An M.G. Locost Build

What We've Learned
Last updated November 20, 2012

If you've never built a car from scratch before, and I'm pretty sure that describes 99.99% of the Internet-browsing population, building a Locost is a great learning experience. Not only do you learn general stuff like cool new fabrication techniques, you also learn a lot of things specific to building a Locost. Unfortunately everything you learn is from mistakes, but it's still a great experience.

The Locost book describes the build process in amazingly poor detail. Not to be critical or anything. I'm sure if you'd already built a car or two you could pick up the Locost book and whip one out in no time. Anything not covered in the book, you'd figure out for yourself. If this is your first build-a-car-from-scratch experience, you still have to figure everything out, but you have to do it without the benefit of knowing in advance what works and what doesn't.

If you figure something out and it doesn't work, you either fix whatever it was you broke, or make it again, or live with the results. Sometimes you're making two or more of the same thing, and you don't figure it out until the second or third or tenth part. The same options apply, but sometimes you can still use the early examples in less critical applications, or where they won't show.

So in no particular order, here's what we've learned so far. We won't be discussing the mistakes we learned this from, but suffice it to say we won't make them on our next Locost. And of course we'll keep adding to this as we learn new stuff, which is pretty inevitable.


  1. Goggles. You won't need them 99.9% of the time, but you won't know which 0.1% of the time flying bits of metal are going to land in your eye. The bits are tiny and you probably won't feel them until later that evening when your cornea has been sufficiently scratched up. But the intense pain will continue for days until your eye's defenses finally surround the ferrous particle and eject it. So you can press your luck and eschew the goggles, or save yourself a lot of pain and possibly worse. Keep your goggles nearby, or at least know where they are, so you're never tempted to do just this one little task without them.
  2. Gloves. Another force of habit thing. You want to get to the point where picking up a piece of metal with your bare hand feels odd. Gloves mostly save you from burns, but metal can be sharp, too.
  3. Ear protection. I have these giant headphone-shaped ear muffs. I'm not as worried about my hearing as I am my eyes or hands, and nothing I'm doing in the shop is half as bad as the thousands of hours I spent at rock concerts standing in front of giant speakers, but the ear muffs are a comfort when I'm grinding metal.
  4. Clothing. I always wear long sleeves in case I'm going to do any welding, but it's good protection for everything, kind of like extending your gloves.


  1. Nothing works better than a band saw. But band saws are expensive and they can be dangerous. The chop saw works fine for straight cuts, and the cutting edge doesn't have teeth so it won't easily sever a finger. Still painful, though.
  2. A Dremel works okay in tight areas, but takes an awfully long time to cut anything. A cut-off saw or angle grinder with cut-off wheel is like having a mini-chop saw in your hands. It won't do curved cuts well, though.
  3. I have a Craftsman scroll saw that I use for all wood cutting. It works okay on aluminum up to 1/8", and steel up to 18 gauge, but you will go through blades, if you're like me and you buy blades made for wood. It makes any kind of curved cut you want.
  4. A hacksaw is a fine tool for cutting out misplaced tubes. It gets into fairly tight places, and unlike an angle grinder, a slip won't take a huge bite out of neighboring tubes. It's slow and takes some effort, which is just the ticket when you've discovered that tube you just permanently welded is an inch too far to the left. You can wail on the hacksaw as hard as you want.


  1. An angle grinder with a grinding wheel is great for taking off metal in a hurry. It's not great for finishing a surface, and it's all too easy to take off metal you didn't want removed, like the next tube over.
  2. The best use for an angle grinder is sanding metal. The 50-grit discs made for metal will sand for several hours, and will remove metal almost as fast as a grinding wheel. But save it for finishing, where it'll do a nice job without taking off too much of the metal you didn't want removed.
  3. Even better is a flap disc. These have a couple dozen small sanding squares arranged in an overlapping ring. They can be used as either grinding or sanding discs. They won't take big gouges out of the metal, and they seem to last forever.


  1. A drill press isn't just more accurate than a hand drill, it also has a lot more power. Use the drill press for anything you can.
  2. With a drill press, it's a very good idea to put a block of wood under whatever it is you're drilling, unless you're positive the table is centered under the bit. Centered or not, always put a block of wood under the metal when you're using a hole cutter.


  1. Other than the drill press, we don't have any machining equipment, but we've encountered many uses over the course of this build for a lathe. Find someone nearby with a lathe. Of course anything you can do with a lathe can be done using alternate methods, but it'll take forever and the results will be terrible.


