Jump to content

A DIY thickness sander

Recommended Posts

In this topic I want to report on my effort in building a thickness sander. The last few months I spend some time finishing a second attempt that finally met with my requirements.


But before I show the final result, I want to tell the story of the first attempt and comment on its construction and what I learned from it.


In 2010 I decided that if I ever wanted to start a scratch build, I definitely needed a thickness sander. I got the chance to have a machine part available that seemed ideal to be converted into a sanding drum. It is an aluminium cast piece with bearings holding a spindle with rubber drum ( 35 mm). The drum was made thicker with some pipe insulation foam and a plastic tube over it. Then a sanding belt was glued on with double coated tape. The next picture shows that part with the sanding paper mounted.




Next I needed an adjustable height table. I constructed a double scissor type jack from some copper pipe and aluminium strip (see next picture).




An M6 thread, mounted in the middle of the copper pipes served as the height control. At the side of the table I added bronze bushings that glided over some vertical pillars mounted on a base plate. The next picture shows that whole construction.




The sanding drum is driven by my drill held in a horizontal fixture.


As soon as tried this setup it became clear that the whole table construction was a bad design because it got stuck at the pillars very often. A little tilting of the table was enough to block it.


So I needed to rework it. I removed the 4 pillars and added an extra M6 thread. These 2 M6 threads are positioned at the sides close to the scissor construction and are synchronized with a timing belt. That setup is shown in the next picture.




That setup did work quite well, but:

- it is (too) large

- needs mounting/unmounting of the drill

- makes a lot of noise

- lacks dust collection and removal

- renewing the sanding paper is impossible without dismantling the drum construction

- height control is intuitive (same turn of the knob will not result in the same change of the height)


That is it for today.

Next time I will show a different design that fits better to the needs of a model maker.


best regards,




Link to comment
Share on other sites

OK Bruce, you may take the front seat.


For wooden model ship building you often need small pieces of wood in different thicknesses.

That inspired me to develop a small drum thickness sander that would be easy to build and operate.


My requirements list is however long:

  • stiff drum to carry standard sandpaper material (no sleeves)
  • easy mounting of the sandpaper on the drum (no gap in the drum)
  • wide choice of grids
  • easy drum exchange
  • drum drive with variable speed
  • support table that can be moved up and down while staying parallel (no tilting)
  • precise height adjustment possible (resolution 0,05 mm)
  • minimum sanding capacity: 100 mm wide and 20 mm thickness
  • sanding down to 0 mm thickness should be possible
  • efficient dust extraction


When looking at this list I realize that I actually want to build my own Byrnes Thickness Sander but in a luxury edition.


From the start I already had an idea of how the whole construction could be made. But before starting the build, I wanted to check if I can make a wooden drum that has a strip of sandpaper wound around it and fixed at both ends in an easy manner. So here we go.


I started with looking for a suitable round piece of wood and ended up with a beech banister of 45 mm diameter and 1 m long. See next picture.




I then cut this beech rod into pieces of 140 mm length. That is about the maximum size I can fix between centers in my Taig lathe.






I turned it down to about 43 mm and then had to do some work at the end sides. The next image shows a 3D model of what I was trying to achieve.




The narrow cut was made with a pull saw (at an angle of about 10° with the end side). It is 5 mm deep where it meets the drum surface. The rounding to the drum surface required a small sharp chisel. The centre hole of 6 mm diameter was made on the lathe and the hole touching the narrow cut is 5 mm in diameter and made in a drill stand.




The groove in the middle was made on a Proxxon drill/mill stand.




The result is shown next and is ready to be finished in a sanding drum.




The sanding belt is 25 mm wide. I bought a box with 4 rolls of 6 m length in the grids 150, 240, 320 and 400.




Wrapping the sanding belt requires a little bit of practice to get the start at the correct angle so that the turns will butt nicely without gap.




Then a small piece of rubber tubing is placed in the 5mm hole and compressed with a screw so that the sanding belt is fixed.




And finally the ends of the sanding belt sticking out need a cut.







I was pleased with the result and decided to go on with this project of building my thickness sander.


best regards,



Link to comment
Share on other sites



Long ago, there were plans for a DIY thickness sander sold by the guild.  I built a machine using the plans.

It had a Maple drum.  I paid a wood turner to make it.  The blocks of Maple were glued with a 1/2" steel rod in the center.

The final size was 11" long with a circumference just at 9".   The sanding media available then was 9x11"  paper sheets.

