Tuesday, December 27, 2016

Give me a Brake

I'm excited about this post because it's been a long time in the making.

When I started this project over 2 years ago, some of the parts I purchased along the way were experimental. For example, I found a go-kart brake kit for an off-road go-kart called a Hammerhead, and the kit was affordable. I think I paid $80 for the entire thing, which included a master cylinder, a rotor, and a caliper with line attached. At the time I didn't really inspect it that thoroughly, and just assumed it was fairly standard as far as brakes and calipers go.
ZJ Brake Cylinder and Caliper (w/ my fabricated bracket)
 After installing my rear axle, with the ZJ rotor attached, everything seemed to be in order....But that was totally not the case. In fact the hub for the rotor was interfering with the caliper bracket which I had fabricated from some angle stock. It took a few dozen tries of attaching, grinding, re-attaching, re-grinding and fitting to make the caliper sit in a spot that would achieve any type of clamping force. But there simply was not enough contact between the rotor and the caliper. In the pic below you might see evidence of grinding and cutting the aluminum hub.
Bracket not so much...
Unfortunate choice of Rotor Hub
The rotor hub was a problem. It was originally intended for a totally different application (drive sprocket), so the section did not really work with the caliper bracket. In the pic below you can see how close the bracket was to the hub, and why it caused problems. As the brake rotor and hub spun, it had the unfortunate tendency to interfere with the shape of the caliper. A new solution was inevitable. Later I learned that these calipers are meant to be attached to an axle bracket, usually not much bigger than the axle itself.
Section of hub/rotor vs. caliper/bracket
It soon became evident that the brake system that in my mind would perform sufficiently, in design was the wrong application. So I took a step back to evaluate where I was and where I wanted to go. I was close to completing the functional parts of the chassis at this point so I was a bit frustrated, which was exactly why I needed to step back and take a broader look.
Not quite ready to test yet...
I retreated for a few weeks, and gave it some serious thought. Thats when I realized that the brake I scavenged from the Rotten Tuna was still in the spare parts box. I pulled it out, cleaned off the 2-years of grime that accumulated on it from sitting in my friends back yard, and not surprisingly it worked perfectly. But the compression ferrules were shot. I needed new ones - and the hardware store does not carry this particular type of cone-style ferrule with a flange, it's an aircraft grade part for AN-3 sized lines.
Ferrules...old vs. new
If you are not familiar with the ArmyNavy standard thread sizing have a look here:
More or less it's a brass 3/16" inside-diameter cone-ferrule with a compression flange.
Then I had the choice of using the traditional MCP nylon tube style lines, or doing something a bit upscale and choosing a braided line. I was able to find a 60" line from 'Allstar Performance' that did the job well (see pic below). Also in this pic is a new 8.25" MCP rotor, with a 4 point hub.
new parts
So there was a bracket that needed to be fabricated for the caliper which was a small project in itself, and then another bracket for the master-cylinder. For the Caliper I chose steel. I had some leftover angle iron...some cutting and welding and then had a bracket.
Steel Brake Caliper bracket in place
New Brake caliper and Rotor assembly
Then for the brake master cylinder bracket I didn't have much steel around but plenty of aluminum so I decided to try brazing. For small areas, that are not-structural, brazing works good on aluminum but I would not rely on it for anything that is a major structural part. Brazing Aluminum rods rely on a temperature of around 800º on the surface of the metal to achieve phase change. The result is a strong bracket that is very light.
MCP master cylinder and DOT5 brake fluid
New brake pedal in position
One thing to mention here is the pedal itself. Usually go-kart pedals are attached to the master cylinder using an adjustable tie-rod, in case there are small adjustments needed. Since this is going to be a binary system, I decided to attach the pedal directly to the master cylinder. This meant that I had to use the same pivot for both the pedal and the rocker on the master-cylinder. It already has a good pivot, although small, so after making a larger hole it fit a 3/16" shoulder bolt easily.

