Radio Controlled Paintball Gun
I recently posted about a laser tripwire audio activated paintball gun. For usability reasons I calved the project in two.This allows me to hide my electronics within the box containing the gun. Now the unit is a carry and drop sort of thing which can be configured before being transported. This makes setup a breeze, and is required on the paintball field when you need to be sneaky. All I have to do now is come by every now and then to refill the paint.
The paintball gun is now just controlled by the audio amplifier. I plug a radio into the audio amplifier and voila! When you transmit on the proper channel, the gun fires.
Some changes were made, and are reflected in the schematic.
Discussion follows.
The updated schematic appears below.
Schematic for audio amplifier relay driver
C1 is a filter capacitor of unknown capacitance. I tried a 22pF initially which was not sufficient. This capacitor prevents the audio input leads from acting as antennae and tripping the circuit in the absence of a sent signal.
C2 is a 220 uF 16V electrolytic which I used to power an 8ohm speaker when I was working on the circuit initially. I didn’t end up having to change it
C3 is a 100nF cap
C4 is a 10uF 16V electrolytic capacitor
C5 is a 1000uF 16V electrolytic capacitor. Before I put this one in, the relay would at first fire strongly, then progressively fire weaker with each shot until it didn’t fire at all. This smoothes out the 12V supply. It ought not to be required, but it helped so I left it in. If anyone can explain the behaviour to me I would be welcome. The 12V source ONLY fires the relay when Q1 is active. It’s the only load in that circuit.
U1 is the LM386N. I didn’t even use any changes to gain beyond the default. If you want to, an electrolytic of 10uF accross pins 1-8 will change gain from 20x to 200x.
D1 – D3 are 1N914 switching diodes. 1N4148’s are what I would have used if I had any. These work.
K1.1 is a relay I scavenged from a computer monitor. It requires 12VDC and switches up to 250VAC.
K2 is the switch in the relay. When on, it connects the gun trigger, firing the eTrigger in the paintball gun
This stripped down circuit is very small fitting into just a few square inches. I use AAA batteries in series for my 12V source, and these provide some capacity as well (I won’t change them every game)
I use a plain 9V for my 9V source. It would be possible with regulators to supply the 9V from the 12V supply, but I don’t have those parts.
The final part of the discussion involves how it operates. I bought some 9500XL motorola GMRS radios which have a nonstandard one pin plug for accessories such as throat microphones.
I used a 3 part pin from a cellular phone handsfree set which I no longer used. This pin allowed me to access the audio signal. Unfortunately it’s not a perfect fit, so it’s not just the audio signal that is triggering the circuit.
On the motorola radio, it has an LED which flashes like a heartbeat to indicate that it is on. The circuit now behaves such that when the radio is recieving AND the heartbeat LED is lit, the gun fires. This is not what I intended to do, but since it works I will not complain. As a bonus, the period of the heartbeat is roughly 1hz, which when coupled with three shot bursts(set on the guns etrigger), it seems sort of like a normal person firing even if the transmitter is held down.
I am donating the laser tripwire to another project. The ‘Cat Won’t Wreck the 600$ Speaker Anymore’ project.
Cheers, MikeTheMike
Pictures:
New Powerbolt 1000 board, Hitched up with Phidgets RFID reader
Greetings my absent commentees,
Even though there is never any chatter generated, I figured I would put this one up here because it’s pretty fun and easy. I recently got the Phidgets RFID, and had the need to change my locks anyways. had ordered the phidgets kit from thinkgeek, and it included a book with a how-to on rfid enabling your door.
I promptly went and bought the recommended lock, but was disappointed to find that it had a PCB revision between the publish of the book and my purchase of the weiser powerbolt 1000, rendering the books instructions obsolete. Not one to cry, I got on the internet, and someone named ziffster had posted a comment about using the door direction jumper as a solution for opening and closing. I used this idea, disabled the front keypad, and used a transistor to activate the locking button. Before when you left the house, you would hit the lock button, and depending on whether your door was right or left handed, it would lock it. Using the changeable logic of the right and left setting allows one to hijack the lock button to then open and close the strike.