I never mastered a technique to have a mechanical paper attachment, so it was a chemical adhesive. 


Things about it that I would do differently -

 Drum = Make it 12" long -  there is cloth backed media here that is 4" wide. I could have 4" 80 grit, 4" 220 grit, 4" 120 grit.  With mine, I had to trim 1" off of the 120 grit.  I had trouble finding a practical adhesive.  Contact cement holds well, but is difficult to undo.  I think that using what I use for my 5" disc sander would do - rubber cement.

Motor =  use a 1/2 hp instead of 1/3 hp motor.  Still make it 1700 rpm for the drum.  Faster burns the wood.  I see no advantage in it being slower

House =  Never enclose the motor.  I did and having the motor in an oven is a poor design.  Have as much air circulation as possible.

The Table =  here I did OK -- 3/4"  AA hardwood plywood.  All 4 edges have Aluminum right angle attached with more than a few, longer length screws - recessed heads.  Longer on the front and on the back is a good thing.


Now, the most important component -  These machines can generate impressive amounts of wood flour.  The size is small enough to float in the air and is readily inhaled.


I made a five sided box  to sit over the drum.  It is made by glueing three layers of Amazon box cardboard together using libral amounts of yellow PVA for each side. The inside corners are strengthened with a 1/4"x1/4" Pine stick  - 8 sticks .  The top has an female attachment for a 2 1/2" shop vac hose.  The inside of the top has Pine stick glued to accept the screws holding on the vac attachment. 

The outside is covered with high quality duct tape.  The cheap stuff has poor adhesion.

If you think that you do not need a cyclone in line trap between your machines and the vac motor filter,  while that may fly for most machines, this one is in its own class.

The volume of saw dust is not to be believed.  Without the trap, much time will be spent clearing the vac filter.  The 5 gal cyclone catch chamber will need checking more often than is imagined.


Because of these machines, having a N-95 mask was no problem for me.  A cousin who is a house carpenter,  had to have surgery to remover a sawdust bezoar from his sinuses.  He avoided

using a mask because it fogged his glasses.  Real world saws have TPI  that produce comparatively large sized dust.   So even the relatively coarse sawdust that does not get to alveoli can cause a problem.  220 grit can get down deep.  ---finer than 220 grit is not a good idea anyway for stock wood surfacing.  Too fine a surface and PVA has no tooth to bond to.

Link to comment
Share on other sites

Jaager, this is going to be a 'mini' sander. But I will take care with the wood flour and provide an adequate dust extraction solution.


In this part I will show how the basic construction is made. The idea is to have the height table mounted on 4 spindles. Turning the spindles will move the table up or down. This can only work if all 4 spindles have synchronized movement and depend on just 1 height control.


As spindles I chose to use a lead screw type known as TR8x1. This has trapezoidal thread with 1 mm pitch. I ordered 4 screw/nut combinations of 200 mm in China (12.59 euro). The next picture shows the spindles and some dimensional data.





To have these 4 rotate simultaneously, I interconnect them with gears. So I make a kind of gearbox that will be sandwiched between 2 bottom plates of 12 mm thick MDF. The next picture shows a model of the gear configuration.




All gears have 1.25 as module, the large one has 48 teeth and the small one has 32. So measured from centre to centre we have 0.5 * ((48 * 1.25) + (32 * 1.25)) = 50 mm. The next image shows all dimensions.




The spindles stand 141 mm apart and that is enough width for my height table to move between them.


The next series of pictures shows the build-up of the gearbox. To start we need 2 plates of MDF that are of exact same dimension (170 x 220 mm), so I saw them together on my little table saw.




Then we need a lot of holes at very precise positions. I made a simple jig with 3 holes of 3 mm in line and at exactly 49.9 mm between centres. Why not the nominal distance of 50 mm? Well, I wanted to eliminate the play between the gears. It was not a total success because when you reduce the play, other artefacts like eccentricity will come into play more dominantly.




The 2 MDF plates are fixed to one another with 2 screws in opposite corners. The the first hole was made in the middle. From then on I used a 3mm steel pin in the jig to position the plates at the correct centre distance for a next hole.




I made 14 holes in total. Then the 2 plates are separated and 10 holes needed to be widened to a suitable dimension to fit a flanged bearing (12 mm) or a steel rod (8 mm). This was done by first mounting a 3 mm drill bit in the drill stand to bring the MDF plate into position, clamp it. The exchange the drill bit to the wanted diameter and drill the hole to exact depth or trough as needed. I must admit that the depth display of my Bosch PBD 40 proved to be very handy.

