Yamaha HS5 Powered Monitor Review

[daverosenthal]

OK, I ordered it. At some point you all need to chip in to buy these as I am not too fond of just having hot dogs and chips for dinners.

Maybe I missed it, but is there an easy way to make a one-time contribution, or is the only option to setup $/mo on Patreon?

 

[amirm]

Maybe I missed it, but is there an easy way to make a one-time contribution, or is the only option to setup $/mo on Patreon?

Click on the donate button up on top and use the forum upgrade. That is a one-time option. You can also start a conversation with me and I will give you my paypal address. Thanks in advance.

 

[JustIntonation]

View attachment 46456

Thankfully the distortion is low where our hearing is astonishingly more sensitive (2 to 5 kHz).

Thank you for the review and measurements!
However, I don't think it works like that where you can say distortion is low where our hearing is more sensitive. Shouldn't that be related to the distortion products?
So for the 2nd harmonic we want it to be low from 1kHz to 2.5kHz (as their second harmonic distortion products falls on 2k to 5k) and for the 3rd harmonic distortion we want it to be low from 666Hz to 1.66kHz as the 3rd harmonic distortion product of those frequencies falls on 2kHz to 5kHz?
And we don't see low harmonic distortion products at those frequencies but rather high HD.

As for future measurements, I'd personally love to see 5th harmonic distortion products as well. To an experienced eye in driver selection this does give good additional information.

 

[MZKM]

I am looking for some smaller desk monitors and would love to see the performance these offer.

iLoud MTM for really small, the slightly larger MTM version even has ARC correction (better than the Neumann KH80 which requires an iPad, no desktop or Android/iPhone app).

 

[HammerSandwich]

Should I return the HS5 to guitar center and get the HS7 to test? Or do we already know what that one does?

Of course you should return it. "Honestly, my results make me question if Yamaha even measured this in a Klippel Nearfield Scanner! Now that I have one in my garage, I certainly expect no less from a legitimate audio company! Now, do you have anything with proper baffle-step compensation & directivity indices?"

Bonus points if you have someone casually record the conversation.

 

[March Audio]

This is a review and detailed measurements of the Yamaha HS5 Powered Monitor (speaker). I purchased it this afternoon for US $200 which is the price that it goes for everywhere. This is for a single unit by the way.

The industrial design of the HS5 is fantastic:

View attachment 46441​

Get the larger ones and I am sure you can charge another $10/hour for your mix.

Fit and finish seems nice with a solid single enclosure encompassing the baffle (front and surround).

Here is a shot of the back:

View attachment 46442​

I have not read the manual but while I was setting it up, turned itself off, ruining my first measurement. Don't know why there is no auto-on-off switch anymore on monitors.

Note that there is no RCA jack. For my listening tests I just used a TRS to RCA and it worked fine. The acoustic measurements were made by driving the XLR connection from the Klippel KA3 analyzer balanced output.

Temperature during the test was a "balmy" 57 degrees F. Testing is 10 foot above sea level.

As noted, the unit is brand new. I thought of breaking it first but figure it would then not make any sound so decided against it.

Speaker Acoustic Measurements
Measurements were performed using Klippel Near-field Scanner which eliminates the effects of reflections and noise in my measurement room. So the results are comparable to "anachoic chamber."

As usual, we start with our CEA/CTA-2034 spinorama graph:

View attachment 46443

The Klippen NFS nicely gets rid of room modes and gives us ruler flat bass response all the way to nearly 400 Hz. But there is some unfortunate peaking that lasts quite a while until 2 kHz or so. There is some unevenness above that as well.

Fortunately our early window directivity (dashed blue) shows a rather smooth curve so some amount of EQ may be effective in killing some of that peaking.

Taking into account reflections in a "typical room" we get a predicted in-room response of:

View attachment 46444

We yet again see the excess energy in mid-frequencies. But also a graph that averages to a straight line. Ideally we see one that is sloping down. Otherwise the speaker is going to sound "bright."

