There is no doubt about the practicality of reliable solderless breadboards, but you must admit that they are not perfect. Of course, they are not ideal for certain types of circuits, but they are less problematic than those with jumpers. Careless people will end up hopelessly entangled in a pile of jumpers with their components, and picky people will spend more time making the jumpers tidy instead of actually making a prototype of the circuit itself. What should I do?
One way to solve this problem is to make a solderless breadboard without jumpers. This is the idea behind [Kevin Santo Cappuccio]'s ongoing "bakery products". The idea is to adjust the standard breadboard so that the connection between any pair of common contact strips and the power rail can be established in the software. The trick behind this is to simulate the CMOS switch chip matrix, especially the MT8816AP. 128 crosspoint switches per chip can handle voltages up to ±12 volts, so there are many circuits that can use these programmable silicon jumpers.
[Kevin] The current version is 0.2, its size fits under a solderless breadboard, and the package is compact. He shared details on how to connect the breadboard contacts, which seemed to be a painful process: pull out the contacts, cut a small label at the end of the gutter, and then bend it down to form the lead printing of the through hole Circuit board. It looks like a lot of work, there must be a better way; [Kevin] is obviously open to suggestions.
Although we have seen cross-point switching used to enhance solderless breadboards before, we found the simplicity of this project to be pleasing. The idea of throwing away all these jump shots is certainly tempting.
This is really a cool project. A clever addition is that there are small screens on each side that show the current connection.
Unless you can't implement a 7-segment LED display or vibration motor or accelerometer or DIP switch or any other "fun and interesting" components.
15mA (ImaxIO) is enough to handle any small signal, when you need more current, you can use some bipolar or FET, after all, it is still a breadboard.
When I was prototyping, I made a single-sided surface mount PCB that contained all the digital chips I needed and spattered CPLD chips that carried "all" signals between the chips. Then just program the connection in VHDL. So if I make a mistake or need to modify the connection, it is just a code change.
For one of those analog synthesizers where you plug in all the cables, this idea seems pretty cool.
Well, once I encounter a breadboard that I trust, I will stop doubting. For breadboards, it *always* requires guessing whether it is a fault in your design or a fault in the breadboard.
To make matters worse, you have to design specifically for breadboard capabilities, and things that actually need to be prototyped, such as switching regs, do not work well.
Analog and high-speed do not like parasitics, digital and low-speed do not require a hand-made prototype stage, if you are careful.
I don't know anything about all these things, but I have always hoped for such a conversion, and every time I look at it, I find that it is not true. Even for simple circuits, you must carefully match the switch to your actual requirements. Then you still have to accept a certain degree of undesirable behavior. They are always true unity gain op amps or some kind of transistors with voltage bias or resistance.
I don't know much about the underlying technology. The MT8816 data sheet I found does not have a schematic diagram of the actual connection. But immediately (assuming I explained it correctly), I saw that the maximum current is 15mA and the typical on-resistance is 45ohm at Vdd=12. It is of course an interesting component that can be used for many things, but the warning is huge. Dominate even. I can't think of a single circuit I have made with a breadboard that can be adapted. Especially because I used the breadboard as a step before soldering the breadboard together... Compared to this switch, the extremely different behavior of the wires made it impossible to start at all.
If I understand how the circuit works well enough to predict the impact of this interconnection on it, then I don't need a breadboard in the first place.
For small signals that maintain low current, this should be more or less like an ideal switch. Of course, the on-resistance will vary depending on your exact position within the voltage range, but you will need to deal with kiloohms anyway, and it should not matter if you add +-75 ohm to this error in your design. Otherwise, you are designing a circuit that cannot be actually constructed due to component tolerances.
Yes, once you have this device, you can go out and invent a project, this project is all +/-75ohm irrelevant small signal. But I have never done such a project. I can't even think of a project of this size. Pure numbers are almost the only thing that meets the criteria, for this you either have so many connections that the breadboard is just a drop in the bucket, or there are so few connections that you would rather just use an ic clip to jumper the cable
In fact, half of the completed code there can place some frequency (or any signal) on the probe so that you can plug it into different rows and tell the board which connections you want to make without a computer.
The next step after I got the basic stuff to work well is to do some of the things you mentioned. There are several ways to solve this problem. I think the most ingenious way is to install a display on the top of the breadboard, but making a display with such a porous display may require Samsung's level of development. Another way is to install a miniature projector and a camera on it, but all the things I have used that try to do such things are too slow and problematic to use, and there is just an absolute trash can fire around.
It's not that any of these things can't be done, I just need to figure out a way to do this without sacrificing experience.
I like the idea of probes. As an indication, each column has an rgb, and each "net" has its own color code. It's certainly not as smart as overhead projectors, but it's easier to drive ws281* strips.
Although in my experience, those off-the-shelf, non-branded transparent plastic plates are rubbish, but it sounds easier to combine the transparent case with the bb830 clip than a display with a hole. Open more LED placement options. But even placing LEDs above and below the circuit board is not a terrible idea (except that the power rail is a bit in the way)
I do like this idea, and must study the chip carefully. I have been looking for something similar-but only in numbers. Each pin should have-programmable resistance plus-programmable resistance to ground-set as input-set as output-set to tri-state all as a shift register block for each pin, so software programmable. But until now no suitable FPGA/CPLD has been found. For a breadboard, the system should cover at least 60 pins 30+30. Since it is all based on shift registers, it can be easily expanded. And all are controlled by software.
By using our website and services, you explicitly agree to the placement of our performance, functionality and advertising cookies. learn more
Post time: Nov-19-2021