JSON Repaint

JSON’s had a tough life. Before I was all that attached with the bike I didn’t exactly treat it with same the respect I do today. It was just a cheap bike to get me from point A to point B. The more I rode it the more I fell in love with it. Combined with damage from the previous owner and other general wear and tear the frame was starting to show some signs of rusting.

Paint damage on the rear dropouts

The plan was to strip the bike of paint, then respray it. This wasn’t going to happen in our apartment, however an excuse to visit Mark in Adelaide presented itself. I took the bike across the border and over the period of several weeks redid the paint work.


At this point I’d become pretty familiar with the process of removing every type of bike part on my bike. Within a few hours I had removed everything which left only the frame. If you are planning on embarking on this adventure and you are unfamiliar with removal of all your parts, get some help or do some research. Some parts can be fragile, some have reverse threads and some can be damaged if force is applied in the wrong spots. There’s also a bunch of tricks to learn for when things don’t go to plan.

An example is the quill stem - often on these bikes they become stuck. If you try to strike the stem directly you’ll likely either damage it or get it more stuck. While it doesn’t always work a good starting place is to leave the stem bolt in, but proud, then strike the bolt with a hammer - this will dislodge the wedge and hopefully allow removal of the stem.

After removing all the parts from the bike frame I decided to strip most of the paint off the bike. Now stripping the paint entirely off isn’t necessary. You only need to strip the paint back far enough for the new paint to stick, along with getting the surface finish to the quality you want it. I went for stripping it back entirely (there were a few bits I couldn’t easily get to so I just roughed these up as best I could). The reasoning for stripping back entirely was to ensure there was no hidden rust - along with removing a nasty paint run was present on the bike already.

I’m really glad I did strip it all the way back during sanding I noticed some weird pattens forming, slowed down an used a higher grit sand paper. To my surprise the red and white paint scheme wasn’t original. I had suspected this given the paint run, however I could never confirm this. I never knew the make or model of the bike and by carefully sanding back the top layer it revealed a uv damaged outline of the original branding. An Apollo Eclipse. This would have been the pink and white model. For me the bike has always been red and white, that’s now part of its history. So while I could return it to original colours, they wouldn’t be original to me.

Sanded back paint revealing pink apollo logo

Sanded back partial exposed eclipse logo

For the majority of stripping I used low grit sand paper however for the fiddly bits I restorted to wire wheel brushes on a cordless drill. Removing the paint is probably the hardest part, once its removed moving up in grit to make a smooth finish happens fairly easily. I went with the sanding approach because sand blasting seemed too expensive for this project and sanding didn’t seem like too much effort (it was a lot of work, don’t get me wrong, but it didn’t really take that long). I looked into chemical processes however youtube tutorials and reviews gave me the impression that it seemed more annoying than just sanding. Everyone has different advice on the process however I settled for about about 400 grit for the final sand working off the theory that while spray paint is thin, it still needs some roughness to the surface to stick well. I used higher grit sandpaper to touch up some parts during painting. If you want to take away anything from this blog post though, think PPE. Not just when spraying paint but when sanding. I generated so much dust when sanding back the frame.

Bike frame stripped of paint

Painting Rust

The next day with all the rust and paint sanded back I went to paint. I thought I’d be able to get away with a day of stripped bike frame in less than 20% humidity but flash rust struck. It wasn’t a big deal however, I was able to sand back the frame with high grit sandpaper again just prior to painting. I gave the frame a wipe down with some acetone to remove debris and now we were ready to paint.


I went with Rust-oleum 2X Primer and Custom Spray 5-in-1 for the paint. This was mostly because I’m familiar with how they spray and work. They probably aren’t the best choice for bike paint, however rattle cans in general aren’t good - this is because the layer of paint is so thin compared to other methods. Regardless it’s something that fit the budget, and that I could easily do.

Selfie of me in full face respirator with bike in background

I actually reached out to Rust-oleum to confirm my planned process prior to painting and they were able to provide a bunch of nice details and provide confidence. First up was priming. The Rust-oleum paints allow re-coat anytime within 1 hour and they recommend coats 15 minutes apart. I setup the forks and main frame near each other so I could do them at the same time. This allowed me to switch between each while waiting. I was able to get 4 coats onto each within the window. The usual rattle can advice applies - thin coats. Sicne the frame is so large I applied thin coats and moved around the frame, usually doing 3 coverages every time I did a coat. Priming was probably the hardest part not to get runs on I got a slight run during this and sanded it back prior to the top coats.

