Graham Audio LS5/9

[vln]

@Strictly Stereo many thanks for the effort of measuring and posting! To me it seems that at least in this room the LS5/9 are on the whole a little bit more balanced? That 5dB difference at 4kHz should be audible in a positive way?

 

[vln]

A few years ago I owned a pair of Dynaudio C1s, and really couldn't stand them. I much prefer any of the Graham or Harbeth speakers I have had since, although they are considerably cheaper. This is not to suggest that the Dynaudios are bad speakers - they just didn't suit my tastes or my listening preferences. Dynaudios seem to need a lot of space, and a lot of volume to sound their best. The BBC designs work best in smaller spaces, in near field, and at relatively low volumes. This doesn't suit everybody, but it is how I prefer to listen to music. Although I don't think I would ever like the C1s with their rather dry and analytical sound under any conditions. You can have all the R&D facilities in the world but your product will still not suit everyone. For me, the Grahams are far more musical and engaging.

As between Graham and Harbeth, having owned Harbeth P3ESR, M30.1s and M30.2s in the last few years, as well as the Graham LS5/9, LS3/5 and LS6, it is ridiculous to say that the Harbeths are in any way better than the Grahams. Both are very good, but I prefer the Graham LS3/5s and LS6s to any of the Harbeths, and only marginally preferred the 30.2s to the LS5/9s, although the latter are in many ways the better speaker. The Harbeths manage to impose their character despite the amplifier and source used, while the Grahams are more transparent to the source and upstream components.

Many thanks for your balanced and informative post. I used to own Dynaudio Contour S3.4 before moving to ATC. Back then not a real like for like comparison, as the ATCs were a) much larger, b) had larger woofers, c) were active, d) were 3-ways. I agree somewhat WRT to the Dyns needing to be played loud to come alive but then people say the same thing about ATC. I too don't necessarily have the requirement of being able to play a Mahler symphony or a Motörhead concert at authentic SPLs level at home, so what you wrote does sound appealing. When I did move from Harbeth C7s to the Dyns before that (does it ever stop?!), the latter sounded much more crisp, detailed and full range than the C7s. Of the 3 speakers mentioned, the Dyns also had by far the best build quality.

There is a huge amount of rubbish said on the Harbeth forum, one reason why I never bothered joining even during the years that I owned Harbeth speakers.

From my own experience (not on the HUG though), I think that's a general characteristic of most single-manufacturer forums which is why these days I avoid them like the plague. Usually too many self appointed police men patrolling every thread to make sure everyone toes the party line.

 

[Strictly Stereo]

@Strictly Stereo many thanks for the effort of measuring and posting! To me it seems that at least in this room the LS5/9 are on the whole a little bit more balanced? That 5dB difference at 4kHz should be audible in a positive way?

I’d say the LS5/9 is more balanced up to about 12kHz. After that they roll off steeply, while the LS6 is more extended. I used the LS5/9 measurements to create some EQ filters for a demo today. I am currently listening to them with the low end (sub 150Hz) corrected and they sound lovely.

 

[Martyn Miles]

Many thanks for your balanced and informative post. I used to own Dynaudio Contour S3.4 before moving to ATC. Back then not a real like for like comparison, as the ATCs were a) much larger, b) had larger woofers, c) were active, d) were 3-ways. I agree somewhat WRT to the Dyns needing to be played loud to come alive but then people say the same thing about ATC. I too don't necessarily have the requirement of being able to play a Mahler symphony or a Motörhead concert at authentic SPLs level at home, so what you wrote does sound appealing. When I did move from Harbeth C7s to the Dyns before that (does it ever stop?!), the latter sounded much more crisp, detailed and full range than the C7s. Of the 3 speakers mentioned, the Dyns also had by far the best build quality.



From my own experience (not on the HUG though), I think that's a general characteristic of most single-manufacturer forums which is why these days I avoid them like the plague. Usually too many self appointed police men patrolling every thread to make sure everyone toes the party line.

I read the HUG Forum from time to time, but have always made my own mind up when choosing.
There are some balanced and sensible posts there, as per other forums.

 

[Ayya Khema]

The Volt BM228.8 (which i use in my own speakers) doesn't beam any more than the 6.5" unit that's in ProAc's SM100s (or other similar sized units) so generalising isn't always sufficient justification.

your volt is a 8 inch. beaming is directly proportional to the driver size

in this case, its not even a generalisation, its physics, the volt will start beaming lower then the 6.5 inch in the proac

 

[Ayya Khema]

Indeed, point being its all about the implementation and the sum of parts and design as a whole.

