ONE PAS, INC.: Super Boards

In theory there is no practice, in practice there is.

We all know that Electronics is a Science, but bringing that science to life is an art. Here we will examine an OP780B ONE PASSuper^TM Circuit PC Board and both the Art and Science of it's use. We hope to share some valuable tips, tricks and techniques that apply to all Super Boards specifically and all prototyping in general.

The first thing to keep in mind is that we are using new technology. This requires us to look at the circuitry in different ways than we are used to. Parts of this technology are quite different than what is being taught in school and used in industry at this time. The most important parts to remember are that the circuitry is both 3 dimensional and "Programmable" by the user. We are all used to "programming boards" by adding to the existing circuits. These boards are also "programmable" by cutting copper traces to remove and or separate circuits. For some applications we use the circuitry just as it is. For others we cut some traces at key points to modify the circuit for our needs. We strongly encourage close examination and use of both sides of the board for both component placement and wiring.

Let's start with Ports and programming. Port 3 is the hardest to find because the circuitry is so dense that silk screening the outline is not possible. Close examination of the holes just above the OP780B silk screen reveals three rows on 0.1" centers that have holes slightly larger than others. The bottom row holes are all connected together horizontally and marked -. The next row up are also connected horizontally but marked +. The third row up is offset 0.05" and not used for Port 3, they are connected vertically to the DIP pin lines in the IC bank above. The fourth row up has holes also connected vertically but to the between pin lines in the IC bank above. These are the signal pins for the servo connectors of Port 3. Pictured are 6 Servo Type connectors implemented by sections of header strip. Each connector is configured as top pin = signal, center pin = +V Regulated, bottom pin = GND. Although there are 14 pins available for servo connectors, we know it is highly unlikely that all would be used. The idea here being that they can be moved along the rows to any place most convenient for the application at hand. As power can enter and leave the board through this port and bypass caps are only effective for about 1 inch of circuit board, bypass cap support C3 and C4 are provided on both ends. In both cases the bottom hole is - and the top 3 holes are +. This allows the use of caps with 0.1", 0.2" or 0.3" centers.

Note that directly below the headers are 4 holes. The left 2 are labeled ON, the right 2 are labeled Crg. To the left of these holes is an area for a wall wart jack J1. Below J1 is a Battery connector. If the board is to be powered "always on" a jumper wire must be between the ON holes. If a remote on/off switch is desired a 2 pin header may be placed in these holes for connecting the switch. J1 is set for a power jack that has an internal switch. When the wall wart is not plugged in, J1 switch connects the battery to the power supply. The Crg. holes are in parallel with the J1 switch. If J1 is not used a jumper must be placed between the Crg. holes to connect the Battery connector to the power supply. If J1 is used the Crg. pins are shorted when the wall wart is not plugged in and open when it is plugged in. The Crg. pins may be used for a charging circuit to recharge the Battery when the wall wart is plugged in. SPECIAL NOTE: as different types of batteries require different types of charging and the power supply may be configured for DC + center, - center or AC use, Extreme Caution must be used for any charging function. This is NOT recommended for novice users.

In this picture the power supply, DB9 connector, Battery connector, on/off connector, 6 servo connectors and a 16 pin dual row header are all installed. The dual row header shows that they can be used even though there is no designated mounting location (much more about this later). We can now see what we have been talking about, actually looks like. The next question that usually arises is something like this. OK I see the power supply and the input and output filter caps, but my Regulator spec sheet says for best performance and noise immunity I should bypass the input and output as close to the pins as possible. So why can't I see any bypass cap locations and how do I bypass my regulator. The answer to this is in the way the regulator's heat sink area and ground plane are designed. It is also the first indication that the bottom of the board is just as important as the top.

In this picture we see the bottom side of the regulator heat sink. First we now see that the heat sink areas on both sides of the board are identical. There is the normal hole for mounting the regulator, we can see the head of the screw. Besides this plated through hole, there are 5 more plated through small holes included in a # 5 die configuration. All six of these holes conduct both ground and heat through the board. While most designs use only the mounting hole for this, we feel this design is more efficient. The mounting hole is placed for the cooling tab, but the 5 small holes are closer to where the heat is generated and help dissipate the heat more quickly. We may also notice that the ground plane is closer to the mounting holes than usual. This allows the use of surface mount regulators in DPACK or DPACK-2 styles as well as the standard through hole type. We can also now see how to bypass both input and output with surface mount caps right at the pin. The spec sheet says "as close to the pin as possible" and it doesn't get any closer than this.

