transistor circuit for counting wheel rotations

[smiddle2012]

Hey guys,
I am new to the forum. I hope this is the right section. Sorry if I ask some dumb questions, I am new to circuit design. This is for my first "no rules" pinewood derby. I wanted to know if anyone here knew of a simple transistor circuit I could build that will count how many times the wheel on the pinewood derby rotates. I was thinking something simple like a piece of foil on the wheel hits a piece of foil on the car, and that is one rotation.

When the car reaches X number, like 5 or 10, rotations, the motor I have hooked up to the car will start. I am going for a back to the future approach, when the car hits 8.8 ft/s well, you know the rest. Anyway, I have built simple circuits before but I am not sure how to do this, or even if I can with just transistors, caps, etc. If this is not do-able, any alternatives to getting the cars motor to start after the car starts rolling, that are still cheap, would be very welcome. Thanks in advance everyone, I love forums like this.
Sincerely,
SM

 

[CocaCola]

Since the wheels are opaque, I would simply mount a photo sensor behind the wheel, and then drill a hole in the wheel...

Or even more primitive use a small magnet mounted to the wheel and a read switch mounted to the car, as the magnet passes the read switch it will make contact and you can simply count the pulses... Using a decade counter (or a few cascaded) you could trigger the motor...

I would do the logic with a micro as it's more adaptable but, if you don't have experience and tools to work with them sticking to prebuilt chips like decade counters is likely a better option...

And last but not least the most basic mechanical solution... Due to the fact that the pinewood derby track starts on an incline and levels out you could simply mount a mercury switch at the proper angle, so that as soon as the car leveled out it made contact and started the motor, no other parts needed as long as the mercury switch can handle the current...

 

[Electrobrains]

Pulse delayed motor starter

Hi SM

Basically, I don't know anything about pinewood derby and the cars (when I googled, they all seem to be made out of wood), but it sounds exiting!

Reed contacts are a bad choice here, because all mechanical contacts chatter and you would have to include debouncing. A better idea would be to use a hall sensor together with the magnet, for instance this super small one:
Mini, cheap hall sensor

Coca-Cola's idea with a photo sensor is good though.

I would take that path, not having to attach magnets etc.
(Not knowing how your car is made) I would choose a very special component, a PHOTOINTERRUPTER, that for sure would work.
You could use a fork type, like GP1S58VJ000F, with a hole drilled somewhere, or for instance a nail put in the shaft. Each time the light is iterrupted you would get a pulse to count.

An even more interesting, modern way would be to use one of the two small Reflecting Photoiterrupters (GP2S60 or GP2S700HCP). You would just need to attach a small ribbon of reflecting material on the shaft (for instance aluminum foil) or paint the shaft black with a white line on it (or white shaft with black line). At the right distance you would read pulses showing the rotating part.

I'll give you one possible solution here. This circuit only counts to 10, but could with some effort be expanded.
I don't know what voltage you are using, but the circuit has wide supply range (3-15VDC). I have tried to make it as small as possible, using SMD components. All resistors (except maybe R1) and the capacitor can be SMD.
If you have more space you can choose similar, but bigger components that are easier to solder on.

How much current does your motor take? The suggested MOSFET is specified for up to 56A (@25°C), but can handle surges up to 226A! In the circuit I downgraded it to 40A. If this type is too weak, there are stronger ones out there.

Another question is how you want to stop the motor (reset the count)? With this circuit you could just remove the supply voltage or use the reset switch in the diagram. In fact it's a good idea not to leave the circuit connected. It slowly will empty your battery (drawing a bit more than 20mA). The main supply current goes to the IR emitting diode and is set by R1. You can try to change R1 and see how the sensitivity of the sensor is influenced: current=(supply voltage-1.3)/R1

I have NOT TESTED THE CIRCUIT, but it should work. If not, there would only be minor modifications needed.

Please be aware of the notes on the diagram!

I hope you will win!

 

[duke37]

Won't you need a lot more than 3V to fully turn on the mosfet?

 

[Electrobrains]

Won't you need a lot more than 3V to fully turn on the mosfet?

No, this is a modern logic level type. It should be totally switched on at maximum 2.5V.
See the size, price and data of BUK9Y14-40B.

It's incredible what they have come up with in the last years. Even this one is not new (2008).

For instance, Fig.13 in the data sheet shows it has 15mOhm at 45A load (Vgs=3.4V)!
Of course, to have that load running continually, you need to heat away 30W somewhere, but I suppose that will not be the case here.

 

[KrisBlueNZ]

I imagine the most important feature of the rotation detector is robustness. These derby cars will be bouncing around and shaking like crazy, won't they?

My first thought would be a bump on the axle and a low-force microswitch with a metal arm that the bump pushes. I was about to suggest a cam on the axle but that would be a problem if you run it backwards! But I expect any of the other suggestions could work too. Please use your own judgement.

Anything mechanical (and possibly an opto solution as well, because of vibration) will not give a clean signal and will require cleaning up and debouncing of some kind. That can be done in firmware if you're using a microcontroller, or in hardware if you aren't.

Re the 4017 counter, I expect you'll need to count more than 9 revolutions before you start the motor. You can cascade two or more of them to divide by any product of any combination of two or more integers between 2 to 10 inclusive.

If you tell us more about your project, we can give more detailed and specific advice.

 

[CDRIVE]

While not a solution for you I couldn't help but post this McGuiver-esc bounceless switch. I imagine he would have made this with his pocket knife and used it to disarm a thermonuclear device that would detonate with anything more than one switch closure within a few mS.

http://licn.typepad.com/my_weblog/2...cury-john-dunn-consultant-ambertec-pe-pc.html

 

[CDRIVE]

No, this is a modern logic level type. It should be totally switched on at maximum 2.5V.
See the size, price and data of BUK9Y14-40B.

These are the most impressive specs I've seen on a LLFET to date. Excellent find!

PS: My apologies for my nastiness the other day. 'Grumpy Old Men' are not just a fictitious demographic. We do exist!

 

[Electrobrains]

PS: My apologies for my nastiness the other day. 'Grumpy Old Men' are not just a fictitious demographic. We do exist!

When I entered the discussions here some days ago, I felt like touching a wasps' nest (especially on that invention/adapter thread).
But I made some mistakes too. I realize it's not nice to jump into a thread with an alternative solution when someone else has spent a lot of time advising and inventing a working circuit.

Thanks for the apology! It proves that also "Grumpy Old Men" can be men of integrity!



For the righteous falls seven times and rises again, but the wicked stumble in times of calamity. (Prov.24:16)

 

[CocaCola]

I honestly feel keeping it simple and using a properly mounted mercury switch is the best solution for a quickie event like this... The design of the track makes it a dead simple solution... And it's only a 3 component solution, the switch the motor and the battery...