Stroboscopic disk sets wrong speed for turntable?


Hi all! I am new here.

I have just set up my Clearaudio concept turntable. Unfortunately, the speed of the turntable was wrong so I had to reset it. In order to do this I have used two methods: an iPhone app called RPM; the stroboscopic disk + the iPhone app StrobeLight.
The problem is that when everything is correct at 33 1/3 (both verified with the RPM app and the stroboscopic disk), the songs are playing one semitone lower than what they should. In order to make them play at the correct note, I have to set the speed at around 34 RMP as shown by the RPM app. Also the stroboscopic disk shows that now I am going faster than the 33 1/3, being consistent with the RMP app. How is this possible? Am I doing something wrong or even the stroboscopic disk is not accurate enough? Could it be the StrobeLight app not being an accurate enough light source for the stroboscopic disk? It is just peculiar that both the RMP app and the stroboscopic disk are consistent in saying that I am going too fast. 

Any help is much appreciated. Thanks!
agforte
I keep forgetting to add one point I wanted to make with respect to the OP. Not only do I agree that the cell phone programs are notoriously inaccurate, but also so can a strobe be in accurate.  In many cases the 60 cycle frequency of the AC line is not stable. That can disrupt the frequency of any lamp that you plug into the wall. This is why I would recommend the KAB strobe. Which uses a battery operated strobe that is finely calibrated at 60 cycles. Plus the KAB device includes a disc that is wider in diameter than nearly any other I have seen.That also adds to its accuracy. And finally, it is not unusual to find that belt drive turntables without motor controllers can be slightly inaccurate as to speed. 
The Roadrunners Tachometers were an excellent product - and as stated earlier we have developed a replacement product for them.  Theres a few mechanical chassis parts left to be manufactured but the circuit etc is complete.  

It works "identical" to a Roadrunner by a tiny magnet tripping a hall effect sensor once per revolution. However it does not display "speed" rather a count which which is 33333 in exactly 1.8 seconds, 45000 in 1.33333 seconds and 78000 in 0.7692 seconds.  

The circuit is based upon 3 very accurate oscillators which output is divided by logic circuits and at each trip of the hall effect sensor the count is displayed on the 5 digit display.   Its highly accurate and will read actual "speed" at the  3 main speeds - there is however the caveat that any number lower than 33333 means that the table is running to FAST and any number higher means that its running too SLOW.  The same of course goes for the 2 other main speeds. Count to be displayed is selected by a switch on the back of the chassis which is slightly larger than the roadrunner. 

https://systems.audiogon.com/systems/4923#&gid=1&pid=15

We will be offering them for sale as soon as the remaining chassis parts are available.  Pricing will be $299

Good Listening

Peter
Thank you Peter! I thought it was PBN and not Sota that was coming to market with a similar product. Is it Sota that bought the rights to the Phoenix design but just for Sota tables?
Could you expound on why your product won't display actual speed? Is it so as to not infringe on the Phoenix design? Is the Phoenix design patented or otherwise protected from copying? Certainly the more intuitive the better and I don't see how anything could be more intuitive than displaying speed in rpm to two digits rather than what amounts to a revolution count without being per-minute. 
I presume you are counting on having perfect pitch to determine if the speed is correct. This is a wonderful gift!In any case, with  the strobe disc, be sure to use an incandescent lamp as mentioned before. As a cross check, use a test lp such as the The ultimate Analogue test LP from Analogue Productions and the hz function on your Fluke DVM. This is more precise than the strobe method and negates any presumption that stylus drag could cause a difference.