MyFurniture #2: How to - Using an Old Coffee Table to Build a Bookshelf

Quick summary of what you may encounter if you keep reading:

What am I doing?

I am building a bookshelf following the same strategy as one of my previous Instructables: a Unique Shelving Unit for my Living Room

Why am I doing it? Few reasons:

My partner needs a bookshelf for her piano room and she wanted it to be in line with the previous build (it looks like she liked it). I promised I would create and document a simpler build mainly focused on the building process. Let’s try to master the process, right?

How am I doing this? My personal “rule of the thumb”:

It all begins with scrap/outdated/fairly used furniture hunting and tear down! I grab a piece of paper and ask my partner what would she like to see in her piano room. Then I come with a concept design that she must approve before I start making a CAD model. The Bill of Materials [BoM] comes in parallel and before I make any potential trip to my favourite DIY shop. From there, I will try to bring any used furniture to its “final shape”, like the shelving panels. This is the most time-consuming job as anything can happen. The rest of the structure will be made from the same material as my previous build and will be developed after I am done with the previous bullet point. Sometimes it can be done in parallel with other tasks. I will continue with the assembly, sanding and lacquering of the structure. I like flush interfaces and smooth finish to the touch. Finally, let the assembly begin!

Still interested? Watch the video and/or keep reading!

One more thing!

I got this question after I published my previous design: would/can you build one for me? Short answer is NO. I can’t. I don’t have the time nor the space. Perhaps in the future (work in progress)!

I have decided to set a symbolic price for what I am doing. You can get access to a fully adjustable CAD model and the drawings from my Etsy store. This will help me to continue building my future, while inspiring yours. If you think this is not fair, then I suggest you watch the video or continue reading because the design is rather easy and fully documented.

The BoM/BoP file and the CAD model for the stackable measurement gauges can be accessed from here.

Let’s go straight to the point:

• Number of shelves: 3 in total.The top one should be able to keep books in place on both sides.
• Overall dimensions: as my partner said, “not too big but not too small either”. Clear. She had a point. She wanted to have a nice aspect ratio between the height, width and depth. As a starting point, the height is conditioned by the number of shelves you want to have, the width by the location in the room and the depth, by the things you would like to store. No worries, you will fine-tune all of these dimensions during the CAD modelling. In my case, I started with ~108x90x30cm
• Space between shelves: enough space to be able to place one of her tallest books. I ended up rounding it up to ~36cm.
• Style: it must match the same style shown in the previous project. This means that I should use at least the same joinery.
• Wood type: not relevant, but there should not be a noticeable clash if 2 or more types of wood are used. Finishing: mat water-based lacquer and/or oil
• Cost: the most expensive component in the last build were the panels (~14.-/panel based on the previous project’s BoM file). Let’s try to play it smart this time.

I followed the same strategy as in my previous Instructable. As you may have seen in the video, I always start drawing in a piece of paper and measuring tape. I jump straight to the CAD modelling part once I have a clear picture of what I want to achieve.

Concept design

• I got the wood for the panels first (an old coffee table), so I had to keep in mind the thickness and how to fix the panels to the main structure, without showing any screws or supports! I decided to use 90deg brackets made of stainless steel (beautiful!) but the panels were so thick that I would end up showing part of these brackets. I decided to create slots. Perhaps this is not needed in your case.
• It always helps to draw (in scale) a few things in the shelve, so it works as a trigger for further questions (e.g. can I fit my favourite trophy?)
• I decided to add “handles” to the top shelve, so they would work as support for extra books if needed. Yes, there are other options, but my partner really liked this one.
• Finally, I was able to foresee how much wood would I need and how many different cuts to perform.

CAD modelling

• Fully customisable design that can be quickly updated via “Modify Parameters” in Fusion360. This means that you don’t have to build this very bookshelf. You can change all the parameters you need and within a few minutes you will get the updated model and drawings. Remember to modify these parameters in a logical way. As an example, too long of a bookshelf might need extra support in order to prevent sagging. However, changing a few cm/inches here and there will not make any difference at all.
• I did not model any fixing method between the shelve panels and the structure as I want to leave it open. You may have a better idea in mind.
• Unfortunately, I was not able to include the number of shelves in the design as a parametrised value and Fusion 360 would show errors. Perhaps I am not doing it correctly. Hence, this part might involve some work from your side.