Full view of Brake Assembly

Wednesday, December 14, 2016

Sprocket and Brake Balance

If I get it right, the weight of the axle will be balanced from left to right. Counting the 'drive hub' as weight on one side is part of the equation, as is tension on the drive sprocket from the chain. The holes I cut in the floor of the chassis are wide enough to adjust slightly. Thinking about it now as I write, it may have been easier to do this on a bench prior to mounting everything, but I'm not that patient I guess.

In any case this pic shows what the stuff looks like without the engine in the way. -CW

Early Changes

Here's an older pic of the early layout prior to attaching things.
  1. The front axle is too wide, and the rear axle bearings are not mounted. 
  2. The steering rack & pinion are not located yet, 
  3. and the rear bulkhead in front of the engine box is in the wrong place. 
I made the wood chassis in a quick short burst during the summer to try and get things off-top-dead-center, and made mistakes along the way. As evident by the multitude of connecting arms laying in the front part of the chassis, I was hunting for the proper length. Like an idiot I ordered way too many of the 'kits' when I could have easily purchased a single rod length at a time. It ended up with 13.5" rods.

The oversized front axle length was a cause for concern. I purchased a 44" aluminum tube with a .375" wall hoping to trim and weld it. After it was welded the guy who did it asked me if I thought it was too long...smart ass. Yeah it was too long! So began the careful plotting to make it narrower.

The result was this spliced tube, and a 5/8" rod. I spliced it all back together and added some set-screws. It needs to be re-welded at some point (Stephen if you are reading this there's another $60 in your future), but for now I think it's OK. It's tough to tell here, but I thought I was good with these steering linkages, but they were way too short. If you look closely at this pic, it shows the threads on the HEIM joints completely exposed. Like I said, getting the right length rod at this point made all the difference. -CW

Monday, December 12, 2016

Engine Stuff

If you feel bored one day, take your engine apart. It's a good lesson in how to put things back together properly, with the correct torque values, and will give you more confidence when it comes time to modify. One thing to note is that the accurate torque values of the bolts on a Harbor Freight 212 Predator are tricky to discover. PM me if you want the proper numbers - the forums may be misleading.
Engine Apart

The Harbor Freight 212 Predator is also very easy to completely mess up permanently. In this picture from right to left top to bottom are the individual assemblies:
  • Torque Converter 'clutch as driven' can be ignored - its just in the picture
  • Case cover with bolts
  • Head cover with pulse valve
  • Crankshaft
  • Carburetor spacer w gasket
  • ARC flywheel and coil
  • Cam (shown here is a custom modded cam)
  • Carburetor
  • Main case with governor and oil sensor removed
  • Stock Piston and connecting rod
  • ARC billet connecting rod and 'flat' piston
  • Stock head with 18lb springs (yellow stripe)

The 212 is 1/5th the price of a real performance engine like the Briggs Animal. So for beginners it's a good choice.

And is based on the original (and legendary) Honda GX200, the Predator 212 has become the 'enthusiast' engine.

 The GX200 is a precise, reliable tool and is the most consistent of any small engine you can buy. Which is why it's used for more than any application in the world. Snow-blowers, generators, concrete mixers, you name it. If you want alot of torque and not alot of weight, this is what works.


If you have the means, or if you just want "the best" then the Briggs & Stratton Animal - World Formula is the engine you want. Purist's will say that its still parts fabricated in China just like the Honda and Harbor Freight blocks (possibly from the same casting companies). But the difference with the World Formula engine is that it's hand tuned by B&S racing. Suffice to say, the most powerful and reliable small racing engine that you can buy. You can option it with or without an electronic starter among other features. With some minor tuning it can generate 18hp at 7100 RPM. This thing is insane. -CW

Sunday, December 11, 2016

Shop Stuff

my romance with the drill press
Once in a while it becomes obvious that the tools being used to build these things can make or break the fun you are having. Back in 2014 I was on the hunt for a drill press. So I got on craigslist and hunted around, to find a 3/4 hp stand up press for about $700. This one had a solid construction and came with a 'speed chuck' and after a few uses proved its worth.

