Fosi Audio V3 Amplifier Review

[maudio]

Last time I tested a sparkos opamp in an amplifier, it actually measured worse: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/


Distortion jumped up by 10 dB.

Maybe because of the Sparkos higher gain, using more power, driving the power supply harder?

 

[Ultratek]

Last time I tested a sparkos opamp in an amplifier, it actually measured worse: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/

Distortion jumped up by 10 dB.

 

[TheBatsEar]

Last time I tested a sparkos opamp in an amplifier, it actually measured worse: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/

Distortion jumped up by 10 dB.

Your test measures Sparkos Pro SS2590 versus Sonic Imagery 990ENH.

Someone posted this link earlier:

Input Buffer for Hypex NC500 and Purifi Audio 1ET400A

CUSTOMIZABLE Input buffer for Hypex and Purifi Audio ✔ HOW IT WORKS ✔ Design Process ✔ Technical Specifications ✔ Spectral density

atm-audio.com

It tests Sparkos Labs SS3602 against a bunch of others, it measured at -110dB highest distortion, all the others measured higher, some even a lot higher.

So it seems enthusiasts should look for the Sonic Imagery 990ENH then?

 

[strawberry644]

You said:
The temperature of element giving away the heat is always higher the better heat conductivity is.

Basically you are saying that a better heatsink will get hotter than a small heatsink. And I won't go deeper that that.

I knew this post will get me in trouble.

What is wrong with being a Mechanical Engineer?. That just proves that I know what I'm talking about. Or at least I studied in it College and passed the subject. Sorry if that offended you.


Back to the point. I'm still not sure if "I" need to apply that extra thermal paste between heatsink and enclosure as I did between chip and heatsink. Due to my low power draw on my particular case. Of course the higher the power consumption the more visible the benefit from better conductivity(thermalpaste).

Some thermalpastes get very sticky or dry (like glue) even after heating up and I'm afraid the case won't be able to be opened. There is no way to twist the heatsink while the pcb is inserted. I think I have valid (non-irational) concerns. And probably the same reason Fosi Audio did not use thermalpaste there.

I think all the points you mentioned are true, none of this is needed the cooling system works fine, but now it works better in my unit.
Im still a overclocker at heart and any small mod I can try is fun.
In my experience the thermal pastes Ive used are not gonna dry out to that state within like 5 years, and by than im sure ill be messing with something else.

 

[Doenerkunde]

Minutes ago I was in the process of pulling the trigger on the V3 in the Fosi store. During the checkout process I was not able to finish the PayPal payment because the input screen wouldn´t react, so I refreshed the site and the price went from 109$ to 131,99$. I missed the old price by just a few minutes.

I don´t really need an amp, but was looking forward to do some REW measurements to compare my Class A/B Marantz amp to this one and see how load dependency would play out with a pair of Elac DBR62. My main reason for ordering was, that many newcomers are put off by the cost of the amp and I wanted to verify if cheap Class-D amps would be a good recommendation. At 109$ I was on the fence, at 131,99$ I don´t think I´ll buy it just to play around with it. Should have just bought it earlier, I blame no one but myself.

EDIT: I´m a fool. The "new higher price" just includes german VAT of 19%

 

[TheBatsEar]

Minutes ago I was in the process of pulling the trigger on the V3 in the Fosi store. During the checkout process I was not able to finish the PayPal payment because the input screen wouldn´t react, so I refreshed the site and the price went from 109$ to 131,99$. I missed the old price by just a few minutes.

I don´t really need an amp, but was looking forward to doing some REW measurements to compare my Class A/B Marantz amp to this one and see how load dependency would play out with a pair of Elac DBR62. My main reason for ordering was, that many newcomers are put off by the cost of the amp and I wanted to verify if cheap Class-D amps would be a good recommendation. At 109$ I was on the fence, at 131,99$ I don´t think I´ll buy it just to play around with it. Should have just bought it earlier, I blame no one but myself.

Amazon and Aliexpress might still have the old price.

 

[Doenerkunde]

Amazon and Aliexpress might still have the old price.

European Amazon is sold out (and only had the 32V power supply), AliExpress has already updated the price.

