Fosi Audio Box X5 Phono Preamp Review

[Skeeter]

I am in the process of moving house, but I will try to do a measurement in the next few days comparing shorted input with one with a real-world source (I have an AT95 I can use). I think the measurements of phono stages here on ASR are good, but the noise measurement unfortunately is really just looking at the voltage input noise floor and unfortunately misses the current input noise contribution which for many RIAA phono preamps is the bigger contributor (especially in bipolar input designs).

 

[Skeeter]

A simple measurement of the Box X5 input capacitance using an LCR meter. Each was set to an average of 128 times.
Also, photos, THD vs. input level, and output impedance.

Does your generator do an inverse RIAA sweep? Looking at your overload sweeps, reference the 1kHz sensitivity, the signal coming out the cart at 20 Hz is 20 dB lower. So if you take it as 5mV at 1kHz, at 20Hz it will be 500uV while at 20 kHz it will be 50mV. This seems to indicate the Fosi has 26 dB o/load at 20 Hz (very good) and 15 dB at 20 kHz. But, why are your 10kHz and 20kHz traces overlaid? I would expect there to be a 6 dB difference.

 

[Heinrich]

(*it would have been good to remain in single value under 100pf...fosi may be taking a little more precaution on a future model)

Maybe some expert could explain to me, why the headroom estimations of @nagster show opposite results. Increasing headroom with frequency, as I would expect if the second (or third?) amplification stage was the limiting factor.

Why is the input capacitance changing in nagster‘s measurements, when switching to MC? As far as I see there is only one physical C present.

(Not as if it was relevant, but running MC with 38dB of amplification is not foreseen?)

 

[Heinrich]

I am in the process of moving house, but I will try to do a measurement in the next few days comparing shorted input with one with a real-world source (I have an AT95 I can use). I think the measurements of phono stages here on ASR are good, but the noise measurement unfortunately is really just looking at the voltage input noise floor and unfortunately misses the current input noise contribution which for many RIAA phono preamps is the bigger contributor (especially in bipolar input designs).

I set a „like“ on the moderator‘s recent reminder. This is a separate, more general topic I would like to discuss more deeply in another thread. I‘ve got a few ideas and calculations for you … ;-)

 

[Heinrich]

Maybe some expert could explain to me, why the headroom estimations of @nagster show opposite results. Increasing headroom with frequency, as I would expect if the second (or third?) amplification stage was the limiting factor.

Sorry, got it already. Thanks to @Skeeter, for the right question!

 

[Skeeter]

Just going back to the shorted input measurement again. I think this in its own right is a good measure because although it misses out the input noise current, it does tell you what other internal noise sources are at play. Amirm mentioned mains frequency components and if you have a bad layout or less than adequate local regulation assuming the phono amp uses an external supply, these artifacts will show up. A 50/60 Hz noise spike in an FFT is almost always due to radiated noise from a transformer or cables. Most other mains-related harmonics are due to inadequate regulation, bad decoupling or layout - eg common impedance coupling.

 

[Heinrich]

Hope I understand the SINAD. In the digital regime all noises are suspicious. Think of codecs with nasty effects that correlate with the input, but in an unpredictable way.

With phono / vinyl the most prominent disturbance from an amplifier is hum, buzz even, thermal noise and in some rare cases exaggerated distortion. Maybe overload recovery is an additional quality parameter in regard to crackle and pops. Otherwise the mechanical limitations of the Media dominante anyway, today by far.

Is somebody willing to open another thread on this topic, on noise especially?

 

[KSTR]

Why is the input capacitance changing in nagster‘s measurements, when switching to MC? As far as I see there is only one physical C present.

My take, from experience, is that this is likely a measurement artefact. With 100R//150pF at 1kHz, the influence of the capacitor is extremely low, the resistor dominates the impedance by 80dB (10000x) at 1kHz. And given the low test voltage of 10mV, even the best LCR meter will have issues.
With test frequency = 100kHz and 100mV level I'm pretty sure the capactance will close in on the value we see in MM setting. There, you can see the effect of test frequency and test level as well. The 20kHz/100mV measurement is the closest one to reality. With 47k//150pF at 20kHz, the cap and resistor contribute about equal parts to the impedance and the meter can easily calculate resisitve and capacitive parts with high precision (7 significant digits is still completely ridiculuous, though).

 

[SIY]

My take, from experience, is that this is likely a measurement artefact.

Agree. The anomaly at the lowest test frequency is a major clue (think overload).

