Iron Clad Solution to TT Motor Vibration

[thrifttrust]

I've been playing around with Hurst 24V, 300 rpm (250 in Europe) motors for a turntable I'm designing. Like others here I'm concerned with the 240 Hz. (200 in Europe) vibration characteristic of this type of motor. I discovered that merely touching the motor to my cast iron drill press vice diminished the vibration significantly, I gently clamped the motor into the vice. The vibration became virtually inaudible and was barely perceptible to touch. Now I'm thinkin: a cast iron motor mount and a cast iron flywheel attached to the pulley? Eh?

This is my initial test setup. The capacitor bank allowed me to trim for the lowest vibration. The best value was near 6 mf as opposed to the recommended 10. Incidentally, the best value also displayed the most symmetrical waveform. I used an old ceramic phono cartridge as a vibration sensor.

This is the resulting waveform. It is a surprisingly clean sine wave. It wasn't this clean everywhere on the motor top plate but the cleanest spot was also the highest amplitude.

This is the motor in the vise. I'm applying very little pressure, I got similar results with the vise horizontal, gripping the motor by the sides.

Waveform with the motor in the vise. A 20X reduction in amplitude!

 

[sliceofhogan]

now what - run for vise president ?

 

[K.A.B]

Hello!

Your observations made are absolutely correct a Swedish manufacturer of the turntables Inertia had the solution with cast iron cup around its motor. I had their Inertia BDT 1 in a few years before I got a Thorens, there is a manual on vinyl engine Inertia BDT-1 Manual - Belt Drive Turntable - Vinyl Engine

Unfortunately I have no pictures but I will see if I can find any service manuals or something that shows what it looked like, I will return.

Anders

 

[Hiten]

Hi,
Thanks for the measurements. I have felt motor vibrations on non serviced garrard 401 (not mine) transferring to chassis and plinth; and have wondered if motor can be put on separate second plinth in tightly firm position. For belt drive wouldn't vibrations be less if motor is rubber mounted and with belt in place? As the belt and motor fixtures will somewhat hold tight the motor in place. Measurements after fitting the motor and belt-platter would be interesting to look at. But still the result of motor with vice is amazing. Thanks again for sharing.
Regards.

 

[user510]

next test is to run the motor clamped in the vise...except put your 'sensor' (accelerometer) on the vise.

-Steve

 

[thrifttrust]

Thank you Anders. It's interesting to see that at least one manufacturer has exploited the vibration absorbing properties of cast iron. That begs the question: Why haven't others done so? It's cheap, adds heft to the TT and sounds great as a marketing buzzword. I checked out the Inertia TT users manual at vinylengine but unfortunately it's in Swedish. I googled it but found nothing.

Anyway, I thought I'd test my much larger 59 mm Hurst 24V 300 rpm motor.

Standing alone the larger motor had a third of the vibration of the small one, and the signal was overwhelmingly 120 Hz as opposed to 240 in the small one.

This surprised me. One of the reasons I purchased the small motor was that I suspected that a smaller motor, with a less massive rotor, would exhibit less vibration.

However, my faith was restored when I clamped it into the vise.

The vibration was reduced only by a factor of 3, and was about 3 times greater than the small motor in the vise. Higher frequency notes came into play this time.

I'll try that Steve. BTW, I finally got around to ordering coils for my E50 motor from AudioSilente. They arrived 2 days after I ordered them. FROM ITALY. Since Simone's English instructions were a little opaque I found your experience, as put forth on your website, helpful. Thank you.

 

[billshurv]

How does the mass affect the smoothness of the rotation? No point holding the body of the motor solid if that actually just causes extra vibrations into the belt!

 

[alighiszem]

It's interesting to see that at least one manufacturer has exploited the vibration absorbing properties of cast iron.

It is rather the added mass which diminishes the vibration. The same energy has to move a much bigger mass. Dynamically balancing the rotor and then adding mass would bring more and put less stress on the bearings of the motor.

 

[sq225917]

The Hurst motor are notoriously noisy, both electrically and physically, the Premotec motor used by linn and the 24v used by rega and massively quieter. Ask anyone who's doen the swap on their TNT deck

 

[thrifttrust]

Here are measurements taken from the vise jaw:

Here's from the top of the vise:

The results are similar to measurements taken directly from the motor. There is some diminishment at the vise top, but that could be because the jaw has a smaller mass and is not perfectly coupled to the vise body, or distance. I donno.

BTW The first post's measurements were taken with the scope set at 20mv/div. All subsequent measurements are at 10mv/div for greater detail. There's a bit of a problem in that the system picks up a little stray 60Hz. Most of it went away when I connected the indicator stand to scope's front panel ground terminal. (green clip lead), but not all. So when viewing the 10mv/div images you should mentally subtract this:

I'm rather pleased with the performance of the sensor (accelerometer). Please tell me how I might improve it on the cheap. I'd like to get rid of the hum. Right now it's just an old BNC cable soldered to the terminals of one of the cartridge channels. Oh, and it could use some gain. I've only got one more step on the scope's knob.

