How to Solve High-Frequency Suckout in Room?

I wish you had one or two of the Stereophile Test CDs where you could play the frequency sweeps. 

They are expertly crafted and you would expect identical everything from both channels other than from room impacts.  

Sounds like you are pretty much satisfied that you have found the issue.   Enjoy your beautiful system.

My integrated can't play in "mono mode" as far as I can tell.  Will try to search for "mono" material to stream; good idea.  Will try that and the mono test tones / frequency sweep tonight.

Pretty sure I have found the culprit by now - I can't fathom what other cause there could be such that the issue exists when playing through a CD player (through an RCA analog input to my amp and not through the amp's DAC module), and follows the right channel output of the CD player.

For me, I think the ultimate test is to try the "worst" example of the problem as a CD, and see if the problem clearly follows then CD player's right analog output when connected to the left analog input of my amp. If that happens, which I fully expect it will, case closed...

This is such an obvious and good suggestion I feel embarrassed to have not thought of it 🙄   If I am not mistaken, you are using an internal DAC, so alternately, what about finding some mono source material?  Should be easy to find something mono to stream. You can probably even find mono test tracks with a frequency sweep. If the image moves as the frequency changes, you have your culprit.

Does your integrated have the ability to play in mono to verity your findings, i.e., that the difference in the left and right speaker sound is from differences in the source material left versus right stereo track recording?

Does your integrated have the ability to play in mono to verity your findings, i.e., that the difference in the left and right speaker sound is from differences in the source material left versus right stereo track recording?

No need to apologize @theaudiomaniac; just glad for some help.

So I only got to #2 on my list of tests tonight (testing with my 25 year old Arcam CD player as the input instead of the internal DAC) but I may have made a breakthrough of sorts.

First thing is it was surprisingly nostalgic hearing the sound I used to hear when I played music - it sounded terrible, but there was a charm to it. Still enjoyed the sound it made. It doesn’t even sound like you are listening to the same version of the music when playing the same tracks streamed to the Gryphon DAC.

Second thing totally confused me momentarily. I immediately heard the missing frequencies - but on the left side instead of the right! After a very brief moment of confusion as to how this could be, I realized I had mixed up the interconnect inputs to the amp between left and right. I then realized, wait....it is the source material causing the issue! It’s the only explanation as to why the problem would exist on the CD player as well, and that the problem would follow whatever the right channel output of the CD player was plugged into. At this point I’m 95% confident that all of my obsessing over this issue was due to a small number of tracks with the slightest flaws in the recorded stereo image of the vocals, combined with the only aspect of my new speakers which I’m not a fan of: a laser-etched sweet spot, regardless of toe-in and setup. My theory is that these particular speakers are more prone than most others I’ve tried to exposing any flaws in recorded stereo images, in the upper vocal range. On most music is not an issue. It’s just a handful of tracks really where I notice it.

In comparison one of the speakers I had auditioned was the Magico A5’s. These speakers in comparison have a ridiculously large sweet spot. In fact it’s almost like there isn’t a sweet spot with those speakers as position doesn’t matter to a large extent. I doubt I’d have noticed such an issue with those speakers. Having said that, I didn’t take to them and I really do like the sound of my speakers with my system, so I think I can live with this "flaw" for the few tracks that have issues with the stereo image.

The one slightly troubling thing still: When I hear this issue, it’s always the same; missing upper mid frequencies on the right.

For reference, three example tracks where I hear the issue are Keep the Car Running by Arcade Fire (the remainder of the songs on that album sound okay), and Cite Soleil and John the Baptiste by The Afghan Whigs (these two tracks expose the issue to the highest degree).

I’ve lost my CD with those Afghan Whigs tracks so I’m going to purchase it again, and put this issue to rest once and for all. I will confirm that even these tracks when played with the CD player, with the interconnects swapped between left and right, produce the issue on the left side and not the right. If that’s the case I’ll be 100% sure my new speakers are just sensitive to stereo image issues in source material.

Yes, I tend to be on the neurotic side of the audiophile spectrum... I’ve completed my upgrades for now so I should be able to settle down and get back to just relaxing and enjoying the music without thinking too hard!

