The learnings of a furniture maker

Having jumped head first into my passion that is furniture making I want to accumulate and share some of the learnings I make along the way. Going from Level 0 to Level 100, hopefully, in skills and craftsmanship will take time, dedication and a methodical approach. Collecting my learnings is first and foremost for myself to first externalise and in turn internalise and visualise my development to make sure I stay on the right path forward.

Having been a game developer for most of my life the transition to working in the physical world, apart from being extremely rewarding, surprisingly has many parallels to both the creative and technical aspects of creating games. Having a conceptual idea, visualising it in sketches and finding a technical solution to realising it into a final product. At the same time the art of working with wood is truly unique and has an endless depth of opportunities to become better; more accurate, more efficient and more creative. I want to capture what feels natural and appropriate for the level I am at that point in time, no matter how basic it may seem. So that I can hopefully progress and learn quicker. 

The learnings will vary in detail and are in most cases blatantly obvious, the page is under constant editing and will never be complete, and at any point in time the learning may be wrong or I will re-learn on top of it. I am using the word "you" because I need to keep reminding myself these things, not because I am stating a truth that you should take as valid.

There is so much to learn and so much fun to be had!

sawing.jpg

1. Measure twice - cut once

Every list of learnings has to start with this old classic. Whilst the advice itself is easy to remember and simple, actually following through on it in practice can be harder. Especially under time pressure. The main thing for myself to learn here is to both mentally and visually check through the intended outcome of the cut before performing it. Especially when angled cuts are involved it is not enough to just measure and validate the right angle mentally, it is crucial to make sure the angle is going to be cut on the right side in the right direction by visually placing or comparing the angle to a reference or drawing.

Too much wood and time is wasted by hasting past the most basic rule of all. I must be dilligent!


2. Sneak in on your cuts

Wood cannot be made longer. Also a basic rule which has a nice approach to make sure you don't end up with a sad face. Sneaking up on a cut, or any operation, means you intentionally make the cut slightly longer and adjust it for the perfect snug fit. Typically when you are making a joint or placing a piece of wood between/within other pieces. It is always better to cut too little than too much.


3. Don't rely on sanding to fix mistakes

Often when making a joint or gluing together pieces I find myself thinking "Oh I can fix that later when sanding". This is not a good idea. It means at least one of two things, typically both; 1. It will take longer time than getting it right the first time around (i.e. measuring and cutting more accurately, aligning and clamping better when gluing and so on) 2. The end result will be less good, when sanding edges or trying to remove blemishes it is very hard to get a straight and good looking edge.

Take your time and get it right the first time around.


4. Use the right material at the right quality

Being able to make the right choices of when and where to use the wide variety of woods and composites available is something I feel will be a trial and error experience over time and I don't aim to write a full and comprehensive list of what materials work where. But more of a mental note of some of the thinking I have done for myself.

More often than not choosing materials is a balance between cost and aesthetic. I love working with hardwood like Oak or Wenge, but hardwoods are not suitable for all projects or budgets. At a basic level I find it is about using high quality material where a good finish is required, and other suitable materials where either the weight, budget or physical appearance require it. Compromising on material quality means a compromise on product quality. 

My biggest learning so far is that Plywood is very exposed to variable quality and the final result is more impacted by the quality of the wood. Finding good Plywood in the north of Norway is not easy so ordering from specialists around Norway and Scandinavia is a must for ensuring a good finish. Plywood typically has different qualities on each side and anything less than BB should not be used for making furniture. I learnt the hard way from using cheap plywood for a piece that ended up warping and splitting only after a short period of time.

ps. For curved project is is ideal to use Plywood that is specially made for being curved either vertically or horizontally.


5. Lower your pulse when work requires accuracy

Especially for me coming from a high pace gaming culture it is crucial to take a second to breathe and lower my pulse before doing an accurate operation. A smooth motion using the whole body in a single motion is the best way to get a good result, be it a cut, routing or anything requiring accuracy. Visualising the action, doing a practice run, slowing down is for me a great way to improve the quality of my work. 


