Like what you see?
Subscribe to the Korf Blog!
- Be notified when new articles are posted
- Access members-only secret preview sales
- Get deals and discounts on Korf Audio products
Expect 1-4 emails a month. We will never ever spam you. You can always unsubscribe—no hard feelings!
The Korf Blog
The inside story: our research,
development and opinions
development and opinions
20 March 2019
Headshell Performance — Nagaoka AL-703
In this series of articles, we measure the performance of removeable tonearm headshells. Today, the subject is a Nagaoka AL-703.
It's a solid aluminium headshell that weighs 12 gram without wires or mounting screws. The exact alloy composition of the body is unknown, the fingerlift is stamped from a typical 5052 aluminium sheet. Both are anodized black.
One neat thing about an AL-703 is a yellow grid painted on its underside. Makes aligning the cartridge a bit easier.
One neat thing about an AL-703 is a yellow grid painted on its underside. Makes aligning the cartridge a bit easier.
That fingerlift makes attaching an accelerometer a bit tricky. Had to put it almost at the front edge of the headshell.
One of the questions I keep getting from my readers is the role this and similar fingerlifts play in the overall vibration picture. To get an answer, we first did the IET measurement of the fully assembled headshell, and then of the headshell without the fingerlift plate.
One of the questions I keep getting from my readers is the role this and similar fingerlifts play in the overall vibration picture. To get an answer, we first did the IET measurement of the fully assembled headshell, and then of the headshell without the fingerlift plate.
But first, a word of caution. IET measurements by their very nature aren't very precise. The overall picture is usually correct, but there exists significant drift from measurement to measurement. To illustrate, I've put together a 3-frame spectrogram animation from 3 separate hammer strikes in the same configuration.
You can see that, while the resonances themselves are more or less the same, their frequency and extent drifts a little bit.
You can see that, while the resonances themselves are more or less the same, their frequency and extent drifts a little bit.
Here's the result with the finger lift in place.
It's a well-dampened headshell. You can drag the vertical line to compare the result to the SME 3009's we have measured in an introduction. The resonances are extinguished quicker, and their Q is much lower. 100 milliseconds, and most of the energy is gone. They are also a lot less "sharp", the energy is spread over a wider spectrum. The main modes are at 300-400Hz, 2kHz and between 7 and 8 kHz.
It's a well-dampened headshell. You can drag the vertical line to compare the result to the SME 3009's we have measured in an introduction. The resonances are extinguished quicker, and their Q is much lower. 100 milliseconds, and most of the energy is gone. They are also a lot less "sharp", the energy is spread over a wider spectrum. The main modes are at 300-400Hz, 2kHz and between 7 and 8 kHz.
That 100 millisecond "cutoff" is very clear on a waterfall plot:
Ok, but what about the fingerlift? What would happen if we remove it? Here you go:
You could argue that the 5 kHz resonance is a bit more pronounced, and a 7-8 kHz is also larger and lasts longer... but really, it's all within the measurement's margin of error. No real difference whatsoever.
Interestingly, this is in line with our tonearm measurements that showed "auxilaries" like antiskate levers, balancers etc having no effect on the vibration picture whatsoever.
The 15 kHz "resonance" is an artifact of DAQ overload. I have found the slight clipping only after all the measurements were finished, and redoing them was impractical. For the future measurements, I would decrease the DAQ sensitivity to avoid such errors.
Interestingly, this is in line with our tonearm measurements that showed "auxilaries" like antiskate levers, balancers etc having no effect on the vibration picture whatsoever.
The 15 kHz "resonance" is an artifact of DAQ overload. I have found the slight clipping only after all the measurements were finished, and redoing them was impractical. For the future measurements, I would decrease the DAQ sensitivity to avoid such errors.
Besides removing the fingerlift, we tried the IET measurement with a cartridge attached to it. This is a rather basic Technics EPC-U24 cartridge, chosen because it has threaded inserts for ease of mounting. Of course, we had to remove the stylus.
The hammer was positioned in such a way as to strike the exposed metal part of the cartridge. What's the result?
The hammer was positioned in such a way as to strike the exposed metal part of the cartridge. What's the result?
Very interesting! The cartridge "ate" all the resonances except for the two main ones! I must say I am looking forward to measuring more headshells to see how much of a unifying influence the cartridge will turn out to be.
So there it is, the first thorough measurement of a headshell. There naturally is a bit of a learning curve, and everything took longer than expected. But the results, in my opinion, are worth it. Measuring and interpreting the data for the next ones should be a bit faster!
Please subscribe to receive blog updates in your inbox!