Fosi Audio ZA3 Amplifier Review

[Talisman]

Im not speaking in RMS values. With an amp like the ZA3 and many other class D amps you have to factor in that the rated RMS power is going to be near the dynamic or peak power.

You don't need to convince me, I've always been of the opinion that having more power for dynamic peaks is a good idea, simply asserting that the LS 50s should be powered with no less than 200watts per channel seems like an exaggeration to me.
At 3m with a sensitivity of 85db the ls 50 can play at over 100db, but at that volume they are literally falling apart with dynamics and distortion and you really don't want to hear them sound that way, the solution with such small speakers is always to pair it with a sub so as not to overload them with sub 100hz frequencies which make it very difficult to reproduce with low distortion, but in that case the need for all that power also disappears.

 

[nvidia_7]

I’m running a subwoofer as well, SVS SB-3000. Part of why I am using the Gustard H16 as a preamp, since I have the sub hooked up to the single ended outputs on that, and the ZA3s to the balanced outs. I find the bass is tighter this way than running it out from the sub outs on the ZA3s.

-Ed

You need to be careful with that Gustard H16 and the ZA3. I just looked up the measurements and the preout voltage is 3.8v with the medium gain settings. With the low gain setting you decrease it by -6dB giving you 1.9v which is still too high for the ZA3 if you ever were to turn the Gustard up to max volume while the ZA3 gain knob was also at max volume. With the ZA3 in stereo mode (26dB gain), the gain knob at max, and your Gustard at max volume in low gain mode, you would be asking the ZA3 to output 37.9v which at 4ohms is 360w. Obviously the ZA3 cant do that.

If I was you I would keep the Gustard in low gain mode, keep the ZA3 in stereo with the gain knob at max (26dB), and never go full volume on the Gustard.

 

[EddNog]

…If I was you I would keep the Gustard in low gain mode, keep the ZA3 in stereo with the gain knob at max (26dB), and never go full volume on the Gustard.

I’ve literally been running it this way other than having the monoblocs on Mono up until today. I didn’t like having the gain anything other than Low because I prefer having granularity with the volume control, especially given I don’t listen at high levels.

Given I don’t listen at high levels, maxing out the H16’s volume is also highly unlikely to ever happen.

-Ed

 

[nvidia_7]

You don't need to convince me, I've always been of the opinion that having more power for dynamic peaks is a good idea, simply asserting that the LS 50s should be powered with no less than 200watts per channel seems like an exaggeration to me.
At 3m with a sensitivity of 85db the ls 50 can play at over 100db, but at that volume they are literally falling apart with dynamics and distortion and you really don't want to hear them sound that way, the solution with such small speakers is always to pair it with a sub so as not to overload them with sub 100hz frequencies which make it very difficult to reproduce with low distortion, but in that case the need for all that power also disappears.

Im going from my own personal experience. I have heard the LS50's in different settings and often they sounded flat and dull to me. It wasnt until I heard them with plenty of power on tap that they had a dynamic "wow" sound to them. I think my experience mirrors what I see on forums and reviews as well. There seems to be a large majority of people who claim the LS50's are dull and lifeless, while another group say they are dynamic and lively. I can guarantee you the difference in opinion mainly rests with what is powering them. Speakers with low sensitivity are tricky. Similar to the SP-EBS73's from Pioneer. You can run the numbers all day and say "you only need x watts to get x dB at 3m" but in practice it never seems to pan out that way.

 

[Doodski]

Im going from my own personal experience. I have heard the LS50's in different settings and often they sounded flat and dull to me. It wasnt until I heard them with plenty of power on tap that they had a dynamic "wow" sound to them. I think my experience mirrors what I see on forums and reviews as well. There seems to be a large majority of people who claim the LS50's are dull and lifeless, while another group say they are dynamic and lively. I can guarantee you the difference in opinion mainly rests with what is powering them.

I agree whole heartedly. I found with all KEF C Series, UNI-Q and Reference Series and on up to todays most modern gear that a really high current capability very low Z drive amp brings out the best in them. I also found some of the smaller models excelled sometimes when placed against the wall for more boomy bass if desired. My favorite amp with KEFs was a, K i n e r g e t i c s, K B A 75 power amp that was linear from 75W@8R all the way to 300W@2R. (I used spaces to avoid peeps from afar web searching and finding the model numbers and then nagging me to fix theirs.) It's happened several times. It's annoying..LoL because I don't have a lab anymore.

 

[nvidia_7]

I agree whole heartedly. I found with all KEF C Series, UNI-Q and Reference Series and on up to todays most modern gear that a really high current capability very low Z drive amp brings out the best in them. I also found some of the smaller models excelled sometimes when placed against the wall for more boomy bass if desired. My favorite amp with KEFs was a, K i n e r g e t i c s, K B A 75 power amp that was linear from 75W@8R all the way to 300W@2R. (I used spaces to avoid peeps from afar web searching and finding the model numbers and then nagging me to fix theirs.) It's happened several times. It's annoying..LoL because I don't have a lab anymore.

