View Full Version : The thread in which I ask the questions needed to make this $&(*& automation work!

10-09-2013, 11:48 AM
Ok we make a captive line of fasteners for a customer. Due to various "reasons" I have to be kind of vague. All those who have worked under those conditions, smile and nod and take a shot. :)

So the upper and lower thingy drop down a vibrating inclined track system, into a small air-punch-press. The force of the punch-motion drives the upper thingy into the lower thingy. I may have said too much. :rolleyes:

Anyway, the upper-thingy comes down the track too fast and I need to slow them down so I get one thingy every two seconds. I thought I had a "servo" designed that would slow the parts down to one every two seconds.. that's why I asked the math question.

So today I learned that DC motors can't be slowed down to far enough to make my wheel turn .875 RPM. Apparently, DC motors become unstable as you fiddle with the voltage/current?

So, first question is there a way to slow my motor down any further, right now I have a pot between the power supply and the motor. I can slow it down to a certain point then it becomes VERY unstable. Lots of surging. Any lower and the motor stalls.

This means... I am going to have to pick a motor RPM and gear it down.

Lets assume 30 RPM at the motor shaft. The rating is 60 RPM unloaded at 6vdc, I'm assuming I'll have half of that once I start attaching gears.

What would I need, gear-wise, to bring 30RPM down to .875RPM at a 3" wheel.

Once I figure out what I need to build my hypothetical gear-train, it's a matter of then transmitting that power to a very tight space, possibly via quantum ****ery or a belt-pulley.

Of course there is always the chance that I may win the Powerball between now and the due-date... if that happens I'm just going to retire and let someone else deal with it. :D

10-09-2013, 11:57 AM
Why not gear the dc motor down before you fiddle with a speed control. Either use sprockets or if you go to surpluscenter.com you can get a 35:1 gear reducer that would put you in the ballpark.

10-09-2013, 12:01 PM
You need a 34.2857:1 reduction. :P

I doubt you'll ever get that with off the shelf pulley's and gears. Does it HAVE to be dead on? Will the compounding error if you use, say, a 35:1 reduction result in a reject widget every 56 parts? If there is no tolerance for error stackup, then the only solution is to go digital....

I'm not an automation guy, but I guess with the control pot and 35:1 you can tune it to get better error control.... and even better, you could add some feedback to a controller to maintain speed.... but that would probably cost more.... I would just ignore my post if I were you. :)

Dr Stan
10-09-2013, 12:13 PM
All one needs to do is calculate the gear ratio. Stepping down 30 RPM to .875 RPM one would need a ~34.28:1 ratio (30/.875). A quick look at the Surplus center gives us a 35:1 gear reducer http://www.surpluscenter.com/Gear-Reducers/Right-Angle-Shaft-Input-Gear-Reducers/?page_no=1&fq=ATR_PowerTransRatio:35\:1

If one used this gear box you would need to run the motor at 30.625 RPM

N/35 = .875
N = .875 X 35
N = 30.625

You could use the same process to determine the RPMs needed if you used a different ratio such as a 40:1

I'm guessing the hp/torque requirement is quite low so you may want to also look here at AC gear motors http://www.surpluscenter.com/Electrical/AC-Gearmotors/AC-Gearmotors/ They have them that run as slow as 1.8 RPM which would give you a starting point closer to the desired .875 RPM. It may also be worth your while to contact an electrical and/or automation distributor to see what they have available.

10-09-2013, 12:15 PM
Here you go. I use this motor to turn a 150 pound pig over a fire. Average time to roast a whole pig is about 12 hours. So my motor runs 12 solid hours. I also bought a controller (POT of sorts) from the same people to slow rotation if necessary. The motor is small but heavy duty and "out of the box" turns at 8 RPM.



10-09-2013, 12:16 PM
You've got a lot of variables to play with here. Is the 3" wheel diameter fixed? How about the motor choice? Does the timing need to be exact? Any idea how much force/torque you need to apply?

If the answers are Yes, No, Kind of, not much, you may be able to use something like this: http://www.herbach.com/Merchant2/merchant.mv?Screen=CTGY&Category_Code=SYNCTIME

If the answers are No, No, Yes, a little more, then you're looking at a smaller diameter wheel, and a bigger gearmotor. For example, a 2.43 inch wheel, with this (1.1 RPM) motor (http://www.grainger.com/Grainger/DAYTON-AC-Gearmotor-3M095?Pid=search) would give you exactly what you want.

There are plenty of other way to skin this cat, you'll have to figure out what is fixed and what isn't.

10-09-2013, 12:19 PM
If you do want to go with a gear train, then 0.875 is 7/8 so the ratio is more usefully 30 : (7/8) = (30 * 8):7 = 240:7

7 isn't a useful number of teeth so change that to 480:14

Many ways to do this but here's an example:

Let's make the final stage 30:14 , leaving another factor of 16:1 to find. (480/30)

Make that from two stages of 4:1. (4*4=16)

e.g. 60:15 + 60:15 + 30:14 (= 4:1, 4:1, 15:7)



Weston Bye
10-09-2013, 12:21 PM
If you need a slow motor, replace the DC motor with a stepper motor and driver and something like a 555 oscillator circuit to set the speed. The top speed of the stepper will be about the same as the slowest usable speed of the DC motor and can be slowed down to step... step... step... step... without loss of torque.

If this seems like a complicated approach, consider it versus the DC controller, DC motor and gearbox solution you have been pounding your head against here. There are many people like me here on this board who can quote chapter-and-verse solutions for you with schematics and stocklists & etc.

Not trying to be critical, but opening your eyes to possible alternate paths.

