Diy Strobe for measuring turntable speed accuracy.

[hagtech]

Actually, you CAN have perfect adjustment for 45rpm with any of these frequencies. You just don't divide the strobe into an integer number of dots (or lines). Put a mark at every correct angle. At the end, you will have a mismatch. Thats ok! The only that happens is that for each rotation there is a shift in position of the strobe pattern. Under strobed conditions, the marks will appear stable, then shift suddenly and be stable again. I don't think this minor problem precludes perfect accuracy.

BTW, you're right. I didn't make the 78rpm pattern perfect. I used an integer number of marks. I figured, what 78s were ever made quartz-locked anyway?

I also chose the 75Hz to purposely stay away from fluorescent or other lighting issues.

jh

[Gatto Murr]

Hi folks,

this could be interesting...

http://snipurl.com/mmbr

regards, Piero

[Gatto Murr]

Actually, you CAN have perfect adjustment for 45rpm with any of these frequencies...jh

Hi, Jim, could you please tell us how "perfect" this adjustment should be?

I'm wondering how much the actual speed could be off before one can detect the pitch variation by ear?
Or -better- when this difference becomes relevant?

regards, piero

[Steerpike_jhb]

what's the xtal frequency? 1.6384 MHz?

for "50Hz" it's 3.27680MHz, for "60Hz" it's 3.932160Mhz.
I put the 50Hz and 60Hz in quotes because the actual oscillator frequency is double that - an LED only gives one light pulse per AC cycle, whereas a conventional or fluorescent lamp gives two (hence you need double the frequency for an LED).
Both crystal types I have found in catalogues - so they aren't difficult to get - it's just 'my' usual store doesnt keep them.

Put a mark at every correct angle. At the end, you will have a mismatch. Thats ok!

This is true, although all my strobe discs built into platter mats or cut into platter edges are integer numbers of bars, and hence are not accurate.
60Hz lighting CAN give absolutely accurate results since it requires an integer number of bars for 33.3 and 45 rpm.

As for ON pulse duration, the shorter the on pulse, the better the contrast of the strobe bars when viewed. It also allows a much higher LED current than the LED can normally handle (hence brighter light). Shortening the on-pulse is the function of the 555 Ic in my circuit. (which I have just realized can instead be done by more intelligent use of the divider's otherwise unused outputs)

Normal AC lighting provides a rather nebulous on duration, but it works well enought, so short on pulse are just a luxury really.
If one wanted to go to extremes, a xenon flash tube could be used, as is done on the strobe lights used for auto-ignition adjustment.

[Gatto Murr]

[quote="Steerpike_jhb... an LED only gives one light pulse per AC cycle, whereas a conventional or fluorescent lamp gives two ...[/quote]

oops... . Of course you're right!


Below another way to measure rotational speed:

http://i1.tinypic.com/nx466d.jpg

These gears are quite cheap on Ebay, but you should put a piece of reflecting tape on the platter rim.
Wonder what could happen with those platters having black-white marks.

regards, Piero

[packratt]

I should go down to the shop and grab the digital tachometer and see what mine is running.

[Gatto Murr]

Hi Steerpike, found an error in the drawing...
Connection of the 555's pin 8 to the +B bus is missing.

regards, Piero

[Gatto Murr]

Strobolight: here we are on the very cheap side of the thing...

http://i1.tinypic.com/nyv09v.jpg

Let's see if I'll be able to put some wires together this Sunday.

regards, Piero

[bauzace50]

Hello all,
Thanks for the enthusiastic developments! Am keeping track, and will be pending to try out a pertinent design. This is surely useful to all parties. Thanks, again.

[Steerpike_jhb]

Here's the final circuit; I built it last night & it works just great. Chip count is reduced to two, plus one general purpose transistor.
(total component cost, with crystal, excluding battery & case US$1.98!)

http://www.vinylengine.com/turntable_forum/gallery/image.php?image_id=3068

I had to up the battery voltage from 4.5 (3xAAA) to 9V (1xpp6) because I could not make the CD4060 oscillate reliably below 5V. The LED on-pulse width is 1.25ms (1.0ms at '60Hz'), which allows a very high instantaneous LED current (=sharp strobe-bar edges), but the average current drain of the whole circuit is only around 14mA. With the 9v battery, the circuit can happily drive several LEDs in series with no extra power consumption.

During the week, it will go onto a printed circuit board and into its plastic case. Pictures, PCB layouts, construction details and design equations etc. to follow...

[bauzace50]

Goodness, Steerpike, this is special! We'll keep up, fully intent on obtaining accurate speed readings!

[hagtech]

That's a fine circuit. One thing I found was that there is some coupling through the power supply from the LED current pulse. Depends on what you have for a supply. I was using batteries. A little internal impedance and you get a synchronous droop on the voltage feeding the oscillator. Probably insigificant. Just nice to keep the output transients from the osc, if possible.

The flops are nice for shortening the duty cycle (on time) of the lamp. I chose the simpler method to get a /4 using three resistors instead. Not as good performance-wise, but simple is good.

jh

[Gatto Murr]

Here is the Poor Man's strobe (last version):

http://i1.tinypic.com/o932hy.jpg

and this is the prototype:

http://i1.tinypic.com/o934lt.jpg

Current drain at 6v is 18mA, period vary from 9999 microseconds at start to 10067 after an hour of continuous running - less than1% at 100Hz.

Like the overall period, the "On" time can be varied, I set it at around 2milliseconds.

All the components come out from the junk box, trimmers are ten-turn pots .

Sorry, no pcb is planned....

regards, Piero

[Steerpike_jhb]

One thing I found was that there is some coupling through the power supply from the LED current pulse.... Just nice to keep the output transients from the osc, if possible

Quite possible there is coupling - but noise in the power line doesn't do anything to this type of circuit, so its not an issue. Now if I was in Hawaii, I'd be out in the waves an not watching for spikes on battery voltages!

Anyways, it's now finished, tested & working just fine.

http://www.vinylengine.com/phpBB2/album_page.php?pic_id=3075
http://www.vinylengine.com/phpBB2/album_page.php?pic_id=3076

I'll mail Jas the 'construction info' for inclusion in the downloads section.

Thanks to Bauzace for the suggestion to make it! A useful little trinket now I have it.

I have not tested it on a synchronous motor turntable, but simply comparing the strobe output to the ac mains on an oscilloscope, the mains is rather drifty - just in the 2 minutes I watched it, it wandered around as much as 0.5%. If this was a continuous error, I imagine it would be inaudible, but since it is continually varying + and -, I do wonder.