Position switch?

[Soloratov]

Slight change in my previous plan of having a timer switch. I am sure now that it would be better to use some type of positioning system instead. Same issue applies however:

6-12v power supply connected to motor and an on switch. Motor spins a 4 1/2" "wheel". This "wheel" needs to stop at a specific point every time so an item can be removed. The problem, it needs to revolve at least 7 times after the on switch is pressed, and then stop at the same start point then immediately ready for another round.

The "Wheel" is light and restricted by friction, and only spinning at roughly 100rpm so momentum from the spinning is almost non-existant. I already tested the design with a power drill, and as soon as I pulled it away, the spinning stopped dead.

Any ideas on start-stop position switching would be greatly appreciated.

 

[Boncuk]

These are the ingredients for it:

- pickup device (Hall-sensor)
- decimal or BCD counter
- power stage for motor

Count the number of rotations and at the desired value deactivate the motor power stage.

Something like this?

The motor starts rotating when power is applied. After the "programmed" number of revolutions the motor stops regardless of clock inputs. When pushing START button the counter is reset and the motor starts rotating again.

Boncuk

 

[Soloratov]

Ok, I think I understand. Now, is this a sensor type that requires something placed on the shaft? Not a big deal....just don't know much about this stuff. Seems like a simple enough plan.

A curious question I had....in my simple world, is this. Could the same idea be achieved using 3 switches? 1 being the power button, and two more SPDT lever style switches. My thought was if I hit the power button, and something is pressing on one of the SPDT switches, the motor will run until the other switch is pressed. Then it could rotate in the opposite direction until the other switch was pressed again. This is all assuming I can make some sort of small ball screw type carriage that sits on either the main shaft or a layshaft. May be a really dumb question, but I have to ask.

 

[MrDEB]

Now I'm lost??

press button A = motor starts, wheel goes in cw direction
press button B = wheel stops then goes in CCw direction
where does the 7 rotations come into play??
need more info
hall sensor or IR slot detector switch would work.
easy way to do this is a pic but a simple cmos or TTL circuit will do same but need more info on what you want.
Also what this is for = school project (hope not again and are you willing to take the advise presented and actually use it.
in the past month, 3 ops have asked for advice and very little if any of the quality advise was accepted.

 

[Soloratov]

Its a design project i have for work, not a school project.

- 8.4v battery is a more probable power source for this project as it will run remotely in the field on a 8.4v NimH pack, with solar recharge.
- Plan to use 12v 100rpm DC motor as drive

7-10 rotations in 10 seconds because that is the number of times the object needs to rotate inside my housing to achieve desired result.

Also, it must be a single START button design. Did some quick looking online at components to get an idea of what is involved and i like the hall sensor idea. Seems easy enough, and effective.
Only two clarifications before I go that route are these:
1. Dumb question=Whats a motor power stage? Not familiar with the term.
2. The magnet for the sensor can be placed on any spinning surface to achieve my desired cutoff?

 

[Boncuk]

- Plan to use 12v 100rpm DC motor as drive
7-10 rotations in 10 seconds because that is the number of times the object needs to rotate inside my housing to achieve desired result.

In your first post you mentioned the position of the disc or what so ever has to be the same at stop as it was at start. Timing the on-time of the motor you wouldn't get the necessary precision to achieve that.

Also, it must be a single START button design.

It is a single START button design. Just the first start will be inititated when applying power. After the cycle has completed you just need to press START for the next cycle.

Did some quick looking online at components to get an idea of what is involved and i like the hall sensor idea. Seems easy enough, and effective.

I suggest to use a TLE4905 (Infineon)

Only two clarifications before I go that route are these:
1. Dumb question=Whats a motor power stage? Not familiar with the term.

The motor power stage is the MosFet transistor (BUZ10) in the schematic.

2. The magnet for the sensor can be placed on any spinning surface to achieve my desired cutoff?

You can use a magnet available in toy shops and mount it a way it will pass the sensor at a close distance (1 - 2mm). Glue the magnet to the rotating part and have the sensor at a fixed position.

To change motor direction use a DPDT-relay which reverses polarity at the motor connections.

The count cycle and stop position won't be affected by reversing the motor.

Here is a revised schematic with a power switch added to prevent motor start when power is supplied to the circuit.

