Fosi Audio Box X5 Phono Preamp Review

[morillon]

No, it is up to whoever claims measurements change significantly. If they have never measured, then they can't make that assertion especially when simple things like mains leakage and equalization is not taken into account.

doubtful...it's up to you to lift it....especially since it is linked to your protocol, AP bench low impedance in just this precis use..

we cannot claim a "scientific" approach and not include doubt... the desire to progress etc. or even self-criticism... it's fundamental, isn't it?
for once it's simple... we're not at CERN
;-)

 

[amirm]

We are not discussing "science." This is engineering. Your back of the envelop stuff doesn't motivate me to do anything different. You can talk about it but don't say this and that impacts measurements when you don't have such. The bar is always high for people to want testing to change. Do your homework or cut back your expectations that a few posts would be the reason to jump and do the homework for you.

 

[morillon]

Audiosciencereview

not from me..and never ceases to be claimed
,-)

 

[KSTR]

Come back when you can show this. Don't do math without considering RIAA equalization.

It's with EQ, of course. As is my math and sim.

I don't care about math or simulation. Bring measurements of actual phono stage including mains noise and RIAA equalization.

Taking values from a proper simulation is a valid actual measurement. And it is extremely close to reality, because if it weren't the life of us today's electronic design engineers would be pure hell and we'd all be fired.

The only point you have is mains hum/buzz generated in the preamp which could be so bad that shorted-input SINAD is spoiled so much that measuring it with proper termination would not worsen things big time.

Perceptually, increased broadband noise, notably in the sensitive ~3kHz region, will still be the obviuos (and only) perceived difference

EDIT: It is clearly possible to indicate whether noise, mains hum or actual harmonic distortion is dominating things. Simply draw a horizontal line with the measured broadband
RMS noise (properly terminated input, of course) in the plot. Any peaking spectal line that comes very close to that line is likely dominating it but if they are all away 10s of dBs then broadband is domanting.

 

[amirm]

Taking values from a proper simulation is a valid actual measurement.

It is not. A simulation is not a measurement. The only way the simulation would match the actual implementation is if all factors are included. Until you measure the real device and cross compare, you don't know if you have done that.

 

[Heinrich]

… ignorantly used in an MM preamp because its voltage noise is low.

The caveat is clearly understood. That‘s why I first was on the verge to provide a spread sheet to isolate the current‘s contribution from two measurements with different resistive generator impedances. Your calculator could serve as the proper complement, once the value is known.

Then I understood also that it won‘t be used by anyone. In case you find some time to showcase the issue with the X5, which frankly reveals the i/n as something around 1,5pA from knowing the opamp, let‘s see.

Btw., I own an old Denon amp that sports discrete FET for MC

 

[KSTR]

It is not. A simulation is not a measurement. The only way the simulation would match the actual implementation is if all factors are included. Until you measure the real device and cross compare, you don't know if you have done that.

Of course it is, quoting Wikipedia "Measurement is the quantification of attributes of an object or event, which can be used to compare with other objects or events." It does not matter if the object is physical or a precise model of a physical object.

And please don't underestimate the skills end experience of seasoned design engineers. In my day job of course I have to validate my designs, comparing sim (rather complex, with all relevant parasitics) and reality. Noise measurements are always spot on.

 

[pma]

Simulation is just as good as are the models used and how much of the non-ideal real parameters of the components and materials used it is able to cover. This may differ quite strongly, from exact to useless results.

 

[morillon]

This is not so much the subject... modeling or not modeling etc..
here.. factual observation "with a given protocol" being easy ...
lucky...
Is the protocol here so relevant or not in this specific case, "load adaptation" ... it's a nice way to question things.
;-)

 

[amirm]

And please don't underestimate the skills end experience of seasoned design engineers. In my day job of course I have to validate my designs, comparing sim (rather complex, with all relevant parasitics) and reality. Noise measurements are always spot on.

You wouldn't know if they are "spot on" until you have both simulation and actual device on hand. You can't make such claim with only the simulation data. As I keep saying, come back with both pieces and then we can talk.

 

[amirm]

Of course it is, quoting Wikipedia "Measurement is the quantification of attributes of an object or event, which can be used to compare with other objects or events." It does not matter if the object is physical or a precise model of a physical object.

Now we are into word salad arguments??? Claim was that SINAD, as I measure it, would significantly degrade with a cartridge input. I am asking for actual evidence of that.

If you are going to look up something, I suggest looking up Amdahl's Law.

 

[pma]

This really depends on cartridge impedance (R + jwL), and on circuit design of the phono preamp input, BJT or JFET (input bias current and also voltage noise with higher impedance). Put the cartridge to the preamp input and measure noise spectrum at phono preamp output and you will see. Want evidence? I can do it and already done it many times. So yes, SINAD measured from AP resistor output does not equal to real life SINAD with specific cartridge.

BTW, I have some examples in my archive.

The noise penalty is 6.5dB with the cartridge.

 

[amirm]

There is no SINAD measurement there.

 

[pma]

There is no SINAD measurement there.

Are you serious????

SINAD = THD+N in dB just with opposite sign (e.g. 100dB/-100dB). If THD (distortion) component is negligible, then N (noise) dominates and SINAD is almost completely defined by noise. THD 1kHz is well below -100dB with a good design, so it is the noise with cartridge that is important. Yes, vinyl distortion is another point, but not discussed now. So it would be very nice of you to admit that the noise profile and thus the best achievable SINAD with the specific cartridge will differ in reality from your measurements. This again is a simulation (your setup) vs. real world conditions. The simulation that must be verified, as you have already said, very well said.

 

[SIY]

Why are you using a Kaiser window for quasi-random noise?

 

[asibbald]

We seem to be getting a little off the topic of the excellent X5 amp.
But didn't Doug Self pretty much answer this discussion (as many others in Audio Electronics) some time ago?

Here's an extract from Edition 1 of his "Small Signal Audio Design":

 

[asibbald]

He also produced a "MAGNOISE" spreadsheet to assist in noise calculations (designer's perspective of course) and quoted some results in his book:

 

[pma]

@SIY - Frankly I am not able to say why I did that four and half year ago. But I do not think that another window would change the relative rise of HF noise from cartridge that is given by input 47kohm preamp resistor noise that is shunted by high inductance of the MM cartridge, 700mH in series with 1.3kohm. That is the reason of elevated noise with the cartridge.

 

[morillon]

the "75"moderne..
flagship in modest...at-95... ( with 2m red/blue or mp-110)
;-)
(maybe observing the current results here mc results finally tell us much more about the engineering of the mm part..maybe ;-) )

 

[Heinrich]

…

Great to see this coming to terms (once real data is at hand). Won‘t spoil your discussion, but is the input capacitance missing in D.Self‘s examples? If taken into the calculation we would have a resonant circuit with increased impedance compared to just a shunted inductor.

The noise figures NF in the table are quite interesting. Except for the FET opamps they are, I would say, low at 2dB or so. NF describes the noise penalty of the amp, all extra noise sources within it considered. Or, as depicted in the table with theoretical noiseless amps, best to be achieved is NF = 0dB. With NF = 0dB all noise is generated by the cartridge alone. And that‘s not nil / nothing.

@pma, to consider the self noise of the cartridge would round up your highly appreciated contribution above, the measurement.

Now, as the penalty is that low, do we care?