I went furniture hunting and I found a beautiful coffee table for a very affordable price and I concluded that I could use most of it for my projects. By making this choice I managed to repurpose something considered as obsolete, also reducing the cost of the shelves by ~11.-/shelve. This means that I paid ~4.- per shelve, and in return I got a thicker panel made from a considerably better material, oak. Not bad, huh?! Now, lets walk you through the process of turning this coffee table into nice-looking shelves:

• First of all, you need to decide which faces will be facing upwards (where you would place the books and what other will see).
• Now, I would suggest you mark only one panel and make a cut. This way, the thickness of the blade is considered on each cut. Repeat this step as for the number of panels needed. A good guidance system is a must! I used a piece of wood that was straight enough.
• Grab a router machine and make slots if needed. Location and dimensions don’t need to be too accurate. I also decided to place on the 90deg brackets with a piece of the structure agains the table top. This way I could verify if the cutting depth was good enough.
• Time for applying paint stripper and to grow your inner patience. In my case I had to apply paint stripper 3 times per face per shelve but the final result was something to talk about. However if you happen to have a planer machine, then you are sorted. No need for this step anymore.
• After finally removing all layers, I decided to start the sanding process by using an eccentric orbital sanding machine and several sanding grits 60, 80, 120, 180.
• Now it is time for braking the edges, making them round or chamfered. I used a router table for more consistent results and I finally decided to create chamfers instead of round edges. I like the miter effect in the corners, very satisfying. Remember to take a wood bit and make some tests first so you can proceed once you are satisfied with the result.
• I also took my time to fix all the imperfections from previous steps, by using some fine wood dust mixed with wood glue.
• Next step, is to go back to sanding but this time using 120, 180 and 240grit. This will remove any imperfections from braking the edges due to the friction between the router table and the panels.
• Oiling time! I would suggest to change working environments, remove any wood dust remaining with a damped cloth and wait for the wood to dry. In my case, I applied a generous layer with a wide and thin brush and I let it soak for ~30min. I removed the excess with a kitchen towel because using a rug would leave a lot of particles. The total curing time of this oil was 9days!
• Feel free to wax the shelves if needed.

I would highly recommend to watch the video for a more-graphical insight.

I thought I found a “better” wood for the structure, so I tried to compare the calculated density of the one used in the previous project and this one, but they were literally the same. You can measure density by keeping the same volume (cut two wood pieces of the same dimensions) and measuring the weight. However, it looked better to me for some reason. This is the process I followed:

Machining

• Prework. Before making any measurement, cut a slice from both sides. These sides are usually ver rough and slightly tilted and you want straight faces in your design. Only after, you can measure and cut all the wood to size.
• Setting the pocket height in the table-saw for the short beams. This is a two-part assembly in which both wood beams must fit flush and with a tight fit in order to take structural advantage. Therefore, in the drawings I have defined the pocket height with a positive tolerance range for certain beams and with a negative tolerance range for the counterparts, avoiding any volume conflict. This is a trial and error process, so I suggest you take a leftover for these tests. Measure, adjust the height of the table-saw and perform a cut based on what is specified in the drawing. Now, measure the depth of the cut, adjust the blade and perform another cut until you are happy with the results. I really like this way of doing it because you already account for any deviations in the measurements and tooling you have.
• Setting the pocket height for the corner beams (4 long beams that will go on the sides). In the drawing I specified that I do not want volume conflicts, but in reality this depends on your craftsmanship. To make things easier, in my case I calculated the average pocket depth for the small beams and I added it to the required pocket depth of the corner beams. This way, I leave what I did as it is and I make up for the errors by adjusting (once again) the height of the blade in my table saw. This is a tiny details that actually makes a difference from the structural/fitting point of view, because you are actually making it work. Take your time and use a caliber (a wood leftover from the same batch) and a square ruler to check if the pocket depth is good enough (flush surfaces).
• Setting the width of the pockets. Mark the width of the pockets and always position your blade on the inside (material to be removed) of the line you just draw. This will account for the blade thickness already. However, blades are not perfectly straight and during the cutting process, they tend to bend/wobble due to the cutting speed and wood resistance. This must be taken into account if you are looking for a tight fit. My advice is to offset the blade even a bit more. If the pocket is too narrow, then you just need to remove a little bit more of material. If the pocket is too big, you can no longer use the beam. And if this happens in the last pocket you made from a long beam, then “swearing” comes as a bonus.

