Image: St Alban’s Fixers. Tony being coached by Mike (and even the members of the public getting involved!). Diagnostic even included some ‘back to first principles’ on AC rectification which explained the voltage of the heating element of the Air Fryer being ~115v. Despite Tony (and myself) being familiar with the theory from Uni - nothing beats practical knowledge!
“No passion in the world is equal to the passion to alter someone else’s draft” - author, H. G. Wells
This quote refers to editing a written draft manuscript, but would just as easily apply to a designer/engineer’s passion to improve upon someone’s product. In fact I’m yet to have a conversation with a designer or engineer who doesn’t have some anecdote of a design they love (or hate). We seem to be ‘magpies’ for details, tricks, techniques and pitfalls to avoid! Repair Cafes are, in that regard, a ‘gold mine’ of product analysis, where you wouldn’t otherwise be exposed to so many excuses to disassemble a thing with legitimate cause and zeal. It’s like a music shop begging you to do a lengthy drum or guitar solo, in-store - how often does that happen? Never, as per Wayne's World!
So consider visiting your local repair cafe as a way to benefit from the hindsight of seeing such a broad range of mistakes and possible improvements you could make - and of course, some good ideas too!
Indeed, just because you design X today, does not mean you might not be inspired to design Y tomorrow, after seeing how much better it should be! (I’d say a lot of Dyson products essentially follow this intuition at their core).
1. Questionable Electrical Design/Manufacturing Bloopers.
One of the first things I saw on my first visit was this battery unit of a polaroid-style camera. If you haven’t spotted the issue - the springs are all at the top, implying all the batteries would ‘point down’, and yet, their orientation alternates (up/down/up/down). This is such a weird departure from the convention (spring to negative/base of battery). It is hard to tell whether it was an extremely quirky design choice or a mistake made in the manufacturing process. The ability to confuse a consumer seems real here, and even though it’s designed to avoid short circuits (with Poka-Yoke details), you could still confuse a user for no reason.
Image: Fixers at Haringey, London.
2. Confusing User Experience Blunders.
Next up is this water boiler, which was in the kitchen. I was told ‘two were broken’ - when in fact two had boiled dry (most likely) and the ‘thermal fuse’ had blown. This is a safety device that prevents fires and damage. The remedy is simple - you allow it to cool down (add more water!) and press the ‘RESET’ button. The trouble is, the button is underneath, and the kitchen staff were none the wiser. As a designer, you might think you add a sticker with ‘reset underneath’ on the side, even if it was impractical to put the button in a more prominent position (as by the power lead).
Image: Haringey Fixers' Canteen / Boiler. Perhaps not best place to 'hide' the Reset Button.
3. Endlessly Complex Assemblies (Practically Unrepairable).
Here’s me after attempting - and failing - to fix a steam iron. We tried circuit testing, chemical mechanical faults, quizzing the owner for accidental misuse, and so on. Nothing was apparently wrong with it - but it would not work. Sadly out of warranty by a few months. (More thoughts on that later and what you can do about it!) Added to which the PCB (circuit board) was ‘potted’ (encased in glue), and the design was very far from ‘designed for repair’. I’d happily concede I’m not an expert on Iron Repair, but I was given sympathetic looks from the other fixer and told that Irons are ‘notoriously bad’ - very few, if any, get fixed, especially the ‘modern’ ones…
Image: Jude at Haringey Fixers - trying to fix an Iron. And failed, for numerous reasons!
…If like me, that ‘gets your goat’, and you wonder why, this is exactly the reason we need more people to attend repair cafes, if they are in the business of making products like this!
For the record, I’m not saying everything can or should be made repairable, but in this case of this iron, which is a Oliso brand - which ranges between £100-£200 range, you’d think this would have been able to ‘factor in’ to its cost to have some repairability, for such a premium product. Is it any wonder then that people sometimes just ‘give up’ and buy a £30 ‘basic’ iron, if they don’t seem to have a big difference in longevity?
