• Please note that Audiokarma will be offline briefly for updates early on Monday morning, September 28th.

Experimenting with Simplex and Shelter 501

watercourse

Member
I am posting the outcome of my recent remounting of my Shelter 501 mkII cart, and what I've learned that has gotten me the best sound yet from this cart on my Simplex.

WARNING: For those of you that believe that the only way to set up and align a cart on a WTL table is via the "approved" method (19 degree offset, golf ball centered in the cup), then I respectfully suggest that you read no further, however, you are of course welcome to do so. By posting this I am not implying that the "approved" method is incorrect, or that the "previously approved" method (the WTL one-point protractor) is better. What follows is simply my experiences using the both the "approved" and "previously approved" methods, and reporting back sonic differences - for collegial discussion and the betterment of the WTL society.

As background, since getting my Simplex, I have used the one-point protractor that was provided in the manual to align carts. A very simple process which only required rotating the arm pillar and adjusting the offset angle of the cart in the headshell parallel to guidelines. The golf ball was never centered in the cup using this method. (NB: readers of the Simplex manual will notice that the requirement of a centered golf ball is never mentioned, so this is ostensibly not a problem.) The Shelter aligned in this manner sounded good, as good or better bass than I'd ever heard from it on other tables or with other arms, but it did have some edginess in the treble which did not make it my primary cart for the Simplex.

Then, on another forum, I was informed that the WTL protractor I had been using went the way of the dinosaurs. I was to simply use the headshell alignment tool to ensure it was at the proper 19 degrees, and align the cart to the headshell and be done with it. Even easier.

Confession: For some reason, I thought that the grommet was at a factory-set location and not to be moved. Since in the past using the protractor, I didn't pay attention to where the golf ball was positioned in the cup, I never needed to move the grommet except for azimuth setting. After emailing Mike Pranka, he gave some helpful responses, which were to adjust the grommet position and arm pillar rotation in small increments to center the ball in the cup, which was easily enough achieved once I moved the grommet. Duh.

First Run: So how did the "approved" alignment sound? While some outer groove distortion appeared to be addressed by the "approved" alignment, there was increased distortion and slight obscuring of detail from the middle of the LP to the inner grooves, and IGD was probably the worst that I'd heard from this cart. No amount of adjustment of VTA, azimuth, tracking force, or damping could address it without giving rise to some other sonic issue. However, there was a pleasant bass fullness that was not there before, and on outer grooves, there was improved treble response, and spatial qualities were better than before.

Since what this was not a clear improvement from prior setups, I went back to the protractor. However, instead of moving the pivot point like before, I kept the golf ball centered (in effect, splitting the difference between the WTL methods). It could not be fully aligned parallel to the protractor guides now that the golf ball was centered, but the guidelines told me which direction I needed to rotate the cart. This meant rotating the cart inward as far as it could go, but still just missing parallel.

Second Run: Guess what? The IGD was improved, but there was less sparkle in the middle grooves, and imaging, depth, and separation were all poorer than the "approved" alignment.

Thinking I could still improve performance, I took some measurements of spindle-to-pivot distance and overhang, and plugged these into the Alignment Calculator Pro at Vinyl Engine. Using the calculator, I modeled the potential differences in distortion characteristics when changing overhang and offset angle to see how I could achieve a more balanced distortion curve with null point somewhere near the middle of the grooves. What made me think of this? Given the deteriorating performance from middle grooves inward, I wanted to see whether putting the null where I began to hear problems might be an option.

It turned out that overhang and effective length were not going to help much (see #2 below) , but changing the offset angle to 20 degrees would give me the curve I was looking for (see #1 below). This meant keeping the golf ball centered, and turning the cart inward by one degree. So this really was "splitting the difference" between Firebaugh's two setup techniques. So I aligned the cart to 20 degrees.

Third Run: And so how does it sound now? The bottom line is that I do have a much better balanced performance from this cart now: nice micro dynamics, imaging, coherence, and snap across the grooves, but with added depth, spatial qualities, greater body, and the musicality and boogie that makes the Simplex a winner.

I know that this is not staying within the WTL box, but so be it. Firebaugh thinks out of the box too.

