Hi Folks,

My name is Peter, I'm a magician living in the UK. 30 years ago I was an embedded s/w engineer working for Motorola in East Kilbride, Scotland but I remember very little from those days!

I'm trying to create a new magic effect but running into some problems I'm hoping someone could help me with.

Overall Effect (note some info about the trick has been redacted as the magic fraternity frown about revealing effects on public forums)
1) Spectator picks up 1 of 4 objects out of sight of the magician and uses that object, again out of sight of the magician
2) Spectator repeats this for the other 3 objects, again only one object in use at a time, all out of sight of magician.
3) Magician reveals prediction that matches the spectator's interactions with the objects.

This has been done many times in the past with a stooge in the audience and some sort of communication/telecoms but but I'd like to do an automated version using electronics. I've split the project up into 3 logical parts, these are...

1. Detecting which object the spectator is using
2. Transmitting that information physically to the magician
3. Communicating the information to the magician

So if we detect that the spectator is using object 2, we have to pass that info to the magician at the other side of the room. This cpould be IR, RF, carrier pigeon etc. The magician can then interpret the info, eg a coloured LED that indicates which object is in use, pulses on a vibrating motor, eg 2 pulses ofr object 2 etc.

The system has to be portable and run of batteries. The detection and transmission unit has to be thin and make no noise. The base will probably be around 80mm * 80mm but it would be good to keep the height under5mm if possible. The receiving unit must fit in a pocket. The communication method must be easily concealable from nearby spectators. A light buzzing is fine but a loud melody for different objects is out. A visible display, such as LEDs would have ot be able to be removed from the pocket and read without being seen by spectators.

I'm not worried about battery life, each performance would only be about 10 minutes and I can take the batteries out in between performances. While it is in no way safety critical it has to be robust as there I don't have a good out if the kit fails to work, I'd have to abandon that trick. There are no great time constraints on operation but the magician needs to get the info within 5 seconds of the spectator lifting the object. Finally it needs to be cheap, under £20 and can not take up a lot of development time. I could see how it could be done with an MCU but I'm not going to make a board, get an emulator/debug system etc for the MCU. Need to keep it simple.

That's the problem, any of you electronic wizards got some great ideas? This is how think I could do it but I've not tested any of it yet. Would welcome feedback or comments.

1) Detection Circuit

Each object will have a magnet in the bottom. The objects will all sit on a stand which will have placeholders for each object. Directly underneath there will be a reed switch (NO) which will close when the object (with magnet) is in the holder.

I'd like to do something with an RFID tag embedded in the object and a proximity reader near where the spectator will use the object but I know nothing about RFID.

2) Transmission Method

Was planning to use a simple rf module but the learning curve for this looked way too steep. In reading up about it I heard about the PT2262 remote control encoding chip and found a product that makes use of it. The remote control has 4 buttons that I will use for the 4 objects. I plan to take out the buttons and replace them with the reed switches. There is a picture of the insides of the remote. I'll remove the buttons and wire up the 4 reed switches in place of the buttons. The default operation is that when a button is pressed on the transmitter, the corresponding pin on the receiver will go high. It will stay high until the button is released. In my situation the receiving pin will go high when the object is removed from the stand, stay high while the spectator used the object and go low when the object is placed back on its stand.

3) Communication Method

This is the weakest part of the plan. It revolves round the magician having to bring something out of their pocket and look at it each time the spectator takes up an object. This involves the risk of a spectator noticing my actions, even if they don't spot the device itself. Current plan is to have a RGB LED wired up to the receiving unit on a long wire. This would allow me to take leave the receiving unit in my pocket and just take out the led itself, thsu reducing the risk of flashing anything to the audience. I would wire the switches, transmitter, receiver and LED as follows.
Object 1 --> S0 --> D0 --> red
Object 2 --> S1 --> D1 --> green
Object 3 --> S2 --> D2 --> blue
Object 4 --> S3 --> D3 --> white (red, blue, green all on) [I think that makes white] This will require a little bit of logic to ensure all LEDs go on when D3 is high.

As mentioned, I don't like this, there is a real risk of discovery with having to look at something. It would be much better to have a vibrate mechanism (not too loud) in my pocket. Then the table would be as follows.
Object 1 --> S0 --> D0 --> 1 buzz
Object 2 --> S1 --> D1 --> 2 buzzes
Object 3 --> S2 --> D2 --> 3 buzzes
Object 4 --> S3 --> D3 --> 4 buzzes

Unfortunately I don't know how to implement that. I could do it an MCU but don't want that level of complexity. I know how to use a 555 to generate a pulse or a continuous stream of pulses. I had thought about trying to use a 4017 to count the pulses and stop the 555 when the correct number had been sent but that was beyond my capabilities for design. Hopefully someone out there knows of a chip that you can set to a value then it can drive a 555.