It seems we are trapped in a vicious-cycle of consumer apathy, not helped by bad designs and/or planned obsolescence. And of course, the manufacturers have financial incentives to encourage consumers to want ‘the latest thing’ even when it works fine, after a short period of time!
4. Be Nice - Label Components.
As demonstrated here with Michael, in Repair Cafe HK, smartphone ‘zoom’ is extremely handy when looking at the tiny text on ICs and other electronic components. I’m not suggesting you label each and every component, (many do indeed have serial numbers and/or rating on anyway) but as an experienced person, you’ll likely know which bits will fail over time - and you’ll be helping out those who might have to do maintenance or repair in future, so it feels like a nice thing to do, which costly a few moments in time and no extra cost (PCB silk screening is the same price, regardless of labelling, right!).
Image: Michael preparing to switch out a capacitor on the air conditioning unit. Using a smartphone to ‘zoom in’ to the small writing. Could you as a designer make this a little easier, with a few notations?
So for example, if you have a capacitor that is doing a lot of ‘heavy work’ and will likely wear out over time (as I was told anecdotally, this is a huge percentage of the fixes of Air Conditioning units, which are widely used in Hong Kong’s sweltering summer heat), then consider labelling the PCB with any useful details. Likewise, labelling fuses. This does not legally mean you’ve encouraged an unqualified repair, but rather you’re helping those in authorised repair places and/or competent people to fix it more easily.
5.A. Excessive Use of ‘Rare’ or ‘Obscure’ Security Screws.
You might be thinking this is a standard security screw. But no, this is a Pentalobe with extra ‘Security’ (the pin in the middle). To put this in context, even Apple (who invented Pentalobes) doesn’t specify ones with extra Security. So what is this extra ‘security’ feature doing on a hairdryer, which retails for under £100?
I would fully respect this ‘high security’ measure if this was something which contained a lethal chemical, super high capacity capacitor, or was perhaps radioactive. But you can take apart a microwave (a magnetron is a seriously lethal thing if misused) with often no security screws at all. Other than this being mains voltage (like a microwave), I can see no clear reason why GHD would use a pentalobe with an extra security hole, which even the most tool-obsessed repair shops don’t usually carry. (Haringey Fixers had to order the bit in specially - which in itself is another good reason to support Repair Cafes, as we don’t all need to buy a Security Pentalobe each!)
Image: GHD's Hairdryer with a seemingly 'overzealous' security screw (which was surmountable anyway), and only adds yet more permutations of the arguably overly long list of security screws.
In short, if you are a design company, and you need to have a Security Screw, please do not make the problem of Repair worse by a. Using a really obscure safety screw (like above), when there is no clear need to have an ‘exotic’ one, and b. As a result, triggering people to buy another screwdriver, or possibly an entire screwdriver set, when the current one is more than enough to deter any members of the public from casual and ill-advised tinkering. A standard Torx is a good deterrent for the rookie, whilst still being accessible for the experienced repair person.
5.B. Create Easy Access to Safety Screws (or if you must - just use standard Torx).
I used to work at Dyson and as a graduate, I wondered why we used ‘fancy’ screws (or Torx to give them their industry name). I didn’t initially see the point in not just using Phillips, after all, it was what I was used to and worked fine, or so I thought…
However, as I progressed in my ‘further education’ there, I realised two very sound reasons Torx was used…
- Firstly, it was easier to put a screw on the drill bit, on a production line, and because it had 6 ‘contact points’ (as opposed to the Phillips 4), it was easier for a production worker to ‘load’ a torque-set-screwdriver, and perform the operation of screwing two parts together. This may seem a trivial difference - after all 4 vs 6, or 90 degrees rotation vs 60 degrees rotation to ‘locate’ the screw on the bit seems tiny. But in production, it really is the difference between seconds which add up.
- Secondly, this is more the point - it is almost impossible to ‘strip’ the ‘head’ of the Torx Screw, because it’s like a ‘cup’ or ‘socket’ shape, as opposed to a ‘V’ profile like the Phillips. Whereas a Phillips Screw can slip - and you’ve damaged the head of the screw. It also means the screw is ‘straighter’ when you sit it on the drill/bit - again, reducing risk of mistakes.