Sides assembly

• Time for cherry picking! These wood beams are not perfect. The advertised cross-section dimensions were 44x44mm, but the truth is that it tends to change from 43mm to 44mm. Therefore, take your time and choose the short beams that match best with the pockets of the corner beams. Warning! Rework might be needed but just slight material removal, as specified before.
• Labelling. Since I did cherry-picking it would be great if you label each wood beam so the chances of accidentally mixing them is reduced. My labelling strategy is not perfect, but it worked.
• Dry assembly. Two more things need to be checked before we drill holes for the structure: flatness and perpendicularity. Place one of the sides on top of a flat surface and assemble it as if you were about to drill holes for the screws. Once done, check perpendicularity and flatness of the structure. If you are happy, you can proceed by drilling the holes!
• Drilling holes. If you achieved a tight fit for all the wooden beams you don’t have much to worry about, but let’s perform a strategic drilling procedure by starting from the two extremes, inserting screws and checking if you have 90deg between the short beams and the corner beams. This way you ensure a perfect rectangle. Finish drilling towards the middle of the structure.
• Repeat the last two steps for the other side.

Connecting both sides of the structure

• As you can see in the video, I placed both sides of the structure on top of wood leftovers (4x), setting an offset with respect to the ground.
• Next, I placed the top and bottom beams on top of the same wood leftovers, resting at the same height as the structure sides. Bring both sides closer if needed.
• Now it is time to lock them in place (horizontally in the video) by letting the tool rest on the structure side too. Follow the same approach as before but don’t forget to clamp them since the goal is to drill holes at the right location. These are simple tools and positioning strategies that help a lot.

Finishing the structure: sanding and lacquering

Before continuing with the last steps, I think it is a good time to check if the panels fit on the designated pockets within the structure and also to have a sneak peek by placing everything in the room that it is supposed to stay.

• It is a good time to sand all small imperfections where the different cuts meet. Take some good time and perform a few passes with the sanding machine and what I believe is a 120grit sand paper since the wood surface was already rather smooth.
• Only after, I continued by disassembling the structure in order to sand each wood bit separately using the same grit as before. I did the corners manually using 180grit.
• Remember about labelling!
• Now it is time for lacquer. Use a thin brush or any other method. I applied 2 layers of water-baes lacquer, sanded every piece individually and manually with a 600grit and applied a third layer. I would recommend you perform the lacquering in a clean environment, not in a dusty workshop.

Ideally you would like to proceed with the final assembly in the room where the furniture it is supposed to stay, but this bookshelf is very manageable, so I decided to assemble it in a more spacious room:

• Place the structure upside down and place two beams across the structure at the same level as the panels are supposed to be. Don’t forget to clamp them…otherwise they would fall, right?
• In the CAD model, I defined a nominal gap of 2.5mm on each side of the panels, so I decided to design and 3D-print measuring gauges in order to set the distance. You can collapse them, adding the thickness of each other and therefore achieving multiple combinations. Models can be downloaded from here, for you, for free.
• However, after testing I no longer had 2.5mm gaps on each side due to overall manufacturing and assembly errors. I ended up having 1.5mm which is not too bad! In the end this gap was supposed to absorb any potential process errors, so it is of no issue if it ends up being 1 or 2 mm as long as it fits
• With the gauges in place, place the 90deg brackets and start drilling holes and placing screws! Repeat this step for each shelve.
• I also wanted to incorporate a nice tip from @NirL (thanks!), by adding adhesive pads that work as feet for the structure. It is a nice addition that avoids any scratching of the floor.

Now, it is time to enjoy the result…

In general, very happy with this good looking, sturdy and unique piece of furniture (I made it!). Now, let’s be a bit more critical about my work:

What I did well

• I believe I have successfully mastered the making of the pockets
• The repeatability of the cutting process is real good. I feel confident.
• The finishing process is very effective, with the outcome I expected.
• The drawings are pretty well defined, helping me during the process with achievable tolerances
• The overall aspect ratio is quite good and looks quite pleasing.
• You don’t see the screws from the front!
• I managed to reduce the cost quite a bit, reusing “used” furniture, although the building time increases substantially.

What I can improve

• The end result clashes quite a bit, although I know the wood of the structure will get tanned with time.
• I realised that the depth of the bookshelf is a bit too big, so some books tend to fall from the sides.
• The pocket depth tolerances in the drawings can be fine-tuned but it depends on the quality of your tools.
• Using pine for the structure is cost effective, light-weight, easy to machine and looks great after post-processing. However, it is rather soft and I would like to use a slightly “better” wood that I haven’t found yet.

As I always say, I am open for constructive feedback! See you in the next one!