I have determined a way to RFID enable the new powerbolt 1000 electronic lock with the phidgets kit for the cost of
the powerbolt lock 100$
the phidgets board $50
a transistor, two resistors, and a connector, plus a few wires. ~5$
Note that I don’t use any relays.
Here’s a few pics of the test phase where I was able to control both the direction and activation of the strike using the 5v outputs from the phidgets board.
-
- RFID on top left, Lock on Right, components dancing in the air
Here’s a schematic of the changes I made.
Next I am going to clean up the wiring, mount it in the door, and get down to implanting the glass ampoule.
I’ve mounted it in the door, now I am programming the door access function. Since my door now requires a computer (I still have a key as backup), I have decided I will add a webcam to take shots when people swipe their cards. I can see who is at the front door then. I can send the webcam shots via ftp to one of my servers (I pay for service for two websites), and then I could see the comings and goings at my house.
As for the programming. I took the RFID-simple example supplied by phidgets for vb.NET. I am using VB express 2008, and I didn’t have to change the code at all to get their program to work.
I then added a thread which performs the door opening and closing tasks by launching threads when the reader initially detects an ID, and then when it loses it the locking function has a longer timer. This way when I swipe in, I can enter the door, and close it behind me with a 2 minute delay, and not have to worry. I will include the code for the threading and you can figure out the rest. (Threads are the only way I found to get reliable time delays in vb express)
Imports System.Threading
(goes before anything else)
Private trd As Thread
(goes after you declare your form)
Private Sub ThreadTaskClose()
Thread.Sleep(120000)
phidgetRFID.outputs(0) = False
phidgetRFID.outputs(1) = True
Thread.Sleep(500)
phidgetRFID.outputs(1) = False
End Sub
(Those subs perform the tasks )
trd = New Thread(AddressOf ThreadTaskClose)
trd.IsBackground = True
trd.Start()
(those lines go in the section where the ID is detected, and also where the presence is lost, obviously the above is where the ID presence is lost. To open you just change the addressof to threadtaskopen).
I am also living in an area where my utilities have meters located within my house. I am thinking about setting up a few more cams to aim at the meters, and give the meter reader people a key. When they come to my house and swipe, the screen from the computer will display the two meters they need to read rather than letting them in.
Making my life easier? Not yet, but once you start taking the plunge total home automation can’t be far behind.
Cheers, Mike
Notes: After working flawlessly for a number of days, it appears that the weiser pcb has malfunctioned. What happened is that my door direction logic stopped working. The strike only moves in one direction now with that malfunction. I have figured out a way around it, but I will have to change the programming now to reflect it. In some ways it will be easier now.
As it is now, one output will always close the door, and need only be pulsed for 500ms as before (The transistor which is activating the door locker). To open the door now though I have to tie the other output to Q3 on the weiser board where Q3 hooks up to the microcontroller on the weiser board (I used the contact at Q3). Now I will have to time the logic to get the strike to open all the way, and I would guess 500ms is okay, but I will figure it out and let you know.
While I was discovering this, I also found that Q4 will rotate the lock in the other direction when +5v is applied. This means I could potentially get rid of the transistor as well, only using two resistors to take over the lock. This would also work well, because it would allow the front keypad to be used. I would probably disable the lock button on the front door, but the way it broke it actually will lock the door for me. If my door opened the other way it would open it, and that would be no good because that button is outside the house.
I will take some photos of this to make it easier to understand.
Unfortunately I did some things wrong, and fried my weiser board. I think I took out the FET used for one of the directions. The way Q3 and Q4 actually work is that when +5v is applied the motor does not rotate. When either become a path to ground, then they rotate, but in opposite directions. This means what I was trying to do would work, but you have to hold the logic high. This means the failure mode when a power outage occurs is such that either you will A) fry the weiser board by trying to rotate in both directions simultaneously (I’m pretty sure thats what I did) or B) You will have a failure which will result in continual rotation in one direction until the batteries give out, or the over rotation protection built into the lock fails, and your gears will be all messed up. I suppose you could build in a protection circuit, but that seems like it’s getting too complicated. The solution I expressed to begin with is pretty good. Use the lock button transistor as previously described, and use Q3 or Q4 depending on your door direction for the other direction. This means the front keypad is still somewhat useless, but you could snip the one trace that is for the door lock button and still use the numbers if you want. I just want the RFID to control it so my options are open.