Yes, for tonight I plan to:

1) Try my old Clarus Crimson Biwire cables which I’ve not sold yet. More out of curiosity as they are less resolving than the Valhalla 2’s (but very good still), so I may or may not hear the issue with these cables

2) Resurrect my old Arcam CD player to see if the problem goes away. If it does go away the problem must be the DAC module in my amp. That said it was a $700 CD player that I bought 25 years ago for $500 so again it’s possible the issue may be present but not audible, comparing against a $5,000 DAC module.

3) Resurrect my old Arcam A85 integrated amp and see if the problem is detectable with it. Again, a lower end amp so I may not be able to hear the issue if it exists. Turns out it accepts spades after all (I had thought it needed bananas).

So far, with 1-3 above, if I AM able to detect the issues still with these tests, it would suggest that either my room or hearing ARE in fact to blame. If I can’t hear the issue it doesn’t necessarily mean anything as it may be that this low end gear “hides” the issue by not being as resolving.

4) I will try the voltage comparison between left and right, when playing a 1KHz test tone, this time with my speakers connected. Will also try at 5kHz where the voltage was the most varied (7mV). Not expecting to see any difference playing at a reasonable volume, since I expect the difference to be even smaller at the lower volume of this test, vs the test I did last night at a higher volume. But maybe adding the speaker load will change things.

5) I will try the 1kHz and 5kHz test tones using a frequency analyzer app on my phone, 1m away from the speaker and 1M from the ground, one speaker at a time.

6) Might save this for another night but I will completely reconfigure my system and living room to be in a different orientation in the room.

If after all of this I’m left at square 1, I will either:

1) See if my local dealer will lend me a high end integrated for a day (probably would be Moon or McIntosh), even though I didn’t buy my amp here, or

2) Bring my amp to my local dealer (who again I didn’t purchase the amp from), or

3) Bring my amp to the dealer I bought it from, which is in a different city, to listen to other high end speakers to see if the issue manifests. I actually prefer the first two options as this particular dealer WAY overdampens their rooms to the point it sounds like an anechoic chamber. I sense all sorts of frequency removal happening whenever I test speakers there.

So it's not the room, it's somewhere in your system.  Further swapping and substitution of cables/channels will reveal where it is.  Proceed by using logic and focus on one component at a time.

@theaudiomaniac , I swapped the cables at the amp end (only) and the problem changes from the right side being too soft in the upper mids and into the treble, to the left side being too soft in these frequencies. This leads me to believe it’s not the room.

That said I am planning to completely rotate my system in the room as a test (setting up my system against the side wall.

Update: I have also questioned if the source material is causing the issue since the problem is not apparent on all music, mildly apparent in some music, and blatantly obvious in other tracks. As above, vocals sound centered but the upper end of vocals seem more fleshed out on whichever speaker is NOT being driven by the amp’s right channel.

@theaudiomaniac I think you might be right, in terms of AC voltage output being close enough. That said, there is a 1.2% variance at 5,000Hz, and close to a 1% variance around this frequency. While small, wouldn’t that be audible

That is at best 0.1db. Left to right, your system could be off 3,4,5 or more db. It will not be audible.

I did read an old forum post that indicated that subtle differences may not be detectable using a voltmeter. The ultimate test is to use a dual channel oscilloscope, inverting one of the channels phases, then using the “sum” function of the scope. Because you’ve inverted one of the phases, the channels should cancel out each other when you sum them, and any variance from 0V that the scope shows would indicate an imbalance. I used to have an oscilloscope that hooked up to your PC but it’s so old it won’t work anymore.

Don’t waste your time. A wide bandwidth multimeter is accurate enough. The difference in time between measurements and thermal differences are smaller than what you are measuring. It is your room. Didn't you already say you swapped channels and the issue is still there? That rules out the electronics.

What you need to do is figure out how to measure room response. There are many resources on the web. It will cost you $100-200. If you have a noticeable difference, then then it will be easy to see them. Measure with one speaker, measure with the other, then compare.

@theaudiomaniac I think you might be right, in terms of AC voltage output being close enough. That said, there is a 1.2% variance at 5,000Hz, and close to a 1% variance around this frequency. While small, wouldn’t that be audible

@djones51 , I think the voltage output should be sufficient to compare the differences (without needing to convert to power which should give the same readings. Unless, of course, you are thinking there might be differences in the resistance between left and right and I should measure that too? As I said the problem was evident to a lesser degree on my old speakers, and I also tried swapping the speakers with each other and the problem remained on the right.