I don't recommend using even the best drill press for engine parts however, use a Milling Machine for that. I remember watching videos of a formula one fabrication shop using a drill press with no locators or punches on an all-aluminum block and thought 'damn that's awesome' but since then learned that was all for show. -CW

Engines, Rods, and Flywheels

When I started this project my goal was to use replacement performance hop up parts to make the 212 Predator into a beast. My plan was

  • Replace the stock flywheel with an ARC billet flywheel for 212cc Predator
  • ARC billet connecting rod for 212cc Predator (not honda clone rod - its too long)
  • Increased compression piston
  • 18lb valve springs
  • Pulse valve head cover - for fuel pump
  • Number .036 carb jet (recommended)
  • mesh style oiled air filter
  • Oil sensor removed
  • RPM limiter governor removed
  • digital RPM meter
  • ARC throttle plate
  • Exhaust manifold
  • ARC billet Flywheel
ARC Billet Flywheel

Which was going to produce (in theory) 8-9 hp. I didn't want the power of the engine to put undue stress on the torque converter or belt. In the past I have seen damage done by not having the correct wall stock or pressure tension on bits of the Comet TAV2 and other versions. What happens is the bell-housing on the 'clutch-as-driven' tends to bend at the keyway. There are lots of issues with rotational velocity regarding the belt and the chain also. Eventually having a 25hp engine will cause things to disintegrate and fly about causing all sorts of mayhem and potential hospital visits. I prefer to avoid all that.

All together now
 There is a tale of woe that I could write here about the first attempt made to install all these parts but I will skip that for now. Instead here's what the engine looks like installed, 4 mounting bolts, and 8 bumper cusions above and below the chassis to eliminate vibration. The whole engine tends to rock a bit under torque, so like in the past I need to install a tension bolt to keep the engine from rocking towards the front left corner when the power is on. -CW

Rear End Development

If you saw the rear axle in my 3D model, it shows that the drive sprocket and the brake are positioned symmetrically opposite on the centerline. This is an attempt to balance the axle while spinning. As it turns out the brake is much heavier than the drive sprocket, so realistically this wont happen...I need to recalculate the rotational velocity of the axle and the moving parts to find a center, or add ballast to the drive sprocket.

The drive sprocket is aluminum in this case and weighs about 3.5 lbs. The brake is heavier and weighs about 8 lbs. Its not a critical item at the speeds this vehicle will move, but if I decide to 'level up' and make a full car out of this design I will need to balance the axle to keep the axle from feeling the effects of torque steer. Without a 'cush drive' on the rear axle, the tendency to torque steer is increased, and having such a short length of chain isn't helping. The longer the chain the more tendency to stretch though, and lose power.

Here is how the back end came together.

its only wood...
\When I first put the engine in place, my idea was to sit the block on top of a plate of steel or aluminum above the axle. As things progressed, I realized I could sink the engine lower and get less height over the rear body cover. What you see here is me removing the bulkheads from the back end to make room for the new engine placement, and axle position. I drilled a single hole here to get started, then made a hasty retreat to regroup and come up with a better plan.

At this stage you can see the idea of where the drive sprocket wants to live (left-hand-side) because the Torque Converter jack sprocket lives there. This pic also shows how low the engine sits in the body, and relative to the axle. The engine bearing sits almost at the same height as the rear axle (just a bit lower) which is perfect for my rotational-momentum forces. If I get it right, the back end will accomplish a minor gyroscopic effect keeping things more balanced at high speeds than normal.

