Drum/ Thickness Sander

[S. Heslop]

I've started working on a drum/ thickness sander to help with making skateboards, of all things, but it'll likely come in handy for some other stuff too.

This might be a bit of a slow project since I have no money (or job!).


The plan so far. I haven't got a motor currently so I haven't dealt with that part in the plan yet.


My table saw isn't anywhere near wide enough for alot of the cuts, so i've had to do it the hard way.


Could still fit some stuff on for cross-cutting though. I'm cutting out the two boards that make the table surface.


Then used the tablesaw and bandsaw to cut out the sides of the table, using printed templates.


Grooves were routed into the sides. The table is going to have a conveyor belt on it, and the grooves are for adjusting the tension and tracking with some slides that fit in them.


The slides were also cut on the router table. I had trouble at first since I discovered that the surface of the router table is no longer anywhere near flat, so the boards raised up enough to make the channel a full millimeter wider at one end than the other. I countered this by building some push blocks so I could force the wood sit pressed against the table and re-cut the grooves.


Another problem I had was mucking up the order of operations, making it very difficult for me to drill the holes that the threaded rod (for adjustment) fits in. I routed a groove all the way across the bottom of the channels.


And fitted in some lumps of wood to fill the gap.


It also turned out that the birch plywood I was using for the sides and table wasn't flat either. I routed a groove in the surface slabs to accept a rod for pivoting (to adjust the depth of cut), which i'd positioned so that the concavity in the bow was towards the center of the two boards. Hopefully the bow in each board will cancel each other out.


When clamping it up I used two boards of actually flat plywood to hopefully encourage the table surface to remain flat once the glue dries.

With glue drying there's not alot more I can do today. Plus when it gets dark I don't have enough light in the garage to film stuff well.

[mattinker]

What's wrong with of all things, making skate boards?

All the best, Matthew.

[S. Heslop]

What's wrong with of all things, making skate boards?

All the best, Matthew.

I got a friend who's interested in making them to sell and asked for my help. Myself I can't even stand up on a skateboard. I'm just not sure how to go about selling them though so i'm not too confident in the idea, but it should still be interesting to give it a go.

I've had a hard time sourcing maple veneer (at about 1.5mm thick) in the UK for a reasonable price, so the drum sander will be for making it myself. But with real high end skateboard decks costing £60, it's probably going to be tough keeping it profitable with the materials and labour involved. But we'll see.

My original goal was to make banjos to sell, and the drum sander will help with that too. Except to make banjos properly i'd need a big lathe, and I can't afford one and probably wont be able to for a while.

[vtsteam]

This should be good! 

[S. Heslop]

It's been a while. Got around to doing a bit more.


Sides of the slab weren't quite flat or parallel so I trued them up on my crappy jointer/ planer. Doing this was guaranteed to nick the knives but it wouldn't fit onto my tablesaw or my router table unfortunately.

I'm also not sure if i'd recommend buying one of these combination tools, they have alot of problems and I haven't really bothered to try sort any of them out since buying it.


I worked out that the gap between the top of the table and those channels should be about 4.3mm, so I used some 4mm brass rods and pop can shims to space the parts. I also made sure it was lined up properly horizontally by checking the steel bar that passes through for squareness.

The sides were clamped to the slab and the screw holes were drilled after it was all lined up.


And that's the sides attached.

[Lew_Merrick_PE]

S. Heslop -- I have made several variations on drum-type thickness sanders over the years on support of my original trade training as a luthier (lute and guitar maker).  The design that I find most useful is to use a piece of thick-wall tubing as the drum and then to spiral wrap (1-1/2 inch or 40 mm) wide strip abrasive media to the drum.  DOM (Drawn Over Mandrel) tubing typically only varies a couple of thousandth's of an inch (call it 0.05 mm) from perfectly cylindrical in lengths up to 20 inches (500 mm) (and your abrasive media will vary more than that), so the only real challenge is getting the tubing mounted on a shaft.  Personally, I make a press-fit plug that has a shoulder to fit the tube that I attache with (strong) set-screws.

I do not have access to my current beast (it is in storage due to a lack of space), but I built it back in the 1990's.  The thickness adjustment is a pivoted piece of laminated (good cabinet quality) plywood that is faced with UHMW polyethylene.  A high-tack conveyor belt drives the piece through the sander.  The run surface of the belt is 40 inches (basically 1 m) to support infeed and outfeed of guitar ribs.  I can hold a bit better than +/-.010 inches (+/-0.25 mm) using this set-up.