Bike frame that’s been sprayed with white primer

Since the bike was lacking any sort of branding, I wanted to give it some. I had a original plan for that - Crossy JSON - however since I now knew the make, model and what the original decals looked like I incorporated some of the design into my plans. There was the option of simply making some decals and slapping it onto the bike afterwards, but I really wanted to try using the decals as mask. I cut out some new decals for a mask and applied them careful. This meant that prior to the decals being applied I had to do some extra top coats in the masked off colour.

CROSSY JSON decal on transfer paper with thick horiztonal lines to replicate  the feel of the original logo

Mask applied to top tube of bicycle

Some red coats went on. The Custom 5-in-1 cans low output mode really helped prevent any runs. I got a slight run where the rear brake sits but it was easily fixed up. The tricky part however was replicating the “original” white gradient on the dropouts. I did a few practice sprays for this to get an idea of what I was doing along with masking off everything that wouldn’t be part of the gradient and in the end it turned out great - I feel like its better than the previous gradient.

Near finished frame with red, decals and gradient showing

The masked off sections were removed prior to the clear coat. This is where I learnt that my masking approach was probably wrong. A negative mask would have been easier to remove (I created a bunch of damage to the paint getting each piece off) and would have made the painting process easier. Regardless I’m happy with the results given it was my first time doing this sort of thing.

I was originally concerned about runs on the clear coat - because that could have caused disaster, however even on high output mode I couldn’t have made a run if I wanted to. I’m sure the temperature helped but I feel like the chemical composition just makes it really hard to run with that paint. One approach for painting bikes it to actually not worry about clear coats, just make the top coat as thick as possible. I however wanted the finish that the clear coat provided.


Prior to reassembling the bike I decided to give its internals a blast of cavity wax. This is intended to prevent rust from the inside. An idea I picked up from Croker vs ROVER. Most of the internals have drainage holes that the applicator can be fed through.


During painting I throughly washed all the parts, rebuilt bearings and placed everything in Evaporust. The Evaporust worked a treat, however flash rust struck again leaving me to have to repeat the process for many parts. The problem is that with the rust removed, washing with water just creates the perfect surface for more rust. I think the solution to this varies per part, varying from an aciditic bath, zinc plating and painting. However for the time being I went with a light coating of WD-40 (which is apparently the original purpose of WD-40). I think if I do this again I’ll have a think about how I deal with this.

Parts cleaned after evaoprust process - they are mostly clear of rust and look very clean

I also found that many parts on my bike were broken. I ordered a replacement seat as the plastic holding it together had cracked. A spoke was broken on one of the wheels. The new spokes that I ordered were too long (they must of been the perfect size before, as the new ones were less than 2mm longer than the old) so I’ve reconfigured the wheel for a 4 cross spoke pattern.

The chain, bell, and cabling were all replaced. Along with the handlebar tape.


Finished bike leaning against tree with a view of a small lake

While the bike is mostly done now I still need to true the wheels a little better, along with draw up some of the spoke tension. I’m pretty happy with the results. There’s some bits that I messed up or could do better but for my first attempt I think I did well. It’s certainly better than when I started which is the key thing.

My new bike: Polygon Path X5

It feels like starting blog posts with “so I crashed my bike” is going to be a thing. So anyway, I crashed my new bike. Down a cliff, in the bush.

I think this means it has enough wear and tear on it for a review, but lets starts with why.

There’s several reasons:

  • Alex has a cool bike
  • I want to be able to carry more shopping home easier
  • I’d like to visit further away cafes without having to carry my backpack
  • Bike camping? Bike camping
  • I’ve been riding a lot more and can justify the investment

So JSON is great and all. I’m still going to do the rebuild and respray. However JSON can be quite limiting. With no panniers I have to wear a backpack if I need to transport anything. The geometry isn’t well suited for my bad back. I want to ride places that wouldn’t be suitable for JSON.

Alex last year found a great deal. The Cannondale Quick 1. A flat bar bike, nice geometry and Shimano 105 group set for $1,600. My initial thought was “I’ll just copy her”, however it seems that the Quick 1 was sold out in most places.

I went exploring other options. Things that were appealing to me were:

  • Group sets that weren’t low end (Shimano makes this so so very complex)
  • Brifters / dropbars
  • Mounting points for panniers
  • XL frame size
  • Hydraulic disc brakes
  • No suspension
  • Aluminium frame
  • Sub $2,000 AUD

This combination was surprisingly hard to find. The two bikes I did find were the REID Granite 4.0 and the Polygon PATH X5. The Granite 4.0 was out of stock in XL and I could only find the PATH X5 online through bikesonline.

