• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

AR XA passes w/f tests

TAPEPATH

TAPEPATH

Active Member
I have a classic AR XA turntable that has been overhauled by AK member marcmorin and have enjoyed it thoroughly. I'm working on another table of mine and thought I'd do some wow and flutter measurements. Since the AR was at hand I connected the meter up and made some measurements. The original specification, for the AR, was for the combined wow and flutter to be less than 0.1% using the NAB parameters. For reasons of ease of reading the meter I chose the JIS setting. For combined weighted w/f I measured 0.07%, with wow at 0.08% and flutter at .055%. The combined w/f, unweighted, was 0.09%. The measured frequency should be 3,000Hz, and the tested frequency was 3,0006Hz. If you read AR's original speed accuracy description they mention the ability to use a Dustbug and have correct speed. I don't have a Dustbug to test that.
Since I had a VPI Turntable power line conditioner at hand I repeated the test but with the AR connected to the VPI for AC. All of the measurements seemed to be very slightly higher with weighted w/f at 0.08%, wow component at 0.09%, flutter at 0.06% and combined unweighted at 0.09%. This was consistent at any setting of the rear panel adjustment or measurement scale. Granted it was extremely small, but still measurable. One nice thing was that I could adjust the playing frequency for exactly 3,000Hz. But my sense was that the AR kept better speed with good old house AC.
 
Register to hide this ad
Hi Ron,
I've collected a number of test records over the years. I used the Stereo Review SR 12 test record.
Ken
 
Thanks for the technical specs you were able to measure. A lot of time was spent in 2006 through early 2007 on the bearing design at the nearby metalurgical college of the state university. Once the numbers were "run" on the various loads in the vertical and horizontal (oil film dynamics) and conversion of the various vibrational modes within the spindle, I then set out to find the "moments" that remained that affect timing. Two of the major moments came from impropper spring reaction and overshoot of belt. Ed Villchur set the ratio of pulley diameter to subplatter diameter to where the system should have run fast, but belt dynamics being what they are allow the table to run at 33.33. (or supposed to) The overshoot is due to the acceleration of the PM rotor from pole to pole and the belt "averaging" that overshoot. Balanced mass loading of the pulley corrects the majority of that. There was still the issue of the micro vibrational chacteristics of the friction surface. That was solved by the implimentaion of the compounds used to drain the vibration. The belt goes from kinetic to static in a much more controlled manner now.
Speed controllers: unless it is a very sophistcated device, I'm still of the mindset that if a speed anomaly occurs during playback (heavy transient causing the platter to slow any) and then to correct that after the fact, you now have two. I believe the music is more organic to the human ear the less correction used.
I'm in the process of designing another simple bolt on device that may just take those numbers lower yet. When I'm happy with it, I'll send you one. Run the numbers again for me.

Thanks

Marc
 
Ken,
Thanks. I'll have to look around to see if I have that record. At least I know where my meter is......

Ron
 
No problem, Marc, I'd be happy to.
In looking at the VPI it outputs a saw-tooth wave centered at 60Hz for 33 1/3 playback. Turning the dial on the front panel raises and lowers the frequency. Since my wow/flutter test was a steady state test and the real benefits of the VPI are present during high load transients in record playback I don't mean to discourage anyone from listening to it.
I just found it interesting that sustained speed accuracy was very slightly worse. It is quite a testament to the character of the AR that it is still a capable and very enjoyable device. I'm listening to it as I type!
Thanks again for all of your effort!
Cheers, Ken
 
TAPEPATH said:
I have a classic AR XA turntable that has been overhauled by AK member marcmorin and have enjoyed it thoroughly. I'm working on another table of mine and thought I'd do some wow and flutter measurements. Since the AR was at hand I connected the meter up and made some measurements. The original specification, for the AR, was for the combined wow and flutter to be less than 0.1% using the NAB parameters. For reasons of ease of reading the meter I chose the JIS setting. For combined weighted w/f I measured 0.07%, with wow at 0.08% and flutter at .055%.
Click to expand...

0.055% flutter is pretty good i think. But i bet you could reduce that wow a little bit more. Is that bearing clean? Is that belt a good quality one?
 
TAPEPATH said:
No problem, Marc, I'd be happy to.
In looking at the VPI it outputs a saw-tooth wave centered at 60Hz for 33 1/3 playback. Turning the dial on the front panel raises and lowers the frequency. Since my wow/flutter test was a steady state test and the real benefits of the VPI are present during high load transients in record playback I don't mean to discourage anyone from listening to it.
Click to expand...

Ken, which VPI was it? The "Turntable Power Line Conditioner" that has two analog dials on the front? I have one but haven't tested it yet.

Surprised to see you got a saw-tooth wave. One would expect a pure sine wave out!!
 
