Project: Avenger Rim drive with BLDC motors & Class AB Amp

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby edw » Wed May 20, 2020 1:35 pm

VA, thanks for the graphs. While it is unclear from the closeup of the graphs whether the 2 measurements were level-matched when running AnalogMagik (I assume they were), it does show a marked improvement as suspected.
User avatar
edw
Senior Member
 
Posts: 330
Joined: Sat Mar 12, 2016 9:02 pm
Location: NY/NJ

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby theeng » Wed May 20, 2020 9:57 pm

VA,

Correct me if I am wrong, but those last two graphs are EMI and specifically conducted emissions (CE) and specifically differential mode EMI. The Y axis should be actually dBuA - dB micro-amps. The chart for the 300 AC 60Hz synchronous motor is characteristic - there is the strong 60 Hz fundamental that you will see for every 60 Hz synchronous motor, and the dBuA at 60 Hz is the motor current draw, but the 2nd fundamental at 120 Hz is very low. The 500 Hz peaks with side bands should be associated with a speed controller - did you take these measurements? The BLDC motor because of how it operates will not show the same 60 Hz fundamental, but it does show the repeating harmonic characteristic of the motor type - albeit very low. But the only take-away is that SOTA (or Anaheim) did a good job with filtering EMII-CE noise.

Neil
theeng
Member
 
Posts: 90
Joined: Mon Dec 09, 2019 9:47 am

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby vienna acoustics » Wed May 20, 2020 10:35 pm

Hi Neil

These are measurements which were taken by using Analog Magik test records and software.
Alike me and some others here, Edw owns this software too, and he has asked to see the measurements with this tool too.

Analog Magik comes with test vinyls, a sound card and software. It provides many measurements (azimuth, speed, wow/flutter, antiskating, VTA, VTF, loading, resonance and vibrations). In essence this software is measuring distortions.

From the Analog Magic site:

QUOTE

Accurate signal retrieval during vinyl playback is highly dependent upon reducing vibrations which the turntable is subjected to. Vibration can come from many sources: foot stomping, motor rumbling, airborne sound waves, or improper turntable isolation, which will all translate into unwanted signals being picked up by the Cartridge. There are many ways to measure and quantify vibrations.

A good way to quantify vibrations is to measure the Intermodulation Distortion Level (IMD) between two frequency signals. Intermodulation Distortions are essentially signals not recorded the LP and are being picked up by the stylus, therefore it would be desirable to have the lowest IMD% number as possible. We have incorporated two low-frequency signals in this test. Note the test frequencies and algorithm in our formula employed in this test is different than the VTA IMD% analysis.

Play the "Vibration Measurement" track on the test LP and the Analog Magik software will display the amount of Intermodulation Distortion expressed in percentage on the computer screen. A good setup should yield a number lower than 2% to 3.0%. If the level of vibrations is too high, to begin with, it will mask the incremental changes which you are trying to measure on the different parameters and render them meaningless, therefore it is important to reduce vibrations to the minimal.

​This measurement is designed to reflect immediate feedback which means the numbers should respond quickly to tapping or other external vibrations.

You can make changes to your turntable, such as going from direct drive to belt drive, adjusting the tension of the suspension, changing the isolation platform, or changing the location of the turntable. You can then repeat the measurement, a lower IMD% number would indicate an improvement has been made.

We wish to say that the test we have incorporated is only one of many which can be used to measure vibrations. In our product development engineer's laboratory, we have ultra-precise vibration measurement devices which are 100x more accurate than what we have provided, but the device cost more than a car. While there are far more sophisticated tools and test equipment available, we have to balance between cost, effectiveness and market demand. We believe the one we have provided will provide a good starting point to help users quantify vibrations.

UNQUOTE

I hope this will help
!
vienna acoustics
Senior Member
 
Posts: 458
Joined: Mon Apr 09, 2018 10:07 pm

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby theeng » Thu May 21, 2020 12:15 am

VA,

I went to their site, and looked at the hookup diagram; there are multiple power sources involved, you have the ART sound card and its switching power supply, then you have the computer and its switching power supply and then of course you have the motor, and the preamp. The setup is supposed to measure only the cartridge output, but that is such a small signal, that noise on the mains from other sources may be interfering. For the lower frequencies <1000 Hz, the output from both look like EMI-CE noise that will push back to the mains supply. That noise will also be picked up by the phono preamp depending on what type of input AC filter it has. If the cartridge was actually picking up the indicated noise you should easily hear it, especially at 60Hz, may be you were measuring a ground loop. Its also curious because their site on a video shows the same 500 Hz peak with side bands just as you measured. For the BLDC motor, the rotor has permanent magnets and the stator that has three separate windings 120 deg spaced to get the quasi-3 phase. The phases are modulated to get the motor to spin, there is a bunch of software involved, and generally hall-effect sensors - one for each phase so that that the software knows where the rotor is in relation to the stator. But the repeating harmonic you see below about 1000 Hz is the result. Once above 1000 Hz, I pretty sure what you are measuring is the cartridge, but not so sure below.