I was extremely nervous about buying a bike without giving it a test ride before however good reviews and a promise that it could be returned in 30 days if it didn’t fit it seemed like something worth trying.

Olive green bike resting against tree with pannier rack and trunk bag installed

I am incredibly happy I did because this bike feels so good to ride. The geometry fits me really well, and it has a Shimano 105 group set, drop bars and hydraulic disc brakes. Assembly was a breeze. Somehow this thing has even survived me riding it off a cliff. Total shipped was $1,834.

The first thing I did was load up the bike with accessories. I’ve splurged a lot here because I want this bike to work for me. I’ve gone with the Topeak MTX pannier rack - allowing for the trunk bag to quickly be put on and removed. However I’ve actually been using Tourbon clip on backpacks. This allows me to stash my laptop and work equipment quickly on the bike and have a decent bag to carry it around the city.

Olive green bike with backpack pannier resting against park bench overlooking the yarra river

For visibility I’ve fitted Garmin UT800 front light and Gardia R300L rear light. I picked these purely because I wanted to play with ANT+ lights. I really don’t recommend anyone spending this sort of money on lights. At least for me the ANT+ integration is a bit clunky and it does feel very much like a gimmick more than a feature. However what does stand out is the Gardia R300L radar. I didn’t realise how much I’d love the radar functionality. It is able to detect approaching vehicles from fairly far away and in a noisy city environment this is super handy when you don’t notice a car sneaking up on you.

Now the bike isn’t flawless, but it’s pretty damn close. The first is mudguards. For some reason the mounting hole you usually find where rim brakes are is 90 degrees from where I expected. This results is mudguard selection being very limited as a lot of mudguards will want to mount here. The other issue is the rims are entirely nameless. I assume that they aren’t tubeless ready as if they were they likely would be advertising it, so if I do go down that route I’ll probably have to rebuild the wheels.

I mentioned bike camping. This is one of the reasons I wanted to get a gravel or hybrid bike. While we have the LandCruiser to explore far away places, I wanted to test the idea of staying at places a little bit closer to home. With VLines fares capped at $9.20 the concept of taking my bike on the train, riding to camp site, and exploring the area without the need of a car seems super compellingly. I’m a bit away starting one of these journeys but I’m getting close, so stay tuned for that.

Restoring new old stock pedals

I ordered some old style bicycle pedals for my bike, JSON. These have a black outer body and a silver shaft assembly. They really suit the bike well and come with toe clips which match the pedals of the bike when I originally received it.

Pedals with toe clips in packaging

Unfortunately when I received the pedals I noticed the shafts didn’t spin very freely and I could feel the bearings biting. These wouldn’t be suitable to ride on. My theory at this point is that these were new old stock and the grease had gone bad. If I could sneak some oil in there maybe they would be fine. There was two problems with this theory - it wasn’t just a grease problem (I should have known this, I think it was just wishful thinking), and removing the dust cover to get to the bearings was a challenging task. The cap was so tightly held in that I ended up breaking the metal body when trying to remove it.

Pedal with cracked housing
I was going to give up at that point however Alex suggested that I should probably ask the store if they would still warranty it. An email was sent apologising for damaging while trying to repair them and I asked if they would still be ok with processing a warranty. They agreed and asked if I wanted a replacement or a refund.

This was a surprisingly difficult question to answer. I wanted this style of pedal, but if all the pedals had the same fault what was I to do?

Searching various places selling these style pedals new they all seemed to share common reviews:

  • No bearing - only plastic bushing - broke quickly
  • Bearing doesn’t spin freely or is seized
  • Pedal broke after short amount of time

I also had a look at eBay for used pedals that would fit my needs. The problem I found was the total shipped price was prohibitively expensive and even if I paid that, there was no guarantee that the bearing races haven’t been badly pitted.

Looking careful I could determine that most of the new products sold likely all come from the same manufacturer with different brands painted on the side. Some models had bearings that didn’t work, some had plastic bushings and some had weak body.

It seems like the pedals I bought fit into the bearings that didn’t work category. Now I don’t know if its because these are old stock and age has gotten to them, or if the manufacturing process is really poor but the bearings out of the box are certainly not fit for purpose.

After a lot of internal debating I ended up with a plan - I would ask for a replacement and attempt to rebuild the bearings.

