Dustcover Blues

I don’t know whether the SS SG phono stages account for this but I’d be surprised if they don’t incorporate some corrective filtering in the 500 to 2000Hz region.

@lewm The last time I heard one, it had no eq. The claim was that it didn't need it- which is misleading. And sure enough, the LP I used for reference (which I recorded) didn't sound right. I heard months later that EQ was being offered, but apparently that is incorrect??

The thing is, a strain gauge has really high output; some of them I've seen can drive an auxiliary level input. The parts needed to execute the EQ are inexpensive and small, and could be installed in a very small box that is part of the interconnect cable, since it can be done entirely with passive components.  So it puzzles me why this isn't done.

 It has to happen in conjunction with transduction of the cartridge signal.

Why is that?

RIAA EQ is +20dB at the bottom, -20dB at the top. SG does not require this so the statement is true. 

@millercarbon 

Not really... The problem is that by the time you get to 20KHz you're not going to be 20dB down. Please look at the actual RIAA de-emphasis curve:

See that flattening out part in the middle? That's what I'm talking about. A strain gauge simply bypasses the flattened out part and continues rolling off the output following the 6dB slope as seen on the left side of the curve- so the highs simply won't be right. Imagine that red line just continuing down at 500Hz instead of flattening. You can see the error that is introduced. In reality it won't be as far off as the red line would be as I just suggested since were talking about a curve using analog filter theory.

To properly EQ a strain gauge, a pole (seen at 500Hz) has to be introduced (a rising response) and then a zero (seen at 2100Hz) going back to flat, so the strain gauge can do its thing.

So imagine an EQ curve that is flat from the bass up to 500Hz, then rises for 2 octaves at 6dB/octave and then goes back to flat. That's the math of it; the actual curve won't do that- it would be smoother of course in a way that is similar to the smoothness of the de-emphasis curve you see above. If this is not done the EQ won't be right.

@mijostyn  Have you heard the cartridge? Theoretically it should out track any other cartridge but his MI cartridges track anything I can throw at them. I can correct frequency response problems but I won't get involved if there are not significant advantages in the design.

I don't know if its the same one but I have heard Soundsmith SG cartridges before. Tracking isn't what I'm talking about- I'm sure it tracks fine as the one I heard did. I'm talking about equalization. I provided the information you need to set the EQ right above. Can you dial that sort of thing in? If you do it in the digital domain, be sure to include an algorithm that includes the phase shift and frequency bump of analog filters.

Because of the way this works it requires no RIAA equalization. One of its many advantages over the conventional approach.

@mijostyn @millercarbon This statement is false. While it is true that the cartridge has an output that seems to complement the RIAA de-emphasis curve, from 500Hz to just above 2KHz (75uS) in the RIAA curve there is a bit where it flattens out. This causes an unequalized strain gauge device to be a little off. So they do need equalization to be correct but not nearly as much as a regular constant velocity device.

Its correct to say that it roughly approximates the needed de-emphasis.

Take a look at the RIAA curves- you'll see a 'zero' at 500Hz (318uS) (where the math says 'flat'- that's why its a 'zero' instead of a pole) and a pole above 2KHz (75uS) which re-introduces the slope. A strain gauge doesn't have that zero so what happens is it has the same slope across the band. This causes the highs to be and sound wrong. Its not glaring but if you know the recording its easy to hear.

My problem is occuring just above the cutoff frequency.

@mijostyn Filter theory says that the greater the filter slope, the greater the hump is prior to cutoff- and the closer to the cutoff frequency. A digital filter might behave differently (some do, some don't) but a simple solution might be to move the cutoff down a few Hz so as to get the turntable resonance out of the filter bump that occurs just above cutoff.

That should be easy- worth a try.

Look at my system page and you can see where the turntable is. It is on a granite slab sitting on a built in walnt equipment/record cabinet (which I made) which rests on a concrete slab. The uprights are 1 7/16" plywood. All the shelves are braced.

@mijostyn Equipment stands are really important. I'm using a Sound Anchors stand with anti-vibration platforms (Ultra Resolution Technologies, long defunct but highly effective; the stand was custom built to accomodate them) for both the preamp and turntable, but I found out 20 years ago that it still had problems- I had to place the whole thing on a set of bearings rather than use its points into the floor. This made a profound difference. Although my 'table (Atma-Sphere 208, which is quite 'dead'; you can thwack the platter as its playing and not hear anything in the speakers) has no suspension, after the bearings were installed I could run the volume up to 110dB (measured) without any ill effects from vibration affecting the turntable in any way.

Have you done any measurements to verify the effectiveness of your stand? I say this because I've seen stands from established manufacturers that actually made things worse!

The bass is due to the boundary effect.

Yes. But it appears you have a standing wave; if you were to break that up the boundary effect might be a bit reduced (that's how it worked out in my room anyway...). From how you've written about this, it appears that if you make just a bit of an improvement the problem would be gone since it only happens with certain cuts at higher volumes. I forgot to ask- what sort of stand are you using for the turntable?

You can see that up against the side wall the bass is way up

My surmise is that is your room boundary effect and may also be complicated by a standing wave (looks like you have cancellation in the center of the room). A distributed bass array might be really helpful. You'd only need a pair of subs to break up the standing wave. IIRC sounds like you already have them. If they are located by the Sound Labs  they are probably exacerbating the problem- they need to be elsewhere in the room.

Once the bass is evenly distributed you may not have this problem anymore.

I never had a dust cover for my Cosmos (it was serial number 0).

I suspect your alcove is a bass node. Have you tried putting the 'table in a different spot?

Whether its Helmholz or not isn't important. Resonance is.

This isn't the first time I've seen this. Since most high end machines aren't meant to be used with the dust cover down (if they even have one) the dust cover may well not have been included in the design of the turntable insofar as its resonance control methods are concerned.

IOW its the luck of the draw that it works or not.

If you simply grab the dust cover with both hands and try to squeeze it can you make the oscillation stop?