The Nitty Gritty
FYI, the measurements were as follows, in case you'd like to input them as well into the model at VE:
  • S-P distance is 226mm, which I measured using the Feickert tool when the golf ball was centered in the cup.
  • Overhang for the Shelter was measured to be 13mm (at 19 degrees, but this will change insignificantly at other offset angles +/- 1 degree from 19).
Of interest, you will notice that Firebaugh uses an ideal offset of 0.5" or 12.7mm in his Amadeus arm design. My "split the difference" alignment (appropriately called "Wilson" in links #1 and #2 below) meets his criteria of max tracking angle error of <5%, but both WTL and my alignments yield max distortion of >1%. Figure 2 in Firebaugh's blog entry implies that based on the max tracking angle error from the model outputs, 2nd Harmonic content (which in the narrative is equated to distortion related to tracking angle error, i.e. one and the same) should be somewhere below 0.05%. This does not agree with the VE models, however, the proof really is in the listening.

Feel free to investigate this yourselves if you have the inclination, or if you run into problems aligning your carts. I've also included links of a few investigations that I used, and their distortion curves compared to "approved".

Happy listening!

  1. Here is a comparison of different offset angles (19 vs 20 deg) and the effects on distortion, with Firebaugh's ideal cart with overhang of 12.7mm and my Shelter's 13mm.
  2. Here is a comparison of different mounting distances of the golf ball.
  3. Here are some other hypothetical mounting distances "bracketing" the centered golf ball, which I may investigate in the future.
 
Last edited:
Register to hide this ad
Further thoughts for discussion

Looking at the links to the VE model outputs I provided,

  1. One can see the minimal effect that mounting distance has on the overall distortion curve. Meaning, golf ball centering is not critical.
  2. Similarly, overhang has minimal impact on distortion.
  3. Given 1 and 2 above, it follows that the arm's effective length also has minimal effects on distortion, given the relationship between mounting distance, overhang, and effective length.
  4. Offset angle, on the other hand, has a significant effect on the amount of distortion, the character of the distortion (curve shape), as well as where the distortion will occur across the LP radius.
  5. Given these model outputs, one might pay closer attention to the headshell angle and setting the desired offset angle than any other parameter (VTF, VTA, and azimuth should be considered by themselves, and is not being discussed here or considered in the VE model).
At the same time, none of the modeled distortion curves should ever replace critical listening. Especially if one accepts that tracking angle error distortion is primarily second harmonic, and therefore can be perceived as "musical" (see Firebaugh's blog here).
 
One thing I noticed with the Shelter 901 on my WTA, don't know how different it is compared to the 501, was that it liked more damping than less. Don't overlook the effects of the amount of damping in the arm to the overall sound of the cartridge. I typically run more on low compliance MC carts and for a high compliance MM cart like my V15V, I run the minimum (ball just in fluid.)

Sent from my cm_tenderloin using Tapatalk
 
One thing I noticed with the Shelter 901 on my WTA, don't know how different it is compared to the 501, was that it liked more damping than less. Don't overlook the effects of the amount of damping in the arm to the overall sound of the cartridge. I typically run more on low compliance MC carts and for a high compliance MM cart like my V15V, I run the minimum (ball just in fluid.)

Sent from my cm_tenderloin using Tapatalk

brminpin: Yes, I have much more damping than "normal", just under 1/3 of the ball in the fluid, rather than just "kissing" the fluid, like with my other two carts. Well, maybe a bit more than just kissing... french kissing?

bacobits: Does the ball kissing the fluid feel better to you too? Who let this guy in the thread anyway?
 
Last edited:
Is it not possible to align according to a more standard scheme (Lofgren or something)? That would seem to cut average distortion by a TON...
 
andyecon: probably have enough adjustability to reach Stephenson, but the "standard" alignments don't have the fullness in sound of either the 19 deg alignment, or my modified alignment.

In other words: distortion numbers only tell part of the story.

If you look at the Firebaugh blog, which discusses distortion as primarily second harmonic type, you begin to understand why listening is just as important.

Firebaugh is not the only one to comment on the nature of tracking error distortion. Read Gilson's 1981 paper here, in the "Optimization" section. Short paper, but full of interesting stuff.
 
Last edited:
Back
Top Bottom