Forum

Any ideas for measuring turntable rotation speed?

Does anyone here know of anyone who has modified a DJ turntable (eg Technics SL-1200) so that you can obtain an electronic reading of the rpm? In such a way that if you scratch the deck you can reproduce the scratching in software.

I was thinking of using a light sensitive diode to detect the movement of the black dots around the edge of the platter. This would be easy to add to many turntables without opening them up or doing anything else.

The only condition is the turntable has to be modified in such a way that it is still usable as a normal deck. I.E no special records or similar

Any help appreciated.

arduino thing ? you may also have a contactor doing on/off

put a band of black/white pattern around the rim, place photosensitive diode as you said, take an arduino or find some other way to read it, write some software to process this into velocity info.
if you want to also find the direction of the movement, you’ll need two diodes slightly offset.
a good idea might be to find an old mechanical mouse and hack it for the sensors. if you’re good with a soldering iron maybe you can even use a hacked mouse directly (i.e. get the info via usb/HID (?) node or something)
see http://en.wikipedia.org/wiki/Mouse_(computing)#Mechanical_mice

Hi
from my own experience, if your goal is scratching (like the real thing), forget about home made binary coded discs, etc. you are better off with some timecoded vynils and a soundcard, i use the MsPinky setup http://www.mspinky.com/, it s got its own vst plugin, stand alone program (even video scratching), max/msp patch and most above all you can get the SDK of the whole thing and do your own coding.
Arduino is way too slow anyway for scratching (beat matching should be ok), if you really want to do some home made photodiode monster you may have a look at the new 32bit MIDIBox core, http://ucapps.de

on what info/experience do you base this? why would an arduino (or 8bit midibox, for that matter) be too slow??

sorry and no offense to the posters, but a vinyl tracking which tracks the platter of the deck is actually broken by design as a skilled dj would never turn the platter forth and back but rather the vinyl itself. (which resides on a slipmat that prevents the platter following every single torque impulse of the vinyl disc)

edit: judging from this thread @midibox.org a PIC/arduino should be sufficient.

Ok fair enough, I don t have a direct testing on Arduino platform, mine was an assumption that coding in C in a small chip may lead to slow performance, my direct experience was on the 8bit Pic used by the MIDIBox core, comparing ASM vs C.
So I apoligize beforehand if my assumptions were wrong.
I still stand on the idea that building an home made coded wheel and the photodiode apparatus is going to take longer time and be more expensive than using timecoded vinyls.
That said, hacking is fun so…
I was looking for MTE s Traktorizer cause it s a good example but it seems like th epage is down.

EDIT: I came to realize that an 8bit chip would proably be faster than my soundcard to analyze the signal…

Thanks guys, I’ll look into all of this.

The fact we can get a lively debate going on the best way to do something like this is why I love the vvvv community.

M4d, good to know about the vinyl.

Cheers

m4d: yes that (slipmats) occured to me a couple hours after the post… :P
io: regarding timecoded vinyls, see first post by OP - “I.E no special records or similar”
i think he wants to passively get the scratching from an inaffiliated DJ or whatnot.

tech-savy solution: vertical camera above decks, feature-tracking of the vinyl center label? :D just kicking around ideas…

Why not use a gray-encoded http://en.wikipedia.org/wiki/Gray_code#Position_encoders label which can be affixed on top of any vinyl record, camera tracked on top or mounted via a small arm protruding from the back side?

if you track by camera, you don’t need a gray code (would only make sense if you read with simple photodetectors), any kind of marker/fiducial should suffice. and since i figure it’s difficult to convince a dj to fix a label to every record he spins, it may be more advisable to try to track the record’s labels. problems are to be expected for featureless labels, though.

Oopsy didn t notice the “no special vinyl”

Camera? Need a rather stable one with high fps.

Well thinking technics 1200 wise, you may try with a photodiode or fototransistor and a led pointing towards the small dots on the platter, with a proper angle so that the receiver can see the reflected diode light, that should do.

where again, we are at the problem m4d pointed out, i.e. slipmats…

maybe midi timecode from a dj software (e.g. dj-decks, traktor scratch pro, you name it…) could be a viable option too? never tried it myself, but it’s probably the most convenient solution as io already pointed out.

ofcourse the hardcore DIYer solution (building your own platters with optical tracking pattern and accompanying photosensitive diodes like in the mentioned thread) is really tempting too, but that depends on your money/requirements/deadline/motivation situation…

I dj regularly and can confirm that any dj worth his salt would never affect the speed of the platter, slipmats are placed under the record so that the vinyl can be manipulated without affecting the platter speed. Timecode is definitely the way to go.