Mad Ax

So - in short, it works! I finished on Saturday evening with absolutely no idea why my LEDs weren't bright enough, and wondering if I was going to have to rewire it all and get another voltage regulator to boost power to the LEDs. It was something of a relief this evening to hook it all up and actually see it working properly. I'm waiting for a few smaller parts to arrive (Deans connectors, resistors for the torches) and I need to change the TIP120 wiring around a little and put some nicer wires on it, but basically the electrickery is almost there. I'm going to try to build a tidy cover to go over the board so nothing gets wet when I open the roof in the rain, which will probably be a job for lunchbreaks this week, as well as working out how the board is going to mount in the rig. I hope my Deans connectors get here soon so I can actually do a proper, proper test on it all. After that it should be driveable again - it's been benched for quite a few weeks while I got all this electrical stuff and the roll cage done, but it shouldn't be too far off another test drive now. Getting the torches properly installed on the cage shouldn't be a big job, although I still have to clean and recommission the 3D printer so I can print some bungs to cover the holes in the bodies. I've got some flexible rubber resin which should theoretically be perfect for that, although I haven't even taken the bottle out of the protective seal yet and it's probably already past its expiry. Here's to a few more productive Sundays

I was busy all day Sunday doomscrolling on my phone while my daughter stared lifelessly into her iPad, but yesterday my plan to finish rebuilding my Mi-8 touring car was scuppered by finding yet more incorrect parts that needed to be ordered, so I had a few spare minutes to have another look at these dim LEDs. Long story short, I'd made a mistake. While a transistor does indeed work like a relay, in that applying a current to the base causes current to flow from collector to emitter, it is a semiconductor and therefore obeys a set of rules that mere mortals are not meant to have the what of. Basically I had installed my transistors "high side", i.e. between 5V and LED, whereas NPN transistors work better when installed "low-side" - i.e. between LED and ground. (I could have installed PNP transistors high-side. I always assumed a PNP transistor was like a reverse of an NPN, in that it stops the current from collector to emitter when you apply a current to the base. But there you go, every day is a school day, etc). What that meant is that my wiring was all wrong, and the board that had taken me literally an entire day to make had to be remade. Fortunately, after doing some prototyping on the breadboard, I realised I didn't have to do all that much - I just had to swap the Ground connection on pin 1 of the LED output headers to +5V, and swap out the +5V wires to the collectors for Ground. Then turn the LEDs backwards. I didn't have time to do any soldering yesterday, but I went up to the workshop at lunchtime today and gave it another look. Here is how it was when I started. Black ground wire going to pin 1 of the LED board (bottom right) and red +5V wires going to the collector busses. And now fixed. Also I had started moving my Arduino-to-LED-board wires from the top to the bottom, which is neater. I didn't have time to finish this at lunch but I came back to finish it this evening while my daughter was failing to eat her dinner. All plugged in for a final test.

It's gonna go sort-of in here, like this. But better. At this point I wasn't entirely sure how I'd make it fit without making it a real pain to get the side panels on and off (since they literally come off 3 or 4 times every weekend) but I've had an idea since and it's genius. I decided that getting everything properly hooked up would be easier done out of the truck, so I took the headache rack off. I need to run 2 power cables off the Deans connector, so that lovely fixed socket I made needs to be modified. This is the headache rack. 2 cables now soldered to the pluge. 1 pair to the ESC, the other pair will go to the voltage regulator which will provide a constant voltage to the Arduino and to the torch bulbs. I tidied up the installation and tied the voltage regulator down. I want to fit Deans connectors to the output of the voltage regulator and to the input of the Arduino board, but I've completely ran out, so that part of the job is on hold. This was pretty much the end of the day. I did have time to fire it all up and check how well the torch LEDs worked, and the answer was, well, not very good. They were visible dimmer than when running off their own batteries (I still have 2 complete torches to compare against) and the current draw on the power supply screen was significantly lower. Hmm... Something was up...

My original plan had been to put headers on the LED breakout board and make up a ribbon cable to connect it to the Arduino board, but I worked out I had space for it to mount end-on and therefore decided to go direct from board to board without a plug. But that sort of meant I'd put the wires on the wrong side and so it looks a bit odd. But it's all there and it all works. Sort of. I stuck the TIP120 on the main board because there was more space for it, and also added some big chunky wires. These are overkill and I'll probably replace them with smaller wires. In a more appropriate colour. Like black. For those who know where this is going. I hooked the whole thing up to the power supply, but couldn't figure out why the LEDs weren't as bright as they should be. Anyway, I figured that was a problem for another day, and instead I should focus on making it all fit neatly in the truck. So I make a backing board.