Next you see both plates with already some flanged bearings mounted.




Then the gears come into play. First the centre gear that will also be connected to the height control later.




Then the four smaller satellite gears are placed. They use no bearings but I inserted a nylon washer between the MDF and the plastic (POM) of the gear to protect the MDF from wear.




Then the 4 large gears that will drive the spindles will follow. The spindles are made to a length of exactly 110 mm on the lathe and then the ends are reduced to 6 mm diameter for 4 mm so they will fit into the flanged bearings. You can also see an aluminum ring to adapt the spindle diameter to the 10 mm hole in the gear. I glued in 2 pins that will fit in a groove of the gear to drive the spindle.




That’s the gearbox. It is covered by the second MDF plate. You can also see 4x a 3 mm pin with a plastic spacer that will keep the 2 plates parallel and at a distance to give the rotation parts a little play.


As it is shown here it all looks simple. Believe me, it required a lot of tuning on all parts to get this whole mechanic move smoothly.


Enough for today, next time the side plates and the height table will be added.


Best regards,




Link to comment
Share on other sites



About your thickness adjuster -  The ideal situation is one degree of freedom.  You may be challenging precision too much with 4 degrees of freedom.  Theory, and shoulda,  and oughta are sometimes at loggerheads with how it really is.  Keeping the table the same distance from the roller along the entire width/length  can be a challenge with any design.  I advise verifying at the extremes and in the middle.  


Link to comment
Share on other sites

Hi Jaager, the 4 degrees of freedom actually reduce into just 1 degree of freedom by the gearbox construction. That will be shown more clearly later.


Let’s go on because it still doesn’t look like its going to be a thickness sander.

Adding the side plates (also in 12 mm MDF) will help to imagine what the final tool will look like. Here is a model that shows both side plates.




The plates are asymmetric because I wanted to integrate the motor in the left plate. The motor is a brushed DC motor of the ‘pancake’ type. It is shown next.




While small it is quite powerful when run at full speed. To mount it a large hole was made and then a recess to accept the flange and bring the motor pulley to the plane where the larger drum pulley will come. The hole and the recess are made with an adjustable circle hole cutter and then the recess was cleaned up with a chisel clamped in some support to get the correct depth.




Then it was time for a test drive.




Now we switch to the height table.

That is also a 12 mm MDF plate that will sit on the 4 spindle nuts. Because The part of the nut that fits into holes of the MDF plate has 10.2 mm diameter and since I have only a 10 mm drill, the nut was turned down to 10 mm on the lathe.

A piece of scrap MDF was used to judge the fit: firm but possible to rotate as is required for levelling the height table.




To get more grip on these nuts for adjusting, I made some profile on the outer edge.




That was done on the lathe with the main spindle turned into a simple self-made divider and my proxxon drill with 3 mm mill mounted in the tool holder. The parts to make the setup are shown next.




The tread is M5, the gray parts are cut from hammer nail plugs, the gear has 20 teeth and the hex key is an ‘Ikea’ key ground down at the end to fit in the teeth and in the middle to fit into the groove on top of the lathe main spindle block.

All pieces mounted and in use is shown in the next picture.



The height table seen from the bottom side. You can see that I added a small clamp to each nut to prevent any movement in Z-direction.




And an image from the side where you can see small metal plates worked in that have a fixing screw in the middle for securing the adjustments made when levelling the table.




As a last addition we have to thicken the height table so that we can use it for a sanding thickness range that goes to 0 mm. A 3mm spacer plate (hardboard) was glued on the height base plate and then an extra 12 mm MDF plate was added and held in position with 2 metal pins of 3 mm diameter. This extra plate can always be removed if you want to sand very thick pieces.




Now we can move on with the construction. The gearbox with its top plate mounted is attached between the 2 side plates.




You can also see a miniature bevel gearbox (90°, 1:1), the black part, that sits on top of the gearbox and brings the height control to the side of the tool.

Details about that construction come in a next part.


Best regards,



Link to comment
Share on other sites

Well done.  

Here's a URL link to one that I built for thicknessing 8" x 20" wood stock for guitar tops and sides down to 2.0 to 3.0 mm thick.  The adjustment mechanism isn't quite so complex.  One possible advantage of my design is that the stock can be tapered across its width by adjusting the alignment of the feed ramp.