Now to find a replacement FET for the motor. Ergh. Do it right the first time, learn from my mistakes.
Automatic Turbo Booster Fuel Controller
Greetings,
Occasionally my friends pop in and ask me “could you make this happen?” Often the answer is no, however, on one occasion a friend of mine who likes to make cars faster than the engineers intended queried thus, and I was able to help him.
He had recently installed an oversized turbo on a Precidia, and had installed a air/fuel ratio gauge to determine if when boost pressure was present that the fuel mixture would still be OK (stoichiometric that is. I think the appropriate value is around 15:1). Unfortunately when his boost pressure was where it needed to be the air/fuel ratio was wrong (off the gauge it was so wrong). There was far too much air, and not enough fuel. In the short haul this simply means less performance than expected. In the long run, the incorrect air/fuel mixture will result in metal in the cylinders being oxidized at a rapid rate leading to engine failure.
He is smarter than most chaps so he looked up the problem on the internet and found some sort of solution. The ECU for the Precidia will supply maximum fuel to a car whose temperature sensor reads cold, and then will adjust down to allow for greater fuel efficiency when the car warms up. So He intended to trick the ECU by having two lines with a switch to the ECU. One would be to the temp sensor so he could still daily drive the beast (because he put in an oversized turbo, boost pressure wouldn’t develop until over the city driving speeds), and the other line would be to a resistor of appropriate value designed to fool the ECU into thinking that the engine was at 0C, and thus would pour fuel into the cylinders.
The switch was inconvenient, as he had to monitor the boost pressure and kick the switch over at just the right PSI of boost (7 PSI). Falsely triggering the switch at the wrong moment caused a chugging which resulted in crap acceleration. This proved to be too tedious for my friend.
So he asked me to make something which would sense the pressure and throw the switch for him. Thus the turbo booster fuel controller was conceived.
I used a PIC 16f819 to monitor the analog output of a Freescale MPX5700 pressure sensor, and then fired a relay using a ULN2803A. For power I took from the remote line of the ignition wiring so it would only be on when the car was on.
I wrote a program in ME picbasic pro which reads the analog voltage from the sensor, converted it to a 10-bit number (defined the variable as a word not a byte), and then proceeded to calculate the value that would correspond to 7 PSI.
In the end the program required some tweaking such that when the relay fired, there would be a debounce time associated (500 ms) with it because the relay would clatter when the correct pressure was reached. We also played with changing pressures slightly and got better performance than when he had the switch.
This also meant his wife could drive his car again on the highway without getting stressed out about having to throw a switch at the right pressure.
The device worked beautifully until one day it got semi munched and some trace broke somewhere. This was OK as my friend had received a new megasquirt on mail order about the same time. Megasquirt is THE way to take control of fuel injected engines and allows full control over all aspects. You can even make fuel distribution maps which correspond to a few variables.
Here is a schematic of the circuit:
U1 is the pic 16f819
U2 is the voltage regulator
U3 is the uln2803a
C1 is 10uF electrolytic
D1 is 1n4148
R1 is 100 ohms
K2 is a 12v DC relay (I steal all my 12v relays from dead CRT’s)
Port 4 is actually the pressure sensor, it actually has six plug, but you only need three to make it work. I probably should have some more caps near the sensor, as in the datasheet suggestions. I did not use them and it worked fine.
Here are photos of the finished project: you will notice the circuit and the final product are not completely concordant. I had thought I would perhaps be controlling more than just the fuel, and I also thought to leave other lines to try PWM output to a filter and then send the voltage to the ECU. Never had the chance.
Component side of board
Circuit Side
Cheers, MikeTheMike
Paintball Chronometer
I have edited this post for length and removed some useless information which was part of my learning process.
