The Steampunk Welding Torch Desk Lamp

Like most of my ideas, which develop as I progress through them, the end result is often not what I originally planned, the theory of evolution applied to design. Also, like most of my ideas, they are based on random changes as something else presents itself. So, if you are expecting a linear "This is how you do it" series of instructions, look away now. What you will be seeing is the Brownian motion involved in going from my original idea, along with a couple of evolutionary dead ends and u-turns. I learnt my design skills at the Cambrian Explosion Design School, the concept being to have loads of ideas exploding only to become extinct. I have bucketfuls in my garage of failed ideas. However, I digress!

The initial idea for this came from getting hold of an oxyacetylene gas torch at a second hand market. I knew I wanted it but I wasn't sure what I'd use it for as I don't have that many old battleships to cut up. It was pretty beat up, but they are designed to take a lot of hammer anyway. Also, being made from brass and stainless steel, they polish up really well and that was when I decided to make a lamp out of it. A quick Giggle search will throw up lots. I was originally just going to mount it on a base and have it as a side light. However, as with all evolving ideas I then came across a Stanley hand drill waiting to be utilised as something other than a drill and hence the idea came about to make it into a desk lamp which I could make rise and lower by using the drill in reverse.

So, I then did what I usually do and sketched out the rough idea. I then did what I also usually do and totally forget about it until something else reminded me of it. In this case, it was Covid lockdown that reminded me I had a lot of time on my hands and maybe I should start doing something with it. And so this began to materialise out of the pile of junk.

A brass oxyacetylene gas torch

A Stanley (not to be confused with the Marvelous Stan Lee) hand drill, preferably the one with two small pinions on the drive shaft to carry the huge torque. The one in the photo isn't the one that I used, which was lucky because I couldn't then have used the Stan Lee pun.

A base of some sort. My base was a Copydex Jointmaster tenon saw guide in a previous life. This turned out to be ideal as this particular tenon saw guide was easily modified to hold the hand drill whilst also being in just the right place for the drill . If you haven't got one of these, any suitable sized base could be used but you might need to make suitable brackets to hold the hand drill in place.

Something to use as supports for the torch. In my case they were Granhult shelf brackets from Ikea. Solidly made from alloy but easy to work with. These were also rescued from my pile of "What can I use those for?" as I'd bought them previously in the bargain bit of "Hi, Kea!" knowing I would eventually use them for anything but a shelf.

Two conduit saddles to be used to clamp the Stanley hand drill in place.

A brass lampholder with a threaded base and suitably size threaded tube to mount it with. I used a switched one so I could turn the lamp on-off at the bulb but you can switch it however you like.

The rest of the general stuff I'll let you in on as I progress with the development of the finished piece.

When I came up with the idea of using the Stanley hand drill to lift the lamp up or down, I made a quick sketch. The drill would be used in reverse and the crank handle removed to be then connected via a linkage to the rear of the pivoted gas torch. To move the lamp up and down, a knob would then be attached onto a spindle that would be held in place by the drill bit jaws. By turning this knob, the crank would then move up or down and that would then control the height of the lamp. It's all a bit like the lifting canal bridge but totally different as it's in reverse and there's no canal or water involved.

So, that was the theory anyway.

Before I started work on the main body, I knew I wanted the cable in the lamp to run through the gas torch body rather than being run down the outside and clipped on. So, that's where we'll start.

The first thing I did was remove the torch nozzle to see what the situation was like inside and also looked at what was going on at the end where the gas pipes would usually go. It was obvious from the start that as it stood there was no way a wire would go through the holes in there. But, I was still determined to do it if I could.

The first obvious problem I had was how to get the cable out of the torch head somehow. It was immediately obvious, once I'd removed the nozzle, that the holes were too small and I couldn't drill them out because they were inside and unreachable as such. So, in the end I decided I would remove the head of the torch and refit it later after I'd drilled the holes larger from the outside. As I wasn't really sure how to remove this or how it was fitted on, I simply sawed through the pipes near the head and separated the head from the body. I then drilled out the holes and bits of pipe with a drill bit the same size as the pipes with the intention of simply slipping the head back onto the body when I'd drilled the holes out big enough to take the cable and the pipes.

