A tiny tabletop bandsaw Part I (prototyping)
Вставка
- Опубліковано 20 січ 2024
- Building a prototype band saw. As small as doable but strong and able to cut solid steel bars and metal sheets.
The start was very promising, but then. %$#*@!!!
.....
I'll be back!
No music this time. Sorry, too much to tell. - Наука та технологія
Love it! I like the 3 wheel version. 120 deg drive wheel contact rather than the 90 deg with 4 wheels.
Interesting to follow your experience with making this saw 👍😀...
I have been thinking about something similar because the saws that are available are to big for my needs.
Thanks for sharing this great video 👍😀
Good luck, this is an ambitious little project, very cool!
Like a lot of the other comments here I had a three wheel saw and had a "love/hate" relationship. It gave the same ability as you are seeking with your four wheel idea but it had very short blade life. And the wheels were even bigger than what you're using. My regular big saw blades are .025" thick. So I think I'm remembering correctly that the three wheel blades were .015 or .017. That's 0.38 or 0.43mm. The failure mode, if the weld didn't go first, was stress cracking at the deep point in the gullet of the blade. And on blades that failed that way the other teeth on many of the blades that did last a bit longer all had obvious stress kinks and cracks and if the one crack were to be re-joined another crack would fail totally within almost no time at all.
These were wood blades though. For metal blades there is a lot more teeth and that means a lot more segments to share the bending of the tight radius. So while I think your wheels are still too small even for thin blades I suspect that a metal blade used on the roughly 6 inch/15cm wheels of the old compact three wheel saw might work fine.
If we stop and think and given the evidence of my many blades with stress cracking at the narrow base of the gullets the issue seems to be the kink angle at each gullet. On a wheel these gullets all end up acting like short straight segments with little bend on the wide part at the peak of a tooth and much of the bend being at the more flexible narrow part at the base of each gullet. Finer teeth means more gullets and less of a bend for each gullet. So small close spaced teeth should survive better than wider spaced teeth. Also a wider blade means less a smaller step with each tooth and a greater chance that the tooth and gullet parts will bend closer to the same and put less stress into the narrow gullet parts of the band.
But I'm still thinking that 80mm might be a little too small in actual use even for a thin blade that is a little wider and has a high tooth count. But it'll be close and it might be enough. Time will tell I guess... My fingers are crossed that it will work out for you.
Good analysis and I agree with you for the most part. In particular, the insight that the saw does not bend evenly over the disc, but does so in sections with more and less bending, depending on the width at a tooth top and tooth valley, shows that you have thought about it carefully. In practice, an 8 cm diameter appears to be too small, even for a thin, fairly wide band saw with small teeth.
But I am going to experiment with 12 centimeter discs, precisely because many manufacturers sell models with such discs, and then you would expect that the thinnest saw blades would last a sufficiently long time. Time to dig through some reviews of those small band saws.
@@Michel-Uphoff It occurred to me after my post that the bend is occurring in little stages as the blade rolls onto the wheels. There's a transition from straight to curved that sets the angle. And each tooth and gullet pair goes through this flexing each time it rolls onto and then off the wheels. And as we all know the best way to fatigue a piece of metal is to flex it frequently. We know there is a point past the elastic point where the damage is real. But this frequency of flexing seems to take its toll on some or most(?) metals even if it occurs often enough within the elastic limit. I feel like that is the key issue with running on small wheels.
Anyway, I enjoyed this mental exercise and I hope you find an option that works. I suspect that the key might be finding the optimum blade tension too. And then there's the tooth and gullet shape. I suspect there is a very good reason why bandsaw blades have generously rounded gullets where hand saws get by with sharper shapes.
Cheers
I have now gone through the specifications and dimensions of a number of small band saws from well-known brands, and I see that there are many saws with discs of 11 and 12 cm diameter. I can hardly imagine that these brands (Hilti, Bosch professional, Makita, Dewalt, Milwaukee) all market saws that devour blades as a hobby, as stated here somewhere.
What is very important is the thickness of the blade. Under a disc diameter of 12 cm it is actually necessary to use 0.35 mm thick blades. Furthermore, the saw tension is crucial, precisely for the reasons you mentioned.
I'm now doing some testing, and so far things are going in the right direction. With a bit of luck I will keep the dimensions of the machine within a 35 cm cube with sufficiently sized disks.
