How accurate is the RPM Wow and Flutter app?

doctor fuse

Super Member
I've been listening to my Dual 1218 recently, and have been so impressed by not only it's keen rhythm and musciality, but also by it's impressive technical quality (quartz-like speed accuracy and low, inaudible wow and flutter, averaging 0.07% on the app).

Yesterday I used it as a changer for the first time. A couple of times the mechanism failed and the tonearm returned to rest and it shut off, after the record fell.

This morning, wow or flutter is audible. Like one slight pitch fluctuation every rotation (wow?).

I did have the volume set very high, for some reason. Not sure if this is making the speed variation more audible. Orchestra record, no piano.

So I checked the speed with the app, and speed was 33.30 RPM (was set to 33.33 last time I checked, but this is acceptable drift for me), and surprisingly, the W and F registered as 0.08% (readings have come in from 0.06-0.09%, so this is basically the same).

Any idea what is going on? I don't think my ears have suddenly got more sensitive. Can anyone explain how the gyroscope app works, for dummies?

PS. I have checked the accuracy of the speed portion of the app, against my Technics DD SL-D2. With the strobe dead still, the app registers 33.33, so it seems to be accurate for RPM speed measurement.
 
Can't comment on the app -- but since you were changing records are you sure you're not listening to the difference in characteristics of different pressings?
 
Can't comment on the app -- but since you were changing records are you sure you're not listening to the difference in characteristics of different pressings?

No, changing records was last night, wasn't listening critically.

This morning I put on a very familiar, favourite LP which I know like the back of my hand, maybe better. It seems like the 50 year old table needs some warm up time, it is playing without audible W and F now, on side two.

But I am confused why the app doesn't register what my ear can hear? Judging by a Rega deck I used to use, w and f is audible to my ears only when it gets over 0.14% or so, maybe closer to 0.19%. But that table had a different thing, a quick warbling on piano notes, especially high notes (I am thinking this is flutter? Confused by which term is for high, quick fluctuations, and which is for slow, low ones).
 
In my observations the speed is very useful, probably not as accurate as a strobe, but certainly good enough to check speed. However the w/f is about worthless and may actually provide incorrect data, which can be worse than no data at all....

VR
Andy
 
The accelerometer (G-sensor) is the primary type of sensor used with these apps. You can get a good hunch about how flaky these apps are by the coordination of the software with the accelerometer. It's the sensor that's designed to feel the change in orientation and, for example, switch the display from portrait to landscape when you tilt your phone. You can sometimes see how inaccurate the sensor is when the RPM app display keeps changing it's orientation and appears to rotate instead of staying fixed. The rate of rotation varies between models and makers, with Apple iPhone devices (pioneers in the integration of accelerometers into smartphones) being largely more accurate and robust.

It is in essence a simple sensor relying on a steady reference frame and a moving mass and on measuring the capacitance differences between the two. Other implementations rely on piezoelectric crystals and their stressing due to acceleration. The implementations vary in accuracy by quite a lot with some being so flimsy and useless they can only detect the most coarse and abrupt of tilts.

For the purpose of detecting acceleration shifts on a linear plain, such as with RPM apps, the G sensor is typically sufficient. It's the sensitivity of the hardware, along with the software calculations that will determine the final level of accuracy. That will inevitably vary between smartphone models and makers and is subject to hardware health. One nasty drop of your phone on a hard surface is typically enough to knock the sensory devices out of snuff forever. The device can only detect axial linear motion and cannot be used to construct a complete digital 3D environment picture.

Enter the gyroscope (2010 with iPhone 4): That's the device that detects angular rotational velocity and acceleration changes. Not all smartphones includes one even today and it can supplement the accelerometer with 3D data. In contrast to classic maritime gyroscopes this is not a mechanical device but rather a MEMS and it can provides navigational data that's obtained from the coordination between the geo locator (GPS) and the accelerometer. Gyroscopes are not required for RPM apps, but some modern versions do employ the benefits and accuracy of a gyro device - at least in theory. In real life, other than for navigational data, the use or unuse of gyros isn't a show stopper.

Such app is "RPM" from the Apple App Store. It mentions (quite clearly) in the documentation: There will undoubtedly be some variation between devices due to the tolerance of the gyroscopes for which there is nothing an app can do. This is as honest as it gets. Other apps for android are way less revealing. In my experience none of them ever display an accurate figure, no matter which phone you use, and the wow/flutter results are preposterous.
 
You sure you weren't listening to an off-center pressing? Hearing it once per rotation leads me to believe that it's pressing-related. The motor spindle and idler wheel go through many revolutions per one rotation of the platter.
 
Wow, tn, thanks for the thorough explanation!

Zvezda - good point. But no, the LP plays fine, and as I said, once I got to Side B, it was fine. I imagine it was a bit of a cold morning, and the motor took a while to warm up? The 1218 is playing its usual flawless self for the past few hours. Now I want to know what I was hearing!
 
I've noticed with my Dual 1019 (and also using the RPM app on my Samsung Galaxy phone) that it takes about 10-15 minutes of running (a bit less of an LP side) for things to stabilize. I also get wow readings between .06% and .08%.on that Dual . For comparison, my Rega P3-2000 with Tangospinner subplatter will usually run at about .08% to .10% although I've gotten readings as low as .06%, my Lenco L75 at .04% to .06% and my Pioneer Pl112D at .11% to .14%. To be truthful I can't hear any wow on any of them. I think I read someplace that most people with normal hearing can't really hear W&F less than .20%. On the other hand I've also read that some lucky ( unlucky?) individuals with exceptional hearing acuity can detect W&F at .05%. Happy I'm not one of them :)
 
I've noticed with my Dual 1019 (and also using the RPM app on my Samsung Galaxy phone) that it takes about 10-15 minutes of running (a bit less of an LP side) for things to stabilize.

