Hero999

Four rechargeable AA cells in series should last over 800 hours with the circuit as you've drawn it. If you want to reduce the power consumption further, then use a micropower op-amp such as the OP290.

If you've tested Q5 out of the circuit and the measurements are as above, then it's almost certainly bad. Don't worry too much about other components failing, just keep plenty of spares.

It depends on they type of sensor you use. A cheap proximity switch will just output a high when the object gets within a certain distance of the sensor; most are adjustable but you need to use a small terminal screwdriver. A more expensive proximity switch will output a current, voltage or digital signal which varies depending on the distance of the object so you can use a comparator with an adjustable threshold to alter the trigger distance. Then you can get plain ultrasonic transducers which need to have a driver circuit which you design. If you want to use a PIC, then you might as well use transducers. I've nver done this my self but it shouldn't be too hard to send the transmitter a 40kHz signal, and time how long it takes for the echo to be received. The speed of sound varies with the air temperature so if you already have a temperature sensor it shouldn't be hard to compensate for it.

I've never hard of foundation paper before. I've just Googled it and it came up with this which is very expensive, you can buy special proprietary toner transfer paper for less than that. http://www.creativegrids.com/acatalog/info_N0234.html

It's up to you of course. If this is just an educational project then it probably doesn't matter how useful it really is. Have you considered the VCO idea I talked about in my first post?

Tandy/Radio Shack have gone from the UK and Canada but still (as far as I'm aware) exist in the US. As you probably know inkjet won't work. A plastic rolling pin should be fine; it shouldn't melt as long as it doesn't come into direct contact with the iron for too long.

Yes, it might work now, as I said earlier, that doesn't mean its life hasn't been shortened. That means its broken, it shouldn't conduct between the collector and emitter without a base signal. Did you try to test the transistor whilst it's still in the circuit by any chance? it shouldn't conduct in the reverse direction, unless your test voltage is higher than 5V which is too high and might damage the transistor. From the results given above, it looks like you've tested the transistors whilst there still in the circuit.. To do any meaningful testing, the transistor must be removed from the circuit before it's tested, otherwise other components in the circuit can influence the result: for example a resistor connected in between the base and collector can turn the transistor on. I think you need to read up a bit more on transistors before continuing.

Do you have the schematic for it? My guess is no, it can't easily be done because the tubes are probably wired in series to the inverter circuit. You'll probably need to rip out the inverter and replace it with a custom made unit.

I don't know where you get the patience from. I thought about responding, then I decided against it as the original poster doesn't seem to be making any effort to try to figure it out for himself. Of course you could add an additional transistor to control the siren, rather than having two relay contacts.

Any of the semiconductors could have been damaged as well as some of the capacitors, especially C13 as it's connected directly across the power supply. Providing this project is not for sale, don't worry about it, just keep a supply of spares.

No RS components are a totally different company to Radio Shack, they are a British based company, Radio Shack is US. RS components are a high quality supplier, Radio Shack, which used to exist in the UK under the name Tandy, sell overpriced crap. Yes, magazine paper works. I thought that the think paper used for RS component catalogues would be too thin for a laser printer but others have used it with no problem. I would generally recommend slightly thicker glossy clay coated magazine paper. From memory the paper used to print Tandy catalogues would have been perfect. The actual optimum temperature varies with the type of toner you have, it's just that many people have found around 180

A quick test for an NPN transistor is to put the positive on the base and check for a 0.6V to 0.7V voltage drop when the negative is connected to either the collector or emitter. To test PNP transistors, just reverse the positive and negative connections. Unfortunately these tests aren't perfect, a transistor might pass but won't work properly because it's not completely broken but badly damaged, for example the gain might be deverly reduced so it no longer amplifies enough. Because the circuit was connected to too higher voltage, it's possible that some of the other components have been damaged, drastically shortening their lives. If you replace the transistor and reduce the transformer voltage, it might work for awhile, then fail again as a result of the damaged caused by the over voltage.