Excellent stuff! I would remain with your 4-wheel configuration, as it gives you the option for a deeper "throat" to use when cutting from wide sheet material. That said, the three wheel configuration does have advantages also. I am interested to see how this goes.
Awesome saw! Thanks for sharing!!!
Great Video! In aliexpress is Diamond coated Steel Wire 0.1 to 0.25mm available. Might be worth a try if a possibility is found to combine the Two ends.
Realistically, you have to view it as a single bend scenario. Smaller radius curves making the blade travel 2 90 degree corners are forcing the blade to compress and stretch to a higher degree than a large single wheel does completing 180 degrees. If you drew the nuetral axis of the blade along the path you intend it to travel and approximated the angle of bend at a single point, considering the largest angle, then even a series of smaller rollers spread out over a larger radius will tend to create an excessive bend angle. You would have to have a huge amount of tiny bearings spread over a large radius, at which point you might as well use 2 larger rollers/drive wheels. You could always investigate different quality blades too, although thats a bit of a witch hunt. it appears to me that the cyclic stressing of the blade as it turned the two 90 degree corners caused it to work harden and fail, due to plastic deformation. You could replace the 2 rear rollers with one large one, and increase the diameter of the 2 smaller ones, this automatically would half the stress on the blade and potentially solve your issue. You could investigate making it fold in half for storage, then in theory you could still have the small storage profile but get away with 2 larger diameter rollers as the drivetrain. Personally i think this would be the best solution, but has its annoyances. If you put the time into the design you could get it to fold away quite neatly without sacrificing too much of the table size. Down side will be getting the blade to track correctly when you unfold it without having to completely disassemble it. I reckon its doable though, and to be fair you are one smart guy, you have some pretty elegant solutions to the challenges you faced in your projects so far, so i definitely think you are capable of it!
Thanks for the input, appreciate it!
Each disk introduces a flex/unflex cycle in the blade per revolution. You had four small disks, so that's 4 flex/unflex cycles, each one to a fairly tight radius. I am not surprised the blade couldn't handle it. In addition to metal fatigue, you may have heat issues in the blade since bending will generate heat. In a traditional two-wheel bandsaw design with only two, large wheels, heat is minimized because a) there are fewer flex/unflex cycles per revolution, b) each flex is much gentler since the degree of curvature is so much less, and c) any given point on the blade has more time between a flex and an un-flex, so it can dissipate heat into the wheel or into the air. The best design would pinch the blades only just before and just after the cutting section, so the blade is straight in the cutting area. But outside the cutting area, the blade is allowed to follow its natural circular curve. The blade would be driven by *pairs* of wheels that contact both the inside and outside of the bade. Several of these pairs--some powered, some idling--could guide the wheel around its natural circle. And since the blade does not have to conform to the curvature of these wheels, they can be as small as you want. This would be enormously more complex to build, but it would minimize flexure of the blade while permitting a maximum throat depth close to the diameter of the blade's natural circle.
So, you are claiming that a large number of small bends place a greater stress on the steel than a small number of large bends. In both cases, the steel bends the same amount and over the same radius per cycle. If your claim would be right, such a saw band would suffer greatly if the bends are minimal, but the number is very large. That would mean, for example, that a resonating tuning fork would quickly break en become very hot. In fact it is all about elastic and plastic deformation and metal fatigue. The total amount of elastic deformation doesn't change between a few or many bending points, in fact is is probably lower when many small bends are used.
@@Michel-Uphoff I'm not sure it works that way either, I agree with you about the first part too, it makes sense that 180x one degree bends will still give you a complete 180 degree curve, whilst bending the blade minimally and maintaining elastic deformation within the blade as it travels the gentle curve one degree at a time. 4x 45 degree bends complete the same curve much more aggressively and I believe it is the greater angle of bend the blade is being subjected too that it pushing the metal into plastic deformation, or at least stressing it to the point it is work hardening and reaching a brittle failure mode, similar to how vibration can cause cracks to propagate. A design may stack up against the calculations and seem workable, only to fail under a cyclic loading scenario, because the constant repetitive bending stress on the material leads to fatigue and work hardening.
I might have been temoted to go in the direction of a micro cold saw. has greater potential to be compact that a bandsaw. less messy to opperate too. can't do sheet with it then of course. I have one of the small proxxon bandsaws and it's not much good as sold. i had to upgrade the motor and do some other stuff to make it useable for anything other than wood
nice, you need more radius, be more progressive in the bend of the blade, you can use a big wood wheel and use rubber of bicicle
just puttin it out there - i have a bandsaw with 5 &7/8ths inch diameter wheels and they are TOO SMALL. they snap blades for a hobby. the bandsaw is basically a pain in the ass to use. bigger wheel, less tight bending, longer blade life. that said, on with the video. hope it works for you!
