Test ATC SM75-150 mid-dome
Copyright 2017 © Troels Gravesen

Two- and three-inch midrange domes were highly popular in the Eighties and both SEAS and Vifa had their go at it. I tried the Vifa 3" dome back then and never liked it much. Dynaudio had the D52 and -54 mid-domes and used them extensively in their commercial products. Why they have gone out of fashion is a good question, maybe because 3-ways generally got out of fashion and thanks to D2905 and other excellent dome tweeters, 2-ways took over although never able to deliver the same smooth transition between a 6" or an 8" midbass driver and the dome. Having a middome adds considerably to cost and crossover complexity.
Today we see a couple of 2" and 3" domes from China and rather recently reintroduction of 3" domes from ScanSpeak. Large midrange domes are claimed rocking in the magnetic gap due to the single suspension, but I've never seen any proof of this - and it would probably be very difficult to do that test. Maybe double laser interferometry could shed light on the issue, but until now it remains unsubstantiated claims. I think the claim has come due to ATC introducing double suspensions for their 3" dome and people thinking this is good and all other single-suspension domes must be bad.

One of my friends lend me his single ATC-SM75-150 dome for testing and here are my results. Would be nice to have two samples, but what I found is well in accordance with other tests of the same dome.

The good question is what a 3" dome has to offer against a lot of really good 4" cone drivers available today? Wouldn't the ScanSpeak 12MU do just as well? Not to forget the price, because a lot of good 4" drivers come at 25-50% price of the ATC dome. Only available source of this dome appear to be Wilmslow/UK @ 387 £. Not cheap at all. You can get two AudioTechnology 15H52 middrivers for the price of a single ATC dome! Unfortunately I do not have two of this to test in a classic 3-way and evaluate against e.g. AT 15H52. I'm almost glad I don't, because it takes a damn lot of work to do so, and we may end up thinking we compare apples and pears. Two ways of doing the same thing - and in the end a matter of taste. Who knows.
I've always found it strange to combine two 15" drivers with a single 3" dome as done by ATC in their larger studio monitors. Despite a fairly low point of crossover there must be some noticeable characteristics from the transition of sound propagating from some 1600 cm^2 membrane area to a single small 44 cm^2 dome. This is one huge transition!

The ATC SM75-150 is a chunky beast of 7 kg due to a massive motor structure.

As can be seen, double suspensions secure stable positioning of voice coil in magnet gap.

Measurements

Above frequency response measured at 0.25 meter distance, normalised for 1 meter/2.8 volts. Baffle 60 x 100 cm, driver placed 40 cm from top. Sensitivity around 89 dB/2.8V. This is a 16 Ohm driver (Re 12.2 Ohm), so sensitivity is really good. Now, this is an unusual smooth frequency response, all the way from 300 Hz to around 3.5 kHz. Some break-up modes appear around 4.5 kHz. More on this from the cumulative spectral decay plots below. Based on simulation it seems a 2nd order high-pass filter produce a 4th roll-off, and at the top a true 2nd order low-pass filter can be achieved. This is an easy driver when it comes to crossover construction and overall implementation. At least from simulation from the large baffle. How the driver behaves on a narrow baffle I can't tell. Probably not as nice as this.

Above horizontal dispersion @ 0, 10, 20, 30 and 40 deg, off-axis. Measured at 0.5 meter distance, normalised for 1 meter, 2.8 volts. With an assumed point of crossover around 3 kHz, this is just excellent.

Above cumulative spectral decay. The lower plot displaying the resonance at 4550 Hz, not something that should bother even with a 2nd order point of crossover at 3 kHz.

Distortion measured at 2.8 V input. This dome should be well suited down to 400 Hz. Green is 2nd and blue is 3rd.

Above impedance plot displaying a huge peak at 320 Hz. Not a trace of unrest from 400 to 20000 Hz. Let's expand this graph some closer scrutiny. 

Above impedance plot with a 50 Ohm scale. Still very smooth with only a small glitch around 4550 Hz. Apparently no symmetric driver is incorporated here to reduce impedance rise towards higher frequencies.

TS data retrieved from the impedance plot. Hard to do a full test from either added mass or added volume.

Just for fun I modeled my AudioTechnology 3WC with the ATC dome. Piece of cake really! Wonder how it would sound.

Above the midrange crossover based on simulation.