Once again, I was keen to get started early on Saturday, as I had a night out with some friends planned and wanted to be done, showered and ready to go out before it got too late. I had chosen the biggest prototype board I had, thinking it would be overkill, but when I came to make the lighting circuit I found myself pushed for space. So I got a smaller board to make all the lighting stuff, thinking it would give me more freedom to mount it later. Here I'm connecting the red power wires that will feed 5V to the collector of the 2N2222A transistors that I'll be using to switch the regular 5mm LEDs on the truck body. Those of you who know anything about electronics can stop laughing now. For the rest of you, I'll explain the joke later - we're doing this saga in chronological order... So here we have 7 transistors. Each will switch a different LED channel. Headlights, tail/brake lights, rear floodlight, reversing light, and the red / blue / green channels for the dashboard LED. Not sure if I explained what that's for yet, but you'll see it sometime in the future when all the LEDs are hooked up. Connecting up some colour-coded wires. Testing it all works as expected. The input pins you see here are the legs of the current-limiting resistors. These stop too much current from going into the transistor base, which could blow the transistor or the Arduino chip. Here I'm applying 5V to each resistor in turn; this opens the base of the transistor, which allows current to flow from collector (which is connected to +5v) to emitter, which then flows (via the pretty rainbow cable) into the current-limiting resistors and then into the LEDs, and then to ground. This worked perfectly and each LED came on in turn. Now stop laughing at the back. We'll get to that bit later.

But in the end I was able to get it all working, including resolving a short that had occurred between 2 of the servo pins at the Arduino header and was stopping either from working. That would mark Close of Play for Friday, and I went in to relax for the evening, ready to start again fresh on Saturday morning.

In theory the Arduino and all the non-roof LEDs could be powered by the same thin 24AWG wire that I've been using for patching the pins, but I wasn't sure yet how I would power the TIP120 and the roof LEDs, so I decided to solder on some thicker gauge wire which will eventually go to the voltage regulator. The underside of the header pins was very tight and I was worried about shorts developing once I've installed the whole thing in a moving vehicle, so I decided to cover it all with hot glue to provide some insulation between the pins and prevent the wires from moving over each other. I would intensely regret this decision later. Hooking it up to my desktop power supply, I got some lights to come on and prove it works. And by very precariously balancing some heavy, expensive and potentially dangerous equipment on top of a very unstable powered truck on an unsteady bench stand, I was able to prove that the ESC responded to the radio and all the inertia stuff was working. In other words, the Arduino was powering up properly and doing what it should do. I then went off to have lunch and watch Raya and the Last Dragon and eat snacks, and while I was lunching, I had a sudden sinking feeling. I had connected the power wire to all the servo pins to the 5V input from the voltage regulator, and so as to avoid overloading the Arduino, I hadn't connected the power wire from the ESC to anything at all. But this was a mistake, because the servos (especially the steering servo and winches) will need the 7.4V output from the ESC's BEC to get the best from them. A high-torque servo on 5V will be pretty useless. That meant I had to disconnect the power wire from the 5V input on the board, and connect it to the power pin on the ESC plug. Which I had covered with hot glue. Aaaagh! I came back to this after we'd finished watching Hook, so it was around 6pm, and it took me a long time to burn off enough hot glue with the soldering iron so that I could actually get the solder to flow and the new power wire to connect to the pin.

Wow. OK. More updates. What a journey. Wow. So this began on Friday morning. I usually work Fridays, but it was my birthday last week and my new employer gives me a day off to take around my birthday. Since it was half term, I had planned to take the family on a trip to the coast for fish-and-chips and playing in the sand, but a week of warm, dry weather broke on Thursday night with non-stop rain, so I got a bonus half-day in the workshop instead, and a cosy afternoon cuddled up on the sofa watching films with the family. I was keen to get started on this rally truck, so I was in the workshop bright and early. The thermometer read 11.4 degrees C - that's significantly higher than it's been recently, positively balmy, in fact. Almost T-shirt weather. This was the joy that awaited me. An Arduino Mega 2560 in Embed form, a 5V 20W voltage regulator that should provide enough wallop to keep those 4 torch bulbs powered, and some TIP120 NPN transistors that can theoretically handle all 3 of the amps required to light those bulbs. I knew this was going to be an adventure. I just didn't know how much of an adventure, or how steep a learning curve, it would be... In order to work out where to place the board and all the peripherals and connectors, I could have used something like Fritzing, but TBH I find it clunky and awkward and I can never get the cables to route properly, so after a few goes I gave up and decided to just wing it. I started with the biggest slice of prototype board I have. I use this instead of strip board because I find it easier. Theoretically strip board is neater, but I struggle to get my head around it, and scratching off the copper when I want to break a line is a chore. I stuck the Arduino at one end and began soldering connectors for the control stuff (ESC, servos, Rx) directly to the pins. It looks messy, but don't worry, it gets neater. No, wait, no it doesn't... Even with my little soldering station, getting into these tiny pins is a pain. Working out what sequence to solder them in helps a lot, but if I make a mistake and have to change one in the middle, that's a major bummer.

