Return to Independent Study

03.22.17 Independent Study, Journal 8

Gene wanted to meet. I E-mailed him but he got really busy. Owens group is leaving for Black Mountain at 12:00 on Wednesday, so I wrote an update instead. Gene was a bit annoyed by the length of the E-mil,  which served to annoy me since I thought I was doing him a favor. I intended to give enough detail that he could choose what needed attention and just send an E-mail instead of trying to work in another meeting when he was clearly busy. Of course, I also wanted to make sure that I stayed accountable for my progress since it is an independent study. I thought that was what I was supposed to do. Not to mention, I can’t imagine any teacher in any context being aggravated by a student working too hard.  I am so tired of being annoyed and of people being annoyed with me. I know the response, “be less annoying,” but I don’t think I can do it.

 

 

 The E-mail:

Sorry, but Wednesday turned into a crazy day!

 

I wanted to bring my stuff this time, but I hate to do that if I have to leave it at school while I travel. If you are willing, you are welcome to meet me at my house where all my stuff is. I live in Old Town, near Tim’s store. If isn’t out of your way to stop on your way to school, that would be awesome. Alternately, I have made a full inventory list of all my electronics. I am going to tell you what I am currently working on and you can tell me what I should bring with me. That way you can answer in person without having to type it out.

 

You suggested that I should practice working with simple designs that don’t require programing. I was not enthusiastic at first because I like the programing, but the more I learn the more I understand what you were talking about and my interest has grown considerably. I feel that better understanding would be very useful for future projects.

Great to hear this approach is working out.

 

I have some specific questions about the power distribution. I feel like I might be doing something wrong. It works, but I am nervous that I don’t understand the circuits well enough. I read a book. I have looked at diagrams. I just don’t feel confident about it.

It takes a while to get confident with electricity / circuits. I always double / triple check and ask others to review…

 

Right now, the robot needs 2 power sources but there are battery slots for 4. There are two rechargeable 12-volt DC batteries and Two thin, flat 5 Volt cell phone charger batteries. I added the space for the extra 12V batteries in case I decided later that I wanted to run LED strips or some kind of sound system. I wasn’t sure how fast the 12 volt batteries would run down. I added the extra cell phone battery because I got a really good deal on them. The batteries are so small and thin that I ended up having an extra space if I stack them. I figured more power couldn’t hurt.

Having a dual voltage, battery based system seems tedious and heavy.  It might have been better to just go with 12 volts all around and then add resistance in the circuit to knock the 12 down to 5 volts where you need it.  What’s running on 12 volts?  LED’s only need 3 to 5 volts.

 

Since I have the extra batteries in place, I want to go ahead and connect them. When I am sure they all work, I will remove the extra batteries and cap off the wires. I am using DuPont connectors for everything so that it is easy to take things apart. Is there any reason this could be harmful?

I’m not sure on this.  I’d need to look at your schematic / wiring diagram.  Can you forward it?

 

My plan is to have an on/off switch and charging port for each power source on the outside of the robot body. Can I use the slider switch system to do this?

Yes.  This is what I recommended earlier.  Make sure it’s a DC switch.

I am also deeply annoyed with the DC power adapters. I decided to order the ports for them, only to discover that there are 2 common sizes and no way to tell which one I have. The difference is too small to be measurable. I read that there are tools at the now defunct Radio Shack to find out, but the other methods I tried did not give me a good answer. Do you have any idea how I figure this out?

There are so many connector options out there, so don’t ever limit yourself to just two options, particularly when you’re working with low amperage.  I often use mono audio connectors (headphone jacks) for low amp dc circuits.

 

I want to connect the batteries in parallel so that I do not get added voltage. I do not think there is ever any need for more than 12 volts of power. The way I understand it, this should mean that if I remove the extra batteries, it does not damage the circuit. However, I am not sure about charging the batteries when connected in parallel. If they are connected in parallel, can I still set it so that I use a switch to select which battery to charge? If I am charging one battery can I still use the other battery? I probably won’t need to do that often since the robot can’t really go anywhere when it is charging, but I might do it when I am working on the robot.   

 

I would avoid this approach.  This is why I recommended the rechargeable Lithium Ion Battery packs that are set up to be recharged via a USB cable and are light weight / small.

 

I have a mountable USB port to charge the 5Volt batteries. Can I split this and use a switch to connect the port to both batteries and select which one to charge?

 

Again, I’m not sure on this.  I’d need to see your wiring schematic / diagram to confirm.

 

This is also a related question and maybe very basic. When I look at the Voltage/Amp rating of a switch, I know that I can’t use a switch with lower rating than my power source, but will it cause a problem if I use one with a higher rating? For instance, it seems like it shouldn’t matter if I use a 12 Volt switch on a 5 Volt source.

 

Higher is fine.  That being said, high amp switches / connectors are often big and bulky and pricey.

 

Unrelated to the power supplies, I still have not started coding the robot because it does not have wheels. It can’t go anywhere yet. I have wheels. I have a coupler to connect the shaft. I do not have a way to connect the wheels to the shaft. I went to the hardware store and tried every nut and bolt I could but nothing fits. I can cut a hex shape out of wood and attach it that way, if I can get a precise measurement, but I have not done it yet. It would be nice if there were a better way. Do you have any suggestions?

It sounds like a bit of the cart before the horse.  Did you ever get your processing based prototype working with your test code?  This is why I recommend a prototyping process that starts simple and adds complexity as you go.  Ideally you would have designed a simple test robot to try your code on, made from cheap materials.  One without all the bells and whistles that has a simple control mechanism and cheap parts to prototype.  I’m not saying this to discourage you or to get you to start over, but this project is already pretty complex and you still can’t test your code.  Let’s discuss this in person when we meet next.   

 

I also need to integrate a servo motor. I knew I needed to do this but I thought I could wait until the second iteration. I am still not sure. I have been testing the IR distance sensors that I got. They do fine for obstacle avoidance, but if I want it to be able to track an object, I am fairly sure I need to put them on a servo so that the robot can swivel them around. I didn’t want to do this yet because I am only just learning to control the motor driver. I didn’t want to complicate it with a third motor. I also did not really work it into the current design. My plan was to wait until the next iteration when the robot will have a round front with space for a swiveling head. Now I feel like I need to go ahead and do it but my test runs could be wrong. They have not been very effective because I have been using a crappy little platform and the parts don’t stay in place. The results suggest that the robot simply can’t track an object effectively with a stationary sensor. I think it could if I included multiple sensors but that adds a whole different complication. If nothing else, it is expensive. I only purchased 2 of the sensors I am using. I have 5 of the regular IR sensors but I don’t think they will do a good job of following a specific object. What do you think? Should I just go ahead and put in the servo for the head? I have one but it is very small and I think the plastic parts will break if it has to hold up much weight. If I screw in a shaft it might break it. What kind should I get?  I have several DC motors from toys but I don’t think they are a good choice for a swiveling head.

Let’s discuss this when we meet.  Once I see your prototype in person I can give you more feedback. It sounds like you could just mount the IR on the front of the robot and let the steering mechanism for the robot lead where it’s pointing.

 

I read up on the Piezo speakers and I purchased a couple from China. They will not be here for several weeks so I probably won’t get them in this version, but there will be a space for them for later.

 

 I have one you can borrow.  They’re quite cheap and easy to find / salvage from other parts / systems.

 

Tell me what you want me to bring and I will bring it. I would rather not bring the robot body. I don’t want to leave it at the IMRC, but I will if the pictures and diagrams are not sufficient.

 

Let’s meet this week.  How does Wednesday afternoon work for you?