Step #2. Drive.
Before I get on with this, a moment of honesty to clear my consciousness. This
improvement step has been driven by pure vanity. When I initially designed dŵr, I wanted to hide all moving pieces out of sight, wherever possible, so that the visual aspect of the device would focus on the rotation of the platter. I wanted then (and still do now) to cover the motor spindle and belt pulley, so that only the belt is visible, akin to some Micro Seiki models.
Alas, time and financial pressure to finish the turntable, as well as inability to do some machining methods, led me to a simpler pod design, where both the pulley and the mass balance, including screws are visible.
In order to see what I was working with, a snapshot of the old motor pod:
Machined in three separate components (not including the pulley) with three different materials. The pod itself is machined out of aluminium (6082), matt anodized. The mass balance is machined out of mild steel (4130), nickel and chrome plated, and the motor mount and the pulley machined out of polyoxymethylene (you see a pattern here
?).
Quiet, vibrationally stable, quite efficient. Not ugly by a longshot, I hope you will agree. Yet, the aesthetics had been gnawing at me.
Breaking down, here is the new dŵr motor pod.
This latest monoblock approach comes in two components (apart from the pulley), and two different materials. The pod itself is machined out of aluminium (6082) with internally grooved adhesion fins for the motor to facilitate cooling, whereas the motor mount and the pulley are machined out of polyoxymethylene (POM).
The vanity element is an isolated aluminium cover which hides the pulley and belt contact, as well as all visible joints, matt anodized, and damped
(*) with rubber foam and silicone drops.
(*) See below for issues with the cover.
This pod has the same outer diameter (120mm), with a slightly reduced internal diameter (by 22%) and an increased height (by 14%) yielding an increased total volume, and approximately the same overall mass as the older pod did with the steel mass balance included.
Power is courtesy of the same type of cable (
Tasker C211), apart from the power connector now being a Neutrik SpeakOn
NL4FC, taking the place of the earlier Neutrik XLR
NC3MXX. I realise that a more appropriate connector set here would be the PowerCon series, yet I did not use it for the sole reason that the PowerCon series can only accomodate cables with a diameter larger than 8mm. I have justified the use of the SpeakOn here due to the fact none of my other equipment use SpeakOns.
Expectations
As mentioned earlier, this has been a pure vanity step. I would have been happy only with having aesthetics near what I wanted. That said, given the thought I put in this (and funds), I would say I expected at the very least (1) better motor cooling, (2) reduced belt slip, and (3) reduced overall noise.
Luckily, we can measure all this.
Results
Nothing to break in here - I started taking measurements approximately 30 minutes after the motor was installed. The cool down period was a bit longer due to the fact that I had to open up the electronics box to install the replacement Neutrik SpeakOn
NL4MP.
Subjectively:
- The motor is quite cooler to the touch after working for a few hours;
- Belt slip has been reduced, which may also be due to the revised position of the belt on the platter (slightly higher up);
- Looks better to me (mission achieved!).
Objectively:
Temperature measurements were taken with the help of a laser, contactless thermometer:
- Measurements took place over a period of two hours (120 minutes) within 15 minute intervals;
- The motor was turned off for at least two hours prior to measurements;
- The latest working measurement is in the 105th minute, after which, the motor is turned off in order to establish cooldown performance in the last 15 minute interval;
- Three points of measurements are considered - (1) drive pod, (2) top motor mount and (3) pulley;
- Each individual measurement point, at every location is averaged between three individual measurements, in the hope to decrease (if not eliminate) the differential error present with the used measurement device;
- Ambient temperature held at steady 22.5 degrees Celsius, controlled via an OpenTherm enabled room thermostate.
Prior to discussing the results, a few words around the importance of temperature measurements. Belt drives, especially in my case, rely on a
soft joint. The intrinsic properties of this type of joint inevitably vary dependent on ambient parameters, including (but not limited to), air temperature and humidity. Variations in these parameters, especially temperature, lead to changes in (1) pulley diameter, (2) belt elongation and (3) joint stability. These variations then translate to speed fluctuations.
And we really don't want that.
dŵr, by design, has an integrated compensator for these changes through the aforementioned algorithm, but if one can control external parameters, there would be less strain to the electronics, the motor will be relieved of quite a bit of fluctuations, which would in theory extend its lifetime.
With that out of the way, onwards to the truths that measurements yield.
1. Pulley
This location is important due to the direct contact of the pulley with both the motor spindle and the belt itself.
Difference are most easily discernable here and I daresay, drastic. A total drop in temperature for the pulley is evident across the spectar, and thermal management is more even. The
average temperature drop is 8.6 degrees (°C), whereas the
maximum observed difference is 15.4 degrees (°C).
2. Motor Mount
Also dubbed the cover in the graphs - I realised quite late in the game some confusion may be had with the actual vanity motor cover. To be clear, I mean the motor mount machined of POM here.
This location is important due to the full contact with the top surface of the motor.
Differences are smaller, but not insignificant, towards a better cooling solution. A total temperature drop is evident, although the temperature curves between the two solutions are quite complementary, both in their rise, and the expected plateau. The
average temperature drop is 3.4 degrees (°C), whereas the
maximum observed difference is 5.6 degrees (°C).
3. Pod
This location is important due to thermal management of the motor itself.
Differences are smallest here - quite minute due to the larger dissipation surface. Also important, the vertical axis is zoomed in to the area between 20 and 30 degrees (apart from the earlier 20 to 50 degrees) so that differences can be more easily observed. Crucial to note is that the plateau is reached earlier, and maintained on a lower position. The
average temperature drop is 0.3 degrees (°C), whereas the
maximum observed difference is 0.8 degrees (°C).
4. Vibrations
What follows is a display of vibrational measurements for a ~600 second (10 minute) rotation cycle (at 33 1/3 rpm), for the system with the old
(left) and the new
(right) drive pods (including different pulleys).
Z axis aligns with the central rotational axis for the turntable.
Onwards an upwards.
P.S. I had mentioned earlier that there were some issues with the vanity motor pod cover. Anyone who has ever cupped their hands to their ears to form a primitive horn can see how that would be an issue. The aluminium cover acted as a horn and amplified the pulley noise to the point of bother during quiet passages. Once I discerned where it came from, a few minutes of scissor work and a bit of rubber foam cured that ailment.