6. Plan as much as possible ahead of time

I love to just start working straight away, the same as I always did with programming. By nature I feel that spending a lot of time planning reduces the amount of time available for actually doing the work. However, this as with programming, is not true.

The more detailed and thought through the building plans can be the better. Best case is having a full cut list of all required parts and pieces before starting, I am not diligent enough to have this at each time. Something I need to improve on, especially for the more complex projects.

A good learning for me is that for all projects, even the smallest, I create a 3D model and a CAD-drawing with measurements so I have time to  both visualise and prepare for a more efficient workflow.

Not only are measurements good to have ahead of time, also a clear plan for what sequence to do things and why that sequence is the right one.


7. Buy the best tools possible within your budget. The first time around.

I, as many others, started out by buying at the cheaper end of the scale in order to save money. Without exception I have upgraded over time, often multiple times and each time buying a better and more expensive option. Over time this becomes more expensive than buying the best tool in the first place.

Buying the best you can get within a category is a good plan, the quality and efficiency of your work is very closely linked to the quality of the tool used. The longevity and predictability of better tools is also a good reason to invest in proper tools for your jobs.

My thinking is that I should buy just so it hurts in the budget, but not go completely nuts. 


8. Invest time in organising your workshop

The most fun and lucrative work is naturally creating furniture. But the end quality and efficiency is directly proportional to how well organised and layed out your workshop is. Spending time creating storage, workbenches and time saving systems for your workshop is time very well spent.

I tend to end up with a mess pretty quickly, especially when working towards tight deadlines. Both the quality and patience decrease as the mess in the workshop increases. 

My experiences is that you do need to get a good feel for how you work and what is needed in your particular workshop for your particular work process before taking too drastic measures.


9. You will never find your folding rule or your pencil. 

No matter how tidy your workshop is, you will 1 time out of 3 end up not finding your folding rule or your pencil. Even if you have a tool belt. Best solution I have found is to have many different storage locations around the workshop with a multitude of pencils and measuring tools.


10. Batch jobs together in a series

When making multiple pieces of the same design I tend to want to complete every single piece as much as possible before moving on to the next one. I just want to see something complete. Of course, this is not the most efficient or accurate way of working.

Context switching is expensive, as it is in computers, so the best way to get good results and optimise time is to batch together all operations of the same type and/or with the same tool. Resisting the temptation of taking too many steps on a single piece.

Maybe one or two exceptions? Glueing and finishing have long wait times and many times I have found it more efficient to get a glue-up done before continuing on a batch on the table saw or similar


11. Function vs. Form

I typically tend towards form over function in most case in life. But there is a time for everything. I swing back and forth, but at this point in time i see function as the defining factor and that form must follow function. Especially when it comes to pieces that have an everyday use, and especially chairs.

But then again, if a piece of furniture is art. Should things be a little crazy some times?

Every piece has its use, so I will go probably go back and forth a few times on this.


12. Don't move big power tools whilst they are on

Obvious, yes. I had to learn it the hard way.


13. Use long continous strokes when finishing

The only way to apply any finish is to use long continuous strokes with the grain. Don't try to fill in patches or start from the middle, it will not look good. Long, controlled and measured strokes always work best.


14. When possible use a template, guide or stop

Trying to freehand an operation typically means you are lazy. When at all possible use a guide, template or some kind of stop to ensure you get an accurate operation exactly where and how you want it. 

Trying to save time by not doing it? Don't, relax and take your time!


15. Clamp your work piece to the bench

Following on from 14. Make sure to clamp your work whenever you are going to do an operation. Especially small pieces. Again, the time you are trying to save or impatience you are trying to please by not clamping the piece will not pay off. You will most probably waste wood and end up having to redo a lot of work.

This also means you have less to think about and free up your hands to focus on the most important action you are about to do.


16. Find a natural stance

Feel uncomfortable? Reaching too far? Holding a two handed tool with one hand? Balancing a piece on something random on your bench? Just resist the temptation, it will not turn out good enough. Take your time and move things around until you, and your work piece, have a natural and relaxed stance.