C meters are not a particularly reliable way to do this measurement- to get accurate and repeatable numbers, you need subterfuge to only tickle the preamp with a low level signal. IME, the best way to go is to inject a square wave via a voltage divider, measure the rise time at the input (using a 10x probe!), and using the Thevenin resistance, calculate C from rise time divided by Thevenin resistance.

 

[KSTR]

C meters are not a particularly reliable way to do this measurement- to get accurate and repeatable numbers

I get quite reliable results with one of the better tweezer-type meters which has the advantage that there is basically no test lead capacitance/resistance/inductance to null out, another factor that affects precision and resolution. And of course fully floating. You can probe the suspected C directly on the ciruit board.

In case of RIAA input, the higher the test frequency the larger one can choose test level to not overload the amp (which is only a problem when that overload actually affects the measurement, this would not be the case here for MC as it is simple pre-pre with maybe ~20dB gain or so).

 

[Muddywaters]

Want to thank you for sharing your technical expertise and insight on capacitance measurement of phono stages! @SIY @KSTR @nagster and @Skeeter

Somewhat off topic but does lower inductance of a cartridge such as some of my older Grado’s 10mH have any impact on noise into 47K? It was noted by SIY earlier, going from 470 to 1000mH was marginal for testing purposes.

When manufacturers specify, (those that do, at least, really can’t fathom not providing it), input capacitance, how do they measure it? Maybe just place a cap in of some nominal value, 50-220 seems common, call it day?

If mods want this deleted or perhaps a new thread created I’m game

 

[SIY]

Somewhat off topic but does lower inductance of a cartridge such as some of my older Grado’s 10mH have any impact on noise into 47K?

Yes. If you have 500mH, the impedance at 10kHz is about 32k shunting the 47k resistor (keep in mind the reactive vs resistive aspect and ignoring the DCR, but that doesn't change the argument). 10mH puts the shunting impedance at 0.6k (and the DCR may dominate).

When manufacturers specify, (those that do, at least, really can’t fathom not providing it), input capacitance, how do they measure it?

They rarely do. They'll stick a 100pF cap there and claim the input C is 100pF rather than accounting for the rest of the circuitry.

 

[Muddywaters]

Yes. If you have 500mH, the impedance at 10kHz is about 32k shunting the 47k resistor (keep in mind the reactive vs resistive aspect and ignoring the DCR, but that doesn't change the argument). 10mH puts the shunting impedance at 0.6k (and the DCR may dominate).

I’m an old stoopid history major, does shunting (like that word) @ 0.6k vs shunting @ 32k mean lower apparent noise?

They rarely do. They'll stick a 100pF cap there and claim the input C is 100pF rather than accounting for the rest of the circuitry.

Thought this might be the case thanks. Probably design dependent but how much might the circuitry impact it in general (rough ballpark). Wide ranging?

 

[SIY]

Wide ranging?

Yes. Can be negligible, can be a pile. Which is why any review of an MM preamp should include that measurement.

 

[SIY]

does shunting (like that word) @ 0.6k vs shunting @ 32k mean lower apparent noise?

Exactly. "Path of least resistance" is a real thing.

 

[NTTY]

Once upon a time, we had true headroom in RIAA stage (THD vs level with 5 harmonics measurement, BW 96kHz):

190mV @1kHz

 

[nagster]

Thank you everyone.
I am especially grateful to SIY and KSTR. Criticism is welcome. I have a lot to learn.

 

[nagster]

This is the AP dashboard when a short plug/AP generator/dummy is connected to the Box X5 input.
204kHz/1uV output only when AP generator was connected.
The dummies are Technics SL-1200mk4 and audio-technica VM540ML.
The RCA cable is genuine and 1.2m long. The earth wire is one I found in the box.

Added.
FFT length is 32K. Window is AP-Equiripple. Average is 3.

Sorry. It wasn't a dummy, it was the real thing.

Added capacitance measurement of turntable + connecting cable.

 

[dagfinn]

This is the AP dashboard when a short plug/AP generator/dummy is connected to the Box X5 input.
204kHz/1uV output only when AP generator was connected.
The dummies are Technics SL-1200mk4 and audio-technica VM540ML.
The RCA cable is genuine and 1.2m long. The earth wire is one I found in the box.

Added.
FFT length is 32K. Window is AP-Equiripple. Average is 3.

Thanks, great work! Could you possibly measure the tonearm/interconnect capacitance?

 

[StevenSGT98]

https://www.ebay.com/itm/2051265354...WQh5JmfS6S&var=&widget_ver=artemis&media=COPY

eBay price drop to $81.99 and free shipping in the U.S.