I highly doubt testing with the speakers connected will show much difference (but I will try later today). This is because I won’t be able to play the 1kHz tone at the same volume as I did without connecting the speakers, and therefore I am guessing there will be an even smaller difference between left and eighth channels.

I did read an old forum post that indicated that subtle differences may not be detectable using a voltmeter. The ultimate test is to use a dual channel oscilloscope, inverting one of the channels phases, then using the “sum” function of the scope. Because you’ve inverted one of the phases, the channels should cancel out each other when you sum them, and any variance from 0V that the scope shows would indicate an imbalance. I used to have an oscilloscope that hooked up to your PC but it’s so old it won’t work anymore.

Thinking my next step could be to rent a dual channel oscilloscope.

And, I’m going to book a frequency response hearing test just to totally rule that out! Although my wife also detected the problem without being told exactly what the issue was first.

I will also try to measure the frequency response playing the test tones, 1m from each speaker playing one side at a time.

Finally, I will also try rotating my entire system 90 degrees in my room.  Just as one final “room test”.  This will require me to dismantle my couch again so might be a while…

I do have a good relationship with my local dealer who I did not buy my amp from.  I could potentially ask them to borrow one of their high end Moon amps for troubleshooting.  I’ve tested Moon gear before and I found their stuff to be very highly resolving, so if the issue is not present with a Moon integrated, I’d think that would prove definitively that my amp is at fault.

The things we do for our hobby!

I thought there had to be a load on the amp to measure this  with multimeter but I might be wrong.  

Your left and right are perfectly matched. No issues

Play the  1 kHz test tone into  speakers don't play so loudly that it damages your speakers. Set your voltmeter for AC Volts. Measure the volts across the speaker terminals. Power is: E²/R where E = volts RMS and R is resistance of load. Both channels should be the same at the same volume setting, maybe slight variation. 

Okay, I tried testing a series of frequency tones without any load / speakers. The voltage of the right channel seems to be slightly higher than the left channel, when playing tones from 800Hz or so through to 12,000 Hz. The difference in voltage is greatest between 4,000 Hz and 6,000 Hz with the maximum difference in voltage between left and right channels being 7mV at 5,000Hz. Across the 800Hz-12,000Hz band, the right channel is always tad HIGHER in voltage than the left side. Outside of this band, the voltages between channels are identical.

Is this difference enough to matter? Interestingly, my ears tell me the problem is centralized around the upper mid range, so that corresponds with where I measured the greatest variance in voltage between channels, at 5,000Hz. But, given the missing frequencies on the right, I would have expected to see LOWER voltages on the right, not higher! Hmmm.

I had my amp set to a moderately high volume setting (20 on the Diablo 300) which is which is about 3-5 higher than my typical listening volume. Here’s the data, with the frequencies with the largest variance between left and right highlighted in bold and underlined:

250 Hz L 694mV R 694mV

400 Hz L 694mV R 694mV

800 Hz L 695mV R 696mV

1000 Hz L 690mV R 692mV

1250 Hz L 684mV R 686mV

2500 Hz L 647mV R 652mV

3150 Hz L 627mV R 633mV

4000 Hz L 600mV R 606mV

5000 Hz L 567mV R 574mV

6000 Hz L 533mV R 539mV

7000 Hz L 497mV R 502mV

8000 Hz L 459mV R 463mV

9000 Hz L 419mV R 422mV

10000 Hz L 382mV R 384mV

11000 Hz L 349mV R 350mV

12000 Hz L 319mV R 319mV

13000 Hz L 287mV R 287mV

14000 Hz L 252mV R 252mV

Is this enough data to tell Gryphon I have an issue?  Or are these differences in voltages between left and right too small to matter?  If the differences too small to matter, well I do find it a bit coincidental that the largest variances in voltages are in the same frequencies where my ear hears the problem (centralized around upper mids).

I did a quick test with my multimeter and will need to re-try the test tomorrow.  Unfortunately I don't have the house to myself and people are sleeping!  I do need to have my speakers connected to do this test right?  I did find that the right side was consistently .001mv lower than the left channel across many but not all frequencies, when my amp was set to it's lowest volume setting - probably meaningless.  Will try at higher volumes tomorrow.

