Maximum rep rate for a standard microswitch.

amdx wrote in news:qlmjai$68p$ snipped-for-privacy@dont-email.me:

Resort it all again and physical file it chronologically and you will always know where to look for something based on when it was made.

Veeder-Root counter, 12V 8 Watts. I tried to post a Dropbox picture, lately the link in my Dropbox window does not highlight to copy. I had this problem then it started working, now having it again, anyone else having Dropbox problems?

Mikek

Rebooted solved the Dropbox problem for now.

Mikek

I right-click on the file on my PC (Win 7) and select "copy Dropbox link."

Microswitch on a drill press? OK, I confess, I've never done that.

But a small microswitch should work at 80 ms on/off time. Do you think it won't?

I don't like reeds, at least reed relays with coils. They are expensive and unreliable. Their rated speeds are illusions, because they bounce and twang for ages.

I l

ng

Optical doesn't require much of a counter. Obviously you have a current li mited voltage source to work with the switch. That should also work with a photo diode as long as the current and voltage don't exceed ratings. The hall sensor can be bought as a unit ready to go but it is a three wire devi ce needing power.

Do you know what the current through the switch is when closed? What is th e open circuit voltage?

If you have 5 volt power available you can get photodiode or hall modules i n a couple of days.

--

  Rick C. 

  - Get 2,000 miles of free Supercharging 
  - Tesla referral code - https://ts.la/richard11209

rotation,

Do you have a diode on the coil of your counter? Might be one build in if the leads of the counter are polarized from the factory. Back EMF can

play havoc with microswitches (and reed switches).

1N4002 or higher would be fine.

John :-#)#

Sanity check. If you used a long-life mechanical microswitch, good for a minimum of 50,000 cycles, at 330 revs per minute and one count per revolution, your switch will be dead after: 50*10^3 / 330rpm = 151 minutes = 2.5hrs If you believe the 1 million operations specification (I don't), it might last up to: 1*10^6 / 330rmp = 3,000 minutes = 51 hrs Be sure to have a box of replacement switches handy.

By coincidence, I recently bribed my auto mechanic with a tiny magnet hall effect sensor tachometer for his lathe: It won't count revolutions, but will measure RPM, which is what I suspect you really want. I've used these on a drill press in the past and found that contact cementing the tiny magnet to the spindle pulley and installing the sensor between the pulleys works well.

The biggest problem was finding 12VDC near the drill press. I used a common "wall wart" power supply because the owner didn't want me to splice into the 117VAC power. Finding a good box was also problem.

You can also get photoelectric tachometer kits, similar to the hall effect sensors. Gluing a reflecting piece of tape to the spindle pulley or the belt should work.

Example: More:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Thanks. So now you disclose what you're trying to accomplish. I assumed you wanted a tachometer for the drill press. So, use a hall effect or photoelectric proximity sensor:

I'm not sure what you have from Veeder Root. If it's marginal, look for an industrial counter or "totalizer":

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I don't think he wants to wrap half the circumference of a drill press chuck with magnets in order to get a 50% duty cycle. A rough estimate:

The drill chuck will be spinning at 330 rpm. The chuck has a diameter of about 2 inches. The magnet might be about 1/4" long.

Circumference of chuck = 1 revolution = Pi*d = 3.14 * 2 = 6.28 inches Surface speed = 330 rev/minute = 5.5 rev/sec = 6.28 in/rev * 5.5 rev/sec = 34.5 in/sec A 1/4" long magnet, moving a 34.5 in/sec will product a pulse of: 0.25 in / 35.5 in/sec = 0.007 sec = 7 milliseconds The "operating time" of a reed relay can be "a few milliseconds" according to:

although it is more often around a few milliseconds dependent upon the actual reed relay.

In my never humble opinion, a 7 msec pulse will be rather marginal unless either a very small reed relay, or much larger magnet is used. There's also the minimum pulse with to trigger the Veeder Root counter, which will be a factor. Of course, with a 50% duty cycle, the pulse width will be half a revolution or: 1 / 5.5 rev/sec = 180 msec which should be more than enough time to produce a count. While that would be great for a coil winder running at 330 rpm, I suspect that most counters will have problems at the low frequency end. Looking at the specs on my cheap eBay counter, it only goes down to 10 Hz = 0.1 count/sec = 6 rpm If the drill press is going to be stopped and started during the coil winding, I would expect some counts to be lost as is slows down and speeds up. This is one reason why many coil winders still use gear driven mechanical turn counters that can be spun up, down, reversed, and stopped, without losing the count.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Hi Jeff, This should be simple although the mounting mechanics were a little chore. I still haven't found my reed switches, but I expect that will work. The is my counter.

Mikek

Ya'll have convinced me. I need two coils about 400 turns each. I'm going to hand wind them. Mikek

to

? I

all

ook

tle

ork.

The coil says is is DC yet the wires don't appear to be colour coded, if

this counter does not have a diode across the leads you will destroy your switch from back EMF.

John :-#(#

You could do optical detection with a 1/2 moon disk, LED and photodiode. Put more than one slot in the disk and you can do fractions of a turn.

(I didn't read the whole thread so someone may have suggested that already.)

George H.

I'm wondering if you need this bit of kit at all. If you know it does 330rpm, you know how long to run it for each desired winding. For just 5 or 10 turns you can turn it by hand. ... unless you really do need 1 turn accuracy.

You might even find that a carbon brush & some thin insulation might suffice.

NT

-----------------------

** That seem realistic to me. Recently had to relace my computer mouse cos a microswitch failed, the left click one IIRC.

Lasted about 5 years at say 50 operations per day.

So about 100,000 cycles.

.... Phil

Personally, I would not immediately elect to use a reed switch for this app lication. There are many better ways to do it, especially if you are comfo rtable programming up 8-bit micros, or even the new Arduino hobby-grade stu ff. I would look at Hall Effect sensors, or something in optoelectronics. You could get rotational speed, a count, and even direction.

Good luck with the project.

• transistor and diode

You would think a electro-mechanical counter would work fine with a switch closure. Any chance the switch is not only bouncing from the contacts, but the mechanical arm is bouncing as well? What is the connection like?

--

  Rick C. 

  + Get 1,000 miles of free Supercharging 
  + Tesla referral code - https://ts.la/richard11209