|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
74.131.169.89
I recently reinstalled my AT-OC9ML/II and decided to measure Arm/Cartridge Resonance using the Hi-Fi News Test Record. Actually, I decided to record the tests and transfer them into my computer so I could look at the waveforms in Sound Forge. By doing this, I discovered something very interesting. The voice announcements do not represent the actual frequencies very accurately.I also discovered that my AT-OC9ML/II mounted in my SME V tonearm has a vertical resonance of 9-Hz and a lateral resonance of 7-Hz. These are considerably different that I previously thought from using the frequency voice announcements. This means that my cartridge has a compliance of about 16-mm/N based on a 10.5-g effective mass tonearm and a 9-g cartridge including mounting screws.
To give you an example of just how inaccurate the voice announcements on the Hi-Fi News test record are, I measured the frequencies visually in Sound Forge and compiled a table below. I adjusted the speed of my turntable very accurately by timing revolutions of the platter so that its speed was accurate to between 0% and 0.04% fast. I made absolutely certain it was not going even the least bit slow. Be advised, this is the old Hi-Fi News test record. I don’t know if the new one is any better. Perhaps someone who owns one can make some measurements.
Best regards,
John ElisonPS. I couldn't measure the lowest frequencies because a cartridge will not respond to frequencies below its resonance frequency.
Follow Ups:
I was wondering why the resonant frequencies for the carts I tested measured higher on the HFN than on the my Shure record (type IV). Unfortunately I wish they WERE higher.
Very user friendly, too.
Hi John!Are you saying after your analysis the vertical resonance is 9Hz and not the 10Hz "announced" on the test? That is what your list indicates.
I'm not questioning your methodology but I wonder how accurate are the frequncies of other tones on various test LPs, not necessarily ones for testing resonant frequency. Perhaps extremely low frequencies are difficult to record and transfer. I don't know.
In the end is it really all that significant? So the RF is 9 not 10Hz?
"but I wonder how accurate are the frequncies of other tones on various test LPs"And you have right to wonder.
The frequency sweeps on the new HFR and even on the Cardas disc are outright crap, with gross deviations in level versus f (!) and with coarse edits along the sweep as if a not-very-good analogue signal generator was used with limited sweep range.
Warps make tracking test signals near to useless. All of my recent dearly-paid test records have warps. Interesting to see the THD meter go up and down ... with the record.
I agree with John that the best way of using these flawed records is to record them and then read into a PC for further analysis. Makes it easier to separate disc artefacts from turntable artefacts.
The pink noise tests on the HFN discs are pretty decent, though. I've been able of measuring frequency response to accuracies of less than a dB and channel separation to 40dB at 1kHz, even when the 'silent' channels contain obvious in-band resonances from the cutting lathe.
That doesn't mean it can't happen.
I have very few warped records, as well, and the few with slight warps are flattened out nearly completely with vacuum hold-down.
It seems pretty significant to me if you want to know the truth and you're paying top dollar for a test record that is seriously flawed. Of course, if you copy the tests to your computer you will know the truth, but you should be able to trust the voice announcements on record.With respect to other test records, my copy of Shure's Audio Obstacle Course for the V15typeV is very accurate. Furthermore, it has discrete frequencies rather than a glide tone. In other words, when the announcement says 9-Hz, the frequency is 9-Hz until it switches to 10-Hz, and the accuracy is very good at about +1%. Generally speaking, the test tones on professional test records are quite accurate.
Anyway, this also means that my DL-103R has compliance of around 12-mm/N instead of 9-mm/N as I previously thought.
If you correct the speed accuracy to the frequency specifications I think everything would fall nicely into place. Either their cutter table was fast, or your table is slow!
This post is made possible by the generous support of people like you and our sponsors: