Designing and Building a DIY Ball Mill

I've been working on this project for a few months now and I am still nowhere near done, since I still have some parts on the way but I thought I would share what I have come up with my research and how it is going, and hopefully I can post an update afterwards.

 Well, what is a ball mill?

A ball mill is a type of machine used in grinding media into very fine powders for applications like dressing processes, paints, pyrotechnics, and ceramics and it's typically cylindrical and rotates around a horizontal axis, partially filled with grinding media (often balls made of steel, ceramic, or rubber) and looks something like this on the industrial scale:

[In this scenario, a pulley system is being used from the motor to the mill to drive it to spin. ]

 How ball mills actually work, is the actual steel/rubber/ceramic balls inside the drum spin with the drum, causing them to go along the wall of the drum, gaining some kinetic energy, which they then transfer to the particles when they drop back down to the bottom of the drum, in a constant cycle, so they are constantly smacking against each other, and crushing the material in between the impact through the constant collisions and attrition between the balls.

 

 

 

This way of working leads us to questions like, what is the optimal rpm of a ball mill, or the optimal ratio between the drums diameter and length. 

To start off, the optimal rpm of the ball mill has to do with the gravitational force on the balls themselves and the centrifugal force from the actual spinning drum. For example, at higher speeds, the balls are sort of stuck to the wall, so aren't very good at moving around and colliding with each other. At slower speeds however, the balls tend to not move much from their position at the bottom of the drum, so there needs to be a some sort of point in between these two extremes where the balls momentarily stick to the wall of the drum and as they climb the wall, gravity prevails, allowing them to cascade down and impacting with each other. Of course, this is dependent on the diameter of the drum itself, so the formula to calculate this critical speed is :

$$N_c = \frac{76.6}{\sqrt{$$Where Nc is the critical speed measured in rpm and D is the diameter in meters of the drum. 

However, optimal speed and optimal rpm are not the same, optimal rpm is about 65 - 75% of the optimal speed.

The other factor to be considered, The drum diameter ratio to the drum length, has two different sides. For somebody who is looking for a higher impact, which is good for coarse grinding, the recommended ratio of the drum diameter to the drum length is 1 : 1.15 respectively. However, if you are looking for more attrition which means a finer grind, a ratio of 1 : 2 ranging to 1 : 3 is perfect.

 There are many types of ball mills that differ from what I have been talking about, for example planetary ball mills which are small jars that are very high energy and rotate on their own axis, ideal for really fine powders and look something like this:

 Other types include vibratory ball mills and stirred ball mills, which I haven't really looked into.

 Another thing to note, is that ball mills can have different types of drive systems, which is the mechanism by which a motor transfers power from itself to the moving parts of a machine. The types include Direct drive, Belt drive, Gear drive and chain drive.

For direct drive, the motor is hooked up to the drum using a shaft coupling usually and it is a much simpler set up in general, needing less parts and factors to consider, not having to consider pulley ratios or gear ratios for example, but it requires a low speed high torque motor to run it, which may be more expensive in some cases.

For belt drive, the motor drives a pulley, which then drives a belt connected to the shaft of the drum (it could be possible to connect it to the actual body of the drum however I am not completely sure) and is a great system for people who would like to play around with the pulley ratios, but it is also generally cheap and is relatively easy to build. A problem with this however is  the belts can wear or stretch.

For gear drive, there's a gearbox that connects the motor and the drum and turns them, and is more efficient when transferring torque than the other options and allows for precise control while being long-lasting. However, the set up is of course more complex and more expensive as well.

 Then, you have chain drive which is like belt drive and gear drive had a child, which uses sprockets and a chain to turn the drum. It could be used since it is stronger than belts but it is more complex to pull off.

 Without further ado, let's go oer my own experience.

I started off with some design ideas, as I had decided to utilize a direct drive system, I made a sketch or two

 

 

 

I then purchased a motor from Aliexpress rated for 12V and 108RPM, and grabbed some wood from the DT department at my school for the frame. 

Then I ordered some castor wheels my drum could spin on, and a shaft coupling to connect the motor's shaft to a bolt I had screwed into the drum.