Once I'd done that, it was a good friction fit and it can be fixed back permanently with epoxy. I did originally try to solder it on but the stainless steel tubes wouldn't take my solder and epoxy is more than good enough anyway.

The next problem was how to get the cables out at the rear end? There was also the same issue regarding the gas inlet sizes but with the added issue of the gas control valves to negotiate. I already knew though that the pipes, valves and rear inlet were in a straight line by pushing a thin wire in with the valves open. So, I first removed the gas and oxygen control valves. This isn't difficult, you just use a spanner to remove the locking nuts (B). However, for safety so they can't come off if loose, they are designed so as to not be removed in one go. They will come up against the gas control knobs (A) and not come off. You then have to open the control knob as far as it will go and then repeat the unscrewing of the locking nut. You may need to do this again and eventually they will come out.

Once the valves are removed, you then simply open up the gas pipes from the rear into the pipe by drilling through them with a drill large enough to take the 3 core cable. At this stage though I also knew that the main stainless steel gas pipes were big enough to take a three core 3A cable so I only needed to do it with one. The one to choose is the one that is closest to the front of the jet. Once that's drilled out, the valve can then be replaced- BUT ONLY ONCE YOU'VE DONE THE FOLLOWING! The valve can close the hole you've just drilled through. That's OK for gas but not advisable if the cable is running through as it tends to make the cable less than safe. So, the first thing to do is to grind off the valve gate end so that when the valve is fully closed it doesn't cut into the cable. This is easily checked by putting the valve back and closing it and looking up the hole to see it's not coming onto the cable route.

To put the valves back in after drilling out the gas pipe big enough to take a three core cable, you simply reverse the procedure, alternately screwing in the knob and then the locking nut.

The cutter head consists of three parts, the main body, a nozzle and the nozzle nut which holds it in place. I modified the nozzle to take the lampholder threaded tube onto which I then screwed the lampholder.

The images above are actually from a different nozzle which I'm converting for another lamp so they are slightly different. However, the method was exactly the same except I cut the nozzle right back to bring the lampholder close in to the main head. I think cutting it back made the head on the desk lamp look better but feel free to leave it full length if you want.

So, once the nozzle is the length you want it to be, simply drill out the nozzle to the diameter of the lampholder threaded tube. Then the threaded tube can be firmly attached to the nozzle. I soldered these together with a blowlamp but you can use epoxy here instead, Then its simply a case of screwing on the lampholder and it's ready for the wiring up stage at the end of the build.

The first job was to remove all of the handles from the drill.The one on the crank was held on a riveted axle so that was drilled out. By chance, the uprights on the Jointmaster were exactly the right separation to match the drill body and just the right height to keep it clear of the base. Each upright was originally a pair which guided the saw but as they were screwed to the base it was easy to remove one from each pair.

But how to hold the drill in place? You can drill a bolt hole through the top but at the bottom end that isn't practical due to the drive axle. Fortunately, once again by chance, the diameter of the main body was about the same as 20mm conduit so it was easy to use 20mm conduit saddles, padded with a strip of copper inside. You can see these clamping the drill on in the first main picture. The thing to note is that the left most one used a countersunk bolt which holds the base onto the post before attaching the drill, which is then held in with the clamp. The rear was originally the same, but here was where one of those little things you don't realise later became an issue. The clamps didn't stop the drill body rotating slightly. So, the solution was to drill through the clamp and the drill body on the left side (top) end of the drill.The axle doesn't go through here so that is fine. Then, a single bolt goes all the way through, as you can see from the visible Allen head. This stopped any tendency of the drill body to twist.

The control handle which you rotate to position the light was off an old water valve where the shaft was removed. However, two issues arose when the final light was put together:

1) The valve shaft spun in the drill chuck. The solution to this was easy, back to the bench grinder and flatten off the sides to match the jaws. This solved the rotation in the chuck problem.

2) The chuck itself kept coming loose. I did consider using epoxy on the chuck but decided it was too permanent and may not have worked. In the end, I tightened the chuck up with the valve in place and then drilled a hole through the chuck and axle to place a clocking bolt in place. You can clearly see this in the main picture above with its rounded bolt. I also wrapped the valve shaft in the rubber from an old bike inner-tube which acted as a grip. It hasn't worked loose but it also lets me remove the knob and shaft if needed.