Example, the VOLT BM228.8 in my 21L cabinet produce better (and lower) bass notes than the ATC50ASLs did. Both ported and one 2/5th the size.
This simulation was run at 20L 40Hz, the final designs 21.5L and port tuned to 38Hz, no humps or port chuffing.

the volt is a 8 inch, the atc 50 is a 9 inch. hardly a world in difference.

ime, after comparing so many speakers, driver size is directly related to bass performace

comparing a 12 inch vs 15 inch is obvious. then 10” vs 15”. ouch. not even worth talking about the bass performance of a 6,5 inch; they sound distorted with no extension, no slam or punch. bass must be felt, a 6,5 cannot produce what i define as bass.

8 inch is ime bare minimum for acceptable bass and even then, for bass, compared to a 15” they sound pretty dissapointing

 

[Cereal Killer]

Agreed. At one point i was going to go with the BM2500.4 and the D2908/714000 for a 2 way 30L compact/ish monitor.

 

[tuga]

This is a job for...a Waveguide.

 

[Paul Burke]

Indeed, point being its all about the implementation and the sum of parts and design as a whole.

Example, the VOLT BM228.8 in my 21L cabinet produce better (and lower) bass notes than the ATC50ASLs did. Both ported and one 2/5th the size.
This simulation was run at 20L 40Hz, the final designs 21.5L and port tuned to 38Hz, no humps or port chuffing.

Looks a decent build. When you say no humps or port chuffing, what alignment did you have them built too? And what port sizes have you used? I’d be very surprised if you actually got linearity from the port in a box that size, it’s practically impossible. I’m doing some testing to try and find out what sizes of ports are audibly non linear at realistic volumes. To me it’s what makes many vented boxes sound poor and cheap from a lot of the usual suspects

 

[Cereal Killer]

Paul, the ports as sized in the simulation, i bought them from HiFi Collective 185x 50mm they’re flared to the exit with a rolled out curve. I’ll dig out the product number when I get home.

I’ve actually tried to make them chuff with some serious electronica and failed even at sofa shaking volume.

This port: https://www.hificollective.co.uk/catalog/051-0013-flared-port-tube-50mmx185mm.html

 

[tuga]

Looks a decent build. When you say no humps or port chuffing, what alignment did you have them built too? And what port sizes have you used? I’d be very surprised if you actually got linearity from the port in a box that size, it’s practically impossible. I’m doing some testing to try and find out what sizes of ports are audibly non linear at realistic volumes. To me it’s what makes many vented boxes sound poor and cheap from a lot of the usual suspects

That's very interesting. Do keep us posted.

 

[Ayya Khema]

Agreed. At one point i was going to go with the BM2500.4 and the D2908/714000 for a 2 way 30L compact/ish monitor.

10 inch 2 way is a nice compromise, my speakers are a 10 inch 2 way: Seas a26. im very happy with them.
im sure your speakers are mighty fine though!

 

[Paul Burke]

Paul, the ports as sized in the simulation, i bought them from HiFi Collective 185x 50mm they’re flared to the exit with a rolled out curve. I’ll dig out the product number when I get home.

I’ve actually tried to make them chuff with some serious electronica and failed even at sofa shaking volume.

This port: https://www.hificollective.co.uk/catalog/051-0013-flared-port-tube-50mmx185mm.html

Good if you’ve managed to get away with chuffing then, flares both ends helps, one gets part of the way there

Did you/wilmslow if they did the design and modelled volume etc go for a flat alignment and know which one was chosen if you did? Sbb4, sc4 & qb3 are the main ones, which will give you a different q and transient response

For a 6.5-8” driver with pretty normal ts parameters at least a 3” port would be better, and have less of a hump on the bottom end as the wick goes up, but as diameter increases length required to tune at the same frequency increases too, quite dramatically, and in a stand mount there wouldn’t be enough space for it, without bends on the port etc but that has its own compromises. Which is what the design basically comes down to, choosing the best bunch of compromises

 

[Cereal Killer]

I did the initial design, spec and made the volumetric choices. The modelling was tested by a old friend who had a bespoke speaker manufacturing company in a virtual state.
Wilmslow built them for me to my exact requirements. Asd they have decades of experience with both the HF and LF driver, i took their advice on the cross-overs.

