As mentioned in a previous post, I'm trying to dive into analog computing and differential equations. I would like to have the Analog Thing kit, but I cannot afford the $500 price tag right now.
My next idea was to build components from the Grappendorf collection.
=> Analog Computer: A modular Analog Computer Kit
That is a cool kit, though it lacks a few nice features like a repetitive run mode. But it is simpler to build, and the files needed for printing it and building it yourself are included in a repository.
The only problem here is that the boards are of such a large design, that they are expensive to print, especially with minimum required quantities — OSHPark requires three minimum per board design. So, I'm just building the components on bread boards, using whatever similar parts I have on hand. Here is my first integrator, following the Grappendorf schematic:
The little chip is an LF356 op amp, per the schematic. I used up some old carbon resistors, as you can see. I had to glue the "run" switch down to the board, since the leads are short and it was wobbling. I'm rather embarrassed about the three huge polyester capacitors, but that was the best option I had on hand, for 1uF non-polarized capacitance. I'm hoping next week to replace that with a film cap or something else small. If I can do that, then I should be able to fit two integrators on one breadboard.
This is a dual supply integrator. Fortunately, I already have handy access to a nice +/- 15V power supply.
Here are the signals for a simple test of the output:
The green line is a square wave input signal, and the yellow line is the integrated output. You can see that the output is inverted.
The basic idea is you hook up your initial conditions voltage to the IC terminal, and your input signal to 1X or 10X. Start recording on your scope, and then flip the switch over to run. The resistance on the input slots are pretty high, with the idea being that your equation or simulation should take at least several seconds to run. This is really the only practical choice in this design, as you have to manually flip the switches on your integrators.
I was a bit confused on how to "balance" the op amp, which is what the little blue-white potentiometer is for, since Grappendorf did not include instructions on that. But what seems to work is to just remove your signal leads, switch over to run mode, and then watch your output signal to see if the line is drifting up or down.
This work © 2025 by Christopher Howard is licensed under Attribution-ShareAlike 4.0 International.
=> CC BY-SA 4.0 Deed This content has been proxied by September (3851b).Proxy Information
text/gemini