What is the thickness of your blades? I understand that for the really small band saws 0.5 mm is the max.
0.65 mm is definitely too thick, I've tested that by bending such a saw blade around an 8 cm disc. That is too much and the deformation is plastic instead of elastic. But with a 0.5 mm blade the bending is well within the elastic region and the blade springs back without deformation. So it comes down to metal fatigue, a much more difficult phenomenon to calculate.
honestly i don't know. i am sure it can be a useful tool if set up right, they made a lot of these small 3 wheel bandsaws back in the day... but there is a reason they aren't popular any more
@@Michel-Uphoff I posted a comment with a link to some 0.35mm blades but it appears to have vanished. A company called Tuffsaws in the UK sells the blades and will make them to custom lengths. I'm not sure where you're located but they're about £13 delivered in the UK. Not affiliated with them, but I have bought blades from them in the past and they were fine.
Interesting, welcome info, so thank you! I will pay them a visit 🙂
I have had one of the 3 wheel bandsaws and had blade breakage issues. Never had much luck with small wheel designs. I wonder if a vertical power hacksaw could be possible. Similar to a scroll saw with guide blocks similar to a bandsaw. Just thinking out loud 😂
Actually, I thought about something similar. I was building a new gadget in the last two weeks and needed to cut a lot of metal slightly above the capacity of my bandsaw. I ended up mounting my jigsaw upside down in a workbench insert. It does work, but it is crazy loud.
Could you not have a series of even smaller discs arranged in a big wide arc top and bottom and run the blade over them in an almost circular shape so there's very little bending. at the moment you have a blade in a square arrangement with r=40mm corners. Imagine you could turn that into a square with say r=100 corners, not by having 4 x 200 diameter discs per corner but by having say 5 or 6 30 diameter discs arranged on a r=100arc. You'd keep the small, square form factor on elevation that way
Very sensible input!
I am considering that in addition to other options; a series of ball bearings arranged in a wider arc, for example. I'm afraid that could also be problematic, but I'll test it. If you look at it a bit more analytically:
Whatever happens, each point of the saw blade must rotate 360 degrees per revolution, the radius can vary. And now imagine that it is around a ball bearing of, say, 3 cm in diameter. The saw will definitely break. Now I put those bearings on a larger circle circumference. In theory, the saw moves along a straight line between the bearings; the thinner the blade, the easier that is. With each bearing the blade will bend a bit equal to the radius of the bearing en that causes strain and fatigue. In total, as much as I wrap the blade around that one bearing.
What will probably help, if the blade is not tensioned too tightly, is that the belt does not immediately follow a straight line between each bearing. The blade will be slightly pre- and post-bent on both the arrival and departure sides, which reduces the bending radius around the bearing. This may be enough to ensure sufficient tension on the saw and avoid undesirable abrupt bends. But I have to test that thoroughly before I go any further.
If possible, I will show the results in a next video.
Unfortunately it doesn't matter for how LONG an arc is sustained. When simulating fatigue, you only consider "amount of deformations, strength of deformations".
So for instance - having 8 bending points of 60mm radius on a loop is almost exactly 8x the fatigue strain of having 4 bending points of 85mm radius...
It doesn't matter HOW LONG the arc of minimum radius bend is, only how tight each bend strains the material before returning to "straight". A five degree bend that's tight against a 60mm radius sees (almost) the exact same strain as a 180degree bend on a 60mm radius
@@Michel-Uphoff Can't wait to see the update!
@AJBtheSuede
I think that's correct. The only difference between, say, 20 short 18 degree bends and 2 longer 180 degree bends around the same perimeter is that there are many more pre- and post-bends due to the number of bends. I can be sure that in a 180 degree bend there will be some portion of the blade in the middle of the bend that experiences a maximum bend equal to the radius of the bearing. The pre- and post-bends are also present there, but due to their small number they hardly count. But with, say, 360 bends of one degree over 360 bearings, the pre- and post-bending at each bearing will almost ensure a perfectly round circle. Then the radius of the bearing almost no longer matters. Unless I put the saw band under extreme tension ofcourse.