Weston Bye
10-09-2013, 12:54 PM
Well, I see that should have read the other thread first.
I now see that your heart is in the right place, WTF, but you should revisit your first impulse.
Yeah, yeah, done is better than perfect and all that, but...

10-09-2013, 01:12 PM
Run the motor with a pwm type speed control. This will maintain a lot of the torque while lowering the speed. Lots cheap on eBay:

http://m.ebay.com/sch/i.html?kw=pwm+speed+controller&isNewKw=1&pgn=1&epp=24&mfs=KWCLK&acimp=0&itemId=&cnm=&cid=&sqp=pwm+speed&_trksid=p2056088.m2428.l1311.R1.TR3.TRC1.Xpwm+spee d

10-09-2013, 02:13 PM
"I need to slow them down so I get one thingy every two seconds. "

I'm going to assume(my bad) that the thingie is something like a screw.
I'll tell you what the operators do in the plant I work for. It accomplishes the same thing and is stupidly simple.
When the want to meter the screws coming down the vibrating inclined rail, they take a strip of alum. about .025 thick-the stuff sold at Ace Hardware- The length depends on the length of the screw.
They bend the strip 90deg and with a little c-clamp clamp one end to the rail. Depending on how many screws they want to feed at what interval depends on how far up or down the rail they place the strip.
They, in effect, make a spring gate. The end of the strip not anchored is bent to allow a screw to pass buy when the weight of the screws behind it force it past the strip.
For heavier parts you can use an actual spring.

Are you using a feeder bowl to feed the screw to the inclined rail?

10-09-2013, 02:21 PM
Yes. It is a vibro-bowl feeder.

Any chance of a photograph?

10-09-2013, 02:26 PM
Agree with macona, use a PWM. I use DC motors on my X and Z feed on the mill I made, and a PWM works really well. Unit cost me about $5.00 and works like a charm.

10-09-2013, 02:28 PM
Lots to digest here, THANK you for all the information. I'm going to digest this for a bit and report back with more questions.

10-09-2013, 02:34 PM
Lots to digest here, THANK you for all the information. I'm going to digest this for a bit and report back with more questions.

The Artful Bodger
10-09-2013, 02:50 PM
If it is a matter of having one thingy every 2 seconds does the motor have to run continuously? Think of a windscreen wiper where the motor runs for one sweep and stops (when in intermittent mode). A 2 second timer is a simple electronic circuit and no doubt easily available off the shelf.

10-09-2013, 03:55 PM
Looking at the problem from another angle for a moment. The vibration for the system is provided by a "shaded pole motor" looks like a motor wedged into a transformer.

I may be able to reduce the need for such precise feed control if I can change the speed of the motor.

115vac 60hz 1.5 amp. Google says I need a "triac" type controller, and I see a multitude of units in the size range I need at many price points. Anyone have a recommendation?

10-09-2013, 04:07 PM
A shaded pole motor is a form of AC induction motor...

So unless you use a device to change the frequency (VFD?) ,it will be pretty much synchronised to the 60Hz supply.

A triac based phase angle power control scheme will not alter the frequency..you will just reduce the voltage until it stalls out.


10-09-2013, 04:13 PM
What I am looking at resembles a dimmer-switch but says "Solid State." Description states it's a variable speed TRIAC type controller for motor control included shaded-poll types.

I see... many variation on the same basic idea. Is this a dead end?

10-09-2013, 04:56 PM
Run the motor with a pwm type speed control. This will maintain a lot of the torque while lowering the speed. Lots cheap on eBay:

http://m.ebay.com/sch/i.html?kw=pwm+speed+controller&isNewKw=1&pgn=1&epp=24&mfs=KWCLK&acimp=0&itemId=&cnm=&cid=&sqp=pwm+speed&_trksid=p2056088.m2428.l1311.R1.TR3.TRC1.Xpwm+spee d

What he said.

Part of your problem with the pot between the power supply and the motor is that you're lowering the voltage for speed control but at low voltages the motor doesn't respond well as you've found. The PWM control that Macona is pointing towards achieves an "average" low voltage by time slicing what comes from the power supply - very fast. The instantaneous voltage when it's ON is still the full voltage but since it's on for for only part of the time it looks like the lower voltage for lower speed that you want.

10-09-2013, 08:18 PM
Rather than timing the ramp, would it not be better to have a trip device and gate that drops the second part into place when the first part gets in position?


10-09-2013, 09:56 PM
A method I have used successfully includes:

A mag proximity sensor or through beam detector that will shut off the feed bowl when the track fills to a certain level

An escapement mechanism that releases one part to the assembly operation while holding the rest of the track in place. When the escapement returns, the next part held in the track drops against the part release.

It's a fairly bullet proof approach and simple to put in place.


J Tiers
10-09-2013, 10:09 PM
Dan and JP are ON this one......

That's the right answer..... GATE out the parts as and when wanted.... the feed stopping per JP is essential for this, really.... you WANT more feed than you can gate out, so there are no "holidays" in the flow of parts.

I'm not clear if maybe you ALREADY gate them with whatever the "wheel " you mentioned is...... Possibly not. It may be the receiver.

10-10-2013, 12:27 PM
I have about four and a half inches of powerfully vibrating track surrounded by welded steel beams, tubing and sensors. I can fit a set of drive wheels in there to regulate the flow, and I can (now) regulate the vibration of the entire system.

Starting over is not an option UNLESS something catastrophic happens to his machine, I mean "total disintegration" as our welder is insanely talented. ;)

One of the "problems" with this forum and I mean this in a good way -- once the ideas start flowing it's hard to keep ya'll on track. Of course had I mentioned the space limitation... Well, is what is.