I did not want the sprocket to be too far away, and I did not want the axle behind the engine. This meant I had to modify the stock design of the torque converter. This isn't the first time having to do this, here's a pic of the previous kart's TAV2 and how it was modified. James M did some trimming on it with his magical milling devices:

Torque Converter Mods
Most Torque Converters come with a mounting plate that is designed to have an axle behind the engine. Frankly I think that sucks. In any case the solution to my problem is to compromise. In order to have a clean path for the chain, the casting material on the torque converter needs to be cut away which you see in this picture. The path of the chain needs to be clear of any sharp standoffs, the chains tend to flop around when braking, and accelerating abruptly or even at the same time. We chop alot of chains without having a chain tensioner - another thing that I will talk about later.

New Brake Disc
Fast forward a bit to the latest mods. You can see that the chassis has some wood structure added to follow the shape of the body, which required some creative trimming. Point being the brake disc as it sits now is PERFECT at a whopping 8.25" and .25" thick. This guy will handle the heat and provide some serious stopping power. I picked up this bad boy from MCP which is in Ohio.

You may notice a bit of oil staining and wood burn marks in the area that the brake sits. This is because previously to this photo I had a different brake setup, which consisted of a $%!& brake system from something called a 'hammerhead' go kart. Needless to say that did not work. No need to bore you with why - unless you have one and are trying desperately to make it work. PM me and I will give you a history lesson. (eventually i may write a post-mortem on that thing).

And here is a detail view of what happens when you buy 'off-the-shelf' parts and need to retrofit.

This isn't the ideal situation for mounting a flange bushing, bare threads in wood with a few washers are not my idea of a solid mount. So more than likely I will need to put a tie-rod under the chassis once I get everything located. For now this is a temporary fix to get everything lined up. Plus I dont want the tie-rod to destabilize the rear end chassis until some of the other bulkheads are in place, it will cause the engine and back end to sag. - more to come on that story later.

 Here's a rare 'top view' of the rear axle and engine layout. I say rare because I dont see too many builders exhibiting the layout of the engine to axle, in a construction view like this.

Here is a word picture of the bits on the axle from right to left. A1.5" aluminum tube axle has the following parts connected to it:

  1. 1" snap-ring for 1.25" axle
  2. 1.25" split locking ring collar
  3. 1.25"-ID : 1.375"-OD x 3" wide Hub-to-axle spacer
  4. Cart Wheel (no keyway)
  5. 1.25" split locking ring collar
  6. Aluminum 8" axle spacer tube, 1.25"- ID
  7. 1.25" flanged set-screw axle bearing (no keyway)
  8. 1.25" split locking ring collar
  9. .25" x 3" key stock
  10. 70-tooth 'split' aluminum sprocket with keyed 6-point locking hub
  11. 1.25" split locking ring collar
  12. 1.25" split locking ring collar
  13. .25" x 3" key stock
  14. 8.25" MCP brake rotor with keyed 4-point locking hub
  15. 1.25" split locking collar
  16. 1.25" flanged set-screw axle bearing (no keyway)
  17. 1.25" split locking collar
  18. Aluminum 6.5" axle spacer tube with notch for keyway
  19. 1.25" split locking collar
  20. .25" x 1.25" key stock
  21. 1.25" ID Martin sprocket Sintered-Metal hub adapter w 4 through bolts
  22. Cart Wheel (no keyway) with 4 holes drilled for the 4 through bolts
  23. 1.25" split locking collar
  24. 1" snap-ring for 1.25" axle
And here is a different view of the same thing:

 In both of the above pictures it shows a chinese hydraulic brake that I found at a local lawnmower fix-it shop, that is common on minibikes and micro ATV's. I mounted it poorly and it didn't give me the brake pressure I was expecting so I changed modes to the original MCP caliper.

