50x2000mm Belt grinder, metric equivalent of 2x72"

[MrCrankyface]

Thanks! It's probably my best casting to date in all aspects of defects. Low porosity, low and even shrinkage, good surface quality etc.
This was old car rims that I've melted down previously into ingots, generated very little slag/dross on the surface when I remelted it for this pour.
I think the pour cup with a slight "wall" before the conical feed and relatively small runners/gates might have helped make this pour good, other pours where I've been sloppy with those has resulted much worse.
I also just heat the metal up and pour as soon as it's properly melted so it doesn't sit too long.
Could the propane possibly have an effect on the melt? Water vapor is a byproduct of it burning so possibly it gets into the melt and the hydrogen escapes into the aluminium, just guessing here though.

 

[Larry$]

Water vapor is a byproduct of it burning so possibly it gets into the melt and the hydrogen escapes into the aluminium, just guessing here though.

True! I do much the same process as you did & I lay the scrap on top of the hot melt furnace to dry before adding it to the crucible. Runner design on castings with thin cross sections have also given me fits.

 

[MrCrankyface]

Took a good look at this foot and realized I had fallen into the 'sunk cost fallacy' and decided to just chop it off.
Even if I could get this variant to work, it would still be subpar to a simpler solution that's just a bit taller.


Found some more square tubing and turned out much better. The foot now extend beyond the center of gravity, keeping it much more stable whereas before it was right on the limit of falling backwards if the tensioner was leaning backwards and the belt sander platten was not inserted.
I think I might make a diagonal brace between the front and rear legs, it doesn't seem needed but is also an easy addition that would add a lot of rigidity.
Not sure if I mentioned in the last post but I've also chopped quite a bit of the I-beam part to both save weight and improve clearances.
Whilst not in great focus, the two "long nuts" along the main beam, above/in front of the rear leg, are new additions for setting/adjusting the two fixes angles for the belt assembly.


I was feeling mentally lazy and made 2 seperate mounts, with the pro of having individual adjustability.
Basically you rotate it around a bit extra, rotate the link/nut into the fork and then rotate it back until it stops.
The center of gravity is such that it will naturally rest at a 45* angle so it takes effort to get it to neutral or sideways position and this in turn keeps the assembly held in place once the nut is captive in upper or lower(hard to see here) mount.
In both situations the screw is held in tension so there's no risk of buckling and makes it easier to have rigid setup.


And in neutral position.
Belt tracking still needs some adjustment, I think I will remove the crown from the drive pulley as it seems to interfer with the tensioner and see how it behaves after that.
Most likely I need to shim out the tensioner a bit too.
Some other random bits and bobs:
-Completely remade the hinges, this time I faced everything in the mill and was quite picky making them all accurate, really helped make it work well.
-Reinforced the tensioner assembly by welding flat stock inside the tube that the tensioner screw rests against. It felt like this was too weak before and would alter the tension as the screw dug into the square tubing.
-Welded in an endstop in the belt assembly outer tube so the inner tube can't go too far back and impact the drive wheel.
-Extended the front foot to match the new height of the rear foot.
-Kinda went over everything and welded more where there was gaps or not strong enough.

 

[Larry$]

Kinda went over everything and welded more where there was gaps or not strong enough.

Constant improvement! (Flight #5 may be caught on land!)

 

[MrCrankyface]

Mocked up a quick table. Not happy with the angle adjustment but it works well enough for now.


Table adjusts well to tilted belt as well.
Table itself can be tilted ~60 degrees.


Finally made a control box. E-stop turns it off and requires resetting it.
L1-L3 are individual glass fuses.
With E-stop out, the green button hits a contactor on the inside which starts the motor.
I was slightly worried it'd start too harshly but it's very smooth and nice.
Need to fix the table so it locks better at the different angles and also adjust the sliders a bit as some of them are slightly binding but otherwise somewhat finished for delivery.
Figured I'd let the new owner decide himself on paint and further accessories.


Edit:
Final beltspeed is around 30m/s and I think it chews through steel quite fast with a 60 grit belt.
I think it could run even faster and cut with less pressure on the belt but don't have a VFD to test with at the moment.