My 1009 SK2 is the same, but it is older and more worn out than the 1218. I should give the motor and mechanisms a thorough rebuild, but I don't quite trust my watch making skills, especially the "putting back together" part.
 
My 1009 SK2 is the same, but it is older and more worn out than the 1218. I should give the motor and mechanisms a thorough rebuild, but I don't quite trust my watch making skills, especially the "putting back together" part.
Ha ha. Yup I suppose it takes buggering up a few to acquire the appropriate watch making skills. I tried with another 1019 I bought on Ebay. Unfortunately the shipper just left the platter on (didn't even tighten the shipping screws). You can imagine what the 8 lb bouncing platter did after several thousand miles on the back of a UPS truck. I tried and tried to rebuild it but I finally gave up and converted it to a manual by stripping all the automatic functions/ Works great now.
 
You sure you weren't listening to an off-center pressing? Hearing it once per rotation leads me to believe that it's pressing-related. The motor spindle and idler wheel go through many revolutions per one rotation of the platter.
Yup, found this out on a record I picked up a week or so ago. Side 1 was perfect...noticed audible W/F on side 2 and looked at the table...the end of the tonearm was moving in/out at least 1/4" on each rotation! I couldn't believe how off-center side 2 was.
 
Going to need a Nakamichi Dragon deck to remedy that problem... ;)
You can manually do it with a Dual. Pull the single play spindle, poke it through the top of the record to center it initially, then bump the record around until it spins true. Floating the tonearm over the LP helps give a visual reference to eyeball against. I usually put a pencil mark on the label to show which direction it needs to go next time.
 
Yup, found this out on a record I picked up a week or so ago. Side 1 was perfect...noticed audible W/F on side 2 and looked at the table...the end of the tonearm was moving in/out at least 1/4" on each rotation! I couldn't believe how off-center side 2 was.

Somehow I always imagined an off centre pressing would be the same on both sides. Thanks for the tips.
 
part of the limit of the app's accuracy is the fact that cell phones are bar-shaped instead of round, so even if you have means of balancing the phone on the spindle as it spins, you're going to induce a slight wow effect just by rotating that phone with the platter. if you're just setting it directly on the platter, it's going to be worse (i've observed this directly). direct drives probably won't care as much, but a belt drive with a lightly or moderately weighted platter will most definitely see that extra unbalanced mass swinging around and you won't get an accurate result...

generally speaking (as a former physics lab TA), i would say the app is "precise" in that it can give very similar results (via the speed vs. time graph) every time that you set your phone and the table up the same, but it's "accurate." on two different phones with the same table, the result varied by like 20% (but the shape of the speed curves was the same).
 
Last edited:
One thing I've noticed in Duals is that the platters all seem to be warped to some degree, due to the cheap pot metal castings they used. Some aren't too bad, but my current 1019 is bad enough that the single play spindle wobbles visibly with the platter, and I've even seen a couple parts units on eBay where the platter has cracked. This shows up on the RPM app as a periodic spike, once per rotation, and throws off the W/F measurements.
 
I've noticed with my Dual 1019 (and also using the RPM app on my Samsung Galaxy phone) that it takes about 10-15 minutes of running (a bit less of an LP side) for things to stabilize. I also get wow readings between .06% and .08%.on that Dual . For comparison, my Rega P3-2000 with Tangospinner subplatter will usually run at about .08% to .10% although I've gotten readings as low as .06%, my Lenco L75 at .04% to .06% and my Pioneer Pl112D at .11% to .14%. To be truthful I can't hear any wow on any of them. I think I read someplace that most people with normal hearing can't really hear W&F less than .20%. On the other hand I've also read that some lucky ( unlucky?) individuals with exceptional hearing acuity can detect W&F at .05%. Happy I'm not one of them :)
Say "hello" to one of those people that can hear w&f down to almost nothing. It's really more of a "feel" than hearing. As pianist for many decades I know the tone and feel of a piano very well and need to essentially suspend disbelief to listen to piano on vinyl but when I do it can be very enjoyable. Digital brings a different set of problems and while speed stability is not an issue, other things can be. Piano at 44.1/48k never feels quite right but 96k and above is fine. My guess is it has something to do with pre-ringing in the filters at lower sample rates. It's interesting because it sounds essentially the same at the different sample rates but definitely feels different. Piano is the instrument that reveals EVERYTHING about a playback system. Too dark, too bright, lack of harmonics and dynamics, messy loudspeaker phase across the spectrum, noise, speed stability, speaker cabinet resonances, everything.
 
Shout out to this great app.

I started to notice a slight pitch change a couple times during each song on my DP80. I thought it might have to do with some new old tubes but running the app I could see the wow and speed accuracy changed about every third or fourth time I ran the test.

I swapped out the DP80 with another I had recapped. I ran the app about 25 times. Every single time WOW was +/-.02-.04% and speed was +/-.08-.14%. How accurate!

Time for a recap on the other.
 
Also, check what the arm is doing, Vertical vibration can also sound like flutter. Is it vibrating on warps?

Want to check RPM? Just time a mark, or a penny, for 50 or 100 turns with a phone timer, stopwatch or whatever. Divide it out to RPM and you'll be as accurate as any other method if you go long enough. Time it as it passes the tone arm post or any other feature.

You can see wow with a strobe disk, but the centering has to be dead perfect.
 
Back
Top Bottom