There’s a bandsaw shaped hole in my workshop and it needs filling.
Why Build Your Own?
Why not? Buying a commercial bandsaw is certainly easier and by the time I’ve finished this project it would probably be cheaper as well but this is about having fun. As long as I get a machine out the other end I’ll be happy and I don’t really mind if it cost more than a comparable commercial machine.
Why Design You Own, Don’t You Know X Has Plans for Sale?
This project isn’t about following someone else’s plans, this project is about learning, from scratch, how to do it myself. I’m happy I can do the woodworking needed, now I want to explore engineering. Having said that no one truly starts from nothing. I have no intention of re-inventing the wheel here so obviously I’ve studied existing commercial and homemade bandsaws.
What is in a bandsaw?
- Blade Guides
- Blade tensioning mechanism
- Drive mechanism
Previous Home Shop Builds
Here is a list of just some of the previous builds that I looked at before. There’s no shortage of builds out there so I’ve tried to keep this list to the more notable ones.
- Popular Mechanics (May 1974, Page 155) had project build of a 12″ bandsaw made from ply. It’s a surprisingly solid looking build.
- Fine Woodworking (July 1987, Page 61) has a generally well thought out plan for a bandsaw.
- Matthias Wandel has built a number of bandsaws over the years almost entirely made from wood, plans can be had for a few dollars.
- John Heisz built a bandsaw out of box section steel salvaged from a garden trellis.
- Paoson Woodworking has plans for bandsaw made from ply.
- Matthew Cremona built a bandsaw mill. It’s an absolute monster and not really relevant here but it’s interesting none the less.
Shafts – Size, Material, Etc
Bigger is better, at least that’s what I’ve been told 🙂
There are various materials that can be used for shafting, the cheapest will be steel round bar but without a lathe it might be hard to get it to fit the bearings. Ground bar is the most
Matthias Wandel uses a bike inner tube and this seems to be a common approach. John Heisz uses clear silicone, this doesn’t feel like it would be good enough but it seems to work for him. I’ve seen exercise bands used as well. I wonder if some sort of rubber paint would work?
Tensioning and Tracking
Tensioning is the act of tightening the blade, tracking is tilting one wheel to make the blade run true and not come off the wheels. The forces involved when tensioning a blade should not be underestimated.
Most bandsaws have a single mechanism that deals with tensioning and tracking although there is no particular reason to do this. Generally this mechanism is located on the top wheel with the bottom wheel only providing the power. A notable exception to this is John Heisz’s bandsaw which tensions on the bottom wheel and tracks on the top.
The tensioning and tracking system needs to be constrained so that it can only move in the directions necessary to tension the blade and rotate in the direction needed for tracking.
The tensioning mechanism needs to be sprung rather than rigidly connected to the upper wheel. A rigid fixing will cause vibration when the machine is running.
I’ve had a look around the web for commercial bandsaws and I think for this build I’ll target something like the Axminster AT2552B in terms of specification. This gives me a budget of around £1000 to work with, while I don’t need a trade rated bandsaw I’d still probably buy one. Approximate specifications are below.
|2,552 mm (100.5″)
|Blade Width Min\Max
|3 mm to 19 mm
|Max Cutting Height
|Max Width of Cut
|Max Width of Cut with Fence
|Overall L x W x H
|690 mm x 720 mm x 1,770 mm
|Table Height on Stand
|500 mm x 356 mm
|5° – 0° – 45°
I have a 2311mm blade that I picked up cheaply from somewhere, I’d like it if the machine I build can take a wide range of blade lengths although I’m not sure how feasible that will be without making the machine ridiculously tall.
Useable Blade Lengths
I initially had the idea to design a saw that could use a wide range of blade lengths. Looking around it seems that most saws have about a 50mm range of lengths they will accept (see here). Lets quickly do some calculations to see if this makes sense.
Assuming the specifications in the table above. The diameter of the wheel is 350mm so the circumference is 1100mm. Half of each wheel is in contact with the blade so the remaining length of the blade is the vertical portion. This gives a vertical length of 726mm which is also the centre to centre distance of the axels.
Wheel Diameter (mm): 350 Blade Length (mm): 2552 Circumference: 350 * 3.14 = 1100 mm Vertical Blade Length: (2552 - 1100) / 2 = 726 mm
Moving the top wheel up or down by 25mm would accommodate a blade 50mm longer or shorter. In other words the blade length range would be 2502 to 2602. Looking at a table of common blade lengths (for example) this adjustment range would cover the ideal blade, the next one longer one and that’s about it.