 

[LeeRoy]

That's xactly what i am doin too! Xcept for 32V 10A PSU, also from alco..audiophonics. I thought it should be enough. Is your Fosi getting hot with 48V?
My goal is to get rid of Class A F7 clone during summer, that's why the smaller voltage. Also i will be using the xcellent Topping Preamp90.
Was it hard to deploy the Sparkos?

I would say that 32v *10A = 320 Watt and 48V * 7.5A = 350 Watt, only 10% more. Not sure it is worth it. I also read some people saying Amperes are more important than Volts to feed the Fosi V3. I would not change the PSU unless you are not satisfied if were you.

 

[Doodski]

I would say that 32v *10A = 320 Watt and 48V * 7.5A = 350 Watt, only 10% more.

It does not work like that. The 48 V DC power supply will result in more power and more dynamic headroom.

 

[ddaudio]

I would say that 32v *10A = 320 Watt and 48V * 7.5A = 350 Watt, only 10% more. Not sure it is worth it. I also read some people saying Amperes are more important than Volts to feed the Fosi V3. I would not change the PSU unless you are not satisfied if were you.

It's worth reading @amirm 's review of the Aiyima A07 where he discusses the amps vs volts question. As with most things, it depends...

 

[ddaudio]

Hi

First post on this forum, a great resource Amir and all.
I got this amp yesterday with the 32V/5A supply. Currently have a Apple TV 4K -> Wiim mini - > SMSL SU-1 -> Aiyima A07 (32V) -> ELAC DBR62
Music I listen to is streaming Apple Music from the Apple TV in lossless, 24bit/48kHz is fine with me and my buddy Nyquist ; ) and my ears are good with even 256kbps alac for most.
The first thing I discovered is that the Fosi V3 is a lot quieter with the dials set in the same position on both amps, for some reason was expecting the V3 to be louder ? Not sure why I thought that though.
Also I don't hear a real difference in terms of audio quality with my ears, about the same I reckon. Same chip I guess, should I be surprised about this, probably not.
Thus due to lower volume than A07, no dedicated power switch, and not much discernible audio difference for my setup am returning it, and will save for something a little better for later on.

I have the A07, and reading this review and thread, I don't think there will be much difference, unless you need enough power to the point where the A07's protection is shutting it down.

 

[ddaudio]

Minutes ago I was in the process of pulling the trigger on the V3 in the Fosi store. During the checkout process I was not able to finish the PayPal payment because the input screen wouldn´t react, so I refreshed the site and the price went from 109$ to 131,99$. I missed the old price by just a few minutes.

I don´t really need an amp, but was looking forward to do some REW measurements to compare my Class A/B Marantz amp to this one and see how load dependency would play out with a pair of Elac DBR62. My main reason for ordering was, that many newcomers are put off by the cost of the amp and I wanted to verify if cheap Class-D amps would be a good recommendation. At 109$ I was on the fence, at 131,99$ I don´t think I´ll buy it just to play around with it. Should have just bought it earlier, I blame no one but myself.

Looking at @amirm 's measurements of those speakers, that massive impedance peak after 2khz might cause problems for the FR of these class D amps - I haven't seen measurements for this particular TPA3255 amp, but @Archimago has measured the output impedance of the Fosi TB10D and it starts to rise at about the same frequency: http://archimago.blogspot.com/2023/04/measurements-fosi-audio-tb10d-upgraded.html

I have the similar Aiyima A07 amp, and I posted very rough and ready REW graphs comparing it to a Marantz class A/B amp in that amp's review thread, here (ignore the absolute values, I was using a not fit for purpose cheap microphone, but the comparison is hopefully still valid): https://www.audiosciencereview.com/...7-tpa3255-review-amplifier.18984/post-1605370 . I think this shows no FR problem in my case, but as you can see if you scroll up a couple of posts from that link, my speakers' impedance is much flatter than the DBR62 from 1 khz upwards.

 

[howard416]

I can't reconcile the measurements of this amp with those for the PA3s.

Looking at this one: https://www.audiosciencereview.com/...er-4-ohm-vs-frequency-measurement-png.293706/

and comparing to this one: https://www.audiosciencereview.com/...m-vs-frequency-balanced-amplifier-png.156825/

The V3 is clearly the better performer overall. Just to take a single reference point for example, with 10 W output @ 5 kHz, THD+N for V3 is almost -75 dB, while the same for the PA3s is only a bit better than -60 dB.

And yet, when we look at this: https://www.audiosciencereview.com/...fier-budget-multitone-measurement-png.293700/

compared to this one: https://www.audiosciencereview.com/...ents-multitone-balanced-amplifier-png.156829/

How is the PA3s doing a better job with the multitone above ~1.6-1.7 kHz?

 

[dr_mick51]

They are very good too and do not usually have the sopping or hardening effect.

TP-2 (APT2560) | Thermal Compound for All Coolers | Pad Thickness: 1 mm | Pad Dimension: 50x50 mm | ARCTIC

ACTPD00002A | The ARCTIC TP-2 is a perfect bridger of unevenness for an even cooling surface | Easy to use and safe to handle | TP-2 | TP-2 (APT2560) | ARCTIC TP-2 | ARCTIC TP-2 (APT2560) | ARCTIC Store

www.arctic.de

Nice, thermal pads. You can try them and let us know how it goes (in and out)

 

[dr_mick51]

Last time I tested a sparkos opamp in an amplifier, it actually measured worse: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/

Distortion jumped up by 10 dB.

@amirm you measured this amp with 3 differrent opamps, right? The original OPA1612, the Sonic Imagery 990 ENH and the Sparkos SS2590 (which are different to the other Sparkos people buy for this Fosi or Aiyima amps.)

Your measurements of the OPA1612 and the Sonic Imagery 990 were the same? The OPA1612 are not exactly "cheap" but not so expensive and those discrete opamps

 

[mozzarolla]

It is because of the Linear vs Logarithmic potentiometer. I also saw Amazon review of people complaining about this.

This is actually one of my favorite changes vs. my old Fosi. I'm surprised people don't like it. Previously, it felt like the volume would go from almost silent to too loud in a fraction of an inch. Maybe I'm more sensitive to that because I do pretty regularly play at lower volume levels (when I'm working at my desk during the day and my wife is working in the next room). So having the ability to precisely modulate volume while keeping it low is really useful. I guess if I only cranked it loud, that extra room at the lower-volume end of the curve would seem wasteful..

 

[alvigg]

European Amazon is sold out (and only had the 32V power supply), AliExpress has already updated the price.
Frustrating that I had to witness the price increase during the ordering process, but it´s an important lesson for me to not postpone ordering hot items. :-(

Amazon US has 13 left at original price.

 

[Doenerkunde]

Amazon US has 13 left at original price.

Thanks, but there was no raise in price, just german VAT that was added during my ordering process.

 

[amm]

I can't reconcile the measurements of this amp with those for the PA3s.

Looking at this one: https://www.audiosciencereview.com/...er-4-ohm-vs-frequency-measurement-png.293706/

and comparing to this one: https://www.audiosciencereview.com/...m-vs-frequency-balanced-amplifier-png.156825/

The V3 is clearly the better performer overall. Just to take a single reference point for example, with 10 W output @ 5 kHz, THD+N for V3 is almost -75 dB, while the same for the PA3s is only a bit better than -60 dB.

And yet, when we look at this: https://www.audiosciencereview.com/...fier-budget-multitone-measurement-png.293700/

compared to this one: https://www.audiosciencereview.com/...ents-multitone-balanced-amplifier-png.156829/

How is the PA3s doing a better job with the multitone above ~1.6-1.7 kHz?

Hello Howard,

Single tone and multi tone measurements are very different. In multi tone test, nonlinear effects that are not observable in single tone measurement such as mixing of various tones (intermodulation products) are also present and measured. There is no simple correlation between the two tests. Even the excitation is not the same for the two tests. Consider in the single tone case, you have certain sinusoidal excitation that provides an output of say 5 Watts into 4 Ohms. In Multi tone excitation, the total power output for combination of many tones injected to input of the amplifier is 5 Watts, so input power level at each frequency component (tone) is very a different. Hope this helps.

 

[PiobGuy]

Last time I tested a sparkos opamp in an amplifier, it actually measured worse: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/

Distortion jumped up by 10 dB.

While completely accurate (as usual) this is an illustration of how audio, when listening for pleasure, is an individual sound preference. Just like with valve amps, despite the objectively poorer performance on paper, some people just prefer that sound profile. Numbers can be "better" or "worse", tonal preference is just that, a preference.