| How to Make Things Work! |
|---|
How to Make Things Work So you spot this really cool radio project in SPRAT or say while surfing the Internet. Immediately a spark races down your spine --- you have to build this! It is something you thought about long ago and now here it is staring you right in the face. Your mind races forward, right past the building process (it was only a mere blur) and now you are ready to apply power. Gulp! Reality sets in: Will this work? Did I build it correctly? Did I miss something? Will it possibly vaporize in a huge cloud of smoke? Well all of these sorts of things can and do happen. BUT there is a way to hedge your bets so "that very special project" will work at turn on! Read on and learn "How to Make Things Work." The Preparation Part There is no real mystery about making things work from the get go, that is if you take a logical approach to the entire process. The first step is to not heat up the soldering iron; but instead to spend some time on the front end developing a plan of how to approach the construction of a project. The very first step in "my plan" is to collect as much information as I can about similar projects and to avail myself of the Internet resources. For each project I create a "Project File" on my computer and collect all of the data in that file. However, there is a trap to simply dumping all data into one file, because as the file grows you will not be able to find what you have stored. The operative words here are organizing the project file. Within my Project Files I create sub-files with titles such as "Data Sheets" another called "Related Articles" and still another called "Photos". If this is a project where I started from scratch and it may end up as a publication two other files are created and include these titles "Article Folder" and "Working Folder" If you think this is being "anal" trust me it is worth the effort especially where some problem may be encountered and you need to refer to a data sheet. You will only need to look at the "Data Sheet" file and not one thousand piece of non relevant data. Usually I start with downloading the data sheets for all of the devices that will be used in the project. Next I download information about special components such as crystals and transformers. I carefully note any cautions about these components especially information about susceptibility to static discharge, voltage ratings, cooling requirements, bypass capacitor values and installation precautions when soldering the components. Armed with this information the next step in my plan is an attempt to size up how much real estate will be required to build the circuit. For this part of the project I use a mock-up process that employs a piece of blank vector board mounted on a wooden fixture. The fixture is just a couple of pieces of 1X4 wood that are screwed together to form a box. The attachment of the vector board to the wooden box is with 3M masking tape. By placing the vector board on the wood box structure this enables passing "through hole components" right through the vector board so their leads are not bent and the components lie flat on the board. It also facilitates space allocation for SMD components. See the two photos below. On the left is the mock-up and on the right is the final board prepped to take the parts. Later where there may be a series of individual circuit boards that make up the project that brings up another "old trick" -- breadboarding. There is a separate tutorial on "breadboarding" see the prior link page.
Using the schematic as a guide, I lay out the parts on the vector board utilizing the schematic as the road map. This accomplishes two things: 1) I have a really good feel of how much circuit board area will be required to build the circuit and 2) I also arrange the components to avoid crossovers and note any critical lead lengths and/or unwanted circuit interactions and finally the need for any circuit shielding. A few times through this process and it becomes second nature. When I have everything where I want it, I take a digital photo and then when I actually build the circuit there is no guess work on which part went where. My next step is to take the board which will be used for the "final build", typically single sided copper vector board, and replicate the layout of the mock-up. My usual process is to use a tool I built consisting of a 1/8 inch drill bit which has a 1/8 inch diameter knob installed on the shank end. The process is to take all holes where leads will pass through the single sided copper vector board for connection on the insulated side and simply give each hole a couple of twists with the drill bit. This removes the copper from around the holes to prevent a short to ground, which is the top side of the board. For those components going to ground the holes are not touched and the components are simply soldered to the ground plane. Many times I simply set the boards next to each other and take a part from the mock-up board and install it on the final board. This is one way to insure having the correct part installed at the correct location. Having the digital photo also is a great help. This process also minimizes downstream problems where circuits are just sort of tacked together (ala Manhattan style) without any consideration of circuit interaction or because of cramped quarters, the possibility of shorts. This mockup process also facilitates troubleshooting any future issues since from the outset you account for any access concerns to board mounted controls such as trim pots and trimmer caps. The Build Part [I should interject that I have some homebrew test tools that have proven to be invaluable. One is a general purpose test oscillator and the other is an RF probe. There is a link ahead of this tutorial that details the schematics for these two devices. Trust me ~ do not start building anything unless you have these two devices and/or comparable commercial equivalents. The Probe is built first so you can use the probe to verify the test oscillator is working. It is all about logical processes. A realted issue is using the proper tools to conduct the Build Part. With the modern components, considering the small size and sensitivity to temperature, this is not place for a blow torch or 100 Watt soldering iron. Here is a recommended list of some of the more critical "tool" items. Headband Magnifier (This is an absolute must!) In addition to the above items also add to the list a pair of high quality needle nose pliers, wire cutters, an Exacto type knife and a set of small screwdrivers including blade, phillips and torx types. Using the proper tools is but another step of the logical process.] In several construction articles I authored, I use a standard approach of building from the back end of a project and use the completed work as a part of the test system. Let us say we are building a SSB receiver. I would start first with the audio amplifier stage and this may be a good time to interject that circuits such as audio amplifiers tend to be generic. If you have a favorite circuit (I do) there is nothing wrong with using that circuit whenever an audio amplifiers stage is called out. The advantage to such an approach is you are working with a known quantity. You can immediately tell once you power it up if it is operating normally. With the audio amp built, debugged and working properly, I would next build the BFO (Beat Frequency Oscillator) and the Product Detector. Once I build a part of a circuit, I run through my normal "check out before power on process." * Assure that the wrong part is not in the right place * Assure you do not have wrong part value (eg 47 Ohms versus 470 Ohms) * Assure you have placed IC's in the proper orientation in the sockets and the same for Transistors and FET's. [In a recent project I had the wiring for the collector circuit mis-wired to the emitter --luckily I caught it before I smoked the part.] * Look for Solder Bridges especially with SMD IC's. * Using your ohm-meter check for "Short Circuits" especially on the power rails * Look for "Poor" or "Cold" solder joints. Also assure all joints are soldered. * Check for "Wrong" connections especially in feeding the power rails * Look for areas where there may be Feedback and or potential for Oscillations * Observe for any damaged or broken parts. WAIT don't power it on yet! Take a 15 Minute Coffee Break and take one last look before you hit the "power on switch". I would rather take time with this "Check Out Before Power ON" process versus trying to find out why I just smoked a bunch of parts. After building the audio amplifier using the above process I would move on to the the BFO. Once completed run the above checks before power on. Now is the time to dig out the RF probe have that handy as you power on the BFO. Using the probe with a DVM check for output from the BFO. If you have a general coverage receiver, tune it to the BFO frequency and listen for an output. If you have access to an Oscilloscope then put your probe on the output port and look for a signal. Still nothing? Then go back over the above list and look for the problem. [The reason to use the general coverage receiver mentioned earlier is that the RF Probe or Oscilloscope will detect that the circuit is working what it won't tell you is what frequency it is on.] Assuming the BFO is working we will now move on to the Detector stage. Again run thorough the check list before powering on the detector. But it sort of depends what you are using for a detector. Some of the more popular detectors include devices such as the SA602/612. Actually this is quite a sophisticated device as it can perform two functions, one being the BFO and simultaneously using other ports on the eight pin device the Product Detector Function. This IC needs only a handful of components to serve both functions and this the reason for its selection. With this IC if the BFO is working --it is pretty much a slam dunk the Product Detector function will be working as well. Another popular detector is the packaged Double Balanced Mixers (DBM). Devices such as the TUF-1 have only four pins so not much to go wrong here, The SBL-1 is another DBM but has 8 pins and again not much to go wrong here as several of the pins are either connected together or are grounded. A third DBM is the ADE1 which is a 6 pin SMD device. The important caution with the DBM is proper wiring and the magnitude of the BFO signal. The TUF-1 and SBL-1 are + 7dBM devices which means the BFO must deliver 1.414 Volts Peak to Peak to drive this detector. The ADE1 is a +7 dBM device and only needs 1 Volt Peak to Peak. This is where using the Oscilloscope comes in handy as these values can be measured directly from the screen. But not to worry, as the RF probe can be calibrated and when used in conjunction with a good DVM can read these values. So OK the BFO and Product Detector have been checked out and now it is time to see if these three pieces are working as a unit. This is where the test oscillator again comes in handy. Using a crystal frequency in the test oscillator quite near the BFO frequency bring the test oscillator near the input port of the Product Detector. You should hear an audio tone coming from the audio amplifier. You may need to peak either the BFO or test oscillator for a loud tone. If you are successful then stop and take a break before proceeding to the next stage. If you are not successful then start first with the audio amplifier to see it is properly connected to the detector stage and that it is working. Now with the RF Probe once again determine if there is output from the Beat Frequency Oscillator. If there is then next move over to the Product Detector and see if there is any signal on the output port. None? Go thorough the check list process. Once all is resolved with these three pieces move on to the IF amp stage. Here again hooking up the IF stage to the three existing stages and using the test oscillator to inject a signal into the IF stage assures that stage is working. I think you get the idea. Build! Test! Build! Test! This logical process approach helps isolate exactly where there is a problem. If you were to simply build the whole receiver and then power it up and nothing happens. Where do you start to find the problem? With the approach I use you do not move on to the next stage until you have the problem resolved and working properly. This really narrows the field and enables one to make things work. The check list is sort of a roadmap too, that enables you to logically look at all possibilities that prevent the circuit from working properly. No magic here just good old fashioned disciplined problem resolution. 73's & Good Luck de N6QW |
|---|