I play paintball, and the basic piece of safety equipment is the goggles. Next in line is the chronometer. Virtually no hobby woodsball players own one (I’ve only met one). It’s always more attractive to spend money on upgrades, new guns, camo clothes, or perhaps the ladies, so it gets overlooked.
I decided to make one which though inferior to the radar chronometers available would be cheaper components wise. It will force me to learn how to use timers and (try) to optimize picbasic pro code. It employs a PIC 16f88 which has enough lines to provide signal for a 7 segment display, two digital inputs, plus a few more lines to switch for multiple 7 segment displays.
The premise for the operation of the device is this:
Paintballs move at a rate of speed commonly measured in feet per second. As an adherent to the metric system, I would prefer meters per second, but feet will do because it is the industry standard. To measure the rate of speed, I am going to place two IR phototransistors 6 inches apart, and have a barrel extension. The balls will leave the barrel of the gun, break the first IR beam, a timer will start in the PIC running at 8Mhz. The ball will break the second IR beam six inches later, and the accumulated time in the timer will be used to calculate feet per second. The result will be displayed on three 7 segment displays to indicate feet per second. Typical values found in paintball are 250-450 fps. 450 hurts and can break the goggles leading to eye injury. Keeping it at or below 300 is considered polite. My friends crank theirs as much as possible for accuracy and range benefits. I am hoping once I get the chrono built I can make my fellow players tune it down a notch. This will improve the game for everyone.
So to the guts of it.
I am using a pic 16f88 to read the input from two IR phototransistors which have IR LEDS blasting them constantly. The ball will block the emission from the LED, this will change what the pic reads from the phototransistor.
The PIC is also used to drive 3 seven segment displays. I haven’t written the code to drive them, but I plan on flashing the LED’s for each seven segment display one after the other at a high enough rate of speed to ensure persistence of vision (ie. each will be displayed enough to ensure 20 times per second) and continue the pattern for 3 seconds. In my circuit, the PIC is generating the signal for the display modules which have a common cathode. I cannot source the current for the display from the PIC directly, and found out I will have to use a UDN2981 to provide the extra current sourcing capability.
I am using 2N3904 transistors to allow the circuit to be completed for individual digits. This provides a convenient way of switching which digit I am currently manipulating by turning two outputs low and one high to activate the bases of the transistors.
I am aware that brass eagle owns a patent for this device, but I have never seen it for sale. The radar chrono’s must be competitive enough cost wise, and performance wise they are nicer because you don’t have to attach anything to the gun. They save time, but I estimate that as a final product these would be about half the price. In addition, I estimate that the accuracy of this device would be sufficient to act as a ballistic chronometer for guns if some minor changes were made (lasers instead of LED’s).
Update: Here is the schematic less the sensors. I did buy some infrared led/phototransistor pairs. For your reference; CdS cells are too slow for this application (msec resolution is not enough). Photodiodes or phototransistors are appropriate. Phototransistors do not require an amplifier chip either. To get even better time resolution, a photodiode plus an op-amp would be the best. For my purposes, phototransistors are fine.
schematic for controlling 7 segment display
Q1-Q3 are 2N3904
U1, U2, and U3 are seven segment displays with a common ground. The actual part I used does not exactly correspond, but for schematic purposes is fine.
U4 is the PIC 16F88
R1-R9 are 220 ohm
R10-R12 are 100 ohm
D1 and D2 are red LED’s used for status. One is for power on, the other will be used to indicate over/under 300 feet/sec
+5v is supplied by lm7805 from a 9V battery.
I will be attaching the voltage inputs from the two sensors on RA1 and RA0
Note this schematic does not include the source driving chip which needs to be a UDN2981. This will be reflected in the PCB. Essentially the lines coming out of the PIC for the seven segment LED units are piped through another chip one for one, so it doesn’t really change much.
Update: I played with my LED/phototransistor pairs, and I have success. I will be able to use digital inputs for the sensors as I get a reading of 4.95V when the light is not blocked, and 0.03V when the light is blocked. These are perfect values for the microcontroller.
