Turntable speed slow with quartz lock

I have an 30 year old Harman Kardon T-60 turntable that runs slow (according to strobe) while in quartz lock mode. I would chalk it up to a bad unit except my friend's unit does exactly the same thing. The coincidence is too great for me to ignore. Any ideas on how 2 units that use a quartz referenced oscillator could both run slow. I have recapped the power supply on mine with no effect.
Two questions - what strobe are you using for reference, and how slow are they running?
Kirkus, forgive me if I'm wrong, but I think what you meant to ask is: "Are both turntables running slow by exactly the same amount?" ;--)

If the answer is "yes" (I'm guessing) it would point to both units being run with (or set to?) the wrong voltage/cycles, or at least not something "wrong" with the units themselves?

Dear Heyraz: Here on page 6 of the TT service manual you can read about: http://www.vinylengine.com/library/harman-kardon/t60.shtml

If that does not help you then maybe the oscillator is out of specs.
Now, you can run that TT with out the quartz mechanism ( off position. ), do you already try it? runs with out speed deviations after you set manually the speed?

Regards and enjoy the music,
If it's like i.e. the Micro DD decks, the built-in strobe is sync'd to the mains frequency, whereas the internal oscillator is not. So if it's by a very small amount, and you're using a mains-frequency strobe . . . It could be the mains that's ever so slightly off, not the turntable.

But if it's way off, then probably the TT is indeed at fault. Crystals do indeed drift with age, and sometimes surrounding circuit components can affect their accuracy as well.
First, try cleaning the speed selector switch with Radio Shack electrical contact cleaner. If you have variable speed controls, clean them too. I would not do anything else until you have done this- this is a very common problem.
Fixed it! Replaced all electrolytic capacitors in speed control circuit myself. Now runs dead on balls steady. BTW, the table ran slow with quartz lock on or off, so it wasn't just the quartz circuit. I suspected C117 in the speed circuit, but replaced all the electrolytics in that circuit (C1XX) while I was at it.