Boncuk

 

[Soloratov]

Thanks Boncuk, you have been a great help! The timing I mentioned of 10 seconds is purely a goal that is achieved by the rpm factor. Perceived time of the user, if its 11, no big deal.
I will most likely go the route of the first schematic you provided. Reversing motor is not necessary, but was an option. Let me get the parts together and give this a try.

 

[Soloratov]

Am I correct in understanding the part U2:A is a NAND gate? If so, how is that wired into the circuit? When I did a search they came up looking like small microprocessors, which makes sense I suppose. I was planning to do this on a breadboard for now, just want to get it right.

I figure looking at a circuit diagram I am connecting to pins 1-3 only. Having a hard time finding anything that resembles this item. Only things I could really find were here:
https://www.allelectronics.com/make-a-store/category/700/Switches-BCD/1.html

Would most likely get all the parts there if I have to, typed in NAND as wll, and again, not entirely sure which one I should be looking for. I know it's a 4011, but those on the site are not labelled as such.

 

[zipdogso]

Nand gate packages do look like microprocessors.

They are usually a number of nand gates in the same package a quad nand has 4 and a dual obviously has 2.

They also have varying inputs for your schematic you need a 2 input.

So look for either a dual or quad 2 input nand.

You will need to refer to the units specification sheet to see which pins to use it should show the 2 input pins and the output pin for each nand gate. Just choose which ever set of pins you want. Generally the first 2 pins are input and pin 3 is the ouput of the first nand gate but I would not like to swear this is always the case it is best to check the spec. sheet.

Here are some on the site you were looking at -
https://www.allelectronics.com/index.php?page=search&search_query=nand&x=0&y=0

 

[Boncuk]

4011 is just the part number. Different semiconductor manufacturers use different prefixes, e.g. TI uses "CD", ST uses "HCF".

You might choose any of those. They are basically 100% compatible. NXP (Philips) offers single NAND-gates (SOT23-5 package (SMD)).

If you use a quadruple 2-input NAND gate (4011) connect all unused input pins either to VDD (+8.4V) or to ground. (Do not connect the output pins of those gates anywhere!).

Also, use a 100nF decoupling capacitor across the power supply pins (pin7=ground, pin14=VDD) of the IC.

Boncuk

 

[Boncuk]

Hi Soloratov,

here are the full schematic and a PCB layout (2 PCBs, No1=2.125X1.525inches and No2=0.7X1.075inches) of your circuit. Both boards are completely single sided. (no wire jump)

Don't forget to mount the TLE4905 on the solder side of the PCB. The center of the tiny square within the IC outline is the center of magnetic sensitivity and must be placed exactly underneath or above the magnet.

PM me for the Eagle files. I'll send them via email to you.

Boncuk

 

[MrDEB]

nice job Boncuk

board layout looks good!
layout is very clear..
no weird 90D turns etc.
I personally have gone with the practice of filled planes. Less etching involved
GREAT JOB

 

[Boncuk]

board layout looks good!
layout is very clear..
no weird 90D turns etc.
I personally have gone with the practice of filled planes. Less etching involved
GREAT JOB

Hi MrDEB,

do you mean something like that?

 

[MrDEB]

YES thats what I am talking about

I leave about .030 clearance around traces.
same as trace width or almost.
mostly going with .025 trace width

 

[Boncuk]

I leave about .030 clearance around traces.
same as trace width or almost.
mostly going with .025 trace width

Hi MrDEB,

it is alright for me using .030 clearance around traces, but it's impossible between IC pins. The fill area decreases with increasing clearance. I sometimes use traces of 0.254mm to fill areas satisfactorily. When filling of e.g. ground is complete there are lots of "blank" areas not filled, thus causing areas harder to etch (because of the big surface), thereby underetching narrow traces.

For a real thorough ground fill use a trace width of 0.254mm and connect short traces from ground into the blank area. Call the short piece of trace ground and the area within it's range will be filled with ground as well, leaving "islands" only where it's impossible to fill with ground.

You might want to try filling non ground areas with any signal name you have in mind. Draw a small piece of trace and call it "fillme". Eagle will do it without any connection to some part. (Don't use duplicate names. You'll get a surprise. )

Using that method every board is etched very precisely even with narrow traces.

Regards

Boncuk

 

[Boncuk]

Hi Soloratov,

here is a zip file including the schematic, PCB layout and BOM.

Perform "ratsnest" before printing the PCB layout.

Boncuk

 

[Soloratov]

You have been so helpful. Thank you, thank you. So glad I found a place like this. Hope I can learn more as time goes on.