pluto and is within the range that DJPRE has.

Distortion with this a-weighting is 0.01% on Pluto. Manufacturer rates it at < 0.005%. Maybe at a different input level it gets closer to that.

DJPRE II's distortion spec is <0.01% and we are getting less than that at 0.007%. So good on them for being conservative.

The DRJPRE II has a clipping indicator. I find out that it lights up at just 0.01% so pretty conservative.

A-weighted signal to noise+distortion ratio ("SINAD") is 83 dB for DJPREE II beating the 80 db for Pluto. Both companies spec 90 dB.

See? It wasn't so bad.

Next, let's look at frequency response:

Here we have three graphs: one for Pluto and two for DJPRE II with its filter switch to Flat and Low Cut. When set to flat, the DJPRE indeed has a flat response that extends past 50 kHz. The Pluto has a low-cut built-in per specification.

When selecting the low-cut filter on DJPRE II, we indeed get a low cut (not as steep as Pluto though) but also a high-cut above 19 kHz or so. Not sure that is a bad thing but it is not what the button says it is doing.

Somewhat related to this, I ran a sweep to see how the phase varies based on frequency and got this:

Here, it is Pluto's turn to nail the phase response with essentially zero deviation at all frequencies. The DJPRe II though starts to have a negative phase shift below 1 kHz and it progressively gets worse down to 20 Hz at some 66 degrees. Selecting the low-cut filter makes things better with a positive phase shift that starts at 50 hz and goes up 20 degrees or so at 20 Hz. I am not sure what is going with the DJPRE II here. I can see a phase compensation at higher frequencies but not lower.

Next let's sweep the input level from 1 millivolt to 50 millivolt to see when overload occurs:

Here, Pluto does better with onset of clipping at 31 millivolt input level. The DJPRE II on the other hand, starts clipping at 22 millivolts. So those pops and clicks will be more obnoxious on the DJPRE II assuming they rise up to such levels.

Finally, let's look at distortion and noise of a 1 kHz tone with 5 millivolt input but without a-weighting:

Both devices are remarkably the same with one significant difference: the pluto bleeds noise from its power supply rectifier into its output. You can see this in harmonics of mains 60 Hz in blue. The DJPRE II actually has higher mains hum at 60 Hz, but it is otherwise much cleaner relative to power supply noise.

I think that is it. It is 1:00am so I am going to go to sleep. Leave me feedback on what you think of these results and I will read them when I wake up.

Conclusions
The ART DJPRE II is a great bargain here. Other than some strange phase shift at low frequencies and lower input dynamic range, it actually beats the more expensive u-turn Pluto in the rest of the measurements.

The main fault I see with the Pluto is the power supply noise. A bit more attention there could have easily landed it on top.

Both are great bargains from what I can see. Personally I find the DJPRE II a more flexible unit so if I had to buy one, it would be that if I could hide it some place.

----

As always, questions, comments, corrections, etc. are all welcome.

----
If you like this review, please consider donating funds for these types of hardware purchases 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).

 

[Wombat]

Thank you. Looking forward to some other 'consumer- level' units being tested for comparison. Input impedance measurements would be good(R+jXc).

 

[sergeauckland]

Firstly, thanks for making these measurements.

I think that both these items show a failing in modern Phono preamps, that of limited headroom. Preamps need an absolute minimum of 20dB above their nominal sensitivity, which needs to be in the range of 3.5mV to 5mV for 500mV output, and therefore with a maximum output of 5V before clipping.

Back in the 1970s, Shure had graphs that indicated that peak recorded levels on LPs were occasionally over 20dB above the nominal 5cm/sec recorded velocity, 18dB was regularly hit. To match more-or-less the maximum level of output from digital sources, somewhere around 2V, then phono stages need some several volts of peak output before clipping. There is of course a difference that CDs are mastered to 0dBFS (all too often) whilst LPs can be cut to any arbitrary level depending on how long each side is, and what level the cutting engineer wants to achieve.

