Fosi Audio V3 Amplifier Review

[somebodyelse]

I think one of the tests of TPA3255 based amps with a simulated speaker load has a similar ~20R at 2kHz giving a peak between 0.5dB and 1dB. That might be audible, but probably not massively so. I haven't sound any frequency response measurements for the Q950 so don't know whether they have a peak or trough there - the boost could make a peak worse, or a trough better.

 

[TimeGoddess]

I think one of the tests of TPA3255 based amps with a simulated speaker load has a similar ~20R at 2kHz giving a peak between 0.5dB and 1dB. That might be audible, but probably not massively so. I haven't sound any frequency response measurements for the Q950 so don't know whether they have a peak or trough there - the boost could make a peak worse, or a trough better.

Would you be able to link the testing?

And the q950 are relatively flat from 1khz to 20khz at the 15 degree listening angle I use. Just a 1db dip at 2khz and a 1db peak at 6khz

 

[somebodyelse]

Would you be able to link the testing?

See Erin's tests of the za3 or Wiim Amp with a couple of dummy speaker loads with ~20R peaks. I don't know the impedance of the dummy load Stereophile use, but they had something similar with their Wiim Amp tests. These all have peaking at higher frequencies too, but I think the Q950 will miss that as it doesn't have the higher frequency impedance peaks of Erin's load, and drops to ~5R impedance towards 20kHz.

I think @pma did a similar test with a dummy speaker load and the A07 but I can't find it.

 

[bgravato]

I want to use this amplifier with my Kef Q950 speakers. But what I'm worried about is the fact that this amp is not load independent. And the Kef q950 has impedance swings up to 20ohm. Would that make this amp a non starter for these speakers?

If that concerns you perhaps consider an amp with a PFFB design.

I think the new Fosi V3 mono blocks implement that (not sure if it's already available for purchase and it will be a more expensive solution since you'll need two of them, as well as a preamp for volume control if you don't have already another way of adjusting the volume).

Another alternative might be the Topping PA5 II (a bit more expensive too) or the new Wiim Amp Pro (quite a bit more expensive, but if you need a streamer it might be worth it, not sure if it's already available for purchase).

There are probably more amps with PFFB design that are affordable, but those are the ones that come to mind.

 

[ninetylol]

I wouldnt sweat it too much, worst thing will be 1db more high frequencies with your speaker while you probably cant even hear a 3db difference in that frequency range.

 

[-Jim-]

I wouldnt sweat it too much, worst thing will be 1db more high frequencies with your speaker while you probably cant even hear a 3db difference in that frequency range.

Especially if the listener is male and over 50!

 

[Count Dacula]

Would 48v 3A be a hair better than 32v 5A?

 

[theronic]

What is the efficiency of the Fosi Audio V3 (48V)? I understand the TPA3255 should be 90% efficient under load, and according to @amirm's measurements, the 48V PSU is 70% efficient under load, or does that include both the amplifier & 48V PSU?

 

[somebodyelse]

Would 48v 3A be a hair better than 32v 5A?

It depends on the application - speaker impedance, power requirement etc. The amp will run cooler at 32V so if you don't need the extra voltage to get the power you need, particularly with 8 ohm speakers, then 32V may be the better option.

 

[Count Dacula]

Thanks. I did the raw math, but knew that wouldn't be the whole story.

 

[NTK]

Would 48v 3A be a hair better than 32v 5A?

[Edit] Forgot that the power supply needs to supply power to both channels. So the per channel max output with both channels driven is ~75 W. Entire post rewritten.

Here are my max power output estimation based on TI's TPA 3255 datasheet.

For simplicity, I assumed the amp is 100% efficient and PS can output total 150 W max for both channels. Efficiency in reality will be lower (and dependent on load impedance), and PS short term output may be higher than 150 W. So there will be some +/- errors in the estimations. I am also using the 1% THD chart instead of the 10%.

With 48 V PS, the amp output is limited by the max PS output power. The PS cannot supply enough current for the amp to reach output voltage limit for 4, 6, and 8 Ω loads. Max power would be ~75 W for 4, 6, and 8 Ω per channel, both channel driven.