Adjustable Travel (mm): 25 Axle Centre Distance (mm): 726 Wheel Circumference (mm): 1100 Maximum Blade Length: ((726 + 25) * 2) + 1100 = 2602 mm Minimum Blade Length: ((726 - 25) * 2) + 1100 = 2502 mm
It seemed odd to me that saws wouldn’t be built with more adjustment but if you stop to have a think about there’s a good reason for it. The inter-wheel distance determines the maximum cut depth of the saw. Assuming you aren’t going to squeeze in beside the wheel the absolute maximum cut depth of this saw would be 376mm since there’s a radius of each wheel blocking the path of the material through the saw. By the time you’ve fitted guards and taken into account the adjustment you’d be lucky to have 300mm of clear cutting height. In reality the saw probably couldn’t handle that so the 200mm given in the spec is entirely reasonable.
Wheel Diameter (mm): 350 Axle Centre Distance (mm): 726 Max Cut: 726 - 350 = 376 mm
My conclusion is that you have to build your saw to fit a particular blade length. You can probably make it flexible enough to handle a the next size up or down but after that you sacrifice too much for it to be worth it.
The specification I’m going by has a blade speed of 800m/min so from that we can determine the rotation rate of the wheels and if we know the rotation rate of the motor the pulley ratio needed to slow down the motor.
The distance the blade travels is independent of the number of wheels in the saw and the separation of those wheels. This means that the circumference of each wheel must be travelling 800m/min. Above we calculated the circumference as 1100mm or more conveniently 1.1m. The rotation rate of the bandsaw wheel therefore needs to be 727 rpm.
Target Linear Blade Speed (m/min): 800 Wheel Circumference (m): 1.1 Bandsaw Wheel Rotation Rate: 800 / 1.1 = 727 rpm
The motor that will be running this bandsaw is a 4 pole and therefore 1380 rpm (from the motor spec plate, often this is sold as 1400rpm) so an ideal pulley ratio would be 1380 / 727 = 1.9. I’ll go for a 2:1 pulley ratio as that it’s easy to find pulleys in that ratio. This will result in a linear blade speed of 759 m/min.
Motor RPM: 1380 Wheel Circumference (m): 1.1 1380 / 2 = 690 rpm 690 * 1.1 = 759 m/min
I’m not aiming, particularly, to restrict myself as to the what tools I can use. I don’t plan on buying any though and right now I mostly only have hand tools. The only large woodworking tool I have is a planner thicknesser which I’m sure will come in handy during this build.
At the moment I plan on making this mostly out of rough sawn pine which I will plane down do flatness. I can easily get 100x45x2400mm (e.g. 2″x4″) timber and I have a load of 18mm ply in stock.
This section is just a collection of interesting parts I’ve found while doing research.
Useful for the lower axle where the wheel is fixed to the shaft and the axle has to spin. There seems to be a naming scheme for this type of bearing but I can’t figure it out. Example.
Used to stop an axle from moving along it’s length. Certainly useful on the lower axle, maybe useful on the upper axle too. There are several different designs of these. I prefer the single split design over the grub screw design as it doesn’t mark the shaft. There’s also a double split design for awkward locations. Example.
Useful for the upper axle since the wheel spins on a fixed axle. There are pressed steel versions that are quite low profile and include grub screws to lock the wheel to the shaft. Example.
Lifting Eye Bolts
Lifting eye bolts are readily available that have openings that match common shaft diameters. They seem to be widely used in the boating world as there’s a lot made from stainless steel. They are easily available on Ebay for a few pounds each and they seem come with a range of threaded shank lengths. They have rated working loads in the low hundreds of kilos and breaking loads well over 1500 kilos.
Fixing the Wheel to the Shaft
For the lower wheel it is necessary to firmly attach the wheel to the shaft. This has proven to be one of the more difficult problems to solve. Most of the other homemade bandsaw builds end up gluing the shaft to the wheel or angle grinding a V in the shaft and putting in a screw. I’m not particular keen on either solution, I’d like something that can be easily removed if needed.
What I want is essentially a wheel hub, the problem is all cars use a splined shaft to connect the axle and the hub, I can no more cut a spline than a key way. Trailer hubs are closer to what I need but they have fixed axles and with bearings in the hub – I need the hub fixed to the shaft. Go karts have almost exactly what I need and there’s even locking connections but they are incredibly bulky and expensive.
You can get shaft mounting collars like this but I think it’s unlikely I could buy them in ones and twos. You can get rigid flange shaft couplings but they only seem to go up to around 14mm, I think they are used for hobby robotics.
I asked about this on the Mad Modders forum and someone suggested I try and get hold of half shafts from either a car or a trailer. They have the benefit of coming with a hub and often already have a bearing fitted.