Audio-Technica launches new entry-level VM95 cartridge series

She's set up and singing sweetly. Did nicely on the torture tracks even with a stiff, new suspension—3rd track was remarkably grit free. I'll let it break in and then take some measurements.

 

Specs from the manual. These are a little different than some I see online. Lateral resonant frequency off the test record using very light hardware in an RB300 is focused around 10 Hz. Vertical is a little higher, maybe 11 or 12 Hz. Makes me wonder about that dynamic compliance spec, but I'll check it again later and see how it settles in.

(thumbnail)

 

I look forward to seeing the frequency response you end up with, jupiterboy.

The reason I haven’t posted any graphs is partly that I’m not sure how much I can trust the 2002 Hi-Fi News pink noise track. In your post #16, I notice the same right channel boost (5dB) above 10kHz. I also get a 1dB boost at 200Hz, plus I think I also need to increase the bass response at 50Hz. If I ignore the treble result, can I trust the other two as being correct? Or are they just another artefact of that recording? I’m hoping your graphs will confirm the problem is with the record, and not with my setup.

 

Yeah, this is a good line of inquiry, so to speak. I've read of other issues with this record. Mine is also a little off center. I do get much better looking graphs from it than from the Cardas pink noise.

I'm going to use an audio app to evaluate the resonance sweep as I've read it is 2 db off before. I'll give it some more time before I post up, but I'm pretty sure at this stage of some basic impressions.

The old AT95 body produced a little extra energy in the 100-200 Hz area. I think the new body is more even handed in this regard, and seems to really like the 800 Hz area. It sounds more a part of the AT line—AT150MLX and 440 series. It is a really good diamond, and a nude for this price is pretty sweet. Right now it is going from a little flabby and congested to thin sounding. It's not going to be a giant killer, but the better separation, the brass threading bits and the nice stylus are going to make this a very solid value cart. Notably, the bass is full AND articulate so far, and I have pretty lean speakers with a acoustic suspension mid-woofer. I've given a quick run at 38K and 62K and like the AT95, traditional loading of 47K or 50K in my case, sounds balanced and solid. At any rate, it is one of the best trackers with my RB300.

 

Jeeze you guys have this stuff sorted out... I just put them on, line it up and go.

 

If you want a nice test record, free from the defects we often get on new pressings, look for an old CBS STR100 in mint condition. You can probably find them quite inexpensively if you hunt around. I paid about $10 for mine.

 

So I would say it is very comparable to the old 95 in terms of frequency response, but so far the sound is considerably different. Make of it what you will, I feel like some of the high frequency anomalies are related to system noise and/or the inexpensive USB interface I use. This is with the micro-line stylus.

 

Green/Red equals??

 

L (G) R (R) channels. This exonerates the test record a bit.

 

Can you overlay computer-generated pink noise on that graph?

 

Possibly, but maybe not with my interface. I might sum to mono and also run a line from a digital source. I've got a three-degree slope applied so it should be flat in an ideal scenario. I've also bypassed the rumble.

 

Thanks for the graph, jupiterboy. How are you loading the cartridge (resistance / capacitance) in the phono stage?

 

That's 50K with 100pF cap total including wires.

 

Just note that the Hifi News test record has some anomalies for the pink spectrum, so the HF region is not that accurate. For modern test records the best spectrum is from the Ultimate Analogue test LP and the white + pink spectra from the Elipson test record.

 

Can you expand on that?

 

Here are my measured results for the AT-VM95ML. I’ve settled on 103k Ohm loading with no “extra” capacitance on the phono boards. Since a 1m phono cable with arm etc. will add roughly another 100pF, I assume this is the actual loading capacitance. It has already been observed that the Hi-Fi News pink noise track is not exactly accurate, so the following frequency plots cannot be relied upon as being correct. Note that AT recommends 47k and 100-200pF for this cartridge. Blue is the right channel.

A3New103kNoCap.PNG

Here’s the response using 52k Ohm, roughly as jupiterboy used. I found it rather dull sounding and this is evident from the following frequency response.

A3New52kNoCap.PNG

I did try adding extra capacitance, but I couldn’t get anything more satisfactory than the first graph shown above. I would prefer it to be a bit brighter though.

The following plots are theoretical (simulated using MultiSim), using my phono stage (a Naim-like MM) attached to a 560mH and 487 Ohm DC resistance cartridge, which are values close to the actual AT body (550mH and 485 Ohm). They give a rough prediction only! Note that reducing the load resistance (lower Ohms) damps (reduces) the treble peak caused by the capacitance. Also, the higher the load resistance (more Ohms), the wider the frequency response. Again, the load capacitance shown includes the arm capacitance, so subtract 100pF to get the capacitance added at the phono stage.

52k Ohm 100pF:

ATa52k100pF.PNG

77k Ohm 100pF:

ATb77k100pF.PNG

103k Ohm 100pF:

ATc103k100pF.PNG

52k Ohm 147pF:

ATd52k147pF.PNG

77k Ohm 147pF:

ATe77k147pF.PNG

103k Ohm 147pF:

ATf103k147pF.PNG

Please compare those with the actual results I measured in the first two plots.

 

One thing is that different test records pink noise do differ, and the Ultimate Analogue test record pink noise is quite the same as its fixed frequencies. Here is a the Ultimate Analogue test record in red, with HFN and Cardas pink noise also in the same figure:


If you compare the fixed frequencies on the Ultimate test record with the pink noise, you will get quite similar curve:



Now I have a different load to reduce peaking somewhat and just looking at the white noise spectrum from the Elipson record it is also quite similar to its fixed frequencies and also to the Ultimate Analogue test LP.

White noise, Elipson.

 

Here's my graph of the original AT-95 vs a Denon DL-110. Contrary to expectations this is a pretty mellow sounding cart with a dip that peaks around -2dB in the upper midrange and treble. Measurements taken with spot frequencies from an old CBS Test Record.

 

Nice comparison. I am however quite sure that with some fine adjustment of load you get within +/-1 dB in the highs with the AT-95 also.

 

As this breaks in, I'm finding 28K better than my first impression would have suggested. Looks, by charts above, that 47/50K are pretty flat.

 

I should have said 38K.

 

Chart above is into a 50K load. Sorry should have mentioned that.

 

What are those graphs showing?

 

It is the frequency response 1 kHz-20 kHz of a cartridge using a converted pink noise track from three different test records, showing that you get different results from different records.

 

So the "most correct" appears to be the HFN one? I'm not really sure what I can conclude from those graphs.