Yes, the bend is less than maybe 5 degrees the center of the bend no longer sees a full deformation strain, no longer stressing the material fully. But that's 72 rollers for a full loop :) With fewer rollers than that, the basic material stress per loop is basically (1/radius)^2 * number of bends. So 20 wheels is twice the strain 10 wheels (of the same diameter). @@Michel-Uphoff
I wonder if an abrasive blade (diamond, carbide) would work better than a chip making blade.
Small BS's crack blades, slightly bigger ones as you are thinking crack and break welds because the weld is not flexible as is the blade. If you make the saw with ability to tension a wide blade high enough you will not need guide bearings.
impressive video Michel , I'm hoping to make a bandsaw for hobby use. Asked advice on minimum diameter of wheels. A company that supplies bandsaw blades recommended 10" minimum , 12" preferable for 1/2" blades 0.025 in thick. fully understand that's bigger than you hope to make. hope that helps.
I presume that's a saw for wood? .025 is too thick for my purpose. 0.02, 0.016 or even 0.014 is the way to go for small wheel diameters. Makita makes a portable metal bandsaw (DPB184Z) with 11 cm wheels, and 0.5 mm thick blades. Bosch professional makes an even smaller one (GCB 18V-63) with one 11 cm and one 10 cm wheel and a 0.35 mm thick blade. 12" would lead to dimensions that are way too large, a minimum height of 70 cm for example. I want less than half of that. I'm busy experimenting and working on it and I think I will get it done.
@@Michel-Uphoff - they were advising for cutting mild steel - with appropriate low blade speed. I reckon most of their customers are industrial , and they appear to only supply .025" ( .63 in proper units )
Your video makes it clear you're aiming for a small , precision machine. I will look forward to seeing how the project progresses.
i wonder if it would help to add 2 more wheels and thus, each wheel would only bend the blade by 60 degrees instead of 90... but the machine gets taller as a result....
Yes, the machine would get bigger, and I'm afraid it won't help much. It is discussed in more detail elsewhere in these comments. The bottom line is that 6 bends of 60 degrees produce hardly (or none) less strain and metal fatigue than 4 of 90 degrees. After all, bending always has to be done around the same radius, so the bending angle stays the same. I think bending in many small steps can be beneficial, because there are now many more lead-in and lead-out areas, where the bend is less than the radius of the wheel. It is possible that, with a certain tensile stress that is not too high, bending does not occur all the way to the radius of the wheel in this case.
Michael, if you only cut small stock the Proxxon chop saw may fit the bill. Enjoy your posts.
Thank you.
I don't want a chop saw, because I also want to be able to saw sheet material up to a width of about 20 cm (and more length).
Could You provide more info on the blade? How long is it and where to buy one like this? Is it a standard or custom length?
M42 HSS bimetal saw band for cordless band saw Metabo MBS 18 LTX 2.5. Dimensions: 835 x 13 x 0.5 mm with combination teeth 14/18 teeth per inch. Suitable for cutting metals from carbon steel to high-strength alloys.
End of ad ;-)
Here in the US anyway Milwaukee tool sells a small subcompact portable band saw that uses 27inch long by 1/2" wide blades This slightly under a Meter long and 12.7mm wide with a thickness of around 1/2 mm. The wheels the blade run on are roughly 5" diameter which equals 127 mm or maybe slightly larger and have rubber tires. Perhaps that information can be of use to you. Anyway it's a great concept Michel, I would imagine many people could make use of such a tool in the home or model shop, even they commercial prototype or small parts world. I know I would find one useful cheers
That's very useful information, thank you!
Everything here is metric, but let me convert it. The blade is 27 inches long, you write, which is 68.6 cm. You measure at least 12.7 cm in diameter at the discs. This means that, if I'm not mistaken, there is (68.6 - (Pi*12.7 + 2*12.7) )/2 = 1.7 cm of free space between the disks. That seems way too small to me. Is the distance between the centers of the disks really only 12.7+1.7=14.4 cm (5.66 inches)?
@@Michel-Uphoff I think we might be talking about two different things. I will measure the diameters of the drive wheels. I looked today and they seemed to be about 127mm in dia. I meant to measure them and got sidetracked by what I should have been doing 😀
very nice workman ship. are there any small bits of plans around for purchase?
When I have finished the final build, I will see if I can publish some clear sketches. But for the time being this is the development of a prototype, so no dimensional sketches yet.