Found the test tone on Tidal.  Will try the multimeter test shortly…

@retiredfarmer I will try getting banana to spade converters; will go to my local shop this week. Then I can try my old amp. I'm not sure the issue will be detectable on lower-resolving equipment however. The issue was barely noticeable (but I did detect it) on my prior B&W speakers which were one model down. Seems that when the mid and upper frequencies are less transparent, the issue is harder to detect. Going to an Arcam A85 from a Gryphon Diablo 300 I expect will be far less resolving (But, I loved that Arcam for what it was, it was an unusual star among other Arcams at the time which were mostly junk).

@djones51 I have a multimeter and can try measuring the voltage on each side! Dumb question though: how do I play a 1kHz test tone, seeing as I normally stream through Roon/Tidal Hifi? My Gryphon dealer is actually not in the same city as me and I’m not sure my local shop is up for any sort of diagnostics.

I like the ides of trying  a different  smp. That should put the room or amp question  to rest one way or another.  For the try even a cheap set of banana  to spade converter s will do what you want. Some midfi low fi store should have them. 

Take it to the shop where they can put test equipment on it. If you had a multimeter you could run the 1khz test tone at 0db on the volume and see what the volts are on each side. 

Thanks @djones51 , I will try that.  I'm not sure if the variation is "big" or not.  I am very sensitive to tonal imbalances, so it could in fact be very small.  Although my wife can immediately hear it as well.

Also, while it is blatantly noticeable on many tracks/artists/recordings, especially if there is not a lot of reverb in the recording, mostly being apparent in upper ranges of vocals (singers sound "fleshed out" in the upper ranges on the left but not the right), on many tracks the effect is very, very subtle.

@djones51 I will try your suggested "imperfect" test.  But I'm curious, what would the perfect test be?

@ghdprentice thanks; actually that is my daughter peeking out! She tends to do that and I don’t even notice...

At least component-wise, it has to be either my amp or the DAC module in my amp.  Can't see how a USB streamer would cause an imbalance of frequencies between channels!

I have a hunch it's more likely to be the amp, not sure why, but I could see it being the DAC possibly.  I am going to dig up my CD player and input it into my amp, to rule out or pinpoint the problem to the DAC module.  Also, I have an old Arcam A85 integrated amp tucked away that I haven't used in years, that I could compare with, but if I recall I think it only accepts banana terminations.

Easy test without buying anything would be download and run a 1khz and/or 10khz test signal through one channel at a time and use a SPL meter on your phone. Make sure you don’t change the volume when switching. The right channel is the one you suspect so I would test the left first get up to 85db with your phone 1 meter from the center axis of the tweeter and mid. Then try the right channel 1 meter from center axis and see what db you get. It’s not a perfect test but it should show if there is a big variation. Should be able to  hear if there is a big variation. 

OP,

Thanks for the photos… I like the one with your significant other (?) peering out> Nice system. 
 

With what you have done, I would also would be at wits end. Happily it sounds like you are working to root cause… that it is it is a component problem. That makes sense. Because, yes, as you say room treatments can dial things in… this problem doesn’t seem to be a room problem. 
 

Okay, so I spent half my Saturday totally tearing apart my living room, removing the couch, and bringing in large cushions, moving blankets, and cardboard (my speaker boxes), to play with various configurations.  I also tried the "excessive toe-in" configuration where you have the tweeter's line of fire cross over a few feet in front of the listening position.  I tried moving my entire system extremely to the left.  I also removed my old speakers from the area.

While some things helped a bit (the thing that helped the most was moving the entire system to the left), the issue ALWAYS remained, no matter what I did.  I even tried changing the right speaker so that it was asymmetrically positioned to the left speaker.  I also compared the same tracks with a decent ($500 Sony WH1000 which are great) pair of headphones just to make sure I wasn't imagining things.  Headphones reminded me of what a perfect stereo image is supposed to be.

So, after all that I realized I had accomplished two things: 1) I don't think my room is to blame, and 2) I found I had a ton of crud under my couch which I cleaned up.