It seems that many modern phono stages have inadequate overload margins, but far better noise performance than vinyl replay justifies. I would much rather see a S/N ratio of, say 70dB but a 25dB overload margin, than 80dB of S/N ratio and less than 20dB overload margin.

Also, I know these particular stages are budget stages, but why don't the more upmarket ones provide balanced inputs as standard? One gets unnecessary balanced inputs on line stages and domestic power amps, yet unbalanced inputs on the only items where they would be fully justified.

S.

 

[Wombat]

This is a review, comparison and detailed measurements of Review and Measurements of u-turn Pluto and ART DJPRE II Phono Preamplifiers. Both fall in the category of budget products with the Pluto retailing for USD $99 and the DJPRE II for just $49 (both include Prime shipping). Of the two, the DJPRE II has more bells and whistles:

View attachment 13107

As you see, the DJPRE II has "low cut" filter switch together with input capacitance selector (100 or 200 pf). It also has a clipping indicator and variable gain switch.

The u-turn pluto on the other hand, is fixed function. No switches, indicators, or anything else other than the power LED.

Between the two, I like the look of the Pluto better. The DJPRE II screams hobby project box. And that bright blue LED is deadly. Needs some black tape to cover it for sure. It is a sturdy box though so it is just a visual preference.

Both units have external transformers. Pluto has its own branded unit that has 9 volt AC at 670 milliwatts. The DJPRE II's transformer also outputs AC at 9 volts and 1000 milliwatts. Interestingly enough both are identical in size so not sure about the ratings being correct. The fact that these are AC adapters means that rectification to DC is inside each unit.

There is not much else to them. So let's get into measurements and see how they do.

Measurements
We are breaking new ground here with our first test of phono pre-amps so expect future measurements to vary based on feedback to this thread. In addition, I am using my new Audio Precision APx555 so bear with me as I refine these graphs and settings over time.

To make the testing easier, I only tested one channel of each unit. That allowed me to test them simultaneously. For input voltage, I picked 5 millivolt. Both of these phono preamps are for moving coil and the research I did showed range of cartridge outputs of 3 to 6 millivolts. I don't see an industry standard for measurement. Let me know if you know of one.

Here is a "dashboard" view of how the two perform using 1 kHz tone at 5 millivolts, with A-weighting filtering of the output (making it more perceptually relevant):
View attachment 13108

I know, lots of data . Let's take a bit at a time.

For starters, I adjusted the variable output of the ART DJPRE II to match that of u-turn Pluto. This setting was just past the "zero" dial (around +1). This is good as it means the DJPRE II has good range of input sensitivity.

As you see, the output is 0.306 millivolts which corresponds to a gain in db of 35.6 or so. This matches the specification of Pluto and is within the range that DJPRE has.

Distortion with this a-weighting is 0.01% on Pluto. Manufacturer rates it at < 0.005%. Maybe at a different input level it gets closer to that.

DJPRE II's distortion spec is <0.01% and we are getting less than that at 0.007%. So good on them for being conservative.

The DRJPRE II has a clipping indicator. I find out that it lights up at just 0.01% so pretty conservative.

A-weighted signal to noise ratio ("SINAD") is 83 dB for DJPREE II beating the 80 db for Pluto. Both companies spec 90 dB.

See? It wasn't so bad.

Next, let's look at frequency response:
View attachment 13109


Here we have three graphs: one for Pluto and two for DJPRE II with its filter switch to Flat and Low Cut. When set to flat, the DJPRE indeed has a flat response that extends past 50 kHz. The Pluto has a low-cut built-in per specification.

When selecting the low-cut filter on DJPRE II, we indeed get a low cut (not as steep as Pluto though) but also a high-cut above 19 kHz or so. Not sure that is a bad thing but it is not what the button says it is doing.

Somewhat related to this, I ran a sweep to see how the phase varies based on frequency and got this:
View attachment 13110

Here, it is Pluto's turn to nail the phase response with essentially zero deviation at all frequencies. The DJPRe II though starts to have a negative phase shift below 1 kHz and it progressively gets worse down to 20 Hz at some 66 degrees. Selecting the low-cut filter makes things better with a positive phase shift that starts at 50 hz and goes up 20 degrees or so at 20 Hz. I am not sure what is going with the DJPRE II here. I can see a phase compensation at higher frequencies but not lower.