  1. If you're new to welding, you can build a Locost, but practice as much as you can before you start on the car. You can't get too much practice. After 500 welds, you'll still be improving.
  2. Two basic rules to ensure your welds will work. Keep your helmet down between welds. Don't let your eyes get adjusted to the light. Second, hold the torch with two hands, brace one hand or arm against something, and use your second hand to help guide the torch back and forth across the seam, pausing slightly at the end points. I know that sounds like several rules, but it's all about ensuring the metal melts on both sides of the seam.
  3. Set up your work so you're always looking at the joint from the same angle. What works for me is to look at the joint from the side, so that the torch will be moving toward me. That way I can see the joint clearly, I can see the angle of the torch, and I can see what the puddle is doing. You do need to take a slight angle on the joint to get some depth perception.
  4. Tack before every weld. Place the tack where you expect the bead to end, not somewhere in the middle. When you're running a bead, tacks are a great place to finish up. Your welds will be both stronger and better-looking.
  5. Get in the habit of dry-running each bead, to make sure you can move the torch easily across the joint and still see what's going on. Nothing worse than getting halfway through a weld and the torch gets hung up on something.
  6. Keep your stickout short. If you're not bumping the nozzle against the work from time to time (and almost always in tee joints), your stickout is too long.
  7. That hollow whistling sound means you've just blown a hole in a tube. The good news is, MIG is great for patching holes. Sneak up on the hole so you've got a good puddle going, then wind the torch around the outside and work your way toward the middle.
  8. If you're concerned about getting enough joint penetration in a particular welding situation, bevel the edges of the joint. This doesn't always work in tee joints, but it can really help with lap or butt joints, especially if you have metal behind the work, like a sleeved tube.
  9. There is some debate over whether you need a 220v welder to build a Locost. It's a good question. What we've found is this: a 110v welder will handle 1/16" mild steel, no problem. It'll weld 1/8" steel to 1/16" sheet with good penetration. It'll also weld 1/8" to 1/8", but not easily, and it'll seriously tax the electrical system in your house. You'll need really good technique to get full penetration in 1/8" steel, but you can do it.

The Donor:

  1. Having a single donor for your Locost is a huge advantage. It doesn't mean every part has to come from the donor, but the more the better. Donor parts tend to fit together better than random parts from random cars.
  2. Save as much of your donor as you can. Don't just grab what you think you'll need. Parts that look old, crusty, or even broken will look way better months or years from now when you need to install them on your car. Seriously.
  3. Take pictures of your donor. Those parts that fit together so nicely in the donor will only fit together nicely in your Locost if you can remember where they went.

Locost Construction:

  1. Keep measuring your frame while tacking it together. Dimensions will change even when it seems like they can't. My favorite measurement is from the rear of each B tube to the forward H tube on the opposite side. These measurements should be around 66", but more importantly should be within 1/8" of each other.
  2. The side tubes on the nose (LB and LC) won't line up square with the frame in both lateral and longitudinal directions. Line them up square with the centerline of the frame (longitudinal). They won't be flat in front, but will instead angle back a few degrees, which is fine.
  3. Fit your donor parts as early in the tacking process as possible. We had a mock-up frame we could use for this, but we still got it wrong sometimes.
  4. Fit your O3 tube at least 1-2" higher than it shows in the book. I think the book has it halfway up the K tubes, but this is too low for just about every known rear axle. Even on Champion's original Locost, the O3 tube had to be "modified" for more clearance in the first test session.
  5. Determine the track width at each end of your Locost before you start building. The rear track will probably be fixed by the donor, and the front should be the same width to maybe 1/2" wider. Because your rear track is likely to be different than the book, the book dimensions for the A-arms won't work.
  6. The minimum rear track for a book Locost is 43" plus the width of one tire. Add 4" for a plus-4 frame. If your track isn't at least that wide, you'll need skinnier tires. Or wheel spacers.
  7. We'd recommend the use of a jig to align your front suspension brackets. It's critical that the pivot axes of the upper and lower brackets are parallel.
  8. We'd also recommend leaving the FU tubes off the chassis until you're ready to attach the suspension brackets. Clamp a bracket to each nose tube at the proper height, then clamp a bracket to each FU tube. Use your jig to align and tack the FU tubes to the chassis, adjusting the brackets up or down so they're at the same height as the nose tube brackets. Once the FU tubes are welded in place, the rotation axes of the upper A-arms will be fixed forever. You want them parallel to the frame centerline. It's probably okay if they're off a little, but it's not optimal.
  9. The lateral angles of the R and W tubes in the rear of the frame isn't 100% clear in the book. I don't think these angles are critical, so long as the W tubes attach to the O tube 4" inboard of each end.
  10. Spend an hour or so every couple of days going over your bare frame with a 3M finishing pad or equivalent. Allowing rust to build up on the frame is discouraging.
  11. We used 16 gauge steel for the rear shock support gussets. This seems more than adequate.

An M.G. Locost Build >> Building a Locost >> What We've Learned


Our Locost
Our Build Plan
Building a Locost
Build Summary
Workshop Manual
Construction Manual
Non-Locost Stuff

Our Build Log