To allow the gas torch to move up and down I needed two support brackets, the aforementioned Ikea Granhult shelf brackets. These already have screw holes which just needed opening out for the bolts I was using. However, when placed onto the base it was obvious they were too far forward. As a result, I shortened the rear flange and redrilled a new bolt hole. Had I needed to go further back, I could have cut the rear flange off totally and drilled a new bolt hole inside the base but didn't need to as the bracket was positioned about right anyway once the rear flange was shortened by about half.

The next task was how to pivot the gas torch? I simply drilled a hole through the body of the gas torch big enough to take some threaded rod. I also drilled two holes in the brackets, making sure they were the same height by putting the two brackets together in a vice and drilling through both at once. I then used some washers and spare tube with the inner diameter the same as the threaded rod to make spacers to go between the gas torch and the brackets to hold the gas torch in the centre. The threaded rod then had round headed nuts put on at each end. Here's where I made a second mistake (only two, what a liar!) that didn't show up until I'd finally added the linkage. I originally put the axle hole to close to the gas valves. I might as well tell it you here though as I didn't realise until too late that I should have constructed the linkage between the gas torch and the drill crank before deciding where to put the pivot rod through the gas torch.

The problem? Gravity. But we'll get that in more detail later.

So, put the gas torch roughly in place but don't drill the pivot axle hole. Then, when it's linked up just hold it roughly in place and see where you think it needs to go in order to pivot correctly. Same goes for the shaping of the top of the brackets. At the end, the brackets stood up too far so I then chopped them down and angled the tops in once I knew exactly where the pivoting axle would be. This was left right to the very end though in case I needed to lift the pivot axle up higher for some reason.

The original idea, shown in drawing (A) was to have the drill crank at the back pulling down on the rear of the gas torch to lift it up or down. The linkage was a turnbuckle with eyelets at both ends. Why a turnbuckle? The answer was I had one about the right size and I thought it would add to the look. It doesn't need to be a turnbuckle at all, it can just be an appropriate piece of metal drilled at both ends, as I used in the other linkages which I'll come to later. If you're not sure what a turnbuckle is, it's used to either tension something or act as an adjustable length link. One of the eyelet headed bolts has a right hand thread, the other a left hand thread. As a result, if the body is turned one way, both bolts move together and if the body is turned the other way they will move apart. The turnbuckle was held in place at both ends by bolts, through the drill crank where the handle used to be and through a hole drilled though the gas torch near the rear.

So, it looks like it should work OK, but it didn't! What was the problem? The problem was that I wanted the lamp to be upright at the top. Putting the linkage there caused two problems. The first was that for that to happen the drill crank end of the linkage would have to be vertically below the gas torch axle. It wasn't, it was behind it. As a result, as the linkage came into a straight line with the body, the lamp was still angled out. This created the second problem. This meant that the pivot point now was in a position where when the lever was pushed the other way, it could actually then go in reverse and instead of letting the torch go down, it would flick the wrong way and push it backwards instead. I think it might be called a position of unstable equilibrium, a position I'm well acquainted, with having fallen over loads of times, though much less since I stopped drinking. Whatever it was, it didn't work.

The solution turned out to be easier than I thought. I just rotated the drill crank by about 90degs. This meant that the linkage never went in to being in a linear position with either the crank or the gas torch body.

This then had the movement sorted. What could possibly go wrong? The previously mentioned Gravity, that's what.

The problem was I'd build a very smooth mechanism, but the gas torch wasn't pivoted at its centre of gravity. I had expected there to be a force from the torch pulling back on the drill crank, but I naively assumed because the gears were reversed that it would provide enough back force through friction. Why I thought this, I don't know because it was actually a smooth running drill. The result was that the gas torch had two stable positions, vertical and horizontal. In between it would happily just drop down. So I then went through a process of trying things.

First I just thought I'd try to increase the friction in the mechanism somehow. My first thought was to use roughened rubber washers inside on the linkage bolts which were then turned tight to provide some stiffness. This didn't work. I then decided to use something similar to that used on vintage motorcycles as a friction damper on the front forks, as shown in the top right side drawing. This can be seen in the linkages at the rear, which consisted of links made from some cheap cutlery knives, the type you can buy that are simply pressed out from about 2mm sheet steel and are the same thickness all the way across the handle. I used three, as you can tell from the diagram, two on the bottom sandwiching one at the top, again with rubber washers and a tightening bolt to control the friction. The result? It didn't work. It still wouldn't hold the non-flaming gas torch in place. I then went through another brainless storming session and quickly decided against a counterweight as making the whole device too heavy and against springs because I could see issues with getting springs with the right tension and positioned correctly.

