Lowther Days, test Lowther EX3
Copyright 2008 Troels Gravesen

Go to Lowther website

I had the opportunity to borrow a pair of Lowther EX3 drivers and my Voigt test caps should make decent compartments for testing these legendary drivers.
Lowther drivers have been around since the dawn of time and basically still consist of a straight-sided, ribbed paper cone supplemented by a small whizzer cone taking care of treble. A serious magnet system with a very narrow magnet gap housing a short voice coil wound on both the inside and outside of the former has always been a feature resulting in high efficiency and claimed fast transient attack. The inevitable huge alnico magnets were later supplemented by neodymium magnet systems when these became available.

When I picked up the drivers I promised myself that I would try leaving all prejudice behind and carefully listen without thinking of all things said about the Lowther midrange shout, etc. I don't think any driver has caused so many people so much trouble based on what can be found on internet discussion forums. These drivers seem to be either loved or hated.

My initial set-up were drivers as-is connected directly to my SET tube amps. No crossover, no equalising circuits or anything. Let's start with the good thing: These drivers present a decent bass from my Voigt horns, better than anticipated from TS measurements. With an Fs around 50 Hz they should be rather bass shy, but this wasn't the case. They go fairly deep all things considered. A well designed exponential horn may do even better compared to my TQWT, but no objections to bass performance as such. Corner placement may help even further.

The problem with these Lowthers is that the overall response raises some 10-15 dB above the upper bass/lower midrange/middle midrange where most fundamentals are found, thus we have a fairly bright sound. You may play some music with a calm intro and think this doesn't sound bad at all, but then some solo guitar sets in, a sax, a clarinet or female vocal and it immediately appears something is seriously wrong. Anything with some energy in the 1-3 kHz range is just way out of proportion. It may sometimes be even painful listening to these drivers on-axis. Having a 15 dB peak at 2 kHz and a 180 deg. phase shift at 3.5 kHz is simply not pleasing to the ear. Quite frankly, it sounds dreadful!

Nor do these drivers - as-is - present a particularly impressive level of transparency, nor transient attack. I've heard just as good or better from many other drivers. Any talk of "timbre" from these drivers is...... far out!


It's hard to review a driver displaying the level on linear distortion seen here. We do need a reasonable balanced presentation of basic notes and upper tones in order to evaluate the overall performance of a driver or driver system. Listening some 30-40 deg. off-axis helps a lot, but this is not how most Lowther systems are meant to be. What's even worse is that non-linear distortion appears significant too; view measurements below.

Adding the equalising circuit to some extent got things in place but still the treble is very, very special. The serious problems between 2 and 4 kHz are hard to ignore. It just doesn't sound right. If you want treble from a fullrange driver, try the Supravox 215RTF; much better despite coming from a driver with a large cone diameter with subsequent problems of beaming. At least it's somewhat coherent, which can hardly be said about this driver.
Maybe a mid-horn, lifting the 200-1000 Hz region up some 5-6 dB might be able to somewhat balance the 2 kHz peak. Ad a decent bass horn + corner placement and we may get an overall balanced sound from 50 Hz to the highest treble. I don't know, but how people manage to live with these drivers in simple horns like the one I've used or on open baffles is a mystery to me.

Lowther drivers appear to be slaves of their history. This is what it has always been and this is how it's going to be. Well, a few developments over time in terms of phase plug, etc., but nothing that seriously attacks the inherent problems associated with the cone layout. A curvilinear cone without the whizzer cone might have been an interesting development supplemented by a supertweeter for the fullrange fans who do not allow any crossovers in upper mid and treble regions.

I had plans for much more writing on these Lowthers, but I feel tired already. I would so much like to write something nice about them, but I simply can't.
I just feel sorry for the guy who paid almost 10,000 DKK for the pair. And he had even bought two pairs (!) having made Academy cabs and everything - but simply had to put them aside. His 300B mono blocks deserves much better.


Left: SPL (red), meter distance normalised for 2.8V/1 meter. Blue= minimum phase.
Pay notice to ~180 deg. phase shift at 3.5 kHz. Right: Free air impedance.

Right: Impedance of drivers in horn; left and right speaker.
Left: The EX3 has two copper plates on top of the polepiece, held in place by the phase plug.
Removing these does not alter the impedance profile as can be seen above. Overall the
free air impedance of these drivers is remarkably flat.

Left: Horizontal dispersion at 0, 10, 20, 30 and 40 deg.
Right: Same as left with octave smoothing.

Left: Impulse response. Right: Step response.

Left: Schroeder plot. Right: ETC plot.

Left: Distortion at 2.8V/1meter, response raised 15 dB. Right: Same with 20 dB scaling.

Left: Cumulative spectral decay, 40 dB. Right: Cumulative spectral decay, 20 dB.

TS data. Added volume used.

Lowther EX3 correction network

LspCAD predicted response

Measured response from equalising circuit. Red is without R1031+C1031.
Measurement merged at 400 Hz with nearfield response.
Removing the 2 kHz peak does leave a significant dip at 3 kHz despite intensive modelling.

Routing for drivers