But you get the idea. The problem with the ZJ caliper is that its meant to be fixed to a tube flange with 2 flange bolts very VERY close to the rotational part of the brake. I had to modify the hub and the brake disc to try and get it to work, but failed. The MCP brake is superior in every way for this application - more to come on that soon. -CW

Saturday, December 10, 2016

The Rest of the Front End

Azusa makes plenty of high quality parts like this 10" round steering wheel. Since I want a vintage look I don't find the standard molded-PVC-over-zinc-plate look appropriate. So i decided to wrap it with something called 'suede-cording' which comes in 20ft rolls off Amazon. If i get some time I will strip off the zinc on the spokes and paint it black, the polished finish doesn't float my boat.

SIDE NOTE: At one point I had a persistent thought about doing a custom wheel (as many CK builders have done in the past) but then realized my time was worth more than having my own signature wheel, and that a few bucks would give me 90% of what I need...stress free. Maybe if I get some extra time in the future it will manifest itself into a custom wheel. For now i'm good with this design.

I managed to find this quick-disconnect steering wheel hub bracket on BMIkarts.com for $33.00 USD. I like it because it's solid, well built, and affordable.
Longacre Quick Disconnect

This feature is necessary to get in-and-out of the front part of the kart chassis without removing my legs. Its also a good safety measure to have a removable wheel in the event that your steering shaft decides to find it's way into your chest cavity. The steering column is a 5/8" rod that is slightly out of dimension so what you see here is prep for grinding off about .1" of extra material that's keeping the flange bearings from fitting.

In any case, this is what the rest of the front end looks like. -CW

Steering Rack and U-Joint

ARC Racing is a Go-Kart racing fabricator and manufacturer of high-performance parts in Albany Georgia. These two steering racks are from ARC, available direct from them (arcracing.com) or from several online re-sellsers like motivationtubing.com and bmikarts.com. The two versions differ from a horizontal thread or vertical post. I guess it depends on how accurately your tie-rods are mounted, mine were not so I chose the vertical post edition.

Here is what the 1411 looks like on my chassis. The module is easy to mount with 3/8" shoulder bolts. Also in this picture is a u-joint. Pay no attention to my shoddy woodworking skills, there is not much glory in woodworking in this case - since I plan to use the geometry for the all aluminum version later. Think of wood as a budget substitute for what will eventually happen.

This is a cut and modified u-joint from a $5.00-USD discounted part (Spiderbox GX-150 steering link). An actual 5/8" u-joint runs in the $50.00-USD range, so this was a no-brainer. I had the opportunity to use a better fitting all-aluminum u-joint from a production mini-van steering column but decided that it would fall into the 'rare and exotic' category for those reading this looking for online discounts. Plus my ultimate goal is to show how easily attainable these parts can be, a special part from a production car would cost hundreds of dollars new which is the complete opposite. -CW

Front End 2 : Details

Modified Azusa Spindle
In order to achieve the 'Ackerman' angle, I used some surveying line to measure from the center of the rear axle, to the center of the kingpins. Then did my best to cut/angle/re-weld the link on the kingping. This picture shows the result - which was a half-decent weld in my humble opinion. The Azusa spindles are easy to modify by comparison.

 To get all these loose parts to fit right was a process, of cutting, fitting, adjusting, re-measuring, and most of all patience. Using off-the-shelf parts without any 3D to use to measure requires patience. This view also shows a few key details, like what the spindle cut looked like prior to welding, and the 'washer plate' on the top of the axle and leaf springs. I feel like the leaf springs could have more curve here, so perhaps in the future I will make new ones from scratch that are more intended for this application. For now these work just fine. -CW

Thursday, December 8, 2016

Front End 1 : Go Aluminum

Tubular Steel
So the front axle of the Stevenson formula calls for a tubular steel "drop axle". This usually is fabricated using a tube bender and a welding fixture, or cut and weld angles of tubing. Sometimes builders use square tubing to avoid having to fab mounting brackets, square tubing weighs a bit more so there is a penalty- but it saves time and is more accurate to have a flat surface to mount things to like leaf springs and trailer bolts. If you do some research you may find that vintage Ford designs of drop axles have various sections, designs, and weights depending on the purpose.