If you can't access the article at the Musical Instrument Makers Forum, please let me know and I'll post a description.




And here is another similar idea that I found a year or two ago,  published in Popular Mechanics, July 1958, p 190.  A dust collection apparatus is really essential and would not be difficult to adapt to this design.  The mechanical parts are available at McMaster.com


https://books.google.com/books?id=KN8DAAAAMBAJ&lpg=PA191&dq=drum thickness sander intitle%3APopular intitle%3AMechanics&pg=PA191#v=onepage&q=drum thickness sander intitle%3APopular intitle%3AMechanics&f=true

Link to comment
Share on other sites

Hello Bob,


Yes I can access that article. It's of a more classic design but well adapted to the lathe.

Certainly enough power to sand wider pieces of material.


The popular mechanics article, I also discovered that some time ago.

It is still inspiring for its straightforward and simple construction.




Link to comment
Share on other sites

There is always the risk to over-engineer things, which is alright, if you have large enough precision machinery, to produce all the parts with critical dimensions to the right tolerances. Otherwise it is good design practice to try to limit the number of critical dimensions, even, if the design in the end looks less sophisticated. Guess, how I know this ...


Having a stage moving up and down perfectly square to the sanding drum is a manufacturing challenge. With four pillars there are two many degrees of freedom that need to be controlled, particularly, because they are not completely free, but connected by the gearing mechanism.


The simplest DIY designs for a drum sander involve a hinged plate, angle of which is adjusted by a simple elevating screw. The angle under the sanding drum determines the clearance and hence the thickness of the material that comes out at the other end. The only critical dimension is that the axis of rotation of the hinged plate is absolutely parallel to the axis of the sanding drum. This can be controlled by having common brackets for both, one on each side, that are drilled as a parcel, tightly clamped together.

Link to comment
Share on other sites

Here is my DIY thickness sander loosely built from the NRG plans that Jaager referred to.  Total out of pocket cost was under $30 as all that I had to buy was a pair of pillow block ball bearings and a length of  1/2 in cold rolled steel shafting.  I had a spare motor.  The drum is maple laid up in laminations epoxied together around the shaft.  A couple of cross pins thru the shaft lock the drum in place.


I do not own a wood lathe, so I mounted the laminated block in the thickness sander frame and pillow blocks and set up an improvised tool rest.  Turned by the motor this improvised lathe worked fine.


The disc sander on the end of the protruding shaft was a plus.  I use it all the time.  


I use belts sold for belt sanders, cut and wrapped diagonally around the drum.  Secured with contact cement this works well.


Sanding thickness is controlled by an inclined plane pivoting on a horizontal shaft with the free end adjusted by a threaded rod.






Link to comment
Share on other sites


as a tooling design engineer I agree with you.   Simpler is always better. Less moving parts with fewer critical dimensions means less that can go wrong.   I often ask myself after I complete a design “Why did I make this so complicated?”  

Kris, with that being said, I am enjoying following this DIY sander. It is way better than the one I came up with. Oh wait I did come up with one. 😜. I like watching ideas come to life. Thanks for sharing. It is how we all learn. 


Link to comment
Share on other sites

I should add my high tech solution for insuring that the drum is parallel with the hinged plate.  I have a piece of sandpaper glued to a flat piece of plywood.  With the sandpaper on the drum removed the sandpaper faced plywood is clamped to the angle plate.  With the machine turned on the angled plate is raised until the sandpaper contacts the drum and removes ant irregularities.


I should also add that the 1 /2 hp motor surplus from upgrading my bandsaw gives me plenty of power




Link to comment
Share on other sites

I built one about 25 years ago. I used 4 rods with a chain sprocket welded on each one. A chain  was moving the 4 posts at the same time for the table elevation.

It was replaced  15 years ago with one who has a table feed under the roll. This way, you do not have to push the wood strip.

Link to comment
Share on other sites

On we go with finishing the height control in this part.

The bevel gearbox I used was bought at Conrad and they have the next data sheet on their website.




As you can see in the picture it has some annoying bolts and nuts that stick out from its body. I remedied that by drilling counterbored holes at both sides of its body. The screw heads just fit in and the nuts are melted in with a soldering iron. After mounting all bolts again, the sticking out parts are ground off. Then it was easy to mount it on top of the gearbox with 2 pieces of MDF (see last picture in the previous part).