Rebuild time

The first step was removing the pesky dust cover. This time I took the approach of removing everything from the pedal assembly first. The rear reflectors can be pushed and levered off while the toe cage can be removed by sliding the reflector out and undoing the two bolts. This making it much easier to work on - plus as I was doing a full rebuild I would need access anyway. It’s a good idea to do one pedal at a time so you have a reference pedal.

Main pedal assembly held in vice
I placed the main body into a vice and tightened it softly - the point here was just to have something holding the assembly, so it doesn’t have to be super tight. I then used some very small side cutters edge to slide/cut into the side. Because of the tight circular shape things like screwdrivers don’t work very well here. The sharp point on the side cutter allowed me to get inside and then I could lever it out. It took a few goes, and does slightly damage the cap - however it’s the least damaged from all the tools I’ve tried so far. Other pedal designs usually have a grove you can use a screwdriver with - these do not.

If you are trying to get these kind of cheap new(ish) pedals working you actually have a bit of a choice here. I found there to be two problems with my pedals - the grease wasn’t great and the bearing preload was set wrong. You can choose not to do a full rebuild if you are happy with the grease provided - you can just set the preload. This is the easiest option and probably fine for most people. If this is the case - skip to the section labelled “preload”.

At this point, lay down a towel or a rag. Something that will stop ball bearings from rolling away. Using a 9mm socket and a spanner or shifter on the other end, undo the locking nut.

Pedal with locking nut removed

Inside the shaft will a slotted washer followed by another thin nut/bearing race. I didn’t have any thin walled socket that would fit down there, however since the assembly isn’t under any load this nut should be fairly loose. You should be able to use a pair of tweezers to undo this nut. At this point be careful though as you’ll be freeing the bearings. Sit the shaft upright so that the end that usually attaches to the bike is on the table when removing this nut, otherwise the body can slide off the shaft and drop bearings everywhere.

View of inside pedal showing bearings
At this point you should be able to see the bearings. Now is a great time to count them, or at least take a picture. We’ll want to make sure we put the right amount back in. I removed all the bearings with tweezers and put them into a nice little pile. For me there was 14 on both sides of the shaft. There might be bearings stuck of the bearing race/nut that we removed.

After those bearings are removed I suggest flipping the pedal over and carefully remove the shaft. Mine has a rubber seal on the bike end. Once again remove the bearings.

Desk with cloth down and pedal parts layed out

Clean all the grease off the bearings, clean the shaft and housings. Once everything is cleaned the assembly process is much the same as the disassembly process. I used red truck and farm grease. It’s probably fine. Spread grease into the bearing races to hold the bearings in place. Make sure you count the bearings to ensure none are missing or you haven’t put too many in. Remember to install the race/nut the correct way, followed by the washer and locking nut - don’t tighten all the way yet. Just enough to keep the bearings from falling out.


Getting bearing preload right is critical, and it can sometimes be the most time consuming part of this process. At both ends of the shaft is a effectively a cup and a cone. By tightening up the nut/race we are removing some of the space the bearings have to wobble around.

Ideally we want there to be no wobble or play. However if we tighten up the assembly too far the bearings will be pressing into the metal, causing a lot of friction and preventing the shaft from spinning. Too tight will cause damage to the bearings and the races. The goal here is to have the assembly as tight as possible without impacting the friction of the bearings.

Tighten up the race/nut with the tweezers. I suggest tightening this up until the shaft doesn’t spin freely anymore. At that point back it off until the shaft spins freely again. Now with the 9mm socket, tighten up the locking nut. When you do this, additional pressure will be added to the race. It’s likely the shaft might have too much friction at this point. Undo the locking nut again, and adjust the race/nut to loosen the race. I was doing probably 1/16 turns when I was adjusting mine, so very small adjustments everytime. Check for play as your doing this. The shaft shouldn’t move in or out, and there should be so side to side wobble.

It’s sometimes hard to strike a good balance. I would favour a bit of play over additional friction on the bearings if thats the only choice.

Reinstall the dust cover. I found it quite hard to put in and resorted to the vice. I didn’t want to use the vice as it’ll be putting pressure on the bearings and shaft but I found it the only suitable way to get the cover in. Reassemble the outer body, straps and reflectors.


Was it worth it? Well they certainly spin more freely on my desk, but I’ll have to have them on the road for quite a bit before I can give a final verdict.

In the meantime, here’s my partner comparing one I rebuilt vs the original.