One difficulty for me is that I am trying to use the infrared LED at a maximal power level. This created one hurdle for me: I needed 1 watt power resistors. By my calculations I needed 40 ohm resistance, and I was very lucky to find two 50 ohm power resistors in a discarded CRT monitor PCB.
Here is the schematic for the sensors; note that I could not locate the phototransistor symbol in the diptrace library, and was too lazy to create one for it. So I just stuck in an npn in which the base is the signal generated by light striking the phototransistor.
Schematic of IR Sensors
R1 is 50ohm 1 watt
D1 is an IR LED operating in the region of 120mA
Q1 is a phototransistor
R3 is a 10K 1/8 watt
This schematic represents one sensor set. In reality there will be two of these in the device, set six inches apart. The light from D1 strikes Q1 and the voltage read at the measurement point is 4.97V when the LED is not blocked and 0.03V when blocked by a piece of plastic. I need to try a paintball, because I noticed that when I blocked with a piece of cardboard there was still quite a bit of transmission.
On the bright side, this means I can now begin on the code portion of this device. I am probably going to update the chip with ICSP as I anticipate a number of revisions during development. I will probably write the 7 segment driving portion first and include 3 digit switching. After that I will implement the timer/velocity code and use the display driving code to put out the result.
I’ve decided to re-do the PCB, and I will incorporate the UDN2981 to eliminate the problems with sourcing current from the PIC. With the 16F88 you can only source 100mA per port. I will be sourcing up to 175mA per digit.
I am still going to use 5V as the source voltage for theUDN2981
PCB:
Note that there are going to be many jumper wires for the PCB. I only make single sided PCB’s at this point partially due to simplicity, and partially because thats all I have. Unfortunately things like the display elements force the creation of wires. I will probably use pin headers for the ICSP connections.
I am expecting the arrival of the UDN2981 within a week. I am quite excited to finally finish this project. I think when I do, I may write to brass eagle about perheps licensing their patent to make kits. There has been significant interest in this project, and at the end of it, it will provide usefull help in a number of arenas;
Using phototransistors as sensors
programming the PIC16F88
driving 7 segment displays.
PCB etching using a Copper Solution SAVE YOUR MONEY
Preface: Anytime you play with chemicals, there are dangerous situations which can arise; burns of exposed skin, vapours which can cause blindness, or lung damage. I personally know someone with visible scarring from an incident with concentrated acids (it wasn’t even their fault, someone else tripped and it got on them; SO unfair). I am not going to be able to make predictions about what is going to happen if you try to generate the solutions listed in this article. The ingredients are wholly unhealthy, and only someone with deep respect for safety should even THINK about etching their own PCB’s with their own solution. I had reservations about it and I have a degree which included years of study on the subject of chemistry. I am able to do the math, and I did the math initially while developing a non mathematical method of solution generation and maintenance.
My basic point: I won’t be held responsible, Do your homework, understand exactly what you are making!
PCB etching is an invaluable capability for the home electronics DIY person. You will reach a point where you want to make 10 of the same item, and doing it with wires and breadboards is not only frustrating, but a waste of time. PCB’s give you many advantages. Stability, easy refinement of design on a computer, and best of all you can avoid have to use wires which criss-cross all over the place (and act as antennae) are just a few of the advantages. The best feature is the ability to perform semi-mass production.
I use computer software and the toner transfer method to prepare my boards for etching (more on the toner transfer method in another post eventually). The software I use is the free 250 pin version of diptrace. I love it! I have looked at Eagle, but I find the diptrace interface to be way more intuitive, and it has also been easy for me to make new parts using my digital calipers for measurement and their part editor to create the new one. If I ever start to get into the commercially profitable end of electronics, I will be buying the 500 pin version of diptrace without a second thought.
Getting into PCB etching leaves one with a number of entry points. You can be a constant consumer by using the iron chloride or ammonium persulphate methods. These solutions will yield excellent results, but eventually you will be throwing the solutions away. The concentration of dissolved metals and the acidic nature of the solution demand that you handle them as toxic. I’ve read that in most jurisdictions you would need to dilute the solution over 1:20,000 times for a safe disposal concentration. Are you going to mix your 1L of waste into 20,000L? I didn’t think so. As someone who works daily in the field of toxicology, I can assure you that it IS A BIG DEAL if you pour toxic waste down the sink. YOU WILL KILL wildlife, DESTROY habitats, and if I ever find you in a dark alley you will taste the iron from your blood before we are through. It is true that you can also pay to have it disposed of, but there is perhaps another way to approach this problem which has some upsides. These will be summarised at the end of the article.