Story is told then. On-axis is not flat as it should be. And we have too much high frequency energy. If you are doing your mix using this with no EQ, you will be creating a dull mix with mid-range sucked out.

For advanced readers, we have more measurements.

Advanced Speaker Measurements
A member post this little graph saying Harman has measured this speaker and showed this for its spinorama:
View attachment 46446

We have excellent agreement until we get to > 10 kHz. It is possible our microphones differ a bit in that region or the speaker samples are different. The important aspects are identical though from flat bass to peak around 1 kHz.

Horizontal reflections are not that bad:

View attachment 46447

Vertical though as is often the case is much rougher;
View attachment 46448

If you need more absorption in your room floor and ceiling would make good candidates as noted.

View attachment 46450

View attachment 46451

If you are a fan of the step function that stereophile published, today is your happy day:

View attachment 46452

Finally the pretty contour shots:
View attachment 46453

View attachment 46454


Distortion Measurements
Got tired of you all asking for this so here is the first draft:

View attachment 46455

The top-graph shows the in-room (in-lab) measurement including reflections. We see that the peaking mid-range is still there.

My measurement mic does not have much travel so I have standardized on 1/3 meter distance from tweeter. To compute 90 dB at 2 meter SPL, I shoot for 106 dB given the shorter distance.

We see that the distortion products are all from the woofer. Shown as a percentage we see this more clearly:

View attachment 46456

Thankfully the distortion is low where our hearing is astonishingly more sensitive (2 to 5 kHz).

During the sweeps, the port, pardon my language, farted like nobody's business. It didn't seem like port noise but some kind of nasty high rate brrrrrrrrrrr noise. Stuffed some cotton in the port but then I could hear it from the front. This at the start of the sweep which makes me think is responsible for those large peaks below 100 Hz. It would have been worse if the port had been in the front. You sure as heck don't want to experience that "fart" facing you....

I was going to post the CSD waterfall but I cannot make any sense out of it so I am not.

Informal Listening Test
Told my wife to turn off the TV sound so I could get ready to listen to the speakers, not realizing it was in the middle of something she was doing. She shut down the TV but I sensed this was not the time to ask her to participate in another listening test. So this is me alone, sighted and all. If you don't like it, skip to conclusions.

I level matched the JBL LSR 305P Mark II against the Yamaha HS5 although it didn't seem to be necessary. The Yamaha HS5 came closest to give the JBL a competition from the few I have tested. It had a clean and rather flat sound. Alas, it has no bass or low end. The JBL sounded like it had a little subwoofer in it by comparison. In that regard, it may be considered a bit "boomy" but I am saying that with the slightest hint of being so relative to Yamaha. If I were to give the JBL a score of 8, then Yamaha would get 5 or 6.

Conclusions
It is a relief to not find the Yamaha HS5 a piece of garbage. Could not imagine 2000 people who bought it on Amazon descending on me to complain. It is a "good" speaker. Just not ideal or excellent as it should be. The excess mid-range and lack of any oomph in 200 to 400 Hz makes it very bass shy sounding.

Yamaha nails the looks though for such a budget product. Overall, the HS5 powered monitor is "fine."

P.S. Text files are enclosed for some of the measurements.

------------
As always, questions, comments, recommendations, etc. are welcome.

Probably spent near $1,000 buying speakers to test and so I am deeply "broke." Don't even know what I am going to have for lunch tomorrow. Don't let me stay in this funk and donate what you can using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

@amirm loving the speaker reviews.

I was just about to ask for a step function to be included in the reviews so we can see driver alignment but you bet me to it

Could you expand the x scale a bit more in future? Maybe - 1ms to 4ms.

Thanks great work

 

[Mnyb]

OK, I ordered it. At some point you all need to chip in to buy these as I am not too fond of just having hot dogs and chips for dinners.

Would not buying speakers in pairs make them possible to sell used ? you buy one single speaker for the test and your pretty much stuck with it

 

[HammerSandwich]

Look carefully at the high frequencies : the H2 curve stops at 10 kHz because the second harmonic ceased to be measured once it reached 20 kHz, and the H3 curve stops at 6666 Hz for the same reason : 6666 x 3 = 20 kHz, where the measurement stopped.