[S. Heslop]

S. Heslop -- I have made several variations on drum-type thickness sanders over the years on support of my original trade training as a luthier (lute and guitar maker).  The design that I find most useful is to use a piece of thick-wall tubing as the drum and then to spiral wrap (1-1/2 inch or 40 mm) wide strip abrasive media to the drum.  DOM (Drawn Over Mandrel) tubing typically only varies a couple of thousandth's of an inch (call it 0.05 mm) from perfectly cylindrical in lengths up to 20 inches (500 mm) (and your abrasive media will vary more than that), so the only real challenge is getting the tubing mounted on a shaft.  Personally, I make a press-fit plug that has a shoulder to fit the tube that I attache with (strong) set-screws.

I do not have access to my current beast (it is in storage due to a lack of space), but I built it back in the 1990's.  The thickness adjustment is a pivoted piece of laminated (good cabinet quality) plywood that is faced with UHMW polyethylene.  A high-tack conveyor belt drives the piece through the sander.  The run surface of the belt is 40 inches (basically 1 m) to support infeed and outfeed of guitar ribs.  I can hold a bit better than +/-.010 inches (+/-0.25 mm) using this set-up.

Shame you don't have access to the sander, i'd like to see it!

For the drum I was going to try making a stack of MDF disks and turning them round. Unfortunately I don't have a lathe big enough to try the pipe method, but i'll keep it in mind in case the MDF doesn't work out. I think the biggest risk with MDF is the stuff changing dimensions over time as it absorbs/ loses moisture.

How did you attach the sandpaper to the drum though? I was thinking about going with velcro since i've seen other people use that, but i'm still skeptical about if it'd hold or not (on the edges at least). The other method i've seen is to physically hold it at both ends with a sort of clamp/ screw and making the cutout in the drum for that might be easy with the MDF disks.

[S. Heslop]

Ran into a problem already. Got the MDF stacks for the rollers glued up, and started turning them true with this setup.





The problem is that it's producing an undulated surface where the looser MDF in the middle... fluffs out.



Still not sure on the solution. I'm thinking I could turn them down further and slip/ glue on a piece of PVC pipe and then turn that to size, or I could dilute down some varnish and try soak it into the MDF to harden it so it turns better. I've already messed with the cutting tool (chisel) geometry and angle a bit, and it didn't seem to make a difference. Nor did sanding it.

Oh one thing I don't think I mentioned is that i'm trying to build this without using the lathe at all, for the sake of the video. I'm also not going to use the pipe and end caps idea for these rollers since I want to try turning in a slight crown to see if it helps with tracking the conveyor belt.

[chipenter]

Sand it with a solid flat block then seal it .

[awemawson]

Make up a 50 / 50 pva glue / water solution and saturate it. Leave over night to dry then try turning it again, or possibly block sanding it

[S. Heslop]

Make up a 50 / 50 pva glue / water solution and saturate it. Leave over night to dry then try turning it again, or possibly block sanding it

I did try sanding it and it didn't do much to help. I'll try the glue thing, although i'm a little worried the moisture will mess up the MDF. I guess if it doesnt work at all i'll just drive the shaft out and try making it again with plywood instead.

[Fergus OMore]

The most obvious thing is 'Sanding Sealer' but if you are windy about PVA, any spirit based varnish left over from other DIY jobs should seal the MDF. Me, a few miles up the road from the Gill, I've used some old 2 part resin for a silly job that has passed its sell by date. Accelerated the gel a bit with a hairdryer.

OK. I'm 'with it' but there are proprietory Sealers for even wet and rotted wood. I did a rotten keelson on a dinghy from Derwent Reservoir with ordinary fibreglass resin and packed it glass in stages as the gel exothermed.

It' s dead easy. Mucky, yea

Cheers

Norman

[greenie]

I did try sanding it and it didn't do much to help. I'll try the glue thing, although i'm a little worried the moisture will mess up the MDF. I guess if it doesnt work at all i'll just drive the shaft out and try making it again with plywood instead.

The sanding drum is without doubt, the most important bit of the whole build.

If you can not get the drum to be stable, then you are going to have variations in the thickness of whatever you put through it. Chipboard as you have already found, is nowhere near stable, plywood will be of a similar thing with the way the grain is layered when making the ply.

My opinion is to make it from steel if possible, that way it is very stable and the only variances will be caused by the temperature. Possibly it might be worthwhile, inquiring about having it made from some metal instead of some form of assorted timber products.

Another alternative to chipboard or plywood, is to make it from a solid bit of every dense hard wood, that way it will be reasonably stable.

Your choice on what you use for the drum, but keep in mind what has been stated.