This requires a fixture, a welder, and a margin of error. The point is to establish jounce clearance above the axle for the chassis, and to allow a half-round leaf spring (or springs) to cushion the ride vertically. The pros-and-cons of doing this are debatable, suffice to say that this technique works well enough for most cyclekart builders.

Leaf springs are relatively challenging to find in small sizes. But this is one of the fun elements of this hobby - finding your own answer to the problems at hand. If you have the ability to have a local shop oil quench some mild steel, it's easy enough to make your own half-round springs for a nominal price. I leave it to you to find out more. One clue I can offer is that horse-drawn buggy supply companies offer a menu of springs of all sizes...and purposes.

And finally there are modern version of half-round leaf springs that are made from carbon fiber.

Air Trekkers CZ Carbon-Fiber Leaf Springs

Below is the concept for the Type 69 axle, an all aluminum tube axle with kingpin joints derived from a square tube section of aluminum. It is strong, flexible, and will be easy to adjust. In hindsight, it could have been a square section which would have been easier to make square to the brackets. At the end though the round shape gives me the flexibility to change the caster angle at any time - an ongoing topic with cycle kartery.

The spacer in the middle keeps the axle from shifting, I am using trailer bolts to lock the front leaf springs to the axle. The trailer bolts are easy to find at any hardware store.

The measurement for the axle was off the first time I cut it, because I forgot to factor in the length of the spindles. So i had to cut down the length to count for the very long Azusa spindles. To keep the front axle from shifting left-to-right, or twisting, there are two 'set screws' that lock the axle to this spacer collar, which frictions against the 'washer plate' on the top.

Choosing aluminum here saved me alot of weight. The steel axle as specified weighed 10 lbs without any hardware. It was strong, but not light. The aluminum version here weighs 4 lbs without hardware, but includes the kingpins and has a measure of flex.


A Tale of Two Cycle Karts

Steel Version 2013
Current Status 2016

I was at this point in a build before when I decided to fine tune things and make it run as good as possible before 'shelling' with a body. The top image is of the Rotten Tuna which was supposed to be wrapped in a Mercedes Grand-Prix shell, but then along the way I realized that it wasn't what I wanted.

The second image shows where I am today - much lighter, much more stable, and a margin of reliability better. Along the way I learned that the standard fabrication techniques that most cycle-kart builders rely on are advanced for mere mortals. In other words the new design applies the idea of using found objects to a greater degree. The one key item that is different here is the front axle. Doing a front axle in aluminum took some special fabricating techniques that I'm not ashamed to admit were above my skill level.

Wheels Part 4: The Answer

If you want lightness, and are willing to accept the fact that Azusa spindles are the best option then the answer is the 20" cart wheel with a 5/8" axle bore. There is no better wheel available for the money. It's pretty obvious what the advantages are. In addition I have found that the hub bore of this wheel is SAE 1-3/8".

I was able to get 4 of these for $200 from Kingston Saddlery.

And they are very robust. I did some additional research while writing this and determined that there is a catalog of wheels generally referred to as 'cart wheels' that come with 1/2", 3/8", 5/8", and 3/4" hubs. Most of these scenarios come with an outside bore diameter of 1-3/8" which is easy to space away from for a rear axle. YES FOR THE REAR AXLE.

Now along the way I realized that often times the motorcycles, cruiser bikes, and lawnmowers that use this wheel have a flange plate for a disc. Either a brake disc or a sprocket can be added to this wheel by using the space between the bearing and the outside bore. Its easy to accomplish this with a machine lathe and some aluminum stock, but finding a suitable part online is tricky.

I was able to find this guy, and it will work perfectly for any key hub, or brake plate.

Martin Split Tapered Bushing (Sintered Metal)

This will shim between a 1.25" axle and a 1.625" hub bore. Also it is keyed, so I can lock this to the axle. This solves all my problems locking a lightweight durable wheel to a 1.25" axle. This is the answer. -CW