That's how I read the graphs before posting earlier AND again while reviewing.

@amirm, this seems confusing to several of us. I understand your earlier post, but the plot specifically shows H2 & H3... Any further clarity would be greatly appreciated.

 

[daverosenthal]

Click on the donate button up on top and use the forum upgrade. That is a one-time option. You can also start a conversation with me and I will give you my paypal address. Thanks in advance.

Don't know how I missed that. Donation incoming; keep up the good work. Small price to pay to help benefit the many people who are going to save $$$ and end up happy with reasonably-priced speakers in the coming years!

 

[amirm]

Would not buying speakers in pairs make them possible to sell used ? you buy one single speaker for the test and your pretty much stuck with it

They sell them one at a time anyway. So I am assuming if someone is interested, they can buy the other one on their own.

 

[MZKM]

That's how I read the graphs before posting earlier AND again while reviewing.

@amirm, this seems confusing to several of us. I understand your earlier post, but the plot specifically shows H2 & H3... Any further clarity would be greatly appreciated.

The plot shows total harmonic distortion, as well as the second and third harmonic distortions, which are part of the total.

The test is showing resulting distortion landing at that frequency, if there is a rise in H2 at 1.5kHz, that means it’s the result of 750Hz distorting (with second harmonics). If there is a rise in H3 at 1.5kHz, that means 500Hz is distorting (with third harmonics).

It is not showing which frequencies have H2, H3, etc.; it’s showing the result of other frequenting causing H2, H3, etc.

 

[MZKM]

Click on the donate button up on top and use the forum upgrade. That is a one-time option. You can also start a conversation with me and I will give you my paypal address. Thanks in advance.

How much do you have to pay to get the Donor badge? I donated $25 on the 8th and feel left out.

 

[JustIntonation]

The plot shows total harmonic distortion, as well as the second and third harmonic distortions, which are part of the total.

The test is showing resulting distortion landing at that frequency, if there is a rise in H2 at 1.5kHz, that means it’s the result of 750Hz distorting (with second harmonics). If there is a rise in H3 at 1.5kHz, that means 500Hz is distorting (with third harmonics).

It is not showing which frequencies have H2, H3, etc.; it’s showing the result of other frequenting causing H2, H3, etc.

No that's not how HD plots work. If there's a rise in H2 at 1.5kHz that means that 1.5kHz is generating a H2 of 3kHz. If there's a rise in H3 at 1.5kHz that means that 1.5kHz is generating (distorting) a H3 of 4.5kHz.

 

[MZKM]

No that's not how HD plots work. If there's a rise in H2 at 1.5kHz that means that 1.5kHz is generating a H2 of 3kHz. If there's a rise in H3 at 1.5kHz that means that 1.5kHz is generating (distorting) a H3 of 4.5kHz.

Which means that the distorsion peak measured at 1500 Hz is heard at 2x1500 = 3000 Hz for H2, and 3x1500 = 4500 Hz for H3, which is, on the contrary, exactly where our hearing is most sensitive !

you see the sum total of distortions existing at that frequency. It is not the THD+N measurement of distortion at that single frequency tone.

 

[JustIntonation]

Oh really? I missed that one! Thanks.
I like it It's how all HD plots should be!
Edit: Had another look at the HD plots and it sure doesn't look like you're correct. I totally looks like a normal HD plot? Look how the HD drops off before 2kHz due to the crossover to the less distorting tweeter for instance..
Scratching my head here as far as Amir's comment quoted above.

 

[JustIntonation]

This is not that kind of test (sweep of frequency versus distortion). Indeed there is no distortion measurement per se! A log swept sine (so called Farina sweep after professor Farina who invented it) is used to excite all frequencies from 20 to 20 kHz. When converted to time domain, it becomes an impulse response which is what is normally used to measure speakers/excite all frequencies. A neat thing happens in that you get a second impulse at negative time that represents distortion. Apply an FFT to it and you get the distortion present at all frequencies.