The axis is now horizontal but has to be lengthened to bring it to the outside of the tool. The next picture shows all the parts to do that.




The steel rod is 8 mm in diameter and will be supported by 2 bearings sitting in a flanged bushing, shown next (different axis in this image).




I have several of these and I will also use them to hold and drive the sanding drum. I only need a 26 mm diameter hole in the MDF side plate. Next it is shown already mounted.




Then mounting all the pieces can start. First the connection with the outgoing axis of the bevel gearbox is made. Then an axle lock is placed (not yet present in the next picture).




Then we continue from the outside again with an axle lock, a nylon washer, a rubber O-ring, the self-made dial and a 2nd O-ring.




And we finish with a nylon washer and a 60 mm diameter handwheel.








Then it is time to add the height table and verify how turning the whole construction feels.




In the above picture the bearings on top of the spindles have yet to be mounted. Once that was done the table can be moved up and down with the handwheel. To be honest, you cannot turn it with just 1 finger, but with some firmer grip it is OK. I use the handle only for large movements of the table, smaller advancements are done with grip on the wheel.


With the table in its upper position you can reach the nuts that support the table through the some extra 26 mm diameter holes in the side plates.




Now the table can be levelled. With a straight metal piece as reference (a square in my case) and some light from behind, that was easily done. Tightening the locking bolts finishes the height table construction.




As some distraction I can also show how the dial was made (I hope ’dial’ this is the correct word to describe the ring with the scale on it).

First some ½ hour sawing the aluminium base material.




Then the same trick with the poor-man’s divider, this time with a gear having 45 teeth, and the proxxon drill to make grooves for better grip.




Then turning the ring around in the lathe and making the scale (again using the gear with 20 teeth). Some improvisation was needed to mount the chisel. A simple tool holder with very sharp chisel mounted on another one that has height adjustment and a 2.5 mm diameter drill between the 2 to get some slant.




The scale is printed (in reverse) on a transparency sheet and wrapped around the ring. It is fixed to the ring with some tape over it.


That’s it for today, see you next time to finish with the sanding drum mounting and adding the dust extraction.


Best regards



Link to comment
Share on other sites

Now let’s go for the last mile of this thickness sander.


The drum or roller is massive beech and has only 2 centred holes of 6 mm at its ends. So it must be supported from both sides by a 6 mm steel axle. I mount these 2 axles with a flanged bushing in the MDF side plates. The 2 holes (26 mm diameter) for the bushings are drilled in one go on the drill stand with the side plates aligned. Each bushing holds 2 bearings so that there is almost no play on the 2 axles.

The next picture shows the side where the drive of the drum is done, shown from out- and inside.




The other side has just a free running axles, shown next from in- and outside.




I changed the flange such that I can easily take it out without needing tools. It comes in handy when you want to change the sanding drum.




Now that the drums can be driven, I took some time to sand the wooden drums so that they are parallel to the height table and have almost no eccentricity. A thin flat metal plate with some sanding paper mounted with double coated tape was used. Patience and regular checks of the eccentricity with the motor turned at low speed resulted in a nice set of sanding drums.






A dust extraction is the only thing yet to add. It is made with 3 pieces of MDF that will cover the top of the sanding drum. The sides are covered with segments cut from a plexiglass pipe (50 mm outer diameter, 3 mm thick).




You can see that I use the 3 MDF pieces to transform the round hole into a slit that covers almost the total roller width. The plexiglass tube segments got small tabs crewed on to have some grip for placing or removing. The bottom of the large MDF piece has 2 grooves (2 mm deep) where it will meet the plexi segments. I added 4 small pieces of beech that are glued in these grooves and then got a round-up with a chisel. They should hold the plexi segments in place and provide some ‘click’ when mounting.

Here the top part is mounted.




And next is shown how the plexi segments are added. They just sit on tiny nylon tube parts sticking out about 3 mm from the side plates. Once supported the plexi segment is rotated up until it click’s against the MDF part.




I almost forget to mention the drive of the DC motor. Its a 15V 10A power supply with a PMW (Pulse-Width Modulation) circuit added that feeds the motor.




Now the thickness sander tool is ready for use. It is shown on my workbench with the dust extraction attached to the vacuum system.




And to complete the circle I add a size comparison of old and new sander.




This ends my story on the new DIY thickness sander. I hope that you enjoyed it and maybe it can inspire you for some more tooling projects.


Best regards,









Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

  • Create New...