Zenish, but it has to be done!


17. Ensure consistency across platforms when CNC'ing

Often when CNC'ing I find that i need to 1. End up with a certain shape on the final piece. 2. Have the shape in Illustrator 3. Move the shape to Fusion 360 (or other application) 4. Export shape to g-code 5. Place work piece on CNC and 5. Send g-code to CNC.

So many times have I forgotten to ensure that orientation and measurements are consistent between all steps. Never start the CNC before you are 100% sure that all things are consistent and correct. And yes, I learned this the hard way a few times, I seem to have to learn this many times... 


18. Sawdust is your enemy

Get a good dust extraction system quickly. Have it on all tools and don't be lazy about it.


19. Have sharp tools, always

Spending that extra amount of time or money to ensure all your tools are clean and sharp is worth it. Nothing is worse than a blunt blade or chisel. It will show in the final work, and it will make your work less enjoyable. 


20. Sketch when needed, 3D when needed

I often find myself deciding wether i should sketch on paper or work in 3D. Some times working directly in 3D has worked well for me, but I need to remind myself to take a step out into the real world and sketch on paper when i feel stuck. Sketching on paper is much more free and allows you to explore variants and directions quicker than in 3D. At the same time working in 3D directly can end up giving more minimal and practical approaches, at least for me.

Being aware of the pros and cons of each approach, and level of skill in each tool atleast is important to remember when working on designs.


21. Track your time in categories

I mainly work on custom orders for customers. This means being able to, relatively, accurately predict how much time will be spent. Typically I underestimate time and end up spending more time than expected, mostly because I am still inexperienced. For me it is key to track time across projects and categories of work so that I can get a better understanding of what takes up time and where I need to adjust my estimates.

Typically planing, sanding and finishing is something I underestimate the amount of time needed for. I need to be better at predicting that time, which means I need to keep track of what I spend my time on.


22. Ask for help

Or just check youtube. There is always someone with a good idea or a neat solution for the problem you are facing. Don't try to reinvent the wheel, look and ask around and most probably you will find an answer from some clever person out there.


23. You can never clamp well enough when glueing

Having good and consistent pressure when glueing is key to a strong joint. Don't be scared to adjust or move clamps in order to get strong pressure across the whole glueing surface.

ps. You always need one more clamp than you own...

pps. Don't forget you can make your own clamps and placeholder clamps or use some waste wood to fix hard-to-clamp glue-ups.


24. Turn the speed of your drill down when screwing

Screwing at too high speeds will force your screw too far and too fast into the wood. It will not look good, possibly break the screw head or make the connection with the wood weaker.


Turn down the RPM and limit the max torque of the drill so that you get a nice, clean and controlled action into the wood.

25. When in doubt use 15 degrees

This is a very dubious and subjective learning that may only live temporarily. But I love 15 degree angles. So when in doubt for what angle to use for a leg or an detail, I choose a 15 degrees over any other.


26. Working under time pressure without making compromises

This learning could probably have it's own book or website and it has direct parallels to any other types of work under pressure. When you strive for quality in what you do the pressure of time tends to act as a barrier or force to make compromises or start rushing your work.

This seldom ends up being the right course of action. Typically I will start rushing my work and try to do many things at the same time. The best way, for me at least, is to become even more organised and structured and even more calm and patient when working under pressure. It may feel like it will take more time, but in the end it will help you get "in the zone" and remove any unneeded frustration or feeling of making compromises on quality when they are not needed.

Time pressure should not be removed, as it is a powerful source of energy, but how to manage time pressure should be practiced over time. Without compromising on quality.


27. Don't force or angle the orbital sander

It is very tempting to apply extra force or hold the orbital sander at an angle to quickly remove a stain or rough section. This seldomly works out well. Either you get a bump in your work piece, or the sander will leave scratch marks. The best way is to apply a firm, but not forceful, amount of pressure and think big and complex thoughts in your mind as the sander does it job.