Taking measurements of the motor shaft is tricky. Measuring the shaft's side is out of the question because the noise due to surface irregularities is overwhelming. But, If you get the needle at the center of rotation you get a clean reading.

Here are shaft measurements out of the vise:

Here's the shaft measurement in the vise:

The shaft vibration is cut in half when the motor is in the vise. Note however, that even out of the vise the shaft vibration is only 17% of that of the motor body.

I have not found 24V Premotecs. I'm open, and I'm not surprised the Premotec outperforms the 59mm Hurst to which it's usually compared. The Hurst is huge. That's why I'm seeking an alternative. (What's with Americans and our "bigger is better" mentality.) I was about to purchase this 49mm Hurst:

Then. on the same Hurst webpage I noticed the little 42mm guy I'm testing here. It weighs in at 135g vs. 193 for the 49mm and 368 for the 59mm. Yet it has more torque than the 49mm. Twice as much in fact, 28.2mN/m vs. 14.1. The 59mm has 60mNm. At first I thought it may be a drop in replacement for the Premotec till I discovered the Premotec is 51mm in diameter. Premotec's spec sheet does not provide weight or torque figures.

As to quality, the Hurst's are fine, as no doubt, is the Premotec. The bearings are well fit and silent. The total lack of 300Hz in any of the 42mm motor's waveforms indicates that the rotor is balanced.

 

[AVWERK]

This is one of my tables that uses a 1.250"x4"x5" steel base and a 2.5" solid post of alum. I machined, with a set screw to secure the motor.
With the aluminum post and motor only, you could feel vibration slightly ( tested it by touching it to my teeth )
With the steel base attached and bolted and again, ahem,, biting carefully on the steel I could detect nothing at all.
The pod is seperate from acrylic base.

 

[awty]

Interesting experiments Douglas.

Ultimately would the main problem with motor vibration is it getting into the bearing/platter/spindle/playback?

So wouldn't it be more important to clamp the bearing case?

 

[Mark Kelly]

Waveform with the motor in the vise. A 20X reduction in amplitude!

Don't mean to be dismissive, but it appears that you have not reduced the energy in the vibrations, just the amplitude. Amplitude is relevant but it is a long way from the whole story.

 

[thrifttrust]

Ultimately would the main problem with motor vibration is it getting into the bearing/platter/spindle/playback?

Indeed.

This started as an accidental observation. I was actually testing heat sinking ideas. I was concerned with how hot the motor got even with no load. With a proper cap value the motor became hot to the touch. With the recommended value you couldn't hold it at all. It turns out that it doesn't take much mass and surface area to keep the heat down but the aluminium heat sink did little to reduce vibration. Surely more aluminium mass would reduce vibration but cast iron is nearly three times denser and several times cheaper.

Regardless, cast iron is known for it's vibration absorbing properties. A common test for cast iron is to tap it with a hammer. Steel goes ding, cast goes thunk. Large spur gears are usually cast iron for smoother running, even though they're softer and more brittle than steel.

I see a cast iron motor housing as one element in a comprehensive system that addresses noise at every stage.

So wouldn't it be more important to clamp the bearing case?

By bearing case I assume you mean the spindle bearing. It has it's own noise issues and they are partially address by, you guessed it. A cast iron bearing housing. My motor test turntable uses a Thorens TD124 cast iron platter spinning in its cast iron bearing housing.

This is one of my tables that uses a 1.250"x4"x5" steel base and a 2.5" solid post of alum.

Beautiful. Indeed, It's mostly the mass.

You should file your teeth so each has a particular resonance so you can do vibration spectrum analysis.

... it appears that you have not reduced the energy in the vibrations, just the amplitude. Amplitude is relevant but it is a long way from the whole story.

I thought I was reducing the energy of the vibrations by selecting and testing a motor that seems to be inherently quiet. It displayed high initial amplitude because it has low mass. There was such a dramatic reduction in amplitude because there wasn't much energy being produced, Next I will explore electrical means of reducing vibration, but please, tell me the whole story.

 

[rcthweatt]

The above illustrates why VPI offered the SAMA(stand alone motor assembly) as an upgrade to the motor mounting they originally used in the HW-19 . The motor was attached to a largish steel plate with lead sheet glued to it, which was mounted on 4 rubber isolators, which in turn were attached to a plate mounted across the back left corner of the base. This did a great job of isolating the base from motor noise(as judged by a stethescope),the point of which, since this was a suspended subchassis table, was actually what? On the other hand, motor noise could be heard on the subchassis, when the belt was in place! Exactly what we don't want. A lot of the motor noise was being stored, rather than drained. Hw-19 owners should try replacing the isolators with coupling nuts to test the effect- if they're pleased with it, a diy SAMA(made of cast iron !) would be a real improvement.

Electrical means- phase angle (adjusted with capacitance) has a real effect, so does voltage, especially the much higher voltage across the phase generated by the capacitor. This is an inevitable effect of capacitive reactance(leading power factor). Getting the voltages close(though not necessarily exactly equal), will definately help.
Rega's control boards for their motors apparently have trim potentiometers for capacitance/phase and voltage, to be set such that vibration is all but imperceptible when the motor is held in the hand.

