Peter Daniel said:
Hi Peter,
Thanks for the pics, presumably of a more modern Shallco switch, which are interesting to see.
The riveted sides of the ring of multi-contacts on those Shallcos which I have are less substantial in thickness so I couldn't afford to reduce this by very much when I scraped them for testing purposes years ago, or it would have impaired the integrity of the fixed contact being 'attacked'.
On the other hand, the actual contact sides of these multi-contacts on mine do have rather more meat on them but this is only possible as they are spaced further apart, this being permitted by fewer contacts per switch.
There is certainly no apparent junction-line part-way down the sides of the 'pill-shaped' multi-contacts, as perhaps could be the case in those helpful pictures of yours.
I still believe that on mine they are all solid silver throughout, but it is no big deal since we are now told by the makers that this is not necessarily the case with more recent switches.
What is more important in my view is the fact that whatever metal is used, there are so few junctions between input & outputs, no dissimilar metal *riveted* joins anywhere, and the quality (and mechanical engineering) of the entire assembly is so robust, with a short signal path, and using high quality metal of some kind of silver-alloy for the wiping contacts.
Regards,
Don't know Peter's opinion but i am seriously disappointed in Seiden. Other than great tactile feel there is hardly any serious comparison to Shallco. One of my Seidens developed some horrible greenish oxidation on the contact surfaces in a very dry climate. Never did any sonic comparisons against Shallco but they just don't last without periodic cleanups.
I'm not familiar with Seidens, I got this one recently, included with S&B transformers and didn't even try it yet.
They have larger switches and picture below shows one used in Denon preamp.
Some additional comments on Shallco: http://www.diyaudio.com/forums/showthread.php?s=&postid=914458&highlight=#post914458
They have larger switches and picture below shows one used in Denon preamp.
Some additional comments on Shallco: http://www.diyaudio.com/forums/showthread.php?s=&postid=914458&highlight=#post914458
Attachments
EUVL said:
Very nice feel. Nice light movement. A snik vs. a clunk of the Shallcos.
Can't compare it sonically to others. It does develop a little noise with motion over time, but turning the knob back and forth can wipe it clean.
I do have some Shallcos that I may try after reading this thread. They have the post terminal rather than the rivit type.
I tend to like very large diameter knobs, so I can turn the Shallcos ok, and the Shallcos do seem more bullet proof. You can also get them custom made. But desoldering 66 TVC leads from the Seiden and transferring them to the Shallco is not at the top of my to-do list.
Bobken said:
Hi Bobken,
I sent the above information to the lead engineer and this was his reply:
~~~~~~~~~~
We made switches for Audio Synthesis about 15 years ago. The contacts
panels used for the 12735-2 referenced in your email incorporated the rivet
type contacts that we commonly use for our attenuators; however, the contact
was a .030" silver headed contact just like we use today with the turrets.
~~~~~~~~~~
So clearly they keep records on what they make. And my eyes are still calibrated pretty well -- my "guesstimate" on 0.030" on the silver "cap" was exactly correct! (I'm sure there's a smiley for patting myself on the back, but I don't have time to go through all 216 to find it...)
I don't know how to explain the apparent discrepancy, but in the end it's not too important -- we both agree that they are the best sounding switches available (at least for applictations where they are used regularly!) and very nearly transparent in a bypass test.
And I have to say that I *really* like the 60 position parts we are using in our latest preamp. When we got into making preamps, about the only company using a stepped attenuator was CAT. They had a 24-position from Greyhill (I think). We upped the ante with our 46-position, and now with our 60 position I think the thing is bloody marvelous. (Now where did that "patting myself on my back" smiley go?)
Cheers,
Charlie Hansen
Peter Daniel said:
Close, but not quite. The second photo of the "business end" shows the silver "caps" welded on top of the copper rivet. The rivet would be useful if one wanted to insert some resistor leads directly to make a hand-soldered attenuator, for example.
But the first photo showing the backside of the "rivets" is not what we have ever used. Instead, there is a solid copper "turret" (I would call it a post) that we insert directly into our main PCB that contains the attenuator resistors and all of the rest of the circuitry, including the input and output connectors. The result was what I liked to call a "wireless" preamp!
Bob Cordell said:
When we were testing the switches we were using a D/A converter made by Jocko Homo that had a variable output with an output impedance of around 50 ohms, I would guess. This fed a power amp with an input impedance of 10 kohms. So not a heavy load, but not too light. (Our current designs have a 1 Mohm input impedance!)
As far as the "psychological" part, I have found that to be silly and a waste of time. I've tried very controlled tests, I've tried tests where I've deliberately tried to fool people, I've tried all kinds of things. In the end, it's not hard to tell what's going on if you have a careful listener that is intimately familiar with the system under test and the specific source material being played. I've found these last two items to be the most important items for conducting listening tests.
As a recent example, we were comparing the sonic difference of taking the 64-bit output of a digital filter and reducing it down to 24 bits for input to a DAC chip. We tried three methods:
a) Rounding.
b) +/- 1 LSB TPD dither.
c) +/- 1/2 LSB TPD dither.
The engineer who programmed the digital filter knew which DIP switches on the rear of the unit would create what type of bit-depth reduction. I had no idea. I told him to see if I could guess. So I listened and got all three correct.
The odd thing was that the audible differences were so obvious. I would have thought that the noise in our circuitry (not to mention the limitations of the DAC chip!) would have made it so that *anything* you did to the 24th bit wouldn't matter. But there you have it. Another case of easily audible and difficult to explain.
Cheers,
Charlie Hansen
- Status
- Not open for further replies.