counter and prox sensor woes

[StarryNight]

I'm not sure what's going on but for some reason my counter has started to miss counts. Thinking it was the sensor I replaced that and no change to the problem. I purchased a new counter and attached the new sensor and... no change to the problem! Then I thought to check the voltage going in to the counter. I've been using a car battery charger as a PS and it seems it only puts out 10vdc. I found a wall wart that puts out 12vdc and tried that with no change either. At this point I'm scratching my head! The only thing that hasn't changed is the motor and PWM. Has anyone else encountered prox sensors failing or counters that skip counts? It seems when I slowly pass something conductive in front of the sensor it works fine but anything at normal winding speed seems to fail badly.

[eschertron]

Can I assume the *first* thing you did was to check the alignment of the prox sensor? In my day job, I use proxes regularly, and have found that many need a gap of no more than 4mm or so from their flag to operate. If you've checked that, then...
maybe the flag is passing in front of the sensor too quickly to pick up every pulse. Have you changed your winding speed just prior to the problem's start? What else could be different in the process itself?

[StarryNight]

thanks for the reply!
ok, I think it's a power supply issue. with the new counter and sensor it seems the car battery charger makes the counter skip 2 or 3 numbers every time the sensor is triggered. with the wall wart it is more accurate, but not perfect. switching from 2A to 10A on the car battery charger makes the counter not work at all (it displays numbers but that's it). I think I need to get cleaner power or something that is variable. I'll keep working on it...

[Jazz P Bass]

A larger (wider) prox 'flag' would help.

[StarryNight]

A larger (wider) prox 'flag' would help.

I'm gonna try that as well!

[StarryNight]

I think I got it working by using a separate power supply for the counter and sensor. the PWM I have might not be right for the motor and I've got some weird stuff going on as I increase the speed. I also added a larger "flag" to the sensor with a 1/8th aluminum plate (instead of a screw).

[Jim Darr]

I think I got it working by using a separate power supply for the counter and sensor. the PWM I have might not be right for the motor and I've got some weird stuff going on as I increase the speed. I also added a larger "flag" to the sensor with a 1/8th aluminum plate (instead of a screw).

[ATTACH=CONFIG]45128[/ATTACH]

StarryNight,

I had a counter that got off once, not the sensor or the flag. Somehow the settings on the back of the counter unit (Micro Wiz) got changed and my counts were off.

In your situation, since the setup was working well at one point the problem must be in one of these areas:

1. The sensor

2. The counter itself or the counter's settings

3. The power supply

4. The speed

5. The flag or trigger

6 The proximity/alignment of the flag to the sensor

It looks like you are trouble shooting the correct items, but keep getting faulty results.

I'd double check any settings on the counter once again. Also, look at the "exact" power supply requirements for the sensor and the counter to make sure they match with what you have.

My guess is that the problem is in either the settings, alignment, or PS.

[Jim Darr]

Quick question.

Just noticed that your pics show two counters???? Do you use both to validate the actual count? or???

[StarryNight]

hey Jim, thanks for the feedback!

1. I went back with the original sensor. Both were functioning equally poorly when sharing the same power supply. Approaching the sensor head on with an object fired the counter way better than approaching from the side. the LED light on the back of the sensor would light up but only half as bright and the counter wouldn't fire. If i went VERY slowly the counter would fire each time, but as soon as i sped up the slightest bit the counter failed.
2. my original counter is a veeder root. i paid way too much for it. I bought the second one off flea bay for $5. there's seems to be a definite different in performance (the VR tracking better).
3. I don't have a good PS set up. I've been meaning to get that sorted out. a variable PS would be nice to have for test purposes. I've burned out various wall warts. the car battery charger seems to be working alright but I will actually loose power periodically and then the power picks back up after a few seconds. I'm running it at 12v and 10A (there is an option to run it at 2A). The motor is rated to run from 12v to ? i think 24v. On the DMM the car battery charger only puts out 10v! I don't think the PWM is compatible. I burned through one a while back and bought another but I don't know enough about them to make an informed decision.
4. I run at lower than normal speeds. I'm pretty sure guys here wind a lot faster. I can't seem to get my machine to spin very quickly. If i open up the pot the power seems to choke.
5. changing the flag was a definite improvement but it still wasn't working until i used a separate PS for the counter.
6. the flag is about 2mm from the sensor. They are perpendicular to each other.

[eschertron]

You'll want a good, solid (on the high side of) 12vdc for the counter/sensor pair. Just thinking out loud...be nice to the motor and run it off a different power supply. It doesn't even need to be electrically connected to the sensor side of the machine, does it?

[StarryNight]

You'll want a good, solid (on the high side of) 12vdc for the counter/sensor pair. Just thinking out loud...be nice to the motor and run it off a different power supply. It doesn't even need to be electrically connected to the sensor side of the machine, does it?

that's the conclusion i'm coming to. two separate PS seems to be helping the situation. At some point i will have to clean up this problem but there's work to be done!

[David King]

Sounds like your motor's current draw is exceeding the PS and or PWM capacity. There's a reason universal sewing machine motors and pedal speed controllers are so popular.
Optical sensors are so much easier and seemingly more reliable than Hall effect or proximity sensors.
Again strive for a 50% dwell time for maximum reliability. Extra counts implies you're getting extra "bounces" at triggering. Most serious counters have a built-in de-bouncing circuit you can try, usually the "slow count" input instead of the fast-count input.

[StarryNight]

my first winder used a sewing machine motor and mechanical counter. It's my backup now. The thing sound like a printing press.

Yes, my Veeder Root counter does have slow and fast terminals. It's a good unit but feeding it dirty power I think was causing the errors.

I agree, sometimes there's an argument for going low tech.

[David King]

You can always go high tech by using a brushless DC motor and speed reduction gearing from a model electric plane or helicopter fro the local model shop. Those motors are incredibly efficient and the electronic speed controllers are accurate, all at a very competitive price.

[StarryNight]

You can always go high tech by using a brushless DC motor and speed reduction gearing from a model electric plane or helicopter fro the local model shop. Those motors are incredibly efficient and the electronic speed controllers are accurate, all at a very competitive price.

that's an interesting idea! do brushless motors usually require air cooling? I don't know much about them. David is this what you had in mind?
DC Spindle Motor Kit 600W 300W 450W 400W 500W Brushed Brushless+Power Supply Set | eBay