I read your posting a second time, Marc, and it rang a bell. I was reading an interview between J. Gordon Holt and Bill Firebaugh, of Well Tempered fame, and he was talking about how he reduced wow and flutter by putting a half twist on the belt. He thought it might change the way the belt came away from the platter. The twist pulled one side of the belt away from the platter before the rest of the belt and may have changed the tension and release pulses the belt makes as it approaches and leaves the motor pulley.
I'll give it a try and see what happens.
 
Hello,
It had been a few years since I looked at what the VPI is putting out and so I connected it up to my HP and it is a square wave, not a saw-tooth. I'm sorry for the error.
Yes, it is the original VPI PLC not the more recent version. My belt is a new one and the bearings are clean and well lubed. Also, the table is very carefully leveled and I let the unit run for a half an hour before testing.
Cheers, Ken
 
Well, interesting results. First I remeasured all of the original tests to have a comparison. Then I put a half twist on the best and let the table play for five or so minutes. The playback frequency was now 3.017Hz instead of 3,006Hz so it plays a little faster with a twisted belt. But, the wow and flutter measurements were a slight bit improved. I repeated the test four times to make sure I had things correct. The weighted JIS combination of w/f stayed the same at 0.07%, the unweighted wow became 0.06% and the flutter was 0.05% and the combined unweighted w/f was 0.08%. So, the primary reduction was in the wow measurement (0.5 to 6Hz fluctuations) which lowered the unweighted combined figure. Probably the weighting wouldn't show any change in the first combination number because of the low pitch error thought to be not very audibly detectable.
So, like Chubby Checker said, "Let's twist again, like we did last summer!"
 
TAPEPATH said:
Well, interesting results. First I remeasured all of the original tests to have a comparison. Then I put a half twist on the best and let the table play for five or so minutes. The playback frequency was now 3.017Hz instead of 3,006Hz so it plays a little faster with a twisted belt. But, the wow and flutter measurements were a slight bit improved. I repeated the test four times to make sure I had things correct. The weighted JIS combination of w/f stayed the same at 0.07%, the unweighted wow became 0.06% and the flutter was 0.05% and the combined unweighted w/f was 0.08%. So, the primary reduction was in the wow measurement (0.5 to 6Hz fluctuations) which lowered the unweighted combined figure. Probably the weighting wouldn't show any change in the first combination number because of the low pitch error thought to be not very audibly detectable.
So, like Chubby Checker said, "Let's twist again, like we did last summer!"
Click to expand...

With the twist you're increasing the tension. You could try experiment with a less-elongated (smaller?) belt. As long as rumble doesn't intrude and you're not making the motor do any funny noises, it's all ok.

For what it is worth, on another forum (pinkfishmedia) there is a thread devoted to w&f tests, "turntable speed analysis". You should contribute on that topic, i'm sure all will appreciate it.
 
Last edited:
TAPEPATH said:
I read your posting a second time, Marc, and it rang a bell. I was reading an interview between J. Gordon Holt and Bill Firebaugh, of Well Tempered fame, and he was talking about how he reduced wow and flutter by putting a half twist on the belt. He thought it might change the way the belt came away from the platter. The twist pulled one side of the belt away from the platter before the rest of the belt and may have changed the tension and release pulses the belt makes as it approaches and leaves the motor pulley.
I'll give it a try and see what happens.
Click to expand...

There are 4 major "moments" a belt goes through when spanning two pulleys.
Static friction is always higher than kenetic. There is also the fact the belt never fully goes into static on the drive pulley. Hense the vibrational mode that occurs there. That vibrational mode is not constant, therefore the amount of friction is not constant. The damping compounds in the pulley remove most of those modes. The change in speed might do more to the music than the W/F figures, but more importantly I think the rumble number(s) may go up. The well tempered belt is much thinner than what you have, and a twist in that won't be near as severe. The belt you have on the table is the best available, but still not up to the ground ones AR produced. Even the Teledyne AR belt couldn't match the original.
 
PM received. Thank you!

I have always wanted an AR TT and I stumbled onto one a few weeks ago. I have been reading a lot of incredible information about it -- a lot of it written by you fine folks, and will continue to do so before starting to restore it.
 
Ultra-Hog said:
PM received. Thank you!

I have always wanted an AR TT and I stumbled onto one a few weeks ago. I have been reading a lot of incredible information about it -- a lot of it written by you fine folks, and will continue to do so before starting to restore it.
Click to expand...

Your siggie line is from one of my favorite comic strips.....I have several of his complilation books. The wife and I had quite a few of those "Calvin" moments with our oldest son...........probably the reason there's 16 years between him and his younger sister :yes:
 
You must log in or register to reply here.
Back
Top Bottom