I happened to work on the DC37 Vacuum, as the second product I helped design, after the DC35 (which tells you how small the company was compared to now, as those numbers are close together!). Anyway, being unashamedly proud that I had done a fair bit of the aero-acoustic and motor mount design, I learned how to do ‘Cut Aways’ in my evenings after work hours. [Please note, I did not work alone, and there was a late-night technician in the CNC workshop who made sure I was alive. Do not work alone using industrial tools - ever].
One of the things I did like about Dyson’s design philosophy was that it did have a ‘logic’ behind it. I was quickly converted to Torx being a good choice, and there was pride in design teams using screws (not glue) as much as possible. Furthermore, there was an ethos of not using more than 2-3 lengths of screw, and ideally T10 (Torx Size 10) wherever possible (and at a push a T8). It really meant you could open a Dyson with just two Torx and a Phillips. Indeed, the second picture shows the Phillips Screw in the motor (which was third-party / not made by Dyson at the time).
This is showing that a full-size Torx T10 screwdriver is needed to ‘reach’ the deep screw boss of this machine. This of course, is not possible with ‘Security Screw Bit Set’ - where the bits and sockets are wide and not long enough - would not fit the hole or reach the screw.
At the risk of this becoming a very niche debate, I would say a lesson to take from Dyson - which I feel legitimates the use of Torx - and needing to buy a long handled Torx Set (see below), is that this is literally all you’d EVER need to open a Dyson (or at least the ones I worked on!).
In summary, My recommendation for Designers / Repair Cafes would be:
- Designers:
- Where possible, make Security Screws ‘shallow’ so they can be reached by small bits.
- If at all possible, only use standard Torx (not Security Torx - with the centre pin, as this is needlessly exotic).
- All internal screws should be either the same Torx type (like Dyson), with as few size variations as possible.
- If you must have a ‘deep-seated’ screw that requires a ‘long’ screwdriver, make it Torx, (or Phillips) - as these are reasonable for a Repair Cafe to have in stock.
- Repair Cafes:
- Stock a set of Security Screw Bits (interchangeable). Ok Quality.
- Stock a few long-handled Phillips Screwdrivers. Magnetic. Good quality.
- Stock a set of long-handled Torx Screwdrivers, as these are the most commonly used ‘Security-type’ screwdrivers.
If Designers can now better appreciate that really the job of a ‘Security Screw’ is to be a ‘deterrent’ for unqualified people to put themselves at risk by opening products they are perhaps not qualified to repair. It is not meant to keep all people happy, all of the time! Legally qualified people should not have to carry an inventory of 50 different types of long-handled screwdrivers to run a Repair Cafe, so please keep ‘Screw proliferation’ to a minimum, and use as few variations as possible.
Indeed, this is not just ‘good for Repair Cafes’. At Dyson, this was considered good business and economic sense. Their maintenance guys didn’t need a load of tools, and when buying screws in bulk, it was cheaper to have less inventory, and this meant the production lines made less mistakes.
Screwy Business?
For those of you who want more musings on the Security Screw situation, some other examples, which you may well want to use when discussing with your team as evidence (should you need more!) that we should curb our over-enthusiasm with making more security screw types, and trying to use the 'basics' - which will be cheaper and simpler anyway, and as mentioned - not really prevent anyone keen to get inside...
To give another example, below is my Glue Gun (which is excellent to use) from Bosch. However, in terms of repair, I think it could be better. In this case, it has a ‘Torx Security’ screw, which is a Torx (6 ribs) with a ‘dot/pin’ in the middle. This means that ‘ordinary’ Torx do not fit (the ‘pin’ protrudes and prevents it). I personally think this is overkill, but the reason is likely that you can technically force a flat-head screwdriver into the widest ‘lobes’ of the Torx and undo a normal Torx, so it is not ‘secure’. This is likely why a Pentalobe was created, but I personally think this is about brands trying to prevent repairs, rather than prevent deaths - see iPhone to give an infamous example, see Wired.