Then there is the overall voltage-reducing it will help, up to a point. For 120V motors, between 70 and 75 volts often works out well. You may note that the motor vibrates least when starting the platter, considerably more when it has brought it up to speed; at that point, the excess torque of the motor produces vibration. Some users run the voltage low enough that they have to spin the platter up by hand, if they use a really massive platter.

Then there are the 'other than capacitor' methods of generating the second, 90 degrees shifted phase, such as the Wien bridge, used by Thorens when they shifted to AR type tables with the TD125. One of the easiest is to use frequency generator program(such as Cognaxon, free download) off a computer via the headphone jack, amplify this(a cheap amp can be used), send it through 20:1 transformers, and on to the motor. The amp probably will not be happy with the very low resistance of the transformer, in which case resistors must be added (4 ohm may be sufficient). Phase can be changed by entering numbers(90 degrees, 90.1, 90.08,etc),voltage levels are similarly easy to set.

You can then record this as a music file and play it back on an MP3(hence "ipod drive"). However, I've found the conversion to WMA, for example, changed how the program behaved, not for the better-I assume this is a phase effect.
This by way of introduction, not by any means the "the whole story"! There's a lot on this site to explore, and elsewhere. Good luck! I can assure that it's worth the trouble.

 

[sq225917]

Don't mistake a wider Q for lower overall amplitude. If you add mass to a motor pod you may redirect some energy but mostly you're just storing it and releasing it over a longer period where it can smear a broader frequency range rather than modulate at quite tight frequency points.

There's a number of Good AC controllers via this site, both Meldano's diy effort and Pyramid's more commercial offerings. Both have adjustable voltage, phase and frequency.

 

[thrifttrust]

rchweatt, at your mention I looked into the HW-19. Judging by the image search alone it must be a popular TT. It also looks like a real playground for diyers. At the heart of them all is the good ol 59mm Hurst.

As you may have guessed, I'm modeling by TT design on the Thorens TD124. I bought my first one (SN 2045) used from a Hi-Fi store in the early 70s for 50 USD. Since then I've never played a record on anything but a TD124. I guess I can't say it's a great TT. I have no basis for comparison. I do know as a sometimes mechanic It's underside is as beautiful as its topside. There was a fly in the ointment in that it ran a little slow. Try as I may I could not figure out why. Everything looked perfect. By the 80s It was slow enough to be objectionable. Then miraculously, Somebody gave me, (It drives me crazy but I can't remember who or under what circumstances.) a TD124 MKII (SN 85982). The only thing wrong with it was that someone had spilled bright red nail polish on the mat. I traded top platters and enjoyed the MKII for the next 30 years. A couple years ago the internet told me that the plastic spindle bearings were the problem with the MKI. Now I have two perfectly running and restored TD124s

The TD124 features I want to incorporate in my TT are a cast aluminium chassis, the cast iron sub-platter, the aluminium upper platter with the instant stop feature that lets the sub-platter continue to spin, and the cast iron bearing housing with a 14mm dia. spindle. I am not interested in duplicating the elaborate drive system with its 1700+ rpm motor. I can't cast iron myself, but I can cast aluminium, so I intend to make my own chassis. From a sculptural standpoint it would be so cool If I could cast my own iron motor mount but I will have to be content with carving one from billet. Thankfully, McMaster Carr sells cast iron in blocks, rods, tubes and discs. They even have a 300 X 40mm disc from which I could carve a sub-platter but that would be quite a task for my meager machine and cost 100+ USD, so for now I'll use the TD124 platters and spindle until I know It's worth it.

While this is an unfunded diy effort. I'm designing it with commercial production in mind. My intent is to build a fully functional, near production ready prototype with the goal of interesting a manufacturer. Short of that I hope at least to have a nice TT for a very low price.

Keep it simple and retro are important design guidelines. Out of the box I want it to run from a remote 24VAC transformer and employ little or no electronics. It may have coil resistors and balancing trimmers. Starting torque may be an issue so It could have a timer and a relay to short out the resistors during startup. I'm kicking around ideas for a mechanical timer.

Then there are the upgrades. First of which would be electronic speed control. In keeping with the retro theme it should have an analog oscillator with a low level phase shifter driving chipamps. Speed would be set using TD124 style under side strobe markings illuminated by a quartz driven strobe light. The only controller knob would be for pitch control.

There's a lot on this site to explore, and elsewhere. Good luck! I can assure you that it's worth the trouble.

Thank you. You've given me much to consider, and yes, I have enjoyed myself and learned a lot from this forum.

 

[russc]

Conservation of energy - energy is neither created nor destroyed.
Large mass, low amplitude = small mass, high amplitude.
The motor mounts would have the same vibration energy to absorb (convert to heat) or transmit to the plinth.
If you used a tub of sand as the mass, the vibrating sand particles would convert the vibration energy, by friction, to heat.

 

[thrifttrust]

...The total lack of 300Hz in any of the 42mm motor's waveforms indicates that the rotor is balanced.

What was I thinking? An unbalanced rotor would shake at 5Hz.

 

[Hiten]

I guess a ready made cast iron enclosure like 'this' can be filled with 1mm steel balls and used.
regards.