Next let's sweep the input level from 1 millivolt to 50 millivolt to see when overload occurs:
View attachment 13111

Here, Pluto does better with onset of clipping at 31 millivolt input level. The DJPRE II on the other hand, starts clipping at 22 millivolts. So those pops and clicks will be more obnoxious on the DJPRE II assuming they rise up to such levels.

Finally, let's look at distortion and noise of a 1 kHz tone with 5 millivolt input but without a-weighting:

View attachment 13112

Both devices are remarkably the same with one significant difference: the pluto bleeds noise from its power supply rectifier into its output. You can see this in harmonics of mains 60 Hz in blue. The DJPRE II actually has higher mains hum at 60 Hz, but it is otherwise much cleaner relative to power supply noise.

I think that is it. It is 1:00am so I am going to go to sleep. Leave me feedback on what you think of these results and I will read them when I wake up.

Conclusions
The ART DJPRE II is a great bargain here. Other than some strange phase shift at low frequencies and lower input dynamic range, it actually beats the more expensive u-turn Pluto in the rest of the measurements.

The main fault I see with the Pluto is the power supply noise. A bit more attention there could have easily landed it on top.

Both are great bargains from what I can see. Personally I find the DJPRE II a more flexible unit so if I had to buy one, it would be that if I could hide it some place.

----

As always, questions, comments, corrections, etc. are all welcome.

----
If you like this review, please consider donating funds for these types of hardware purchases 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).

For MM not MC.

 

[SIY]

Turn off the time/date stamp. And you can change what the titles say and their font for clarity.

I think that to get accurate measurements of noise and SNR for an MM stage, you need to have a cartridge in series with the generator. Having a 50R source impedance alone can cover up issues with current noise. I keep a few old cartridge bodies around for just this sort of use.

RIAA accuracy is a critical measurement, especially channel-matching.

Since there's an RIAA curve, it's important to do headroom as a function of frequency. For clarity, you can also normalize it against velocity.

I may think of more suggestions later; my apologies for not posting examples (I've measured a lot of RIAA stages), but I'm about 1000 km from my home lab at the moment.

 

[SIY]

...why don't the more upmarket ones provide balanced inputs as standard? One gets unnecessary balanced inputs on line stages and domestic power amps, yet unbalanced inputs on the only items where they would be fully justified.

This x1000. It's pretty easy to make the cable/connector swapouts with most tonearms and run the system properly.

 

[amirm]

And you can change what the titles say and their font for clarity.

Thanks. I know that. But alas, the tools in photoshop are so much better for this use so I am sticking with that for now.

 

[amirm]

I think that to get accurate measurements of noise and SNR for an MM stage, you need to have a cartridge in series with the generator. Having a 50R source impedance alone can cover up issues with current noise. I keep a few old cartridge bodies around for just this sort of use.

I thought about this when I was testing. I tested with 600 ohm source impedance but at least in frequency response it made negligible difference. I will ponder what to do about this.

 

[amirm]

Since there's an RIAA curve, it's important to do headroom as a function of frequency.

You mean find the clipping point for each frequency? I am trying to think how to present a three-dimensional set of data in a 2-D chart.

 

[sergeauckland]

Since there's an RIAA curve, it's important to do headroom as a function of frequency. For clarity, you can also normalize it against velocity.

Yes and no. The RIAA curve is also applied on the cutting side, so overall cutting/replay headroom should be independent of frequency. That's why if headroom is even given as a specification point, it's normally quoted at 1kHz, as at other frequencies, the reduced headroom one side of 1kHz on reproduction should be compensated for an increased headroom on the cutting side and similarly for increasing headroom.