28. Draw clean single-stroke lines when marking a measurement

The whole idea of doing a measurement and subsequently marking it with a pencil is so you can perform an accurate operation in just the right spot. I typically wiggle my pencil back and forth a couple of times with a blunt pencil to quickly mark the measurement. Leaving a squiggly line approximately where I intended to. This does not make sense.

Marking with a single stroke of a sharp pencil means there is no ambiguity about the mark you just made. I know this sounds silly, but the amount of times I have to look more than once at my measurements just to "read" them is quite embarrassing. 

ps. It is a good idea to also find a standard approach. Are you going to cut "on" the line or leave the line on the wood? Is this where the inside or the outside of the tool should go? Try to be consistent when possible to reduce error.


29. Hook your dust extraction up to a wireless switch

The amount of times you will turn on an off your dust extraction during a day is pretty insane. Best way to save time is to hook your dust extraction power up to a remote switch that you carry around so you can instantly turn it on and off without having to wander across your workshop.


30. Remember to consider how the customer will use their furniture

I get quickly carried away by how a piece of furniture should look, and forget to consider all the different aspects of how it will be used. (See 11.) I've mainly learned this through making things too heavy or hard to move around. 

As a reminder for myself I need to ask more questions and think more through how the customer actually will be using the furniture and what is important for them apart from how it looks. Are there some important constraints? How can the piece of furniture improve or complement what it is they want to achieve?


31. Kickback is real

Kickback is when a tool ends up shooting out the wood because it gets trapped or tight against the blade or moving part. Typically the most dangerous thing when working with a table saw, apart from cutting yourself on the blade. This happens most often when cutting thin slices of wood or the riving knife is removed (some saws have a riving knife that is taller than the blade itself and when making vertical cuts partially into the woods height it has to be taken out) causing the wood to squeeze in behind the blade.

First thing to do is always stand a little to the side of the blade on the saw rather than directly behind it. And when cutting thin slices try to have a waste board with zero clearance so the thin slice of wood does not get trapped between the blade and the opening around the blad.  

Having seen and heard wood fly around the workshop this is definitely not just a myth, but something that should be actively managed as a risk.

ps.

Kickback seems to have a tendency to happen on the last cut in a series as you rush through to pick up the pieces you just finished... Rushing is never a good thing. I think I may have mentioned that a few times already!


32. Cutting tapered legs

Tapered legs are very nice! Especially when going for a slim and minimal look. Cutting the legs will require a jig of some kind. I like this one (although the stop at the back is a little shallow).

For legs with tapers on to opposite sides, having tried a few different approaches I have found that the easiest is to cut the first side with the wanted angle, say 15 degrees, then flip the wood over and cut the wood at double they angle, so 30 degrees.

Making a tapered cut is a bit more fiddly, partly because of the danger in learning 31., because you need to a) hold the work piece to the jig (or use a clamp, if you remember to be patient like in learning 15. b) move the jig along the fence of the saw and 3. make sure the fence is parallel to the fence at all times. So a bit more concentration and care is needed for tapered cuts.


33. Have an air compressor to blow away dust

As in learning 18. sawdust is still your enemy. No matter how much dust extraction you have there will always be a build-up of dust around the workshop. To get high quality finishes and in order to stay organised it is great to have an air compressor to quickly and efficiently blow away even the smallest particles from the tools and surfaces you are working on.

As a bonus it is actually quite fun to see that evil sawdust fly away!


34. Use tabs when CNC-ing contours through the work piece

When CNC-ing contours that go completely through the wood and isolate a part of the wood without support to the stock piece it will typically fly lose and get crunched in the bit or break. The easiest way, other than fixing the wood with double-sided tape or vacuum to the waste board, is to have your 3D-program insert small tab that the CNC will leave at certain intervals around the contour so that the "island piece" is still connected to the stock piece with a minimal tab you can sand away later.


35. There is no "undo"

Being a computer guy I have trained my brain to work fast and quick with many iterations. This often means basing progress on making many quick changes, testing and undo-ing if not correct. Goes without saying that this does not work in the world of wood... 

This means I need to rewire my brain to take more precautions and either mentally or through working with prototypes and waste pieces ensure that I don't need to to "real world undos".