Are you telling me if I get a good deal on a non-functioning KBA75 I cant send it to you to get repaired?

Jokes aside, thats some serious class A power. I just checked the used value on those. Thats quite a few ZA3's.

 

[Doodski]

Are you telling me if I get a good deal on a non-functioning KBA75 I cant send it to you to get repaired?

No can do... No smoke in my living quarters and I have no lab gear or PCB rework tools etc.

Jokes aside, thats some serious class A power. I just checked the used value on those. Thats quite a few ZA3's.

They run cool too. Very nice unit. Used prices of a working example ~$700 plus are reasonable.

 

[EddNog]

Im going from my own personal experience. I have heard the LS50's in different settings and often they sounded flat and dull to me. It wasnt until I heard them with plenty of power on tap that they had a dynamic "wow" sound to them. I think my experience mirrors what I see on forums and reviews as well. There seems to be a large majority of people who claim the LS50's are dull and lifeless, while another group say they are dynamic and lively. I can guarantee you the difference in opinion mainly rests with what is powering them. Speakers with low sensitivity are tricky. Similar to the SP-EBS73's from Pioneer. You can run the numbers all day and say "you only need x watts to get x dB at 3m" but in practice it never seems to pan out that way.

My setup is quite dynamic and lively, thank you very much.

-Ed

 

[312elements]

Im going from my own personal experience. I have heard the LS50's in different settings and often they sounded flat and dull to me. It wasnt until I heard them with plenty of power on tap that they had a dynamic "wow" sound to them. I think my experience mirrors what I see on forums and reviews as well. There seems to be a large majority of people who claim the LS50's are dull and lifeless, while another group say they are dynamic and lively. I can guarantee you the difference in opinion mainly rests with what is powering them. Speakers with low sensitivity are tricky. Similar to the SP-EBS73's from Pioneer. You can run the numbers all day and say "you only need x watts to get x dB at 3m" but in practice it never seems to pan out that way.

I know what you mean (I think) and @EddNog if it sounds good to you, that’s the only thing that matters as far as I’m concerned. I have a pair that I’ve since moved to my studio and they make for a great conversation piece with my clients. I find them to be a visually attractive speaker, but not my favorite to listen to. The c298 had plenty of power (purifi modules) but even with that they never really became the speakers I hoped they were going to be based on the measurements.

 

[EddNog]

I know what you mean (I think) and @EddNog if it sounds good to you, that’s the only thing that matters as far as I’m concerned. I have a pair that I’ve since moved to my studio and they make for a great conversation piece with my clients. I find them to be a visually attractive speaker, but not my favorite to listen to. The c298 had plenty of power (purifi modules) but even with that they never really became the speakers I hoped they were going to be based on the measurements.

Probably best for a separate thread as it’s off topic, but these are an upgrade for me from an old pair of JM Lab Electra 926 floor standers. I much prefer the sound of the LS50 Meta & SB-3000 combo, but a lot of the reason for me is that this setup is far easier for me to adjust for the room. I couldn’t get the Electras to sound right in my room, but the combination of being able to place the subwoofer separately, and adjust the bass end of things with the DSP really worked out perfectly for my listening space in ways I could never pull off with the big Electras.

You can always change your components, but you can’t always choose or change your listening space.

-Ed

 

[sejarzo]

In fact this is the case, see the smsl A300 review, where in PBTL the load dependence performance deteriorates drastically.
https://www.audiosciencereview.com/forum/index.php?threads/smsl-a300-stereo-amplifier-review.46742/

However this amplifier does not seem to use true PBTL but rather excludes one channel to provide all the power to just one, so it may not be a problem in this case

The way I read the A300 review, it operates in BTL, not PBTL, when placed in mono.

 

[GM3]

Erin's corner has a very good review of the amp. It may explain why some users of digital amps have reported experiencing different results to what you would expect after just seeing one set of measurements/review.

 

[312elements]

@nvidia_7 in the above linked video it was suggested that the power ratings for this particular amp (and presumably others) can be misleading. I feel like you were alluding to this as well. The way I understood the measurements were that the stated and measured watts would be more accurately described as a peak wattage as opposed to a sustained wattage.

Annoying to see this guy advocating for a high pass filter. He really came off ignorant regarding that.

 

[staticV3]

Annoying to see this guy advocating for a high pass filter. He really came off ignorant regarding that.

Please elaborate.

 

[312elements]

Please elaborate.

High passing speakers should be done in the digital domain.

It can certainly be done in the analog domain and can be done well and there are solutions already available for that. Those solutions costs more than this amplifier.

