ATC SM75-150S Dome Mid Measurement Data

You assume wrong - it's for the whole speaker including the bass unit.

Yes, it's comb filtering caused by lack of time alignment of the 435 with the supertweeter. The designer himself runs these without the supertweeter for that reason. Sadly Japan insists on them.

The rise in 2nd harmonic is due to the non-linear compression and rarefaction characteristics of air at high levels and is perfectly normal at over 100db AFAIK. It get's up to a whopping great 0.5% at one point. I wonder how bad the ATC's would be at those realistic levels?

 

Yes basically. The JBL's give you more of the recording, the ATC's more of your room. There is no right or wrong here - it's down to personal preference and you obviously like the influence of your surroundings. A lot will depend on the types of music you listen to IME.

 

I don't think that 2nd HD rise can be fully attributed to the distortion of the air.

The disto plot Shinobiwan took of the ATC mid was at 96dB, so not far from the 100dB of the plot you just showed. 2nd HD never rises above 0.2%, and 3rd HD is even lower across the board.



Also I think whether or not you prefer having wide dispersion depends not just on you and the music you listen to, but on the room you are listening in - does it have enough diffusion, absorption etc..

 

Sadly Shin's plot isn't calibrated. Which part of the response was at 96db given the uneven response Simon?

I would say that even the peak at those extreme levels is well below what is considered to be audible so the point is rather moot.

Given the choice of a controlled directivity monitor, and a wide dispersion monitor in a treated room, I'll take the former with a little room treatment. It's the only realistic way IMHO to get the recorded acoustic accurately in a typical living room.

 

The ATC only varies by 2dB from 450Hz - 4KHz so if it was calibrated at 1KHz then the majority of the plot is at 94dB, if it was calibrated above that, or using broadband noise then it is as stated 96dB. Either way I don't suspect a dB here or there is going to change a trend in the response.

What I notice is that the JBL has a rising distortion from 1.5K upwards which is missing on the ATC at similar levels, so thats why I don't think it is all down to non-linearities in air.

I agree though, that at these sorts of levels it is pretty much inaudible, if not entirely, for 2nd HD. 3rd HD and higher harmonics are another matter, but again both are extremely low above 500Hz.

When it comes to dispersion I think you have to consider frequency, too. Wide dispersion in the high frequencies is not such a bad thing since they are more easily diffused in a normal room. As we go lower in frequency I think we need more controlled dispersion. Kind of the opposite of what naturally occurs in fact!

 

Hi Titian,

thanks for the write up. It sounds like your impressions were very similar to mine. Very good speaker with some reservations. I look forward to hearing how you get on at home. When will you be able to borrow a pair?

 

I think you are right - I've seen some other measurements taken at 96db which also show the rising 2nd harmonic, but again at theoretically harmless levels. Interestingly the levels are a lot lower than some of the older Pro units.

Having said that, I'm not a big fan of the 435be. It needs a lot of correction to get that flat FR - in that context it really doesn't compare with the ATC nor the TAD drive units. The advantage it has over the ATC of course is the crossover placed at 800hz rather than smack bang in the presence area.

I've never understood why you would want to spray HF all round the room if the aim is to reproduce the recorded acoustic. I do agree with the bottom end though - the best results I have had by far have been had with subs firing into the rear walls.

 

I'm getting on very well at home and I won't borrow any pair as I'm "happy" with what I've got. At least for my priorities concerning the sound characteristics, I haven't heard anything which would make me hot for changing what I have.

 

These distortion figures are impressive and probably inaudible at anything like normal listening levels.
FWIW The low compression Tannoy tulip waveguide HF in the 215 dmt (if manuf figures are to be trusted)never gets above 0.3% 2nd HD at 107 dB @10k and 0.1% @ 1k or 5k, though is frequency response is not as good as either the ATC or TAD.

 

Because we hear sound integrated over time, and if the reflections are on a landslide with rising frequency, the power response will be rather 'dull'. Even and consistent off-axis response is equally important though and that is something horns do very well since the dispersion of each driver can be matched at the Xover point.

Ideally I would choose both wide and consistently even dispersion with rising frequency. In a good acoustic.

Have you any vertical off-axis response measurements for a horn system? With a direct radiator the center to center spacing is very important for good vertical off axis performance, and with a horn it would be near impossible to get the center to center spacing close. But maybe with a horn the rules are different and it is only the edge of the mouth that has to be close to each other. Do you know?

 

Ah sorry I didn't realise we were talking about direct radiators - of course the power response is crucial with any loudspeaker.

You are going to need an awful lot of drive units and then of course you'll run into even greater problems. I'm more of a realist.

It's a problem. With my speakers I can get the centre to centre spacing close and have a really ugly loudspeaker (see Cessaro), or compromise. The compromise will cause an off axis null at the crossover point, the best solution is to use high order slopes to minimise the magnitude. That's the route I've gone down.

 

As Simon has mentioned, the ATC were tested at 90dB/2m or in otherwords 96dB/1m.