On a website if your margin is off by 5 pixels, you just change a variable and it is all good. If you are off by 5cm's when CNC'ing a logo in a sign (yes, i did that...) it has bigger consequences.


36. Spray painting on vertical pieces needs a careful approach

Dripping is never a nice thing with either paint or spray paint. When spraying areas that are vertical I have found that the coat needs to be extra thin in order to reduce that amount of dripping caused by gravity. 

Generally you should spray less than you think and with more coats in order to get a good result.


37. Don't trust self-tapping scews

Having assembled my new CNC there were several pieces that relied on self-tapping screws to form the thread in an aluminium piece. I managed to break the head of quite a few screws in the process, no matter how carefully I went about it. 

Never trust self-tapping screws, always tap the thread with a dedicated tool


38. Dry-fit and pre-build larger pieces before mounting on site

When building large pieces that are to be mounted and fit at the customers home or business it is important to complete a full build in the workshop before doing the final installation.

Small adjustments are fine, but discovering the need for major fixes or changes on site will take much more time and not be executed as well as in the comfort of the workspace. 

The time spent not pre-building before going to the customer is not worth it. 


39. Factor in additional time when doing on-site installs

When doing on-site installs it is important to factor in extra time as the work environment and work flow will not be optimal. In addition to the inefficient workflow it is, nearly, unavoidable that you need a couple of extra trips to the workshop either for tools or detail work.


40. Leave additional space for uneven walls and angles in tight fitting furniture

Quite often you will want to make some furniture that fits exactly within a space in someones home. This is easier when walls and floors are straight and level. But they never are, it is better to leave a centimeter or two and factor in a decorative finish that you mount after in order to get that edge-to-edge finish.

At least make accurate measurements and check level across the whole space and surrounding area where the furniture will be, so that there are no surprises.


41. Be careful when sanding plywood

The veneer layers on plywood are normally only a couple of millimeters, and typically alternate in colour. If you sand too much you will sand through the outer layer of veneer and reveal the darker layer behind it. This is very hard to fix (I tried with colouring pencils!), so I just have to be very careful when sanding on plywood in the future.


42. Finding the right hardware is not easy

Making anything hinged or sliding has its own challenges. Finding just the right hardware to use either it is a hinge or a slide can be quite hard. Having tried and failed a few times the best options I have found so far is Häfele or Blum. They have high quality products that allow for a nice concealed look when needed.

43. Wood moves more than you think!

Working with wood is a wonderful thing! Understanding how wood behaves and compensating for it is wonderfully complex. Having learned the hard way that wood will move, especially in our arctic climate, a good reminder is always to consider how the wood will expand or shrink.

The most typical mistake is joining two pieces of wood with the grain direction perpendicular to each other. For example a table with a bread board (plank across the short edge). If these are plugged/glued together the plank across the table will constrain the movement of the wood and cause it to crack.

Best way is to join the pieces of wood in such a way that will allow the wood to grow and shrink.

Do not ever think that "ohh the wood will probably not move that much", it will! 


44. Estimating production time in advance

As with any craft estimating the time it will take to produce something ahead of time is a magical art in itself. More often the time estimate will be too optimistic, the larger the project the bigger the consequences.

The mistakes we have made in estimating production time have typically come down to two things. 1. General optimism and 2. Underestimating complexity. 3. Forgetting how long it takes to finish/oil/paint.

For point 1. this is difficult to change, I would never recommend changing an optimising and "yes-attitude" towards how much time things will take. But it is good to be aware of this tendency and try to compensate for it without bloating time estimates. For point 2. it helps to have some tracking and history, as previously mentioned, to compare estimate with previous jobs. And taking time to plan and visualise the work, also as mentioned before, will help get better estimates. 3. Just make a rule that you have to factor in much more time than you think for the final finishes.

In general I also find that "work fills time", so having to long time estimates can cause the work to take longer than it needs to.

I think this will be one of those constant learning exercises where I will always tend to estimate too little time. As I always did in the world of computers and IT.