My personal opinion is that most appliances should not get carried away with excessively niche screws, as frankly, their function is to keep out the ‘General Public’. Unless it is truly a device one should never attempt to repair, even if a professional. Besides, if you were sufficiently motivated - Amazon/eBay/AliExpress/RS/etc. sells all of these things to the General Public within a few months of a company like Apple ‘inventing’ them, so it’s a game of ‘Whack-A-Mole’ and seems pointless at best, and cynical at worst!
Keep It Simple Stupid.
When visiting UCL’s Repair Cafe ‘pop-up’/’open day’, 21 Feb 2025, I spotted this baffling combination of fixings: This is a pretty fancy toaster (ranging between ~£100-£200), and which had failed just after its 1-year guarantee. If you look closely on the right - you can see a Phillips ‘+’ screw, which is fine, but above it a Security Screw (triangular). The technician didn’t have this bit, and even though it was ‘shallow’ (easy to access), it meant we had to hunt around to ask if anyone had one; nobody did. So he resorted to carefully inserting a ‘hex’ (as of course a hexagon can fit in a triangle) and very gently unscrewed the screw. To my earlier point - KitchenAid did not ‘keep out’ a determined and resourceful repair technician, but simply wasted their time. The owner, having spoken to them, was not even remotely confident to attempt the repair solo, so a standard Torx would have been sufficient to dissuade her. The Triangular Screw was ‘overkill’ in my opinion, and from an engineering perspective, seems inferior to the Torx anyway in terms of manufacture (DFMA).
I hope this is a useful ‘rant’ on the subject of screws. I suspect there are many in the industry that may extol the virtues of a Pentalobe, Tri-Wing, or Pig-Nose over a more common Phillips and Torx, but I hope this convinces you, dear reader, that one can create ‘reasonably safe’ appliances with much less fuss. Anyone who wants to ‘jailbreak’ can, and will, and anyone who’s arguably not competent will be dissuaded with a Torx and a warning sticker. So please be considerate to your maintenance, repair, or indeed, recycling centres - and keep it as simple as possible.
You Can’t Keep A Good Engineer Out.
I have recently also visited St Alban’s Repair Cafe, and met some great engineers and technically-inclined folks. As previously mentioned, most engineers can often ‘work-around’ obstacles, and when trying to fix this moderately priced toaster, once again we see the seemingly excessive use of ‘triangle’ screws used for ‘security’.
Image: Fixer at St Alban’s Repair Cafe, showing how a ‘security’ triangular screw, can be overcome by a simple ‘flat head’ screwdriver. Makes you wonder ‘why bother’ - especially as it does not actually keep anyone ‘safe’ from a toaster, with its open slots!
Given how fundamentally dangerous a toaster is (it is what’s known as ‘grandfathered-in’ meaning the safety rules came after the product was invented, so it was deemed ‘common knowledge’ how to safely use a toaster) - it further illustrates the futility of keeping someone out of the electronics, when you can stick any object (including fingers) inside a toaster slot - where live wires reside.
In summary (perhaps I’m at risk of labouring the point here), but when you consider, rightly or wrongly, I’ve yet to see a Repair Cafe fixer not get into an appliance, with myriad Security Screws offering some ‘barrier to entry’, but not the a meaningful extent that warrants so many variations.
I think as designers and engineers, we should perhaps simplify matters and use Phillips - for the general public, and Torx - to imply you need to know what you’re doing. Given that anyone can buy service screws/screwdrivers online, it seems a bit of a ‘hustle’ to make us buy more and more bits for no real benefit to safety.
So, if you are in a position to resist creating a new screw, please just use the basics where you can. Repair Cafes will thank you for it, and it’ll still be compliant.
Contents:
Part 1: Repair Like Your Life (and Career) Depends On It! (link)
Part 2: Why Engineers and Designers Love Volunteering at Repair Cafes. (link)
Part 3: Five Design Issues, Every Engineer Should Know About. (This article)
Part 4: Twelve Things I Learned At Repair Cafes. (link)
Part 5: How Did We Get Here? (link)
Comments