IMHO a 75mm port isn't necessary, the port in the link above is tapered along it its length by a few mm which definitely helps.. It's also located 45mm short (centrally) of the back HF unit which caused the inner end of the port to be virtually flared.

 

[sunbeamgls]

No.

Do you need to drive this car to know it will fail?

But you've missed the point - of course the car will fail if you drive it beyond its capabilities. My point is, why do you think an 8" is better at all the things it is doing compared to a 6.5"? The 6.5" mid-range output may massively outweigh any perceived bass limitations.

 

[sunbeamgls]

thing is, a 8 inch midbass will always be much better in the bass then a 6,5 inch. imo, 6,5 inch based speakers are toys and cannot ever be played loud: no matter how good the mid will be, bass will be compromised (distortion, imd, extension limited). id also say i like big mid sound, 6.5 inch mid will always sound “small”

while yes the 8 inch beams and is a compromise desing to make it play up to 2.5khz/3khz, its still much less compromised then asking a 6.5 inch midbass to do bass.

i wouldnt be surprised that from 160hz and up the dynaudio would be slightly more refined then the graham, but the bass on the graham will be undoubtly much better, can play louder, better extension, ect. for me bass is very important

and i also think 8 inch sound small for bass vs 10”. then you compare a 10” with a 15” and you tell me if its even worth mentioning the “bass” of a 6,5 inch woofer. so yeah, the confidence 20 priced at 12 000 euro is insanity beyond words

lets just say id buy the dynaudio contour 60 (2x 9 inch woofer) over the confidence 20 without a doubt

As for harbeth shl5+ vs graham ls59, ive had both and they are more similar sounding then not, i just prefered slightly the shl5+

But listening to the Grahams will tell you much more about their bass and many of the points you make about being able to go louder and deeper just because it has an 8" driver you'll find are not necessarily the case. There's a lot more to speaker design than the size of the cone.

Some of the most feeble bass I've heard has been with those 15" paper drivers that don't actually have any excursion capabilities. Talking purely about the diameter of a speaker cone as an indicator of its abilities is not borne out in practice. Its far more complex.

 

[sunbeamgls]

What one can't say is whether the Dynaudio 6.5" driver will perform better than the Volt 8" driver.

Nor vice-versa. The output is the result of the whole speaker as a complete system, not just the drive unit. I'm sure I could build a box to massively compromise either driver.

 

[sunbeamgls]

the volt is a 8 inch, the atc 50 is a 9 inch. hardly a world in difference.

A 26% increase in surface area is not worth considering?

 

[tuga]

But you've missed the point - of course the car will fail if you drive it beyond its capabilities. My point is, why do you think an 8" is better at all the things it is doing compared to a 6.5"? The 6.5" mid-range output may massively outweigh any perceived bass limitations.

The car's performance, or its capabilites if you prefer are limited by its topology (3 vs 4 wheels, and narrow wheelbase and high centre of gravity).
And you can arrive to that conclusion just by looking at it.

The midrange spans a range roughly distributed between 250Hz and 2.5kHz.
As far as I know, the advantages of using a 6.5'' instead of an 8'' driver are that it will go into breakup a bit higher in frequency and will not be beaming as much at the top of the passband.

 

[Ayya Khema]

But listening to the Grahams will tell you much more about their bass and many of the points you make about being able to go louder and deeper just because it has an 8" driver you'll find are not necessarily the case. There's a lot more to speaker design than the size of the cone.

Some of the most feeble bass I've heard has been with those 15" paper drivers that don't actually have any excursion capabilities. Talking purely about the diameter of a speaker cone as an indicator of its abilities is not borne out in practice. Its far more complex.

i had the graham ls59 for about 8 month in my room and they do decent bass.

when i speak about hifi i mean comparing good driver vs good drivers. when it comes to bass its the one thing that is really not complex, a good 12” or 15” absolutely destroy a 8 inch for bass.

however, a 15 /12 inch needs a totally different speaker desing: ie implemented into a 3 way, or crossed to a horn for a 2 way and that brings a very different sound or set of compromise then a bbc 2 way. just integrating a 12/15 to a midrange around 300hz is often audible, or a horn have its own set of compromise in terms of coloration. i find that at about 8 inch, it can start to give satisfactory bass. never heard a 6,5 inch do bass without obvious serious limitations/coloration/distortion