 As my drum, I got a cylindrical container I had lying around. for the actual motor stand,it was just a piece of wood that I screwed into the frame.

 

 I had my friend press-fit a ball bearing for me into the whole where the motor shaft would go through. I also ordered some steel balls to grind the media. A problem I encountered was that the shaft coupling I had ordered was too small for the shafts I was using so I had to drill it out to fit them. In my case, my drum's diameter is approximately 10cm so in my case, using the equation, my optimal RPM (according to the formula) is 157-183 RPM, but this is purely theoretical, since this doesn't tend to work in reality and so the actual optimal rpm would be around 60-90.

This is a video of it all set up and working (but definitely not optimal speeds which is why I am getting a motor speed controller soon and ill 3d print a case for it):
 

 

In the end, I made countless mistakes, like making the motor stand at a very bad height, and having zip tied the motor down, letting the drum move around like that probably due to the bad placement of the wheels but also the drum not being an even cylinder throughout its length, which let it wobble around, and also not having the right rpm setting. However I have 3d printed a new motor stand (just a block) and I am going to try figure out a way to have the motor be stable. Hopefully I will revisit the project, since I have learnt a lot from it, in terms of designing, and also the mechanics of how these great machines work.

Thank you for reading.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

How I set up a Secondary Router with Pi-Hole for a Clearer Wifi Connection

 

Where did I come up with this?

 

About a week ago I went to a car boot sale and found a TP-link router for 7 pounds and had an idea for a project! I wanted to see what I could do with this new router, as I have heard that privacy routers were a thing and that would be very useful, considering the amount of ads websites display in this day and age. Also, the router would be set-up in my room, meaning that the WIFI connection was also stronger and I have no ads to skip.

 

 

The research process

I went ahead and started researching the idea and found out about pi-hole (https://pi-hole.net/), a network-wide ad blocker that acts as a DNS sinkhole, meaning it blocks ads, trackers, and malicious domains. Instead of using browser-based ad protection,pi-hole works at the DNS level, preventing unwanted content across all connected devices. 

 

The setup

 

What I used was my main computer (running Ubuntu) to actually run pi-hole and route the traffic from my router, through the pi-hole DNS server, and of course the router. I set up the router to create a whole new network, supplying bandwidth from my current router that provides WIFI to the whole house, but has a different subnet, so the devices on the first network don't interact with the ones on the new one. Here is a simple network diagram for this:

   

 How I did this was connecting a LAN output cable from my main router into the WAN port of my new router, then connecting to my new router's network, configuring the DHCP settings and running a few commands in cmd in order to ensure the IP addresses assigned on both networks are different, preventing any interference with data transmission. I then set the the DNS of my router as the pi-hole DNS, allowing all the data on the network to flow through there, meaning no ads.

Testing if it works and it's effectiveness

 

This is the actual pi-hole dashboard where you can see the data and requests being filtered through it, acting like a sort of filter, taking out the particles you don't need and leaving the fluid to flow through, and this is it working and processing queries! 

                          

 I went ahead and tried out an ad blocker test at https://adblock-tester.com/ and managed to score 100/100 points which is amazing.

                            

I also tried out a network speed test to see if I was getting a good amount of Mbps, and it was great.

    

Final thoughts on the project

 

Here is what the actual router looks like with some stickers I put on it:

 

A problem with this setup is that pi-hole only allows data through the network, as long as my main computer is on, which is not all the time, so I needed to add a secondary DNS server, so I added Google. A way I could solve this problem is by investing in a new Raspberry pi that can run this set up 24/7, so I am on the lookout for one right now to fix that (I could always use the one I have but it is a bit overkill). I also want to set up a VPN to make it even more secure, but for what I have right now I am satisfied, that will probably be my next addition after getting a raspberry pi to host this.

Overall, I think this was a fun project that really allowed me to expand my knowledge about how routers behave, and how networks can be manipulated as a whole, a lot easier than I thought. I believe this was an invaluable learning experience considering the effort it took. I recommend you try this for yourself as it is really fun but also rewarding in the end.

 

If you have any questions, feel free to ask away.