If you're talking about music I was listening to when I got back into Tamiya as an adult, then that would have been around the time I was renting a shared home with my cousin, and I mostly would have been listening to: Infected Mushroom - Vicious Delicious Future Sound of London - Lifeforms Fluke - Progressive History X although my tastes are fairly eclectic (if not exactly mainstream) so it's possible there was some rock, metal and britpop in there too. If you're talking about when I was first obsessed with Tamiya, then... CDs? What are CDs? Back then I'd have been listening to cassette tapes, such as: Mike Oldfield - Tubular Bells Jean-Michel Jarre - Oxygene Erasure - Chorus These days, on any given Workshop Sunday, my phone gets plugged into the workshop amp at 9am and doesn't get unplugged until I come in around 5 or 6pm. Just this weekend I listened to, among other things: Be Svendsen - Live at Mount Nemrut (Cercle) Leftfield - Leftism Beastie Boys - Hello Nasty

My first thought was that they look painfully like Sweep M-chassis wheels. I can't be sure, because there's not enough contrast around the rim, but I can just about make out the spoke shape behind the front right wheel and it suggests they might have the same tight radius at the outer end of the spoke, like the Sweeps. Looking at the carbon fibre and incredibly neat packaging, I'm assuming Grahho is the same person as Grahoo, who posted some fabulous carbon fibre custom builds on here around 10-15 years ago? If that's the case, it wouldn't be beyond his ingenuity to graft a set of M-chassis hubs into some bigger buggy wheels. Alternatively, whoever makes the wheel for the Sweet tyres may also make a buggy wheel in the same design.

Absolutely. I have no experience myself but my mum had brain surgery 25 years ago and has also had spinal surgery. A few years back I randomly started reading Do No Harm by Henry Marsh, a British neurosurgeon. I was blown away by what can be done. A thoroughly moving book.

Wow, that's tough. I woke up feeling bad this morning that I had to pay for my own birthday meal out last night, so I'm now penniless until payday. But that really puts it in perspective. All the best with your recovery, and we hope to see pictures of your new project here as soon as you are able (and no sooner)

At one point I started designing a 3D printed 1.9 touring wheel, and although it was very early stage the wheel came out fantastic in resin. I need to redesign it to grow directly from the foot, so it doesn't have all the support stuff on it, but I think it should be good enough for drifting. Sadly that was years ago and I haven't had the enthusiasm to go back to it since. Potensh could print it in a hard plastic, or there are some tough resins out there these days.

Yeah, I've thought this for years. Cars should look like cars, and wheels should look like wheels. Proper spokes and detailed mouldings. I intensely dislike flat and dish racing wheels, on buggies or touring cars, and Tamiya's efforts really aren't much better in the buggy world. I guess the Dark Impact style wheels are kind of scale, in that they look a bit like an OZ Racing wheel, or a modern F1 wheel with a wheel cover, and in some futuristic buggy racing series they could be a spec wheel to aid brake cooling or aerodynamics, but right now I want a wheel that looks like a wheel. The early 3 piece wheels were great, but too small for today's buggies, so let's have something like that, or a Fox style wheel, but for modern rubber, please. I use JC Racing's 5-spoke wheels on my Cougar. They completely change the look of the car. I detest the original Schumacher conical wheels with a passion

So, what's next? Wiring up the Arduino has got to be a priority, I think. The new Mega Embed board arrived last week, along with the voltage converter and high-amp transistors, so I can safely power and switch the torch lamps. The design is probably going to take a while, but I should be able to plonk it all on a prototype board and wire everything up so I can stick the board to the inside of the body and plug in a load of servo leads to run everything. And then we're done. Details and paint, which will take no time at all. Probably only a few years. There are so many details still to add...

Well, that's really not bad for a day's work, especially with 3 hours removed from the middle! I now have solid steering, a winch servo ready to plug in and route up, a roll cage that properly closes and a transmission that shifts.

Fitted. Steering geometry is perfect.

The standard servo mount is 7mm deep. The closes I have is 2 pieces of 3mm aluminium sheet. Which meant making the same part twice. I'll let the pictures speak for themselves.