[greenie]

I've started working on a drum/ thickness sander to help with making skateboards, of all things, but it'll likely come in handy for some other stuff too.

This might be a bit of a slow project since I have no money (or job!).

A bit overly ambitious with a motorised in-feed roller when the readies are a bit on the short side, why not scrap that idea and use a bit of glass that's stuck onto the adjustable table, to allow the product to be pushed through with the least resistance. Much, much cheaper if you use a push stick to shove the product through it, eh.

Good to see you have allowed for a cover over the sanding drum, make sure you attach a vacuum cleaner to it, to suck ALL that dust away.

[Fergus OMore]

The sanding drum is without doubt, the most important bit of the whole build.

If you can not get the drum to be stable, then you are going to have variations in the thickness of whatever you put through it. Chipboard as you have already found, is nowhere near stable, plywood will be of a similar thing with the way the grain is layered when making the ply.

My opinion is to make it from steel if possible, that way it is very stable and the only variances will be caused by the temperature. Possibly it might be worthwhile, inquiring about having it made from some metal instead of some form of assorted timber products.

Another alternative to chipboard or plywood, is to make it from a solid bit of every dense hard wood, that way it will be reasonably stable.

Your choice on what you use for the drum, but keep in mind what has been stated.

All very interesting but********************

Just over the hill from where Mr Heslop lives is a huge American owned firm  which made rollers from rubber- which was actually blown onto a mandrel and used in the spinning industry- after being ground -precisely.
The foreman was and maybe still is a member of TSMEE.

My belt sander was from Picador and was aluminium or Chinese metal( I forget) and my present ones are actually plastic.

But here's the rub( pun intended), a belt sander is NOT and never has been accurate. There is a lot of information which was written about it elsewhere- probably Model Engineer- but it ain't going to produce FLAT surfaces.

Sorry and all that

Norman

[greenie]

All very interesting but********************

Just over the hill from where Mr Heslop lives is a huge American owned firm  which made rollers from rubber- which was actually blown onto a mandrel and used in the spinning industry- after being ground -precisely.
The foreman was and maybe still is a member of TSMEE.

My belt sander was from Picador and was aluminium or Chinese metal( I forget) and my present ones are actually plastic.

But here's the rub( pun intended), a belt sander is NOT and never has been accurate. There is a lot of information which was written about it elsewhere- probably Model Engineer- but it ain't going to produce FLAT surfaces.

Sorry and all that

Norman

I'll agree with you that a BELT sander is not very accurate AT ALL.

BUT your forgetting what is being made here, it aint NO belt sander, it's a drum sander, it's got a rotating drum which does the sanding.
If the rotating drum has been made round and straight, the sanding medium applied so that it is very constant in it's thickness, then the machine can and will reproduce the very same results constantly. Far superior to what a belt sander will ever be able to reproduce.

You can get a drum sander that is very accurate, to with-in the variances of + or - 0.001", I have made one that will do that quite adequately each and every time I use it. These measurements can be proven with a micrometer time after time, so don't try and tell me that it can't be done, -- it can.

[vtsteam]

Simon. just lurking here in the peanut gallery, hoping you're doing what you want.   

[S. Heslop]

A bit overly ambitious with a motorised in-feed roller

Absolutely, but it's the part of the project I care the most about. I've made it so long so I can use a sanding belt as the conveyor surface. There's alot of information on drum sanders where you just push the wood through, but the few people that have built conveyor belts are tragically vague about how they work. Plus i've read a few people saying they were difficult to get working and they eventually gave up trying, and that sounds like a challenge.

As for the stability of MDF, it's really the ideal for stuff like this since it doesn't have a directional grain. If the drum is properly sealed it should hold up fairly well, and if it does warp a bit you can true it up by just sending a bit of sandpaper on a board through (or with a sandpaper conveyor you could just raise the table till it touches). If I used solid wood it'd expand along the grain to make the drum oval shaped, and truing it up would be futile with it going back and forth with the seasons.

Simon. just lurking here in the peanut gallery, hoping you're doing what you want.   

Haha, thanks.

Yeah i'm trying to build a thing that should be repeatable by anyone else. Mostly I hate seeing videos where a guy uses, say, parts from an old tablesaw to build something else, and I imagine it's equally irritating for people when they see me using a metal lathe in a video about woodworking. Plus it's fun finding ways around problems.

[vtsteam]

Plus it's fun finding ways around problems.

[Spurry]

As for the stability of MDF, it's really the ideal for stuff like this since it doesn't have a directional grain.