Put another way, at any point on the distortion graph that I showed, you see the sum total of distortions existing at that frequency. It is not the THD+N measurement of distortion at that single frequency tone.

Are you very sure this is correct? The distortion plots you posted look an awful lot like normal HD plots where a rise at for instance 1.5kHz in H2 means the actual H2 distortion product falls at 3kHz and a rise in H3 at 1.5kHz means the actual H3 distortion product falls at 4.5kHz.
The HD plots just really looks like that, including what you would expect with the drop in HD below 2kHz due to the crossover to a less distorting tweeter..
And including the 20kHz cutoff! Why otherwise would the H2 plot stop at 10kHz and the H3 plot stop at 6.667kHz?? This is really a normal HD plot, how else can it be?

 

[amirm]

But after that, the frequencies are shifted so that they are properly displayed in the graph.
Look carefully at the high frequencies : the H2 curve stops at 10 kHz because the second harmonic ceased to be measured once it reached 20 kHz, and the H3 curve stops at 6666 Hz for the same reason : 6666 x 3 = 20 kHz, where the measurement stopped.

As I explained there is no specific measurements of harmonics. Each of the falls out of the special sweep signal at a different time with their own impulse. Here is what it looks like:

The graph then simply shows the signal in red on the left as the "fundamental" (converted into frequency domain) and then one line for each of those follow on impulses. At any single point then on frequency graph, we are seeing the decomposition of the fundamental and what harmonic distortion exists at that frequency.

If you zoom into the above you get this:

By definition, there cannot be any second harmonic at 10 kHz and beyond. So the math generates that result. Ditto for third and fourth harmonic.

So what you are observing is the simple meaning of harmonics of signals show in a limited spectrum, not that there is some other logic as explained. We simply have a single separated into its subcomponents: fundamentals and harmonic distortions at each frequency.

 

[amirm]

How much do you have to pay to get the Donor badge? I donated $25 on the 8th and feel left out.

Oops. Somehow I missed that. The system software relies on a callback from Paypal saying payment was made. Well, occasionally Paypal would produce these obscure messages saying it couldn't provide the notifications. After a bit, it stopped answering altogether. So now I have to catch the emails and manually upgrade titles. Sometimes I miss one or the other. Sorry about that. Fixed. Much appreciate the donation.

 

[JohnYang1997]

Pls measure hs7. I have been using one for 2 years. It's nice and actually sounds better than hs8 which measures flatter. I would score hs5 5, hs7 8, 305 7, hs8 7. 305 is a bit rough and hard sounding. hs8 is smooth but seems mids are a bit recessed(too much difference between drivers) in perception.

 

[JustIntonation]

As I explained there is no specific measurements of harmonics. Each of the falls out of the special sweep signal at a different time with their own impulse. Here is what it looks like:
View attachment 46600

The graph then simply shows the signal in red on the left as the "fundamental" (converted into frequency domain) and then one line for each of those follow on impulses. At any single point then on frequency graph, we are seeing the decomposition of the fundamental and what harmonic distortion exists at that frequency.

If you zoom into the above you get this:
View attachment 46602

By definition, there cannot be any second harmonic at 10 kHz and beyond. So the math generates that result. Ditto for third and fourth harmonic.

So what you are observing is the simple meaning of harmonics of signals show in a limited spectrum, not that there is some other logic as explained. We simply have a single separated into its subcomponents: fundamentals and harmonic distortions at each frequency.

Thank you very much for explaining. But I simply do not get why this would generate anything else than a normal HD plot, even though the method of obtaining it is different.
The plots themselve look like normal HD plots, I would expect different results if the plots would show the actual frequencies the harmonics fall on. And I also don't understand why there cannot be any harmonic above 10kHz for the second and above 6.667kHz for the third. Of couse there can be? (and there are) Only a standard limitation of 20kHz of measurement will make the H2 stop at 10kHz and H3 at 6.667kHz (that is, the generating freq, the harmonics themselves are measured up to 20kHz).