Generally, if the headroom quoted at 1kHz is enough, i.e. over 20dB, ideally over 25dB there is unlikely to be anything off the record that has a higher cutting velocity and therefore cause a playback issue. Note also that (very) few cartridges will track anything above 20dB above 5cm/sec, so a 20-25dB overload margin is what's required.

S.

 

[SIY]

I thought about this when I was testing. I tested with 600 ohm source impedance but at least in frequency response it made negligible difference. I will ponder what to do about this.

If you want an MM cartridge body, I'd be happy to donate one. A 600R source is OK for the DC part, but there's that 1H or so inductance which pretty much dominates at mid to treble for noise measurements. I have a calculator on my website which illustrates this nicely.

 

[SIY]

You mean find the clipping point for each frequency? I am trying to think how to present a three-dimensional set of data in a 2-D chart.

You can measure clipping at, say, ten spot frequencies, then plot it. There's an amazing variation between preamps in this, which is very dependent on how the EQ is set up, as well as the headroom of the basic gain block(s).

To Serge's point, it's not just what's cut on the records and what the cartridge can track, it's what happens when there's mistracking and when there's ticks and pops. A great phono stage will handle all of that without an issue. A good preamp will clip and instantly recover. A more usual preamp will take a few tens or hundreds of milliseconds to recover and sound bad with anything but pristine records.

Did I mention that measuring phono stages properly is a lot more complicated than line stages or DACs?

 

[Blumlein 88]

http://www.schiit.com/products/mani

Surely one of these will get tested by Amir.

 

[amirm]

You can measure clipping at, say, ten spot frequencies, then plot it. There's an amazing variation between preamps in this, which is very dependent on how the EQ is set up, as well as the headroom of the basic gain block(s).

Thanks. Do you mean this? Here, I am plotting the gain vs frequency that results in 1% THD:

The red one is Pluto and the brown one, the DJPre II. No RIAA pre-compensation is applied, I am assuming clicks and pops come unequalized .

 

[RayDunzl]

I wonder if that looks good or not...

Are the gain numbers just arbitrary or related to the input value?

 

[amirm]

Are the gain numbers just arbitrary or related to the input value?

They are real gains relative to input. It is basically saying that distortion occurs at much lower gains at the low and high frequencies. The RIAA is responsible for one end of this because there is that extra gain there which causes more limited headroom. I am still pondering what the other end means. Maybe SIY will tell us more.

 

[svart-hvitt]

Thanks, @amirm . Now we need someone send you a handful of $$$$$ phono amps!

How much does a vinyl lover need to put on the table to make phono amplification of top-notch quality!?

 

[sergeauckland]

Thanks, @amirm . Now we need someone send you a handful of $$$$$ phono amps!

How much does a vinyl lover need to put on the table to make phono amplification of top-notch quality!?

Very little. A decent enough MM phono stage can be made from an opamp like the 5532, for pennies. The real cost comes from the switches, necesssary for adjustable loading, and a nice box to put it in.
Just like DACs and line level amplifiers, there's no correlation between price and audio quality.

S

 

[SIY]

What I'll usually plot is Vin at clipping vs frequency, so you can determine the headroom.

An argument can be made that it's less critical at low frequencies because of the cartridge's velocity response, so potentially one could normalize the plot against frequency. I want to think a bit about the best way to do that which would convey the information most usefully.

 

[sergeauckland]

The issue with overload against frequency is the RIAA equalisation that's applied on the cutting side and the opposite on the replay side. This theoretically makes the overall result constant velocity, and as RIAA accuracy can be easily within 1dB and typically rather better than that, measuring the overload at 1kHz gives a decent enough measure of what's important.

Bearing in mind also that cutting levels at HF are typically well controlled, as due to the RIAA boost, and the fragility of the (very expensive) cutting head itself, excessive HF is a no-no, and with 'real' music as opposed to electronically generated, HF drops off naturally, there's not a problem at HF.

At LF, there's more of an issue as LF can be boosted on cutting, for 'artistic' effect, but also the arm/cartridge LF resonance always boosts bass output, so the post RIAA equalised headroom is important, especially at LF.

S.