Prototyping : 2D wiring

This page is about one of my favorite prototyping techniques. Until now, I have not come across it on other webpages so probably all other engineers feel ashamed about it. I don't. This method is cheap, simple and faster than having made a board by a company. You only need a board, the parts and a soldering tool. Plus some imagination, but all engineers are endowed with that in abundance.

If you show a 2D wired board to an accountant, (s)he will ask from what garbage can you collected it. If you show the same board to an engineer he will like it: it's simple and it gets the job done. I guess all of my audience is more or less in the engineering class so look around and see which parts you can use.

I could just show some pictures and keep it like that, but that's not my style. So I will explain about it using a simple project: a variable current LED tester. The steps taken were:

I already had a LED tester. It was a simple 9V block with a 330 Ohm series resistor. Simple and efficient. Yet unpredictable since the current drops when the battery gets older. So now I need a LED tester with a presettable current that can run on 9V AC and 6-12 V DC.

For hi-res pictures consult my album at http://www.mijnalbum.nl/Album=N6A3EV7X .

2D Wiring : the circuit

This is the circuit. It is an easy circuit and that's good since we want to learn from it. A short circuit description follows:

Operation of the opamp is easy: An opamp changes its output such, that the voltage of the (-) input equals the voltage on the (+) input. For the rest we see the opamp as a black box (which isn't too difficult if you ever saw one).

Suppose I set the voltage of the LM 358's (+) to 1.1 Volts. Then the LM358 will start changing it's output (pin 1) such that the voltage on the (-) input will be 1.1 Volts as well. Now we all know Ohm's law. For those of us that forgot about it: F = m . a

Oh wait: that was from an english bloke. Ohm's law was U = I . R which is identical to I = U / R. We know R (47), U (1100 mV) so I is easily deduced: I = 1100 / 47 = 23 mA. To get this done, the LM358 will need to set a voltage of between 1.6 and 1.9 Volts on the base of the BD135. Yes, you read it right: BD135. Eagle only has the BD235 but the 135 is good enough for this as well. And I only have 135's in my toolbox.


2D Wiring : the case

I happen to have some nice cases which I bought from Reichelt last year: the Eurobox type. Cost per box: less than €3. Still, with an external size of 135 x 95 x 45 mm, you can pack quite a lot of thingies in- and outside. The case has 7 mounting posts and accepts a PCB sized (roughly) 80 x 75 mm and it comes in five vivid colours (blue, green, red, grey and yellow) plus black.

The case consists of two identical symmetrical shells so you can mount two PCB's, one in each shell. That's close to a full euroboard (costing €4 each) in this €3 box with lots of room for actuators and indicators.

As you can see, the case has plenty of ventilation holes as well. So it is perfect for housing an open frame switch mode power supply, to mention only one application.


2D Wiring : The circuit

Above we already saw the circuit of the LED tester. The first step after drawing the circuit is making a PCB out of it. Not for really making that PCB but for getting an impression about component placement. Normally you would route the PCB traces for optimum PCB size. here, you will route for least amount of PCB curves. Just see what groups of components end up where.

To the right is the result of it:


2D Wiring : Requirements

To the right you see the table where I do my wirings. The tools required for 2D Wiring are:

First make sure the PCB fits the case. Then drill the holes for the mounting screws. Please drill all holes, even if you will use only two screws. The hole in the board will signal to you "no components here please!" which is necessary since the post under the hole will lift the PCB if there are components there. Or you must cut off the unrequired posts with a sharp cutter.


2D Wiring : Components placed

To the right you see how I placed the components. Placing was not optimal, but that's the kind of thing that happens when you are busy running wires on the underside: you tend to get a bit sloppy on the topside. The IC socket is much too close to the three pinned header just to the left of it. Due to this, the resistors that needed to end up near the IC socket had to be placed in strange positions.

Nah. Better luck next time. This is a messy technique by nature so some messyness is part of the deal.