This was done outside. The average level was set prior to measurements using a regular cheapo radioshack SPL meter and white noise. White noise has less low frequency energy than pink but even so I placed a steep 8th order high pass on the driver at 150hz to protect from any possible damage during this level setting. For the distortion test I removed this high pass and instead started the stepped sine sweep at 300hz.

The ATC super is an extremely low distortion driver and especially so when considering its modest size and virtually unassisted loading. Those figures I took at 96dB/1m compare favourably to those you have shown even taking into account the extra level.

The measurement units on the y axis of our graphs are different but consider that -60dB below the fundamental at 0dB is 0.1% distortion and -50dB is 0.3% and -40dB is 1%.

 

Mike when are you going to give the Acourate and PC route a go? It would be perfect for the horns. Per frequency time alignment and phase correction would greatly smooth the issue not to mention create a more 'sorted' speaker. But then of course you may feel something is missing with the digital route. Its relatively cheap to try out considering the rest of your gear - about £240 for Acourate, the PC you already have and a suitable soundcard could be anything from about £200 up to whatever you feel the potential of the system justifies. So if you call it £500 then that will at least give you an insight into what's possible, forget the DEQX as I've had one myself (a tricked out model from Overkill at that) and there's an extra layer of smoothness and refinement missing.

If you want a hand generating good correction filters I'd be more than happy to do those for you and show what needs to be done so you can hit the ground running. Its quite easy though so I wouldn't expect you to have much trouble at all once you'd familiarised yourself with the programs operation.

The improvements have been so stark in comparison to what I've tried and heard before that I can't help but think you'd notice an improvement yourself.

There's really little to loose, hell if you don't think its an improvement I'll buy the Acourate license off you for what you paid, I'll find someone who wants a copy in DIY land, and you'll have little trouble selling a pro soundcard on over on ebay etc.

 

Thanks for the offer Shinobiwan. I'm really unlikely to trial a PC based setup to be honest. I'm currently finalising cabinet design and that should be finished by early in the new year. I'm old fashioned. I appreciate the hifi attributes of all the DSP units that I have used over the years, but an analogue signal chain sounds more natural to my ears, and more enjoyable to listen to. Maybe I grew up with too much distortion! I just seem to like my musical meat to be of the organic variety.

With all of the DSP units I've used previously, the time involved in getting the sound I wanted extended to many months and sadly I'm too busy these days to spend so many hours on my audio system - I admire those like yourself that are more committed. Add to that the fact that I only use Macs now - laptops at that - and the whole thing becomes rather messy to implement.

But I never say never - maybe one day in the future when I have more time on my hands I'll play around with one again. Or I might find a user who wants to try TAD's before buying and we'll be able to help each other out!

 

For me the problem with PC bassed Xover is I don't want to have my computer running all the time while I listen, and more to the point I don't want to use it as a music server... I like picking my CDs and putting them in the player! Just like others like putting vinyl on a TT. Also if the computer breaks down (as it invariably does) I would have no music, which is just when I need it!

If someone produced a DEQx that was just as capable as Acourate then I might be tempted some time down the line. Speaking of which, have yo seem these? http://www.kscind.com/digmoda/

IIRC they can do FIR filters as well as IIR, and also have driver correction so it could be quite capable. A much neater solution thats for sure! If I didn't already have a digital crossover / DAC / Preamp then I might well have gone for those. So yeah, the problem with a PC setup for me is the spreading out and separation of the hi-fi across so many different systems. I want something more integrated, and dedicated.

 

Those look great.

There seems to be a strong trend in the pro industry to have the amps and DSP's in one box. The last one I saw was the new L-ACOUSTICS LA8 amp. They are 4 x 1000W/ch with onboard DSP. Weigh about 12kg. Insane!

 

Ahh yeah I heard of the LA8. I think both have a big market but are for rather different applications. Obviously the LA8 is for live sound, where it is really convenient to control all of the system, on the fly, from a single point.

The KSC plate amps I think are more aimed at studio monitoring or small sound reenforcement since you would have to connect to each pack separately to control them. Very very good for studio monitoring though where you would set it and leave it alone. With driver correction, room correction and all that built in it is ideal for custom installations. I still can't quite convince myself to put a digital amp with all its signal chain complexity in my system at the analogue stage, but I'll probably get over it with time. The future is bright

 

I would just stick with the old amp power supply technology. If you don't have to lug them around, they are much easier to work on if anything goes wrong. The new generation of "light" amps can be a pain to fix, at least that's what the repairers have told me (makes sense).

So does this trend in on board DSP's mean there should be some nice external DSP loudspeaker controllers coming onto the market. I'm trying to get one of the BSS 336 units from the venue where the LA8's were installed (to experiment on the JBL's).

p.s. It's a real buzz being able to walk around a venue with complete wireless control of a mega sound system via a little VAOI laptop (Digidesign Desk )

Back to topic!!

 

The LA8 is made for L-ACOUSTICS by Camco. http://www.camcoaudio.com/

They have moved away from using rebadged Lab.Gruppen amps for some reason.

 

Not quite;-)
I'm currently using an xta 226/TDX2 a very similar spec to the BSS unit on my DC's with very good results, to my ears at least once levels are set correctly and with only subtle eq ing the unit remains transparent-I'm not going back to active analogue xovers.