Yes, temperature setting is critical. I found, that if it's too hot, the toner can smear and some of the magazine print can transfer, especially if too much pressure is used, too cool and the toner will never soften. Yes clean the board until it's really bright and shiny. Make sure that you don't touch the toner on the print-out: any grease on the toner might stop it transferring to the board. Is it always the same part of the board which isn't transferring well? I've found out that it can often be difficult to transfer around the edges, you should never track too close to the edge of the board anyway. If everything else is right, then the chances are you're not applying enough pressure to all parts of the board. Try putting a rolling pin on the other side of the board, roll it back and forth as you iron, then repeat with the board rotated by 90

The circuit failed because the input voltage was double what it was designed for. Look at the original schematic. The transformer's centre tap is connected to 0V so the voltage at either winding (with respective to 0V or neutral) will be 24VAC. The voltage at one winding will be 180

Using a non-tapped transformer and bridge rectifier won't make any difference so long as you reduced the secondary voltage to 24V. Which transistor failed? Simply blindly replacing it won't help, the chances are it'll fail again. There is potential for Q5 or Q6 to go in to thermal runaway, especially if R9 is adjusted incorrectly. Did you correctly follow the instructions, setting R9 to its lowest value and increasing it until the quiescent current is 25mA to 30mA, wait 15 minutes and adjust if required? I'd recommend doing this with a speaker or 3R9 to 8R2 dummy load resistor connected. Adding emitter resistors to Q5 and Q6, also might be a good way to prevent thermal runaway..

Thanks for posting the PCB. It's still a good idea to check the schematic here against the original because lots of the schematics here have been copied incorrectly..

No I don't think anyone does and if they did, you'd need to be sure that it's for the latest version and that you can get hold of all the components with exactly the same footprint.

I've never seen a 2200΅F capacitor, I presume it means 2200

Use a parabolic reflector as I said above. You can take the reflector from a torch and replace the bulb with the temperature sensor. The compensator should be near the sensor but not enclosed within the reflector. It doesn't matter what type of sensors you have as long as they are as identical as possible. I'd probably use precision thermistors but a temperature sensor IC will do. I got the idea from a book: Forrest Mims Sensor Projects, page 42 & 43. I wouldn't recommend using the exact circuit in the book, rather than the 741, I'd use an LM311, LM393, LM339 etc. and probably make a few tweaks to the circuit such as adding hysteresis. If the sensor is designed to very directional, then simply sweeping the sensor from side to side will give the user an indication of where the obstacle is. Well if the fire is behind them, then at least they're already walking in the right direction. ;D To be honest, I don't see much point in a fire detector. How many times does one go walking and encounter a fire? If the user is at a bonfire party then the fire will be so fierce they'll be able to smell the smoke and feel the heat on their skin. And as I said before, a fire detector will be activated by sunshine and other heat sources such as incandescent lighting. This could be partly solved using a blue light sensor to disable the heat sensor in sunshine but then it will become unable to detect fire.

I'm building a flat pack piece of furniture, I have all the diagrams and instructions, the trouble is I've lost this screw and don't know what to do.

Either of those sensors will work, each has its advantages and disadvantages. If you use only one temperature sensor and set it to sound an alarm only when it gets really hot (i.e. >50

It depends on what you need to do I suppose. Do you want the light to just flash once when something is placed near the sonar's output or do you want it to stay flashing as long as something it near the sonar's output? Bear in mind that the sonar circuit needs to be isolated from the mains flasher circuit for safety reasons.

You can also buy ultrasonic proximity sensor modules which turn on an output when an object is placed within a certain distance. For the thermistor circuit, you actually need two thermistors, one behind a parabolic reflector to sense radiant heat and the other sensing the air temperature, to compensate for ambient temperature changes. A comparator could be used to activate an alarm when the thermistor behind the parabolic reflector senses a higher temperature than the compensator. An LM324 could be used but the LM311, LM393 or LM339 are more suited to comparator applications. What else do you want to do with this project?

You can get pretty simple proximity sensors designed specifically for this type of thing - Google it. There are different types of proximity sensors, it's up to you but I'd recommend ultrasonic because they have the greatest range and are pretty accurate. Why use a PIC? Heck, I'd do this the analogue way because it's easier and cheaper: a VCO connected to a speaker or piezo buzzer, could monitor the voltage and increase the frequency of the tone or pulsed tone as the object is approaching. Don't bother with an expensive VCO IC, it's easy to make one from a couple of op-amps. A momentary switch could be placed in series with the whole circuit so it's only active when needed. I know the analogue way isn't normally the cheapest but this is an exception, so unless you have to use an MCU then I'd recommend it.

Yes, there needs to be some hysteresis to prevent oscillation, just connect a 100k resistor between the op-amp's output and the + input. This will make the turn on voltage higher than the turn off voltage, see the data-sheet linked above for more information. I wouldn't use the LM324 anyway, the LM339, LM393 or LM311 would be much better because they're specifically designed for use as comparators. If you only need one op-amp/comparator then I'd recommend the LM311 which can drive a small relay without an additional transistor. Don't get me wrong, it'll be fine with the LM324, it just wouldn't be my first choice.