The corners that will be cut to affordably accomplish this goal will negatively impact sound quality.

If there really is a ton of demand for a cheap analog high pass filter then they should bring a new product to market instead of crippling an otherwise decent amplifier.

 

[Talisman]

If there really is a ton of demand for a cheap analog high pass filter then they should bring a new product to market instead of crippling an otherwise decent amplifier.

This is a discussion that we had already addressed, and to which I had already responded, exactly as for the "pure direct" button which activates a relay that changes the signal path without degrading the quality, in the same way a pure direct signal can be implemented and activate a secondary circuit with crossover and high pass filter with a relay. There is no need to paralyze anything

 

[hmt]

Erin's corner has a very good review of the amp. It may explain why some users of digital amps have reported experiencing different results to what you would expect after just seeing one set of measurements/review.

Doesnt look so good. It almost seems that it was optimised to look good on the ASR measurument protocol. No 180W "Hifi watts" delivered here.

 

[nvidia_7]

@nvidia_7 in the above linked video it was suggested that the power ratings for this particular amp (and presumably others) can be misleading. I feel like you were alluding to this as well. The way I understood the measurements were that the stated and measured watts would be more accurately described as a peak wattage as opposed to a sustained wattage.

Annoying to see this guy advocating for a high pass filter. He really came off ignorant regarding that.

Yes, what he stumbled upon without realizing was class D's tendency to rate power at extremely short bursts. Take a look at the data sheets below. The first being the Hypex NC500. Its rated at 700w at 4ohm and is quoted as such from vendors like Buckeye amps. But wait, the real, continuous power rated is only 100w (4Ohm load 1200W SMPS). And notice the requirements for "sufficient cooling". Most amp makers using these boards are just slapping them into thin aluminum enclosures with no fans, no heatsinks. Thats a far way off from the advertised 700w. Now look at an amp from Icepower, rated at 2000 watts at 4Ohms. Except its not. The data sheet shows it can do its 2000 watts at 4Ohms for just 20ms, not seconds, milliseconds. The rated continuous power is 290 watts. A far cry from the advertised 2000 watts.

I don't remember off the top of my head the exact duration the Hypex can do its advertised 700w, but if im remembering correctly it was something like 3ms (milliseconds). Now depending on frequency, that may or not be enough cycles to reproduce the rated power when needed in audio reproduction. Amplification specs are still rather unstandardized and easily manipulated. You really have to dig into the data and sift through the fluff.

 

[312elements]

Yes, what he stumbled upon without realizing was class D's tendency to rate power at extremely short bursts. Take a look at the data sheets below. The first being the Hypex NC500. Its rated at 700w at 4ohm and is quoted as such from vendors like Buckeye amps. But wait, the real, continuous power rated is only 100w (4Ohm load 1200W SMPS). And notice the requirements for "sufficient cooling". Most amp makers using these boards are just slapping them into thin aluminum enclosures with no fans, no heatsinks. Thats a far way off from the advertised 700w. Now look at an amp from Icepower, rated at 2000 watts at 4Ohms. Except its not. The data sheet shows it can do its 2000 watts at 4Ohms for just 20ms, not seconds, milliseconds. The rated continuous power is 290 watts. A far cry from the advertised 2000 watts.

I don't remember off the top of my head the exact duration the Hypex can do its advertised 700w, but if im remembering correctly it was something like 3ms (milliseconds). Now depending on frequency, that may or not be enough cycles to reproduce the rated power when needed in audio reproduction. Amplification specs are still rather unstandardized and easily manipulated. You really have to dig into the data and sift through the fluff.

I’m unsure as to how much this matters in the real world and more importantly how to test it. It would seem that if you were to record the output of one of these amps at a high enough level and do the same with a level matched class AB amp like the benchmark for example the wav files would be more extended (at times) in the AB amp. Would that be your expected outcome as well? This is likely one of those things that doesn’t matter until it does? Likely not noticeable until you were pushing the amp to its limits? Does the amp clip when it’s called for more power than it has to give out or does it gracefully just give it all it can?

 

[phoenixdogfan]

High passing speakers should be done in the digital domain.

It can certainly be done in the analog domain and can be done well and there are solutions already available for that. Those solutions costs more than this amplifier.

The corners that will be cut to affordably accomplish this goal will negatively impact sound quality.

If there really is a ton of demand for a cheap analog high pass filter then they should bring a new product to market instead of crippling an otherwise decent amplifier.

The the amp should not include a low pass filter either. If there is a high pass before the amp, the amp itself will never see the lows to pass them on to the sub, so a low pass without some sort of high pass makes no sense either. My OG SB 2000's included as low pass, and it was workable and listenable. I used it for a full year before I got a miniDSP 2 x 4 HD, and it was very listenable with my LS 50's. And it was analog on both legs.