[78-L] Co 113a Alignment issues

[Ron L'Herault]

Great to read this.  I like the way you think.

Ron L

-----Original Message-----
From: 78-l-bounces at klickitat.78online.com
[mailto:78-l-bounces at klickitat.78online.com] On Behalf Of Iñigo Cubillo
Sent: Monday, January 16, 2017 3:16 PM
To: 78-l at klickitat.78online.com
Subject: Re: [78-L] Co 113a Alignment issues


Thanks for your enthusiastic feedback, guys!
I love to tinker with gramophones, and my aim is always to try to make
modifications to correct design faults WITHOUT SERIOUS MODIFICATION TO THE
ORIGINAL.
With this machine, I have a personal fight... Not being truly exponential,
its sound is particularly meaty, round and full, and I like it very much. So
it is a constant pain for me to see its bad alignment, due to its short arm
and bad offset.
Now, answering your questions, for a given distance between the spindle and
tonearm pivot, there is only an optimum tonearm length, together with an
offset distance, to obtain the minimum possible tracking error across the
disk. Will not insist on this, for there is ample literature about the issue
on the web. The maths and geometry of this is very simple.

In my photos you may see that there is in fact one only point at the middle
of the record with zero tracking error (see photos named NULL  ERROR).

Usually, the ideal thing is, for a given spindle-pivot distance:
- a tonearm slightly larger; the surplus distance when you place the needle
pass the spindle is called OVERHANG, and should be positive. So the needle
must go beyond the spindle when you place the tonearm in line to it.
- a certain inwards angle to the soundbox, so it points backwards (i. e. the
side opposite to the needle) to a plot at a certain distance to the right of
the pivot; this is the OFFSET distance.

With these variables optimized, you can achieve a (-K) tracking error at the
innermost and the outermost grooves, and the opposite (+K) error at the
middle of the record. Ideally, there are two points (at roughly 1/3 and 2/3
of the record) where the error is null.  The longer are these distances, the
better the adjustment and the smaller the maximum error values. Ideally, the
tangential solution is a very large tonearm and spindle distance, which go
to the infinite...

But if you also fix the tonearm length, far from those optimum-aimed values,
with an "underhang" instead of overhang, which is the case with the 113a,
you have a problem. Still, you can try to modify the inwards angle (or
offset distance at the back) twisting the vertical plane of the soundbox to
minimize the error. That's what I intended to do with the "wedged" silycone
joint. This is simple to imagine: I created a rubber joint with variable
thickness so the diaphragm plane twisted inwards looking to the spindle. 
Adjusted "at the naked eye" I got a fairly good tracking, but the joint was
so thick at the back, and too compliant, that it caused other problems.

So I prefer to stay with the less bad solution: trying to optimize the
tracking at the inner grooves, for there is the hardest strain on the record
grooves.

Of course, when talking about the "tonearm length" one means the ideal line
going from the pivot to the needle. The inwards angle is the ideal angle
between the diaphragm plane and this line.

The tonearm length could be modified changing the vertical zenith angle of
the needle against the record, but this is not advisable, for the ideal 65
degree angle must be, more or less, respected for the better tracking.

One can also change the position of the soundbox at the tonearm neck by
adding a purposely made trunnion, with adequate length and curvature to
provide a certain angle (or offset distance) so to improve things. Playing
freely with this trunnion you can modify the tonearm length, etc. This is
still to be tested at home...

My pain is, for instance, when seeing a portable like HMV-101, which has a
fairly good adjustment (Long tonearm for the pivot-spindle distance, well
adjusted offset). So it could be done; why Columbia engineers did not make a
good design too?
HMV 102 has a short tonearm and a good set of parameters, but the
maximum/minimum error values are well noticeable. This has been well
optimized, but for a set of values (spindle-pivot distance and tonearm
length) very small, to make the case shorter, I suppose, due to the larger
size of the horn mouth. Compromise...

Hope to entertain you,

Inigo Cubillo 

_______________________________________________
78-L mailing list
78-L at klickitat.78online.com
http://klickitat.78online.com/mailman/listinfo/78-l