I will now refer you to an excellent resource on the subject.
http://members.optusnet.com.au/~eseychell/PCB/etching_CuCl/index.html
This fine fellow has done all the hard work (Mathematics) determining how to use a copper solution for etching. An understanding of high school chemistry is all thats required to understand it. If you did not take chemistry in high school, I will present three subjects within chemistry which you would need to understand; Stoichiometry, Molarity, and Oxidation/Reduction reactions of ionic compounds. These can be covered on your own in a few weeks to a few months time depending on how much time you can devote to it.
I will now give you my meta instructions on using the copper method, this will simplify things for you, and can allow you to operate without performing calculations. To use this etching method I insist you obtain:
concentrated HCl, AKA Muriatic acid: can often be obtained at hardware stores as it is used to clean concrete
Hydrogen peroxide, 10% would be ideal, higher percentages can be violently unstable, 3% from the drugstore will work
Pure Water, Obtain deionized water which is often sold at automotive stores for use in dilution of coolant.
Ice:I get it from a restaurant I used to work at.
Sodium Hydroxide: Often available as soapmaking ‘lye’ is a white powder. Do not use DRANO it has other adulterants.
optional: Copper chloride, highly recommended as it will speed up your initial solution generation.
When working with chemicals always ALWAYS wear complete PPE consisting of (at a minimum): face shield (I use a paintball mask), respirator (to help handle acidic fuming, a gas mask would be best), elbow length gloves (available from commercial kitchen suppliers), rubber or nitrile gloves (scientific/medical supply), closed toes liquid proof footwear (rubber boots), and a lab coat which goes to the knees.
You will look like a dork, but at least you will be a safe dork.
Method with copper chloride:
1.Dissolve 50g CuCl in a minimal amount of pure water to saturate the solution.
2.Add two volumes (to CuCl solution) concentrated HCl (25%-30% is best) SLOWLY while stirring with a glass rod.
3. Add some solid copper chunks (100g)
4. Add 50 ml hydrogen peroxide SLOWLY while stirring observe the reaction for excessive temperature (>50 degrees Celsius) and cool on ice if necessary.
Method without copper chloride:
1. Combine 3 volumes HCl with 1 volume hydrogen peroxide(3% or 10%) by SLOWLY adding the acid to the hydrogen peroxide while stirring constantly. If the temperature exceeds 50 degrees C, cool on ice.
2. Add 100g solid copper and monitor the reaction for excessive temperature(>50 degrees C and cool on ice if necessary)
3. Add more solid copper if required to reach an acceptable concentration of copper ions.
you should see the solid copper dissolve, and the color of the solution darken. To convert the copper ion to the proper charge required for etching add more peroxide, this will brighten the green color of the solution and make it primed to etch your boards. Don’t use pennies, they have other metals inside them these days, and in Canada it’s also illegal to deface the Queen.
As you etch a board the solution darkens, the copper ion concentration increases, the acidity decreases.
Most often you need to regenerate the copper ion to the correct charge. You can do this two ways: bubble air from an aquarium bubbler (very, very slow), or add hydrogen peroxide (very, very fast) it’s up to you.
You will also need to add acid from time to time though not as often.
The end result is that your volume of etching solution increases over time. With a copper based etching solution you can be a very good person and pass half your volume on to another person to get them into etching their own PCB’s, or you can deal with the solution without throwing it down the sink initially.
To deal with the solution, you will need to add sodium hydroxide pellets to your used solution. These are just as dangerous as concentrated HCl and the same precautions should be used. Also note that when NaOH is added to water, heat evolves so be CAREFUL. Constantly mix the solution as you add pellets slowly allowing them to fully dissolve before adding more. Eventually you will get a solid precipitate.
