The desire to automate everything in the world haunts me.
In the production of boards using photoresistive technology, it is necessary to illuminate (expose) the pasted photoresist with ultraviolet light. The easiest way to do this is to use a UV lamp. The minus of the lamp is that the light falls on the photoresist at different points at different angles, and the larger the board, the more this angle changes from the center to the edge of the board. As a result, the illumination at the edge is slower and the boundaries of small elements are blurred. In some situations, this is critical.
For example, in practice, I had to make such a payment:
Its length is about 30 cm, and in order to illuminate the photoresist relatively evenly, I had to move the ultraviolet lamp during the illumination process. This noticeably complicated the process and therefore lost confidence that the board would be made with high quality.
After being convinced of the “necessity” of this undertaking, I want to add that in part I just could not sit still =)
Before, on the Internet, I came across a solution in the form of a matrix of ultraviolet LEDs. I went the same way, but by adding automation of the device on the AVR. I wanted to have the following features: install and change exposure timereceive sound signal at the end of the process, be able to suspend exposure (just in case), save settings to EEPROM. And what was not planned, but additionally succeeded, is to do brightness control.
Thus, a full-fledged device should have turned out, which I launched and went to prepare a bath for developing.
The production process of the LED board is standard, but I’ll leave a few photos, after all, I didn’t have to make larger boards.
Initially, the board was conceived with two LED circuits: if necessary, the entire board works, if not, only the inner part. Over time, I realized that it is difficult to save electricity on this, and the convenience falls due to unnecessary options. I combined the circuits, and now the entire board is lit. If I had come to this earlier, the design would have been easier.
When the board was illuminated, the lamp moved to 4 different points for a more uniform illumination (once again confirmation that something needed to be changed).
With help drilling machine 8 holes with a diameter of 3 mm were made in the board. External 4 are needed for attaching the legs, internal – for attaching the control board.
After tinning and soldering the LEDs, the board began to look like this:
By and large, the control board has nothing remarkable: an ATtiny2313 microcontroller for implementing logic, 3 buttons for start-pause-stop + backlight time / brightness settings, a buzzer with a built-in generator, an LED for displaying the current state and a roll (voltage stabilizer) to reduce voltage from 12 to 5V for controller operation.
To display information, I used my own project shift register display. I posted the project so that I could repeat it, but you can use any other developments, as long as the 74HC595D was used as the interaction interface. With another connection of the shift register to the indicators, it remains to correct only the display of characters in code.
To attach the display to the board, I designed and printed on a 3D printer corner. The display on it stood up like a native! It has 2 power lines and 3 wires for data transmission to the 74HC595D shift registers.
When the device was tested, it turned out that it is so much brighter than the lamp that I used before that during the manufacture of the board with its help either overexposure or underexposure occurs, depending on the manufacturer of the photoresist. To solve the problem, a brightness setting option was added to the firmware of the control board, implemented using a PWM controller. The duty cycle of the pulses changes – the brightness changes. The settings for the operating time and PWM duty cycle are stored in the EEPROM memory of the controller, so there is no need to reconfigure these parameters when turned on.
On this, in general, you can finish by adding a video, but I will additionally indicate what was done incorrectly.
The place behind the roll was not left by chance, but, oddly enough, it was not intended for the radiator, but to bend this very roll and the board looked a little more elegant. The only problem is that I completely forgot about the bright display, which consumes a lot. And as a result, the roll does not just heat up – it heats up like an iron. Because of this, I had to install a radiator on the roll, after which it already heated up less. Since the water on the radiator does not boil (yes, I checked it), and the operating temperature of the bank can reach up to 150 degrees according to the datasheet, it was decided to leave everything as it is, although if you touch the radiator during operation, it burns unpleasantly. Correct, of course, would be to use DC-DC.
Photo of a working installation. The display shows 3 minutes 10 seconds. After this time, the installation will stop shining and squeak with a beeper, they say, “I’m done.”
Video brightness settings + device operation. Legs 35 mm high have already appeared here, on which the installation is placed. From the unusual, on the video the display is wrapped with masking tape. This was done only so that the then phone could capture information from a very bright display. Human eyes are made of better quality, so this crutch is not needed in real use.
Thank you for your attention!