[marknoir]

This topic was probably beat to death, but I couldn't find specifics

I recently aquired a nearly new AT OC9ML II. I've heard that it is a superior tracker, and that is what I was after. I was tired of hearing SHHRSHR on sibilants and inner groove distortion. And - it delivered amazingly in that respect, no distortion, and all sibilants are as clear as a whistle. Problem - at least in my rig - the one that was described before - sound is thin, no bass to speak of, but PLENTY of highs. Actually, unbearable to listen to. Any suggestions on taming the beast? With TX103 transformer it is much brighter than when run directly into MFA Lumi, but still no cigar. My taste lies in Shelter 901 tonal balance...

My rig:

Oracle Premiere Mk IV, Adanalog air-bearing MG-1 parallel tracker, either MFA Luminescence or Hagerman Trumpet phono stages, S&B TX103 MC transformer.

[nrenter]

Have you played with your loading?

[Wayner]

The AT OC9ML II is spec'd to have 20 ohm load across the coils. Have you done this?

Wayner

later added: nrenter beat me by seconds, asking the same question.

[marknoir]

Specked 20 ohms is for transformer. TX103 has 5.6 ohms primary. Of course with transformers loading is tricky. I did play a bit with secondaries, loading it down with as low as 1k (!), which corresponds to about 2 - 3 ohms across the coils. Volume dropped by about 10 db, but thinness andhighs are still there...

[Wayner]

I think you have to include the load of all devices in the chain, SUT, interconnect...etc, to get the proper loading. A guy should have a LC meter and a pretty accurate (and deep) VOM to do this shit. Don't give up, if it were easy, anyone could do it, and it aint. The other thing that will change the load value is induction. The LCR formula needs to pay respects to all parties involved.

Some phono preamps have selectable loading for resistance and capacitance, and that just keeps getting to be like the hound chasing the rabbit.

Wayner 

[analognut]

Well, I have an AT-OC9MLII which I got new about a year ago, all literature & freq response graph included. The spec says nothing about a transformer. It says "Recommended load impedance 20 ohms". The cart seemed to take 100 hours or more to become fully broken in on my SME V/ Sota Star (vacuum) TT. I tried 8 or 10 values between 20ohm and 220ohm and decided 20ohm sounded the most natural. I could easily hear a small difference between 20 and 39ohm.

The frequency response graph shows it to be up by +0.75db at 20Hz, gliding down to become flat at 30Hz. There is a -0.5db dip at 35Hz and two more even smaller dips between 45Hz and 65Hz. My testing of a pinknoise track from a test LP into software in my computer did not show this to be the case. I found much larger dips in the response at the afore-mentioned areas. Don't remember how much, but my ears were hearing it as well when compared to my DL-103R.

Resonse graph and my testing agreed on the rest of the spectrum. From 65Hz to 6500Hz it was perfectly flat. At 7kHz it is up by +0.4db. From there it goes up to exactly +1.0db at 9kHz. At 12.5kHz it reaches a peak of +1.5db and stays nearly constant at that level on up to 20kHz. My ears were hearing this as well (at least up to 12.5kHz, which is as high as I can hear). This is quite enough to cause an unnatural-sounding high end.

I remedied the situation by running the output of my phono preamp into a 31-band graphic eq. It is an old Behringer Ultra-Curve DSP-8000 I had laying around. Dual 24-bit processors and +/- 15db of gain/cut in 0.5db steps. I then output the eq into the computer and again played the pinknoise track into a real-time spectrum analyzer in the computer just as I did before. I'm sure many of you can see where this is going and honestly I didn't expect the sound to be anything that would sound really great. I actually figured I was just wasting my time, which I had lots of on that particular day.

All that was left to do was for me to adjust the eq so the response I was getting at the computer formed a straight line. For pinknoise, if amplititude is plotted algebraically against frequency, which is plotted logarithmically, the result is a straight line when all frequencies are present at the same volume. I was able to get the response almost perfectly flat across the entire audio spectrum. I am talking less than 0.5db. Separate adjustments for L and R channels. Playing that first LP was astonishing. The sound was so good I most definitely left the eq in the signal chain and have no plans to remove it. In fact I can't imagine using my TT without it now. As far as I know there isn't a cart made that has perfectly flat response, unless it's one of those multi-thousand dollar jobs that I can't afford. 

[Jeff K]

From my instruction sheet. Purchased in November '09