With 32 V PS, the amp output is limited by the max PS power output to ~75 W per channel for 6 and 4 Ω. The amp power output would be ~55 W max per channel at 8 Ω, limited by the PS supply voltage.

 

[antcollinet]

With 48 V PS, the amp output is limited by the max PS output power. The PS cannot supply enough current for the amp to reach output voltage limit for 4, 6, and 8 Ω loads. Max power would be ~75 W for 4, 6, and 8 Ω per channel, both channel driven.

You are assuming the amp can use the full PSU power regardless of speaker impdedance. This is not the case. The maximum current limits the power to I^2*R. With only 3A, that will be:
4ohm = 36W
6ohm = 54W
8ohm = 72W


Much lower than this if both channels are driven. With only 3A of current, there is only 1.5A available for each channel. This puts a limit on power of speaker impedance x 1.5^2:

4ohm - 9W
6ohm - 13.5W
8ohm - 18W

I would not use any 3A power supply.

 

[NTK]

You are assuming the amp can use the full PSU power regardless of speaker impdedance. This is not the case. The maximum current limits the power to I^2*R. With only 3A, that will be:
4ohm = 36W
6ohm = 54W
8ohm = 72W


Much lower than this if both channels are driven. With only 3A of current, there is only 1.5A available for each channel. This puts a limit on power of speaker impedance x 1.5^2:

4ohm - 9W
6ohm - 13.5W
8ohm - 18W

I would not use any 3A power supply.

It is only true for class A/AB/B designs, but not for class D. (Reference: https://www.irf.com/product-info/audio/classdtutorial2.pdf)

 

[antcollinet]

It is only true for class A/AB/B designs, but not for class D. (Reference: https://www.irf.com/product-info/audio/classdtutorial2.pdf)

View attachment 396464

Oh, yes. Similar to a switching buck voltage regulator. Energy storage and release in the output inductor. I should have known this. Thanks.

 

[Spiky]

How bad is it if you have 3 Fosi V3s on top of each other? There is a 48 V power supply for all of them, so you can feel warm after a movie. Thanks!

 

[ninetylol]

How bad is it if you have 3 Fosi V3s on top of each other? There is a 48 V power supply for all of them, so you can feel warm after a movie. Thanks!

heat is never good for electronics. you will see how bad, when it fails someday in the future.

 

[Spiky]

heat is never good for electronics. you will see how bad, when it fails someday in the future.

Yes, that's what I thought. Then they will stay nicely next to each other, as they are now. Thanks for the confirmation

 

[VientoB]

If Fosi make a V3 mark 2 there are a couple of improvements I would like:

1) Two sets of inputs. One should be XLR.
2) Reduce the high frequency variability with impedance
3) add a lower gain setting - my volume knob never goes above halfway and is usually between 10 o'clock and 11 o'clock.

Then it would be perfect.

 

[elshaddai]

If Fosi make a V3 mark 2 there are a couple of improvements I would like:

1) Two sets of inputs. One should be XLR.
2) Reduce the high frequency variability with impedance
3) add a lower gain setting - my volume knob never goes above halfway and is usually between 10 o'clock and 11 o'clock.

Then it would be perfect.

I would add an RCA sub out and an optional, adjustable High Pass Filter on the main speaker outputs.

 

[Matt1974]

I obtained this for the princely sum of £72 including taxes and delivery, with the larger PSU, and I think it’s my favourite thing I’ve ever bought. All thanks to this site and the enthusiasts contributing to it!

These new class Ds have incredible treble, extended, smooth and detailed.

Most problems I thought I had with this amp have eventually been chased down as being due to another component. I have a tendency to blame the amp due to its extreme cheapness.

I hesitate to talk about burn in but it does seem to be better after a couple of months ordinary listening. Of course during that time I have been fussing with it, tweaking speaker position etc.

I think it’s more dynamic when the volume knob is around 75%, with volume reduced from the source. However, once you think you’ve identified some characteristic like that you are subject to confirmation bias. I think it’s flatter, as well as quieter, when the knob is lower. I’m not in fact sure exactly what the knob controls.