The Korf Blog

The inside story: our research,
development and opinions

3 September 2019
Headshell Performance ⁠— Listening, Part I
We've done a whole lot of headshell measurements. Now is the time to see how our observations translate into subjective reality.

Listening to all the headshells would be too time-consuming, and probably require a panel of listeners as the single person would tire quickly and become desensitized to small differences in sonics. Taking just the best and the worst is, I think, a sensible compromise. It would also allow us to do the test twice: with a moving magnet cartridge, and later on with a moving coil.

But first, I have a confession to make.

I just cannot write about the sound
There are people who can listen to a system for a short while, and then come up with a couple of lucid paragraphs of prose that describe precisely what they have heards. You read it, nodding in agreement: yes, I see what that was about. I understand and I would recognize the same traits were I to listen to the same system.

I am most definitely not one of those people. To me, writing about sound is sort of like dancing about architecture. The futile attempt to express specifics of one art using the totally unsuitable tools of the other. I get writer's block just thinking about relaying what I have just heard.

When I do try to describe the sound, nothing but platitudes come out. Resolving or not, full-bodied or thin, bright or dull, coloured or neutral. These are, perhaps, useful expressions, but they don't get you there. They don't do justice to the magical moment when it all comes together and you're enveloped in gorgeous sound, a passive listener no more.

So I can't help but do some measurements again. They are a lot easier to present. It is perhaps better to let professional writers describe the sound. Of course, I will keep trying, but please don't expect Stereophile-quality reporting from me.

There, I said it. Now let's go and compare the two headshells that had the best and the worst IET measurements in our series (click the pictures to see full reports):
.
CNC aluminium + steel rods
14.5 grams
One low amplitude resonance at 480 Hz
2mm stamped aluminium alloy sheet
7.3 grams
Peak at 300 Hz, ringing for more than half a second
Measurements
For consistency's sake, we're using the same Audio Technica AT7V. It has a new stylus now—the suspension on the old one collapsed from many months of misuse.

The rest of equipment is exactly the same as in the previous post: GrooveMaster II tonearm mounted on the TechDAS AF 3 turntable. We're playing back the vertically modulated 20 Hz to 20 kHz sweep from a test LP, and recording the resulting vertical acceleration with 2 accelerometers at once: one at the headshell, and one on the armtube.

The first chart shows the measurements taken at headshells. Blue line is the stamped aluminium one, orange line is an Orsonic AV-101b.

(please ignore the 150 Hz peak, it's the 3rd harmonic of 50 Hz on a noisy DAQ channel)
Interesting that the main mode at about 280 Hz is so much higher with OEM headshell. This isn't particularly surprising, because we had a huge peak at exactly this frequency in the IET measurement. Is it just at headshell, or will the effect persist at armtube?

Orsonic's crossbar's own frequency is exactly the same as accelerometer's. This was a bit surprising.
The role of the headshell is a lot more important than I originally thought
Oh my. The effect on the main mode is even more pronounced at the armtube. The role of the headshell is a lot more important than I originally thought: the poorly designed headshell doesn't just add its own colorations. It also excites the arm's resonances.

It also looks like having arm and headshell resonances close together is one of those really bad ideas. The arm's measured performance is essentially halved by using the primitive OEM stamped aluminium headshell.

Subjective Evaluation
With the Orsonic headshell, there's an instant transformation
I've started listening with the OEM headshell, to quicker get over with the unpleasant. It isn't actually terrible, but the combination is completely devoid of magic.

It's a playback of a stereo recording, all right⁠—but the scene shifts all the time, there's very limited depth, and in the intense loud moments everything just collapses into a heap. The OEM headshell seems to emphasize the worst parts of AT7V's sonic signature: brightness becomes glare, sparkle transforms into harshness, neutrality becomes emptiness. There's no "darkness" in the silence, just grey sonic dusk.

With the Orsonic headshell, there's an instant transformation. The effect is akin to finally putting on the special glasses when a 3D movie is being shown. The strange artifacts transform into a whole new dimension. The stereo picture is palpable, deep and immovable. The composure remains even in the loudest tutti. And the former annoyances, you scarcely recognize them at all.

Yes, the AT7V is still on the bright side, and there's no artificial warmth anywhere. But the total result is wonderfully even, highlighting the recording and letting the technology step aside.

So yes, headshells matter sonically. Rather a lot, I would say. And it looks like the simple IET measurement can predict their subjective quality to some extent.

Next week, we'll do the same exercise, but with a lot more demanding Moving Coil cartridge. Stay with us!
Please subscribe to receive blog updates in your inbox!
comments powered by HyperComments