Have you considered using the green MDF. It's so much more stable than the normal brown stuff. I would hesitate to say it's waterproof, but water resist is how they advertise it.
Pete

[vtsteam]

Simon, random thoughts per your own design intention to true a wooden roller:

Traditional, non water-based polyurethane varnish won't dissolve the MDF binder and is thin enough to penetrate deeply and harden fibers uniformly. Polyurethanes are probably the most vapor resistant of any coating. Including epoxies, and more widely available.

Sanding to remove contours using a block against a turned wood piece is difficult to achieve. A rotary sander arranged like a shaft grinder to keep a fixed distance from the axis of the turned part will remove only the high spots if the pressure is light enough. See John Hill's redneck shaft finisher thread for the general principle.

Finally, if you do decide to give up on the MDF, a traditional way of using conventional wood in a turned design that must remain dimensionally stable is to rip strips into 45 degree beveled staves and glue together into an octagon section, then turn to round. The grain orientation should be consistent in the staves. For large turned sections, the joints should be splined.

Hope that helps as FYI suggestions only! 

[greenie]

Absolutely, but it's the part of the project I care the most about. I've made it so long so I can use a sanding belt as the conveyor surface. There's alot of information on drum sanders where you just push the wood through, but the few people that have built conveyor belts are tragically vague about how they work. Plus i've read a few people saying they were difficult to get working and they eventually gave up trying, and that sounds like a challenge.

 Plus it's fun finding ways around problems.

Simon, here’s something else to ponder over.

With the infeed roller using a sanding belt to feed the timber through the drum sander, will this not create one sanding surface working against another sanding surface ? One action of trying to pull/push/drag the timber one direction whilst the rotating drum is trying to do the exact opposite action. Therefore creating a double sanding action and negating the effect of the infeed belt altogether, as it does require a bit of pushing pressure on the timber, to actually hand push the timber through the machine, against the rotating action of the sanding drum.

So one light weight piece of timber sitting on top of an infeed belt, without any downward pressure to hold it against that infeed belt, then the forward movement of the timber will stop and allow the drum sander to just either spit it straight back out, or, allow the front edge of the stationary bit of timber to be chewed away by the drum sander, whilst the infeed belt is chewing away at the the underside of the timber, until there is NO more grip/resistance from the belt underneath, and then just spit the timber straight back out anyway.

What does a normal blade thicknesses actually use, to push/drag the timber through the machine to "negate" the effect of the blades pushing/spitting the timber back out of the machine ?
Possibly some type of similar set-up, might be required for this design of yours as well.

Either more infeed rollers will be required to add downward pressure onto the timber work-piece, whilst it is on the infeed belt, to counteract the forces of the rotating sanding drum, or, forget that idea as others have done, because of the above problems.

The alternative to an infeed belt is to glue a piece of glass over the rise and fall table, this will virtually eliminate any resistance/drag on the timber work-piece, from the surface of the rise and fall table. By you using a push stick to control/maintain the infeed pressure, you can now have a definite feel for what is happening, when the timber is being mutilated by the sanding drum. 

It’s your choice on what you do, just a few more ideas for you to mull over.

regards  greenie   

[mattinker]

I'm really not in agreement with Greenie, your sanding belt drive sounds interesting, I can't see any way that if you set it up right it will double sand, the belt is there to grip and not slide over the wood. I avoid MDF at all costs, it's full of formalin based glue which is pretty poisonous! If you must use MDF, Andrews sugestion of PVA wood glue diluted with water is a very good way to go, dousing it once in water has not affect at all on the MDF.

I'm following this with great interest!

Regards, Matthew.

[Fergus OMore]

This MDF thing is interesting.  The largest man made forest is just north of me.
As a natural byproduct- it had been designed to supply pit props in WW2, there is an Austrian firm called Egger.
It set up a huge MDF factory to absorb the supply- the coal mines had closed- it is only 20 miles from me- and Simon.

No one can miss it at Hexham. It bursts into flames a bit lile a volcano.  One would have thought that - well, one would have thought.

Ahh well

Norman

[S. Heslop]

Yeah i'm not convinced about the double sand thing either. The conveyor belt is going to move fairly slow, there'll be more surface area with the part flat against the sandpaper belt, and you're only going to be making fairly light cuts with a thickness sander to begin with. But we'll see.

One thing I have thought about is, since I want to use this to primarily smooth and thickness 1.5mm veneers cut on a bandsaw (a bandsaw i've yet to make), which will probably end up being fairly warped at that thickness, I might still need some sort of infeed/ outfeed roller to hold them flat against the table as they pass through.

Also I do really like Kielder forest. I never knew that it supplied the Hexham chipboard factory, but I suppose that makes sense. I'd mostly assumed it was used as fuel.