The idea here is to force the copper ions out of solution by precipitation through changing the pH to a high enough basicity. The copper will combine with hydroxide to form solid copper hydroxide. You can then allow the precipitate to settle and pour off the remaining solution which should be clear and highly basic.
Neutralize the basicity with HCl, and monitor pH using strips (available for pool maintenance) until you see pH of 7.0 then you can pour the solution down the drain as you will only have salty water remaining. Salty water of sufficient concentration can be harmful to wildlife, but is many orders of magnitude less harmful than a concentrated dissolved metal solution, if you dilute it as you dispose of it you will be fine. So turn on the tap as you pour it down the sink.
Tips on etching: I like to heat my solution (increased collision kinetics) and raise and lower my boards repeatedly (helps wash away layer of Cu 1+) to increase etch speed. I also find that I can monitor the etch better this way and remove the board as soon as possible. I heat my solution in a glass pot OUTSIDE to avoid poisoning my GF and cat with acid fumes.
NOTE: This solution is very aggressive. I had a beautiful stainless steel sink which got very minor drops on for a very short amount of time and it is ruined. DON’T bring this solution near anything you think is remotely valuable.
To dispose of the drips which coat your board as you remove(and to stop the etching) have 15-20 liters of water in a plastic container next to your etching solution. When the etch is complete remove the board and allow as much as possible to drip back into the etching solution, then sumbmerge the board in the water bath.
Dispose of the water bath by running the tap in your sink and pour about 1L per 15 seconds. By reducing the concentration using the water bath it will not harm stainless steel, and you will be achieving the desired 1:20,000 dilution so as not to be toxic (this is an estimate, and depends on board size).
Additionally: to get good etch times, include ‘copper pours’ on your PCB design. Not only does this save time, but it also means less solution waste.
As promised: The upsides
No expensive recurring costs. HCl and hydrogen peroxide are cheap and easily available. Orders of magnitude cheaper than re-buying etching solutions.
A solution that grows: Instead of becoming less effective, it remains effective and you get more of it. Bonus: You can give the extra to a friend. When you show up with a bright green solution for free, their smile will be payment enough.
Friendly disposal: Disposal of toxic substances is not a fun task, but it is the reality of etching circuit boards. With the copper chloride method, disposal is not too difficult or too toxic. You can bundle the precipitated CuOH in plastic bags and send it to the landfill. Alternatively people use it as a fungicide in gardens (in small amounts), and I’ve heard you can mix it in concrete with no ill effects. If you know of any other uses for Copper hydroxide please pass them on to me.
Happy PCB etching.
Cheers, MikeTheMike
Laser Tripwire Audio Input Activated Paintball Gun
I play paintball with a varied group who seem to have nothing but hundreds to thousands of dollars to invest in guns. I personally bought a cheap Tippman Triumph eTX for 50$ on clearance. This gun was fun, but is bulky and red, so I spent some more money for my current weapon a spyder electra 08. (This was a horrible mistake. The triumph is a far superior gun that costs five times less, with non electric versions selling for 35$). I still use my spyder and suffer from short range.
To even the playing field I knew I would have to make some items that would allow me to mess with people who have superior firepower, and it would also have to be an item that is not for sale, because they would just buy it. So I considered a variety of choices…
In the end, I made a laser tripwire/radio activated paintball gun using a Tippman Triumph eTX, a CdS cell, and a lm386 opamp.
Firstly, I consulted a dark activated circuit found here as my inspiration:
http://www.uoguelph.ca/~antoon/circ/actrelay.htm
Except I made some changes, and a mistake. I had a circuit board left from a previous experiment with relays and a ULN2803A which is one of my favorite IC’s because it has built in protection diodes, has multiple lines in and out, and can do highish voltage(50v), and decent amperage(600mA per darlington) on the relay side ofthe chip
I rigged the circuit up as follows.
BT1 is a 9V Battery
U1 is an LM7805 to step down to 5V
R1 is a 1K resistor
VR1 is a 50k Potentiometer
VR2 is a CdS light dependent resistor which ranges from almost no resistance to meg ohm.