Just some thoughts, but I would be curious to know what Audio Magic might think of those scans.

Regards and stay well

Neil
theeng
Member
 
Posts: 90
Joined: Mon Dec 09, 2019 9:47 am

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby vienna acoustics » Thu May 21, 2020 12:22 am

Neil

You made some wrong assumptions.

The sound card is powered by the portable’s computer usb not through a switching power supply. The portable computer is powered through the battery and not connected to a power supply during the measurements. also this computer has solid state hard disk (not spinning ). During the measurements the phono is powered through Torus isolation transformer and the Controllers/ motors through Keces, ultra low noise, linear power supplies connected on another Torus isolation transformer.

In regards with the Hall sensors, as you will see at the first thread’s page construction photos, are removed and only three cables are connected. The Hall sensors are completely out of this project’s realization.

Lastly both measurements (for Hurst and BLDC Motors) were made exactly with the same setup. The only difference were the motors and controllers. At my system there is no switching power supply.

This Rim drive is very special.
!
vienna acoustics
Senior Member
 
Posts: 458
Joined: Mon Apr 09, 2018 10:07 pm

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby theeng » Thu May 21, 2020 1:17 am

VA,

I stand corrected in the power supplies, although I was merely making an observation that was prefaced by may. Otherwise, by your setup, its hard to see how motor noise could have been picked up. But I have done EMI-CE testing for years, and those cuts sure do look just like them, and the difference between the motors is as expected, but I will leave it at that. As far as the Hall effect sensors, when those were taken out, that should have forced them to 'sensorless' control. This is mostly an open-loop control scheme vs the closed loop provided by the Hall effect sensors. Which is better, for this application that is essentially fixed speed, obviously what is likely an open loop control is working very well.

I do not doubt the outcome of your efforts, I was merely observing the data you provided and trying to make sense of it.

Neil
theeng
Member
 
Posts: 90
Joined: Mon Dec 09, 2019 9:47 am

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby vienna acoustics » Thu May 21, 2020 1:23 am

No problem at all Neil

It’s not a sensorless system, it has the Roadrunner which provides two options for either an auto correction (through a very sophisticated algorithm to avoid the negatives of such a solution) when the roadrunner is connected to the controller, or the user can keep it disconnected and by montoring the speed can apply incremental manual speed corrections.

I am not an Engineer like you, my background is related to shipping business. In this thread I tried to present the project with measurements too with the means I have (and most users can have) in order for me to be more objective and minimise the subjective personal sound experience.

The important fact is that the BLDC motors with Bill’s controllers and amplifiers are making a huge sound improvement and I can say that this is a successful project.

As I have posted earlier, Bill Carlin had made two amplifiers for this drive system a class d and a class AB biased close to class A. To-date I am only listening to class d and shortly I will post my impressions on Class AB.
!
vienna acoustics
Senior Member
 
Posts: 458
Joined: Mon Apr 09, 2018 10:07 pm

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby theeng » Thu May 21, 2020 4:55 am

VA,

After I wrote my last e-mail, here I am some 3-hrs later because I awoke thinking I am idiot - isolation transformers cannot remove the fundamental because if they did, you would have no power, and in general, isolation transformers are not very effective for differential mode noise because it magnetically couples between the primary and secondary's of the transformer; https://en.wikipedia.org/wiki/Isolation_transformer, . So, even with all the isolation you are using, differential mode noise from the mains is leaking through. That to me explains why the graphs look like EMI-CE at the low frequencies - and I suspect that 500 Hz peak is noise from the controller. Differential mode noise can be filtered with inductors and capacitors to create a cutoff frequency above which at very defined slopes attenuate/filter the noise - aka just like a speaker cross-over.

Sensorless control is a term used for permanent magnet motor control when there is no direct feedback from the stator/rotor, there is still manual control and using a secondary signal you can get a feedback control.

But, none of this diminishes the benefits of the BLDC motor.

Take care,

Neil
theeng
Member
 
Posts: 90
Joined: Mon Dec 09, 2019 9:47 am

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby vienna acoustics » Thu May 21, 2020 5:17 am

Neil I suffer sometimes too, from persistent thoughts which keep me awaken.

This 500Hz peak was only present with the Hurst Rim Drive and has dissapeared with the BLDC
!
vienna acoustics
Senior Member
 
Posts: 458
Joined: Mon Apr 09, 2018 10:07 pm

Re: Project: Avenger Rim drive with BLDC motors & Class AB A

Postby vienna acoustics » Fri May 22, 2020 12:37 pm

I have just ordered two SEN-09269 3 Axis Sensor Accelerometers and have purchased the ARTA labs software in order to perform more accurate measurements between the Hurst and BLDC rim drives.

They will be needed for my next project too.
!
vienna acoustics
Senior Member
 
Posts: 458
Joined: Mon Apr 09, 2018 10:07 pm

PreviousNext

Return to General Audio

Who is online

Users browsing this forum: No registered users and 7 guests

x