Finally, I went back to wondering about my amp.  I swapped the speaker cables on the amp end only - moving the right cable to the left amp terminal and the left cable to the right amp terminal.  And this time, I REALLY listened.  I realized that the problem did in fact move to the left side, from the right!

With this in mind, I now suspect my Gryphon Diablo 300 amp has an issue.  I'm using the internal Gryphon DAC module, so it's possible that is the culprit and not the amp. 

Has anyone heard of something like this happening with an amp before?  What could cause it?  Frustrating as I adore the sound my system is producing other than the lack of stereo image.  On the other hand, I feel like I'm getting closer, in that I'm getting closer to isolating the cause of my issue.

On a related front:  Why sort of meter should I consider investing in to measure the frequency response of my channels?  And what test track should I use for measurement, to prove beyond a doubt it's not in the source material? If I go back to my retailer/Gryphon with this issue, I feel it would be good to be armed with some "proof".

When you're doing excessive toe you're doing, time intensity trading, like they talk about in video. You could give it a try as it will move the reflection to the side wall on the right. 

I should have been clearer, but when I s said “higher frequencies”, I actually mean upper mids as well.  You can notice that upper part of vocals is toned down on the right, which is annoying.  I tried putting thick moving blankets along the left on that island half-wall, and it did actually help even out the left and right.  But, I don’t really like all the missing higher frequencies with this approach! 
 

So, I’ve concluded that rather than create a balanced suck out on the left, I need to increase the higher frequencies on the right.

I still need to try @milpai ’s idea and block off the gap on the right of my right speakers.  But I think I’d need to use a non absorbing material to preserve the higher frequencies and have them reflect toward the listening position from the right.

Thanks all for the info.  I found this great video discussing narrow vs wide dispersion speakers and it covers some of the principles that we’ve discussed here.  
 

Wide vs Narrow Dispersion

The wall is reflecting too much mids from about 1-4 khz, that makes the highs when you listen to that channel alone seem suppressed. Effectively they are in comparison.  You will even get some mids from the left channel reflecting off that close wall too. You need left / right room response measurement to nail it down.

Thank you for your response and it's always nice to see pictures of the gear in the listening space.  Your gear is miles above mine so please take this with a grain of salt.

I experienced a similar problem as you are having, switched speaker positions and the imbalance (for lack of a better description) stayed on the same side of my room.  And when standing close to the speakers there was no imbalance.  

I have no idea what this means, but in my case I found that if I stand near my listening position and face away from the speakers, the sound was identical from both speakers.  I eventually decided it may be a hearing issue and by adjusting my speaker position I was able to compensate for the difference.

I have been following your last few threads and remember in a prior thread you emphasized you wanted just a TAD more bass but did not want to use digital sound processing to achieve it.  I've never used any DSP so consider this coming from a real amateur.  But seeing your set up, it's hard to see how you will fit RELs in the room without a major reshuffle of your gear and audio furniture and to also take care of the treble problem maybe you should consider it.  A lot of people swear by it.

I am only speculating, but it seems to cure the treble suck out on the right you are going to need to find a way to create an equal suck out on the left without DSP???  Again, just guessing out loud.

You could try extreme toe in where the speaker axis crosses about 3 feet in front of your listening position. You could also pull the speakers out into the room more, In live rooms where you aren't able to do a lot a treatments speakers with narrow dispersion would do better if you aren't married to those B&W. 

@jetter it 100% did occur with my first speakers, just not quite as much. These two speakers have a much different presentation in the mid and higher frequencies. My old speakers sound muddy by comparison. I think this is why the problem is more pronounced with my new speakers.

To rule out that I had a problem with my gear, I did try swapping speaker positions, swapping the cables, leaving the amp speaker cables as is and swapping the cables at the speaker end. No matter what I could still hear the issue on the right side.

Oh, and I even thought maybe I needed to get my hearing checked!  But I was somewhat relieved when my wife heard the same flaw that I did, and I didn’t even explain to her what the issue is other than I told her to pay attention to the stereo image.  And I don’t hear any issues wearing headphones etc.

It's curious that it did not occur also with your first pair of speakers.  

@erik_squires , in my cases I don’t know that your explanation is what his happening, since there is no wall to the left of my listening position (I’ve uploaded photos of my room to my profile).

