Plan changes and design problems
Welcome back! In this post I hope to explain to you the new circuit I have been working on and the progress made in building it, as well as a calendar of my blog posts that I will try to stick to. Let's jump right in!
Circuit Design
Why:
First off let's address why I have the need to build this LVD (Low Voltage Disconnect) circuit. In my previous post, I explained how I planned to use a solar panel to trickle charge my golfcart's batteries. Another thing I found is that I need a way to shut off this charging process when the voltage from the solar panel drops below 36v. Originally, I planned on solving this problem with the ambient light sensor provided in the kit, but there are a few problems with that. In a system where I am using an ambient light sensor, I have to come up with a table that connects light level to voltage, and then have the SensorXplorer shut off the charging circuit when the light level goes below a certain point. This circuit is inefficient, as it uses a second variable (light) to find voltage. In the end my goal is to know the voltage, so I decided to cut out the middleman and set up an analog circuit that looks directly at the voltage. (This also lets me use the Ambient light sensor elsewhere, allowing more creative designing).
The Circuit:
Now for the explanation of this circuit.
This circuit can be divided into two stages, the 20v stage (Yellow) and the 30v stage (Green). The 20v stage is where we check if the voltage solar panel voltage is at the required 20v for the input of the boost converter. When the solar in reaches 20v, the resistors R1 and R2 divide that voltage to 0.6v, which is the threshold voltage for the transistor (Q1). Transistors work as voltage-controlled switches, using a low voltage to control a larger one. As you can see the transistor doesn't control the power directly though, instead it switches a relay. The reason for this is that transistors tend to get hot when switching higher voltages, something a relay doesn't have a problem with as it is mechanical. This setup was what I thought would originally work, and it did, sort of. What I didn't expect is that this boost converter doesn't work well with fluctuating voltage. When I hooked it all up and tried varying the voltage to simulate cloud cover, the circuit switched off like it should, but when the power came back up the boost converter stopped boosting! Apparently, there is something holding a charge (Either inductor or capacitor) in converter that has to be discharged at the output before being able to boost again. Thats where the second stage comes in, with an extra relay and resistor to discharge the output when switched off, and to reconnect when switched on again. Having tested this circuit though many problems occur, such as relays not turning on, and a single relay oscillating when the other is disconnected. More troubleshooting to do, so I hope to get back to you on that.
A Note About the Ambient Light Sensor
Now that the ambient light sensor is free to be used elsewhere, I have thought about what problem I could solve with it. I liked the idea of automatic lights, but as I don't drive at night much, that isn't really a problem for me. (Though imagine an RGB golf cart ) Instead, I got a cool idea from a friend about using the color of the environment to change the song that is playing from the golfcart sound system (see more on that later). We will see how that goes, and if it doesn't work, we'll see about the RGB.
Scheduling
Over this competition I hope to post once every other week, but as my education takes quite a bit of my day, I may not be able to post on the same day every week, but I do hope to keep to that week. Here are a few rough dates of what I hope to post.
- 7th of October: Finishing LVD circuit, hopefully* the testing of Exploration station kit, attaching of solar panel to golfcart
- 21st of October: 1st drafts of code for SensorXploror and ColorSound Idea, starting designs of backup sensing circuit and code.
- 4th of November: Finishing code for SensorXplorer, wring of sensors and outputs in golfcart
- 11th of November: Concluding post, finishing touches.
Note: Unboxing of sensorXploror will be done when it arrives.
*I say hopefully as there are issues with the shipping of my kit, to whoever is working on that: Please contact me so I can give you the correct address! I have tried contacting e14phil with my information but have gotten no response. Thank you!
Thanks for reading and I hope you have a great day!