Cartridge frequency response specs

It seems that sometimes, but not always, a cartridge will be given two frequency response specs, one with db and one without - and at the Dynavector website, the descriptions even vary in this regard among their own cartridges, e.g.

The Karat 17D2 Mk II: 20Hz-100kHz; 20Hz-20kHz (+/- 1db)
Te Kaitora: 20Hz-50kHz; 20Hz-20kHz (+/- 1db)


10X4 Mk II: 20Hz-20kHz (+/- 2db)
XX-2: 20Hz-20kHz (+/- 1db)

Why is Dynavector (or Denon, etc,) giving us the non-db specs, and why just for some DV cartridges and not for others - and what is the significance of these way-above-20kHz frequencies as far as our listening experience goes, 'vibes'??
Specs are ADVERTISING. (Sorry to shout.) There are many ways to measure a component, and you can assume that manufacturers use whatever method makes their products look best--and that may mean using different methods, or reporting results in different ways, for different items in their line. Specs don't exist to help you choose among different products. They exist to make each product look good (and a little better than the products in the line that cost less).

So stop reading spec sheets and start listening. I know that's hard to do with cartridges, but it's the only way.
They are giving you the overall fequency response (wideband) in one spec(with no db tolerance shown)and also the linearity in the audio band(with db tolerance shown)with the other spec.
Whether the wideband response is significant, is a matter of debate.
As an information retrieval transducer, the audio band linearity of a cartridge is important. But it gives no insight into the sound characteristics of the cartridge. Or tracking ability, or resolving of detail, or many other things.

Anyone remember "Quad" sound from the 70's? They had to come out with super high freq. cartridges to pick up the rear channel data which was encoded in ultrahigh (beyond hearing) freqs. At least in one scheme. Response specs without dB's to tell you the max. variations can be misleading or useless. I guess it is nice to go up to 50k, like on my Grado, it may hint at some refinement, but what exactly it proves is a mystery to me. But analog master tapes can go higher in freq. response than CD's so maybe that is a good reason to go beyond 20k, whatever harmonics are there, maybe it will delight housepets. Maybe transient response is enhanced by HF extension beyond 20k in a ctg?
The frequency response of most cartridges vary DRASTICALLY as the load in the phono stage, tonearm cable and interconnects are altered. IAR covered this in Journal 5 to a very great extent. Moncrieff went so far as to take 54 high end cartridges and actually chart the frequency response and noise floors with various terminating loads. In all but a very few cases, the manufacturers recommended loading ( in terms of resistance and capacitance ) did NOT produce optimum results. By optimum results, i am talking about the most linear frequency response and / or noise levels. He did provide recommendations and charts for those cartridges tested though, so knowing this info can help one to extract the best performance from them. Only problem with this is that most of those cartridges are no longer made and nobody else is going to put forth the time and effort to do this with newer offering. Not even "analogue guru" Michael Fremer. For all the good that he has done trumpeting the benefits of the vinyl groove, i just don't think that he has the know-how or test equipment to undertake such a task. Sean
That is a very good point, Sean. Just out of curiosity, in low ouptut MC carts, did the high end roll off with higher ohm loads than specified, or lower ohm loads? I was thinking about experimenting with different loadings for my DL103, and would like some initial info on this before I start. I have read that in many cases listeners preferred much lower ohm loadings than spec'd by the mfr. I currently load my DL103 at 40ohms at the step-up transformer. Any advice on this?
A Denon DL-103 ? Yeah, you can start by throwing it out...

Did i have you going there for a moment ? Hahaha. Just let me know where you threw it to so i can come by and pick it up : )

Actually, the DL-103 has come in several different versions over the years. Moncrieff reviewed two of the 103's, the 103D and the 103C. From what i can tell, the C is a spherical stylus but i can't find the info as to what shape the D is. Maybe someone can help us out with that part.

Both ranked VERY well, especially for the money involved, with the 103D coming in ahead of the 103C. They also make ( made ) a DL-303, which was a revised version of the 103. However, Moncrieff's tests show that it did not do as well as either of the 103's mentioned here although it did not do nearly as bad as some other extremely reputable cartridges.

With the 103D, Moncrieff states that output level starts falling as you go below a 100 ohm load. For that matter, the same thing happens with the 103C and the 303. He basically recommended running a lower resistance than what the internals of the cartridge itself measured i.e. appr 36 ohms. He thought that about 20 ohms was a good balance between tonal balance, low distortion and output level. While he did comment that going lower on resistance to about 15 ohms would lower distortion measurably, it did this at the expense of reduced midrange liquidity and neutrality. Depending on the gain of your phono stage and step-up transformer, you might not even be able to go low enough in resistance and get a decent signal to noise ratio.

Moncrieff also recommended loading the cartridge with appr 1400 pF of capacitance. According to his notes, this tamed the high end peak that the cartridge displayed and presented the most linear "average" frequency response. After looking over the supplied graphs, it looks like it did lower the noise floor quite a bit but also created a bit of lower treble / upper mid suck-out. The high frequency peak is still apparent but is drastically reduces in amplitude. As such, you might want to play around with the capacitive loading but 1400 might be TOO much for your liking.

The other comments that he makes about this cartridge state that it works very well at low to medium level recordings and that it has excellent transient response. The one major flaw that he comments about is that as output levels increase, so does distortion. This is not uncommon though, as some cartridges go haywire with a loud passage. One of the most beloved and highly reviewed MM cartridges is VERY guilty of this "crime".

For the 103C, he also recommends trying to run down the resistance as low as possible while still retaining enough output to drive the phono stage. This cartridge clocks in at an internal resistance of 39 ohms, so it is not that different from the D series. Judging by his comments, i'm thinking that somewhere around 20 ohms or so should be the magic number. He recommends loading it with 400 pF of capacitance, which looks like a good match according to his graphs.

The 303, which was supposed to be a "new and improved" version of the 103, does some things better but a lot more things worse. It measured 41 ohms internally but ran best at about 30 ohms or so. Moncrieff recommended a whopping 2200 pF on this model, but going by what i can see on the graphs, it looks like that is a little over top and drops the top end down too much. Then again, he did the testing and looked at all the results, so i'm sure that he knows what worked best.

Given that all three of these cart's work best with a lower load resistance and some added capacitance, you might want to drop your load resistance and invest in some decent caps of various values in progressive steps. Try working with a 200 pf, 400, 600, 800, 1200, 1400, etc... You should be able to find something that gets you in the ballpark. Once you find the appr value, you can then delve into caps in that region divided by smaller increments for fine tuning. Once there, you'll just be adding more icing to the cake.

Hope this helps and answers your question : ) Sean
Sean, buddy, that was super excellent info! Thanks! I will play around with that because I am getting a tinge of what sounds like distortion at really tough dynamic high-freq passages. It may be that distortion that you mentioned that could be tamed by slightly lower load. I am just running it "free" on the capacitance right now, but I'm going to try some caps to see what they do. Thanks, again. That was really, really useful info!