It wasn't far off going-out-for-a-meal time, but I decided to start work on the next big thing, so that I had some momentum for when I came back later. The next big thing was fixing the steering servo mount. I noticed when I was bench-testing it the other week that the steering servo wobbles all over the place. It's a heavy, heavy rig with grippy tyres, so there's a lot of force on that plastic servo mount, and it's obviously not up to the job. With the entire front-end stripped down and the servo mount removed, I saw the problem. It looks like it's sort of still hanging on there, but it isn't - that adjustable end is completely snapped away. thus I came back after lunch and decided to think about the winch servo. I've always planned to have a winch on the front of this rig, partly because it might be helpful if I get stuck while out on a tough trail section somewhere but mostly because winches are just cool. Fortunately there is exactly the right amount of space to get two servos in between the frame rails, in front of the transmission mount, and behind the front crossmember. Perfect!

So the solution, I decided, was to use the bumper as a template to make two holes in a block of wood, so the spacing of the crossbar is exactly where it needs to be, then re-solder it. except, when I tried it, it was still too tight and wouldn't fit. Gah. Obviously the best way would be to solder it while it's fitted in the bumper, but the torch flame is hot enough to melt aluminium, and I didn't want to risk that. But I did find a piece of scrap U-section that was big enough to be a heat shield. There. Done. Literally as I applied the heat and the solder softened, it dropped neatly into place and I was able to re-solder it. Except after it cooled it wouldn't fit again, so I decided enough was enough, and I overdrilled the holes in the bumper. After that it was a perfect fit - easy to slide in and out, but enough tension to stay in place while driving. Perfect.

The result being, it won't fit in the bumper. In theory, the crossbar should rest on top of the bumper, but it was very marginally too narrow. In retrospect, if I'd put a very slight bend in the crossbar, it would have given some flex to the legs and also added some tension to keep it in place when fitted, but, I didn't think of it until it was too late. So, to get it to fit, I decided to try cutting the crossbar, then re-soldering it. Here it is chopped: One option was to put a sleeve over it, but I don't have anything the right size. This aluminium tube is too big and rattles around.

When that was done, I turned my attention back to the roll cage. I've decided to use shock chord to attach my new banner to the roof when I finally get around to ordering it, which means I don't need a slide-on-slide-off option for the roof. So I can braze on some rear crossbars to add more rigidity. Here you see again where my brazing skills lack some finesse. When the joint gets just about hot enough for the solder to flow, the rod is soft enough that it won't retain its shape. The vertical part has sagged (I actually caught it with my glove and moved it roughly back into shape before it completely fell) and it's twisted out of shape near the cross bar. So there's always some manual re-forming to do afterwards. I might check with the solder supplier if there's a lower-temp option for brass, or consider moving to steel rods for making cages. Finished the top cross bar now. nasty wobble A bit of brute force and ignorance puts it right Now for the lower crossbar. I very, very carefully measured and cut this one because it must be exactly exact, otherwise the roll cage won't fit in the bumper.

Next problem - the body is too big! I had to file the bracket out. It didn't take long. Getting it all back in was a pain. After that I had all manner of issues getting it to work properly. The gear servo runs off a special "aux" channel that I've configured off the arduino, which basically just takes whatever input it gets and passes straight through with no change. That means I can use the endpoint adjustment on the transmitter to fine-tune the servo travel, without having to plug it into the laptop - that means I can make adjustments on the trail if the engagement isn't right. The new servo horn was slightly different in size to the old one, but the old one wouldn't fit the new servo splines. Everything was binding up and stopping travel, and the rod was jamming before the selector selected. We were pretty close to a teddies-out-the-pram moment before I relented and bent the rod with some pliers. I was concerned it would snap, but it was OK. All working smoothly now and selecting positively.

More updates! Sunday was a split day, as I had to come in at midday to get showered and changed for a meal in town, then back to the workshop in the afternoon for more work. I wasn't really sure I'd get much done, but as it happened, it was one of the best days I've had in a very long while. To begin with, I wanted to fix the shift servo. Annoyingly, the servo has been largely redundant since the transmission was installed because the gearing has been too high. It's always in low gear, and it never changes. Then I installed the new reduction gear a few weeks ago, and managed to fry the shift servo the same day - so I haven't been able to test it properly. I placed an order a week or so ago for some parts from Aliexpress, including a cheap micro servo. I went for a metal body servo which promised to be higher torque. It's an unusual fitment, the servo being clamped in place by a metal bracket. Getting at it doesn't look that hard, but in the end half the truck had to be disassembled. Note: tilting rear isn't on the build plan, it just happened to work out that way. Final version has too many screws and won't do this. Kinda makes for easier servicing, though! Getting the servo out is a pain. Slacken the grub screws, then undo the long screws that hold the bracket to the transmission. Here's the new servo. Note the lugs have been cut off the original servo, which is a hacksaw job on the metal body servo...

That's awesome! Great work so far