The final solution? I decided to move the axle pivot point closer to the centre of gravity of the gas torch, as previously mentioned. You can actually see the spare hole which was the original axle point in some of the earlier photos. Yeah, success!

I did leave the damping mechanism in place though, because it was contributing to the friction and because I liked the look it gave to the finished desk lamp. Plus, if anybody asks what it's there for I can smugly reply "Isn't it obvious?".

So what else could go wrong or need fettling? How do you control the range of movement of the gas torch so it doesn't carry on past vertical or keep dropping down until the bulb hits the desk? I first considered putting a bar across somewhere across the front of the uprights at a suitable height to give a rest stop for the gas torch, with something similar at the rear. However, again by chance and observation, I realised that the bolts on the drill crank and where the links joined the torch body could be extended out to go past the sides of the support bracket. You can see these in the two photos above, with the second one showing how the crank one butts up when at the lowest point to stop the gas torch getting any lower. The rear one does the same at the rear when the gas torch is vertical.

I did have one other problem in doing the linkages that I finally solved. This has been a problem I've had all my life- how do you drill stainless steel? The knives I used to make the rear linkages were cheap, but they were still stainless steel. In the past, because I didn't know any better, I've just used a drill bit as fast and as hard as I could and got nowhere. Sadly, I had to refer to the universal manual on such things and resorted to the internet. The answer? Do the opposite. Use a sharp drill bit, set the drill on absolutely as low a speed as possible and first put some oil on the drill location. The reasons? Stainless steel work hardens. This means if it warms up it gets harder, so you do your best to not to let it warm up and the oil cools it. The result was I was drilling through them without any problems once I knew how. This means you can't really use a two speed drill without any speed control, you are doomed to failure... and so is your drill bit.

Before bothering to wire it up, you might want to paint it and polish parts. I only actually painted the base and did that once I'd finished messing about with the drill mounts and the bracket holes, etc. It doesn't take long to strip it down and I didn't want to paint it first only to get it messed up drilling holes. But how could I decide on a colour scheme?

As a child I owned a Mamod steam engine, so I already knew once I'd seen the polished brass gas torch where the colour scheme would go, plus red along with green are common traditional colours for industrial steam engines. As for what happened to my own Mamod, I wish I could say it became a cherished family heirloom, ready to be passed down the generations, but like the majority of my mechanical toys it got taken apart. I must have frustrated my parents no end because I remember from about 6 upwards taking apart model trains, tin toys and basically anything that moved. Even more frustrating for them, I didn't do much putting back together. However, it did pay dividends in the long run. I reckon I can now probably take apart a Boeing 747 given enough time. I never did learn how to put things back together though, but I'm an expert on making piles of bits.

This bit is easy if you know what you are doing. If you don't know what you are doing, then don't do it but get someone who does know how to do it for you. I'm not even going to try to explain it here because I've seen some interesting home wiring jobs before now done by people who didn't know what they were doing but thought that they did.

So, I'm just going to make the following comments:

1) This thing is metal. Metal loves to carry electricity through it to your hand and then let you know you've been electrocuted by killing you. So, it's important that you use 3 core, not 2 core, cable- two for the power to make the thing light up and one for the earth to make sure it doesn't light you up in case there's a short somewhere. In my case, I used a lampholder with a built in earth connection. If not, I would have attached the earth elsewhere to the gas torch body through an earthing bolt. I haven't run an earth to the base because everything else is connected through by metal plus the base itself doesn't carry any cables anyway, but if you decide to mount a switch on the main body I'd run a separate earth link there as well, just in case.

2) I used a lampholder with a built in switch. You can either do the same, use an inline switch, fit a box and switch to the main body or use no switch and switch it at the mains. Alternatively, fit it with a smart bulb and voice control it through a smart speaker.

3) Use an appropriate sized fuse in the plug. Mine uses a 3A fuse. Also a good idea to use an LED bulb as that exposed bulb would get really hot if you were using an incandescent old school bulb.