Next when it comes to prototyping there are two tricks we have found extremely valuable. First is the use of a Dremel tool. Attachments include engraving bits for cutting copper traces on a board quickly and accurately. Cut Off Wheels for cutting entire rows of traces and even cutting entire boards down to smaller sizes. Sanding drums for smoothing sharp edges, rounding corners, etc. Grinding stones for notches and other irregular shapes. Micro carbide drill bits for adding additional holes or enlarging existing ones.

Second, using many different colors of hook up wire makes things much easier to follow, but stocking many spools of wire is too expensive. Good quality multiple wire cable is a great source. It can be bought by the foot or salvaged from obsolete parallel printer or serial port cables. Look for Belden if you can. Cut the outer sheath lightly down it's length. Then remove it, the bare wire drain and the foil shielding. What we are left with is multiple hook up wires, each with it's own unique color code. The Belden wire we see here is standard red, black, white, green, orange, blue. It repeats all but black with a black stripe, all but red with a red stripe, etc. When they run out of single stripes they go to double, black & white, black & red, etc. In this picture we see Port 1 using a 0.2" screw terminal block and Port 2 with a 0.1" Molex friction lock header. Remembering that our port bypass caps can be 0.1", 0.2" or 0.3", on the far right we see a 0.2" disk for C5. On the far left, a surface mount chip for C6. Port 1 is being driven by a ULN2803 Darlington array and surface mount LEDs display that all outputs are active. Port 2 is wired as an 8 bit I/O with no indicators. In the area between the wires and the LEDs we can see that the between pin lines have a copper strip on the top surface connecting the top 2 holes in each line. These strips were cut using... you guessed it... a Dremel tool and engraving bit under each of the LED places before they were installed.

Now we look in closer detail at adding LEDs to both ports 1 and 2 plus we throw in a few wrinkles. We want to drive Port 1 with a ULN2803 as above. It is an active low device. That is to say it is an open collector that takes the driven line to ground when it is on. When off the line is brought to + Drive Voltage by the device being driven. We want the LED to be on when the transistor is on and off when it's off to show us that the circuit transistor is actually doing what we need it to do. Again Port 2 is a data I/O. In this case we want to monitor the actual data so we need the LEDs on when high and off when low. Our first step is to cut each copper trace on a between pin line that will have an LED over it. Next we install "S" shaped jumpers that connect each port pin 0 thru 7 to the DIP pin line just below and to the right of it and to the between pin line just below and to the right of that.

These jumpers connect three holes in a row on a downward left to right pattern. The wire comes up through the port pin hole, then down through the IC pin hole and back up through the between pin hole. Cut the left over wire on the bottom just long enough to solder. Cut the left over wire on the top just a little higher to act as a post that we will position the LED against to hold it in place while we solder. Place the LED against the wire observing polarity (cathode for Port 1, anode for Port 2). Then using a dental pick (another great trick) put the point down through the hole at the bottom of the LED. This wedges the LED between the wire on top and pick on bottom, holding it in place. Now place a heated soldering iron tip against the wire and hole ring to heat the connection and add just enough solder to wet the wire, hole and top side of the LED. Let it cool, then remove the pick, then solder the bottom side of the LED. When all 8 wires and LEDs are in place, it will look like the picture.

Next we install the current limit resistors for the LEDs on the bottom side of the board using the fold back function of the between pin lines. In this case we are using super bright LEDs, only because we have a ton in stock from a great "surplus deal". Port 1 is driving +12V loads and Port 2 is at +5V logic level. We are over limiting both sides to save both power and our eyes. As the board is now upside down, Port 1 is now on the left and Port 2 on the right. Remember we must think 3 dimensionally. We will use 1.5K resistors for Port 1 and 475 ohm for Port 2. Because Port 1 is active low, those resistors go to the high side of the power bus. Port 2 is active high, so those go to the negative side.