My questions to the community

1. Is this feasible within the constraints mentioned at the start?
2. Are there better ways of doing any or all of it?
3. Is there a simple way of generating pulses from pins D0 to D4. eg if D1 is high, generate 2 pulses, if D3 is high, generate 4 pulses etc.

If anybody wants more information just ask.

Best regards,

Peter

 

those RF modules can work but need additional electronics.
i would rather consider ESP8266, it is way more powerful product and can be used not just as WIFI module but also as a controller
https://www.amazon.com/TBD-Controls...e=UTF8&qid=1500497639&sr=1-4&keywords=esp8266

 

Thanks panic mode, I'll have a look at the ESP8266, thanks for the link.

Best regards,

Peter

 

Hi panic button, it looks a great device, cant believe how cheap these things are. Its a great suggestion for my project as it would allow the transmission of data between two devices and I could program it to pulse the buzzer the correct number of times. I probably wont go for the ESP8266 for this project as...

• It needs access to a wifi network and I travel round many different venues. I don't just work at one or two locations. I think it would be too complicated to put the wifi password into the two units when I arrive at the venue. You could set one up to act as an access point and the other to be a station for that access point but I worry that it might not be reliable enough, too complicated etc. I've not tested it out but it fills me with dread.
• It adds another learning curve. You need to have an Arduino dev kit or similar and all the problems that comes with a new development system. Again the thought of getting it to work fills me with dread. I'm sure I can do it but I just want something that works out of the box. I don't have access to a dev system and not keen on getting involved with one.

I agree that the bare rf modules are way too much work for someone like myself. This is why I still prefer the remote key fob and remote control unit solution as it works (alledgedly) straight out of the box and all I need to do is connect the reed switches to where the fob buttons were and attach LEDs (with resistors) to the output pins of the rxing unit.

Thanks for bringing the kit to my attention though. Its amazing how much is out there!

Best regards,

Peter

 

That's very simple. You have an infra red LED on each item and when it is picked up it transmits a set of 2 3 4 or 5 pulses and then a gap. You have an IR receiver and it gives you a buzz via a motor on your body.
You can also have a zapper that gives a 1,000v zap each time the LED blinks. This way you will never make a mistake.

 

The only problem is anyone filming the trick with a camera will see the blinking

 

Wow, good save with the IR stuff on camera. I had no idea about that. Thanks Colin55.

Thanks also for your thoughts on how to make it work. I really appreciate you taking the time to read my description and come up with ideas.

I had though about IR for the transmission but was worried about line of sight issues. I don't have any experience working with IR other than say a remote for a tv. A lot of the remotes I've used have required tight sight alignment to the sensor. I'd always end up swearing and pushing the buttons even harder trying to get them to work.

I felt an RF solution (this product) was preferable over IR as there were no line of sight issues + and I could make use of an existing commercial solution. (for the transmission part)

What do you see as the advantages of IR over RF here?

You also spoke about having the transmitter in each object. That would certainly cut down the complexity but the objects are all fairly small, everyday objects that need to work. An example would be a sharpie pen or a whiteboard marker. While I can glue a magnet to the inside bottom of a pen, I'm not so confident of getting a whole circuit in there and making it unlikely to be noticed by a spectator. I'm sure it could be done, James Bond (more likely Q) has done it, just not sure of my ability to do it.

I liked the zapper comment, made me laugh! However it also did bring up the very serious issue of misinterpreting the data. I've rethought now about how the buzzer system could work. When I was just thinking about LEDs as an indicator (only as I could do that rather than pulse a motor) I would be able to recheck which object was lifted several times if needed. I might get distracted, forget etc. If we went with buzzes from a vibrating motor then there is the possibility of misinterpreting the number of buzzes or again forgetting them. Each object is likely to be out of the holder, thus keeping the D0 (for object 1) pin high on the rxing unit for at least 30s, sometimes longer. While D0 is high I could pulse the motor pause, then repeat until D0 goes low. I've been reading up and thinking about how to do the pulses and I think I have some workable ideas. Will post them up soon.

Thanks again for the ideas Colin55