4) I didn't really want the lead to just come straight out at the back but nor did I want to have it flexing. My ideal choice would have been to have fed it somehow down the brackets into the base and then out at the rear but I decided that would have maybe been a flex to far. My latest idea is to eventually find a spring coil type wire protector and feed it through that. We'll see what pops up. In the meantime, it's working fine as it is.

I'm actually really pleased with this but I have a couple more things I may add on to it eventually.

1) A shade maybe? I'm 50/50 on this as I like the idea of the bulb being a flame and it fits well into the design without a shade. However, I had thought of making a shade either from half a can or from a cheap metal dish with a hole drilled through it.

2) Seeing the Mamod steam engine made me realise what this was missing - cylinder and driving piston! How could I have designed this without one? This will definitely be happening when I can find the right brass pipe for the cylinder. The plan will be to make a cylinder out of a brass pipe and then just feed a threaded rod into it with a couple of washers on to act as a piston. I'll then cover the exposed bit of threaded rod with smaller brass tube and fix it to the base at the front and the drill crank with a couple of brackets, as shown in the drawing above. Alternatively it could end up going across the damper linkage but I think it'll fit in better in terms of the theme to be at the front. Hopefully I'll remember to update the photo when I've added it.

And so, this is finally the end! I hope you've enjoyed my random design journey from initial idea to its conclusion and I hope it prompts someone else to have a go.

Tatty bye for now!

I decided to add the cylinder and piston using what ever I had lying around and let it evolve. In the end it involved some general plumbing connectors along with the serendipitous find of an eye bolt and an old spark plug. So, it's now developed from Steampunk to ICEPunk (Internal Combustion Engine Punk).

As this another evolutionary stage, I first decided the original connecting points wouldn't work with what I has so I decided to use the Equal Tee connected to the main gas torch axle. I first round off the threads on the middle tee using my trustee bench grinder and then drilled across it with the same size hole as the threaded rod I used for the torch axle. It then attached the female socket with the reduction socket on the end. The reduction socket was also, by chance about the same diameter as the eye bolt shaft, it just needed slightly drilling out to let the rod ride smoothly in and out.

As you can see from the photos there was a length of copper pipe attached to the other side. This represented the cylinder, but was also there to give space for the eye bolt shaft to move in. Also, with continuing luck, the copper pipe wasn't too far off the diameter of the spark plug. My grandmother claimed to be Irish so I'm at least a quarter lucky. I'm not too sure the remaining 75% contributed much luck though based on past experiences and my continuous u-turns and holes in the wrong places. All that needed was the aforementioned bench grinder to be fired up again and the threads taken off the spark plug until it became a good friction fit.

Once the piston had been made, the brass tee was fitted to the gas torch axle and the eye bolt was connected to the drill crank.

Voila, the new end, an ICEPunk Gas Torch Desk Lamp!

I was never happy with how the flex simply came out of the end of the torch. I had always intended for it to be hidden but couldn't originally come up with a good way to do it. By chance, when I was looking for cable protectors, I came across these, which are designed to carry the cables for an electric lock from a door frame into a door. They turned out to be ideal. They are very flexible, have an industrial look, are just the right size for the flex I was using and finally, with a stroke of good luck, were the right outside diameter to fit the original gas pipe nut.

The only change I needed to do was to cut off the rounded edges of the end clamps. These were designed to hide the cable entry holes into the frame and door, but in my case I wanted them to be open ended clamps at both ends. I simply then drilled a hole through the stanchion support base to take the cable under the main base and out of a hole at the rear, into which I first placed a grommet to protect the flex. The clamps were then screwed to the stanchion side and the other end went through the gas nut and was held in place with a cable tie around it to stop it pulling back out of the nut.

So, job done, right? Wrong! The issue was that when I'd added the nut, it then interfered with the Stanley drill mounting and stopped the lamp going to vertical. The solution? Yet another hole for the gas torch axle to lift the end up slightly. This now works perfectly and I don't mind the extra hole, I'll eventually probably just fill it with a length of screwed rod and put a dome headed brass nut on each end.

And now, finally finished, I can say I am really pleased with this. I think the new flex cover works really well and is in keeping with the rest of the lamp. I am also pleased to be able to say that, at long last, this is finally...

THE END!

(or is it?)