We first turn our resistors upside down and lightly tin one side. We then turn them right side up and place the tinned side over the fold back surface mount pad. We can move it around and get the best angle with our dental pick, then use the pick to hold it in place. Next touch our heated soldering tip to the tinned side to reflow the solder between the resistor and the pad. Let it cool, remove the pick and solder the other side to the power bus as shown. Buy now it should be clear that each between pin line can be used with either bus, or both. Similar techniques may also be used to create voltage dividers, RC, LC or RLC networks using the between pin lines.

Now we will look in closer detail at using dual row headers. As stated earlier we place them between the IC Banks. We can see in this picture that the banks are offset from each other. The DIP pin lines in one bank line up with the between pin lines in the next and vice versa. In this case the bottom header pins line up with the DIP pin lines in the Bank below it and the top pins line up with the between pin lines in the bank above. If we don't need the between pin lines in the upper bank for any other function, we can leave it as is and call it good. In that case we can also use the between pin lines to mount pull up or pull down resistors, LEDs or bypass caps to any of the pins as needed. This will also allow the signals to be available above any DIP IC that we may mount in the bank above the header as well. On the bottom side, we may want some signals from an IC in the lower bank on the header pin. In that case again we leave it alone. But more than likely we may have to separate some or all of the header pins from the IC pins. In this case we see the white lines extending down from each header pin to the hole below and more lines from those holes to the holes below them. These lines tell us there are copper traces connecting these holes together on the bottom of the board. We turn the board over and cut the traces between the first and second holes down from the header pins. This leaves us one hole for each header pin connection and one hole for each DIP IC pin in the bank below the header. And the header pins and IC pins are now isolated from each other.

In this picture we have shifted the header 0.05" left and 0.05" down. In this case the bottom header pins line up with the between pin lines in the Bank below it and the top pins line up with the DIP pin lines in the bank above. Again the same rules apply as in the case above, only reversed from top to bottom. We will now look at the between pin lines and see the copper traces that run through the IC bank power buses. These are the lines that must be cut in order to use surface mount ICs on these boards. If these lines are not cut every other set of surface mount pads are shorted through the IC from one side to the other. The most convenient place to cut these lines is right in the middle of the power bus between the square pads of the + and - bus connection holes. Placing a metal straight edge over one side of the bus to use as a guide for a Dremel cut off wheel makes it very simple to zip through the whole row. This would be for those who use only surface mount ICs and don't need or want the between pin lines for other circuit needs. If we only want to cut some of the lines, the engraving bit is the better choice for exact individual line use.

Now this brings us all of the way back to Port 3. We can now examine how porting across an IC bank affects the bank and IC usage. In this example we have our original 6 servo ports on the left hand side, on the between pin lines. And a 16 pin dual row with the bottom pins also on the between pin lines. If by some twist of fate we need the servo signals to also be on the dual row header, we're good to go. If not we could cut the lines in the middle as described above. This leaves the bottom half of the line for the servo signal and the top half of the line for the bottom dual row header pins. This gives the header pins a way to use pull up resisters or bypass caps if needed. Even with the left hand between pin lines used, we may still use up to a 28 pin DIP IC in this bank. We could also use a 16 pin DIP on the left side and up to a 20 pin SOIC on the right side at the same time. In this view we can better see why we must cut the between pin lines to use surface mount. We must also remember to cut these lines if we use the OP40B adapter boards for SSOP or 0.025 pitch surface mount.

This now brings us to use of the OP40B boards. Jim Donelson of Florida was kind enough to send us the pictures below. He decided to cut down one of the adapters for use with a 16 pin SSOP IC. Need I say he used a Dremel cut off wheel? He shows how to modify a header for use from the top of the adapter down through the OP780B. It looks like he first cut down the board, then solder the chip in place and then installed the headers. The most interesting part is that his finished job leaves us with a 16 pin device that takes up just a little more room than an 8 pin DIP. Great job Jim and thank you again for the input!

ONE PAS, INC. P. O. Box 108 Woodstock, IL 60098-0108