2D Wiring : Solderside

And this is how the solderside looks like: plenty of solder. This is a picture and it says more than a thousand words. So better not spend too many extra words on it. What it comes down to is:

It takes some experience to get things done. But after that, when the thing got done, the result is a very cheap circuit that does work. And in a sense it is better than any wire based circuit since each wire behaves like an antenna for sending and receiving noise.
For small circits like this, no decoupling capacitors are required. The power/ground 'traces' are so very fat that a capacitor would not do much extra. Still, it won't hardm if you do.


2D Wiring : Just in case

Here is a first indication of how things look in the case. I chose the yellow case. You see clearly that this is a shell type case and how they snap together. It is a strong embrace and it needs some pratice to split the two halves. Which is a Good Things since this means that the parts do not come apart when the box hits the floor.

Ample ventilation. Good support for the PCB. Two plastic front/rear panels that are retained in slots, yet are easy to remove (to be drilled or otherwise machined). What you see here is the rear panel. It has two open frame style barrel connectors for accepting AC (left) and DC (right).

These barrel connectors are very fine for prototyping:

  1. Drill a hole of 5 mm
  2. Widen the hole with a conical reamer
  3. Insert a female barrel connector
  4. Mate it with the male connector
  5. Mark the hole positions with a CD marker
  6. Drill the holes with a PCB drill
  7. Mount the barrel connector


This is the AC connector. It has a 2.1 mm barrel so that the AC will never fit on the DC connector (which has a 2.5 mm barrel). AC has no polarity so both wires have the same colour (for the colourblind among us: orange).

See the open frame barrelconnector?


And this is the DC connector with a 2.5 mm barrel. This will mate with the 2.1 mm barrel, but that's no problem since the rectifier will take care of a possibly reversed polarity.

The multitude of spare holes you see here is a leftover from an attempt to mount another connector, but it failed. So I reverted to the good old open frame style barrel connector. Always works.


2D Wiring : error checks

I connected AC power. The LED lit. That's a good sign. I checked the voltage across the diodes: OK. I wired the potmeter in and everything went dead. I made an error wiring the connector... Swapped two leads and the the reference voltage was OK.

Now see if the (still empty) IC socket receives the right signals. Yes. The variable reference is on (3). The feedback voltage is on (2). Time to insert the opamp and wire up the test socket. I check the solderings of the LED-Under-Test connector. Only one solder joint. Two are required, so.... And the collector of the BD135 transistor is still floating as well. Hmm. Time to power up the Weller again.

2D Wiring : final version

When I was almost done wiring, I noticed the vast emptyness of the circuit board. And only half of the LM358 was used. That's a waste. So I changed the circuit for the better: the unused opamp was used for converting the LED current into a voltage for easy measurement. We now have 100 mV of voltage for each mA of current through the LED.

What the circuit does:

The second opamp's amplification factor is determined as follows:
  1. A test LED is connected to the circuit with one leg loose
  2. The DMM is set to current mode and the current through the test LED is determined
  3. The test LED is now reconnected in the normal way
  4. The DMM is set to voltage mode and the output voltage is measured
  5. Turn the potmeter such that the measured voltage has the right magnitude
  6. Calibrated
Suppose the LED current is 14 mA. Then the output voltage should be (in my case) 1400 mV or 1.4 V.


2D Wiring : Finished

And this is how the final PCB looks like. If you would not know of this webpage, you would think: "What a nice circuit board!". But you know about this webpage, so probably you still think the same. It works. It does as expected. But if you see the messy tracks down below...

Still, this is just a prototype. It's purpose is to get fried for the sake of progress. It won't fry, therefore it is too well made of course.

Taken all this in mind, 2D wiring ain't all that bad. Below is a picture how the LED tester looks like with the lid closed (and a blue LED mounted) and a DMM attached. The exterior reveals nothing of the 2D wiring spaghetti down under.

The DMM shows 1.274 Volts. The LED current was something like 12.5 mA. Good enough.

The testsocket is a part of a wirewrap SIL header. It has one anode and three kathodes so that any LED module upto 300 mil pinspacing can be tested.

Total costs of this project: close to €10.


Page created on 12 Nov 2008 and

Page equipped with FroogleBuster technology