Q1 is a 2N3904
R2 is 100 ohm
D1 is a red LED I had laying around
Q2 is and npn darlington transistor w\built in diode for protection. This is a functional unit from a ULN2803A IC
K2.1 is a 12v DC relay.
K1 is the switch that K2.1 throws. The normally off position switches to on when the relay is powered.
I did perform a cardinal sin, NPN transistor circuits like my 2N3904 should have the load in the collector portion of the circuit, not the emitter. Nonetheless, it still works, so I didn’t change it. This portion of the circuit then detects light at the CdS cell, and if no light is detected, the transistor fires which then fires the downstream relay which joins the trigger switch in the gun The potentiometer provides sensitivity adjustments for different ambient light levels This worked OK. The device itself is not really useful for me BECAUSE IT IS TOO DANGEROUS. I play in an area frequented by other non-paintball users who are not wearing masks, and I therefore cannot have an autogun set up. I wired in a switch to disable the laser tripwire portion in case I go to a dedicated play area where an autogun would be considered safe.
This got me to considering what a safer method would be to have automation, without the likelihood of shooting a non-player.
We use radios during play for tactical comms so I got to thinking, I could use my extra radio as an activator if I could read the audio as an activation signal. Alternatively, one could use the radio audio signal to quell the activity of the autogun while non paintball players are present (I didn’t do this because I still think an autogun is too dangerous)
This is my first attempt at using an op amp, so no doubt I have done some silly things, as I cobbled it together from various sources.
See Forrest Mims book, and this webpage:
http://www.uoguelph.ca/~antoon/circ/relay4.html
BT1 is 9V
U1 is lm386n
C1 is a filter capacitor to stop the input lines from acting like antennae I don’t know the value, it was laying around and when I tried it, it worked. At first I used a 22pF which was insufficient.
C2 is 10uF, and sets gain to 200X for the lm386
C3 is 220uF, and I used this to test the circuit with an 8 ohm speaker.
C4 is 100nF
C5 is 10uF
Q1 is 2N3904
D1,D2,D3 are 1N914 switching diodes
K1.1 is a 12v relay which switches K2
K2 ties a +5V signal to part of the original tripwire circuit
R1 is 100 ohm
D4 is a red LED
Q2 is the functional unit from the ULN2803A IC and has a protection diode built in
K4.1 is a 12v relay that switches K3
K3 switches the gun trigger contacts when power is applied to K4.1
The important part (as far as I can tell) is using the diode to rectify the voltage to DC after it leaves the lm 386
I then learned my lesson about NPN transistors:
Your load should be included in the collector portion of your transistor circuit I could not tie my emitter in to the same place as my laser tripwire because of this! This meant I had to add a relay, and use the relay to trigger the same section of the circuit which the tripwire portion did.I tried adding an extra transistor instead of a relay but it wouldn’t work for me.
This was silly. I now know that I could have optimized this circuit because I made some mistakes which came from not understanding how to use NPN transistors correctly.
I would love to have criticism of the circuit, especially of the lm 386 portion as I really don’t know what I am doing.
Cheers, MikeTheMike
P.S. Here are some photos to bring some life to the circuits
This is a picture of the trigger portion of the eTX opened up. You will see two green wires leading from the trigger. This is the only point where I modded the gun. This is convenient because I can set the gun to single shots or three shot bursts with no additional messing around.
This is a picture inside the box I mounted it in. You will behold all sorts of bad practices like long floaty wires, things which are not secured, and of course some uninsulated connections just waiting to short out. I say too bad for you. It’s my lasertripwire radio activated paintball control box. EAT PAINT.
The front of the box. I took my GF’s camera box. She won’t be impressed. You will notice I use pin headers and munched connectors to hook a lot of things up. I find the connectors inside dead monitors, and I bought a bunch of pin headers cheaply in the past.
The CdS sensor is also visible. I experimented with a reflective inside of the tube or a blacked out inside. It’s easier to use when it’s reflective
The other side of the box demonstrating the rest of the connector
My brother built the box to secure the gun and hide the red. He’s a good brother.
hes a good brother
