Optimal loading for the Orpheus

Hickory, the thing is that if you are looking for a load anywhere from 0 ohms to say 2000 ohms, you can do fine by just inserting a loading resistor of the desired value -- since having a 47,000 ohm resistor in parallel with it won't reduce its value by any significant amount, as your calculation illustrates. BTW, the formula you give only works for 2 resistors in parallel -- there is a more general formula for multiple resistors in parallel.

I have often stated that a good place to start loading a given cartridge (getting into the ballpark) is to multiply 25x the cartridge's internal resistance/impedance. The optimum will vary, but not more than +/- 50% of the 25 multiple. Raul and I agree that the best procedure for discovering the optimum load is to listen to the bass while starting at the low end of the loading range and working up to higher resistances. If you do it this way, the bass will go from looser to tighter and then as you go too high (in load) the bass will begin to weaken (thin out and lose strength) the mid/high won't change (except in relative terms, to the bass) that much throughout the procedure.

If the load exceeds a cartridge's optimum by even a small amount (say running a cartridge at 800 ohms where it should be optimum around 100 +/- ohms) it will be hard to detect because it will sound pretty much the same (bright) from 800 all the way up to 47,000!

Dear Hickory: What Nsgarch posted: +++++ " the optimum load is to listen to the bass while starting at the low end of the loading range... " +++++

is a very good key about and I totally agree, of course it is not the only one.

Regards and enjoy the music.
Raul.

Thanks for the great advice about the variation in theresponse of the bass to loading. I have tended to listen to the highs, e.g. avoiding excessive brightness and edginess with high loadings. I have often started with the 20x - 25x priciple for loading relative to the output impedance. To be honest, this has been one of the best predictors for getting fairly close. Also, I rarely fool with resistors in parallel. I have broad selection of resistors (albeit different brands hence different character), mainly in the 40-1000 ohm range that I can usually define a good and fairly narrow window for loading. Sometimes I will use resistors in parallel just to get to a better defined loading. In the end, I have gravitated to ordering Vishay's resistors that appear in the middle of the desired window. These have been the most neutral and transparent from my listening and generally very satifactory.

Hi Hickory,

Nsgarch mentioned concentrating on the change in bass response and you concentrate on the high end. Both extremes are critical. Because I expect a good MC cartridge to deliver an open and airy top end, I am particularly concerned with loading that maintains enough top end response while avoiding excessive brightness and sibilance. I tend to, like you, concentrate a bit more on the change to top end response.

Just to clarify things, "high loading" is where you use a lower value resistor. Whether you are using one resistor or several values, these resistors are in parallel with the signal from the cartridge and act as a voltage divider -- a high value resistor mean LESS current is flowing through that path and more is flowing along the original path. Hence, a very high value, like 47k or 100k acts almost as if that path is not existing at all.

Another thing worth mentioning is that changing the loading of the cartridge has a similar effect on tonal balance as changing VTA of the cartridge. A high VTA tends to sound somewhat like lower amounts of loading (i.e., a high value of loading resistor). So, to some extent, you can juggle small changes in VTA with changes in loading to get an optimum combination. This adds to the complications, but, it also adds to the possibilities that an ideal combination can be found. If you have a tube phonostage, you can throw into that mix changes in brand/model of tube to further complicate things.

Great points, Larryi. The VTA/SRA and loading have always presented a vexing challenge since they tend to track/compensate as you indicate. Thus, among the challenges of analog is getting all of the variables (e.g. loading, VTA/SRA, VTF, crosstalk) 'optimized.' I seem to be constantly fiddling with all of these, probably never really getting it exactly right (whatever 'right' is), but gravitating toward a sweet spot, to meet my aesthetic appeal. However, this is also one of the joys of analog - you can personalize via the tweaking.