It works because, relatively, you have excess reflections on the left compared to the right.  You could use a freestanding panel, or wall mounted, whatever you'd like.  Thicker the better here. Between the speakers consider diffusion.

@erik_squires and @audphile1 , I was thinking about your suggestion to put dampening on the left. I was thinking that to the left of the couch, maybe I need a freestanding panel/divider with some dampening installed to it. Do you think this might help, if the panel was as wide as the couch and maybe slightly higher than the speaker?

I don’t know why this works but if I hold a pillow to the left of my left ear, it sort of centres everything, which is why I think your suggestion could be on the right track.

theaudiomaniac, I also experiment as there is too much mid, what's the cause?

Narrow directivity speaker like Dutch and Dutch 8c, Kii3,  would work better in that room.  Or a horn speaker maybe JBL 4367. 

@nyev 

You will need a cardboard that is ONLY as tall as the speaker. And it will be only 2 feet wide to accept the first reflection of the speaker. This will direct the sound waves towards your seating position, rather than to the back of the room where they are currently going. Hope this helps.

@milpai good idea. I will try your suggestion to block off the “jut-out” on the right side, but I’ll need a LOT of cardboard, and it will be tough with the 12ft vaulted ceilings. I won’t be able to block it off practically but I can partially and I will see if it helps.

@ghdprentice , I’ve uploaded photos of my room. I know the room needs to be treated, but other than the one issue mentioned here, there are no other glaring issues with the sound. Before you say it might be the L-shaped couch, I thought it was too, and tried simulating equivalent dampening on the left side, using a large amount of outdoor couch pillows and many heavy and large blankets. This did not help. I’ve tried putting large pillows on the right hand wall, and large couch seats, as well as against the island on the left. Tried covering the TV with a blanket.  I’ve tried all of these measures at the same time and nothing has helped. Lastly the second set of speakers is temporary, as I’m selling my old ones. The issue in question existed slightly before I got my new speakers. But with more detail coming through with my upgraded speakers the problem is more pronounced now.

I think @milpai may be on to something, I will try blocking off the gap on the right side.  But that is obviously not a solution from a practical perspective.   If I walk towards my speakers, the stereo image snaps perfectly into place with high frequencies coming through on the right,  just as I get to the gap on the right.

I do think treatment in the left would help as others have suggested but there is no place or wall for the treatment to go…

@ghdprentice I will upload photos later this evening.

Regarding placement and toe-in, I’ve tried many different positions and various toe in and the right side sounds soft in the high frequencies no matter what…. Its noticeably worse on some tracks, and with others barely noticeable.  My wife can hear it as well (she thought it was off balance but it’s not actually).

There is nothing on the left side so no ability to install dampening.

I did try shifting the entire system to the left, moving the right speaker away from the right wall, and it seems to be the only thing so far that has helped.  However it’s still noticeable.

I’ve not measured it but it’s pretty obvious and only on there right side.  Others can hear it too….

Will upload photos later tonight!

OP, put some photos of your system an venue under your UserID… create your virtual system. Then we can see what is going on… then help.

I think this might be happening because the first reflection point is close to the wall on the right side and the wall reflects the sound again to the front of the room rather than towards you. I had similar issue with the soffit running on the left side of my room. As soon as I identified that and put a panel there things improved pretty much and I got great stereo image. If you have got some tall cardboard boxes, try to line them as it extending the wall "towards" the speaker and see if that is what the issue is. If that is the case, then a GIK gobo might be your solution.

Well @erik_squires posted while I was writing my response. Looks like @erik_squires is also thinking along the same lines. 

How much of a toe in towards your listening position? Where do the tweeter axis cross?

Here’s what I’m thinking…it isn’t the wall on the right that’s an issue. It’s the absence of wall on the left that’s causing the highs frequencies to just get dumped into that open space. Have you stood in that opening on the left to see how it sounds vs your listening position just as far as the highs are concerned?

I may be completely off track here though…

Put absorption on the left.  GIK mondo traps which are around 4" thick would work.

Have you measured it? Perceived lack of highs could be from too much mids.

My drawing sort of worked!  I should also note that the high-frequency suckout doesn't seem to take effect if you stand in front of the speakers before the right-hand wall gets closer to the you.  If you continue walking backwards, you can hear the missing frequencies kick in when the wall on the right starts.