Edit2:
Some other lessons learnt:
-Tension wheel should probably be less crowned, it's VERY sensitive at the moment.
-Drive pulley should not be crowned, it definitely seems to interfere with the tracking.
-1.5kW feels like plenty of power for this type of grinder, even pushing the piece relatively hard against the platten it does not slow down the least.
-Speed control feels like it could be valuable for when you want a less aggressive experience.
-Some kind of dust evacuation/sucking would be great as it seems to throw dust in all directions.

 

[MrCrankyface]

After testing that belt grinder for a bit I realized how extremely handy they are so might as well make my own now, a bit simpler in design though as only the table will be tiltable, in one axis.


Some chopping and welding of whatever I had in the scrap pile, at which point I wanted to have rollers and the drive wheel available to test with.


So time to cast those. Tried doing several parts at once and also packed the moldmuch lighter than usual to see how the result would come out.
I can't see any signs of gas entrapment but plenty of flashing and "oddities" like the cylinders looking slumped over and such.
Most likely the molten aluminium has bulged out the mold in places.


With some cleanup it's still highly useable parts. The big pulley is pretty much finished whilst the 3 small ones need a bearing countersink on the other side and then be turned to final dimension on the outer diameter.
Now I should be able to start mocking up where the motor should sit and weld the last parts of the frame together.

 

[Larry$]

I can't see any signs of gas entrapment but plenty of flashing and "oddities" like the cylinders looking slumped over and such.

I often get enough shrinkage to ruin the part. Looks like you had almost no shrinkage. Surprising given the mass of the parts. I also usually get some hydrogen bubbles. You are better at controlling the process than I am, nice job.
Still, I enjoy being able to cast workable parts.
What was your source aluminum?

 

[MrCrankyface]

Thanks! This casting was ~95% car rims, some of the ingots used had been recast several times which I think showed up especially much in the smaller cylinders where the surface doesn't polish up as nice, seems evenly dispersed "artifacts" so probably hydrogen inclusions?
Will take some better close up pictures.
Also almost ruined one of the cylinders because a part of it had shrunk or the mold had "moved" when casting, luckily I could just turn it smaller and use it as the crowned "steering" pulley.

I suspect the large pulley comes out so nicely because it has a lot of center mass with reducing thickness the further out from center you go, I would at least like to think that this causes the outside to freeze up first whilst the riser can continue feeding the hub as it slowly freezes up towards the center.

 

[MrCrankyface]

Surface definitely isn't top quality but can't feel anything with my nail. Would be interesting to look at it under a high strength microscope maybe.


Got most of the frame welded together plus the tensioner arm.
Took me 4 tries to get this right somehow, each try resulting in having to cut things off.
First time I attached it on the wrong side, second time it was correctly attached but some tubes were too long(misread my own drawings), third time with correct tube length on the right side but wrong rotation of the platten etc ... Rough day ...


Here's the tensioner assembly which wheel I managed to save by turning it down a bit smaller. Weaker crown than the last grinder.
The belt wants to tip the wheel towards the left/up direction and the smaller screw going through both the aluminium piece and square tubing resists the belts pull.
The bolt threads into a cylinder inside the square tubing to both allow it to angle around and spread the load better than my previous design.

 

[MrCrankyface]

Had to tweak quite a bit to get the belt to track reliably.
With the tensioner aiming straight back it would track great in reverse and not at all going forward.
Quick enough test was to cut the tube, bend it upwards and tack some bars on, suddenly it's much more responsive.
The flat drive wheel was also not a real great hit, it did seem to track a lot better once I added some layers of tape to act as a crowned wheel.
I added just a few layers to the "front" wheels as well and it definitely helps.
I also added a gas strut I had laying around to provide the tension, and it provides quite a lot...
I turned the heads of two M6 bolts down to be more spherical and it worked great as ball joints for the strut.
You can "ping" the belt like a string, if it keeps together I will use it like this but if the belt comes apart I think it might be too much tension.


Finally I turned up some thick tubes to go through the frame, they're internally threaded to M10 which lets me bolt various tables/jigs to the frame.
First up was just a basic table that can tilt to 60 degrees or so.
Not much to say about it, upper bolt act as pivot, lower acts as a angle lock/slot. With both locked down it seems incredibly sturdy.