[a_j_p]

So I figured I would wait until after completed to post anything about this, mostly because I have found limited threads about building FS bikes; and a good portion of the ones I can find don't seem to have an inspiring following. I will also preface with this being my first FS build I was not looking to build something amazing, but rather to see if I could do it and start to learn the process. I figured I would post what I did at least if that is somehow helpful to anyone else, and I welcome whatever comments you may have as I generally learn quite a bit through them.

The frame is 150mm, progressive rate, single pivot with a rocker mounted to the top tube and intended for a 160mm fork. I am not the best frame builder nor would I say I have an amazing grasp of bike geometry so I will be brief with this part, but the high level summary is illustrated in the below CAD + LinkageX3 screen shots:



After talking to a buddy of mine who has built many full suspension bikes, I settled on the tubing / steel selection as shown below. I was advised to use a minimum of 0.058" tubing (if not 0.065") anywhere that I pierced a pivot point, and 0.035" straight gauge anywhere I was going to tie in the shock mount.



For bearings, I chose to stick with Enduro MAX bearings because they have a reasonable selection of sizes, are specifically intended for bicycle frames, can be found easily/cheaply, and have higher static loads than typical radial bearings. I used a single bearing at each side of every pivot, with the exception of the main pivot, where I stacked two on each side.

Physically incorporating the bearings into the steel parts of the frame, and being able to jig things properly, in my mind was going to be one of the hardest things to figure out how to do. I opted to experiment with this first before purchasing any tubes/parts/etc. to make sure I had a feasible plan... In the end, what I ended up doing was purchasing 17-4 bar-stock in necessary sizes that I would turn into bearing housings. I started by cutting to length, pre-drilling a smaller hole than needed for the bearing (i.e. a 1/2" hole), and then welding this to the respective tube. After the welding was complete I would ream to final diameter via endmill + CNC.
This process was a PITA and required a separate fixture for every angle and tube size where there was a bearing, but it did work well. I have to imagine that there is a better and/or easier process to achieve this, but without spending a decent amount of money on special reamers (I already own the machine + endmills) or slitting the bearing housing and using a bolt to clamp tightly around the bearing (which I wanted to avoid) I couldn't think of a great solution... Maybe someone else has better insight here.

I will also mention, regarding the bearing housings, that to keep the width minimal for clearance around the cranks I did 'squish' the ends of the tubes to ovalize them. This probably wasn't necessary, but it was easier for me to machine soft jaws to ovalize the tubes and guarantee I had clearance than risk interference issues between the main pivot bearings and the front chainring.



After getting the front triangle welded together (same process and 80-20 jig I have used for several HT frames) came piercing the holes for the pivot points. With what I have available, I decided the best option was to make aluminum tube cradles (with constant tube centerline height regardless of tube diameter) and place the frame onto the mill table for both locating & boring the holes. Getting the frame oriented in a position where I was confident I could locate the hole positions was tedious, but did work.

For actually boring the holes, I started with a 1/16" undersized hole saw, and followed that up with an endmill to final diameter. I opted to use an endmill because I have never had a holesaw actually cut it's intended diameter (actually in this case I needed clearance to slip in a 5/8" diameter pivot axle and the 9/16" hole saw almost gave me that) and I wanted the fit to be as tight as possible to minimize my chances of pulling out of square when welding. This actually worked out very well but if anyone has a simpler way to do this, please advise.
apparently I can only add 10 images to a single post - will continue this in a reply below

 

[a_j_p]

A shot of one of the pivot axles welded in:



Moving onto the rear triangle, I did put together some scrap 80-20 into a new jig for locating the yoke (all other jigging was done in my standard frame jig). Because I used 17-4 for the bearing housings, I opted to press in the bearings and leave them there for the remaining welding; I also opted to just mask those areas during painting vs. remove the bearings. Once I had the yoke welded to the pivot, the rest of the chain stay-to-dropout portion was very similar to what I have done on several hard tail frames.



The upper portion of the swingarm was a little more difficult, but mostly in trying to figure out the process I wanted to assemble/jig things. I opted to make the rocker and attach it to the frame, along with a 3D printed spacer to hold it in the correct position. From there, I could bolt the bearing housings in place and miter the tubes to fit in place. Actually for this jigging process, when I pre-drilled the 17-4 stock that I would turn into bearing housings, I drilled them to slip-fit over the bolts (in this case ~10mm) and used that to hold them in place while I got the miters correct. I was anticipating issues after welding/reaming the housings to fit bearings that potentially one side may be longer than the other, but it turned out to work great.



The shock mount was next. The frame was once again bound in the jig and the shock was pumped up to max pressure, bolted into the rocker and used to locate the shock mount position on the downtube. I also placed an aluminum block between the shock mounts and bolted everything together in effort to maintain correct width and prevent distortion during welding. Once tacked to the frame, the shock was removed but the aluminum block I left until the welds were fully cooled.


again I will continue in another reply since I have run out of allowable images to upload

 

[a_j_p]

The final item was one the bridge between the upper swing arm tubes. This likely was over-complicated but seemed like a reasonable solution that I could implement. The most difficult part of this was keeping clearance between both the tire and the seat tube when the frame is fully compressed. In my CAD model, if I located this piece correctly, I would have ~7mm clearance of both when fully compressed - however in real life, I was unsure what I would actually get. The bridge itself was made from 5/8" 4140 barstock, where I machined the ends to a hollow 1mm wall thickness, and then machined the center portion instead resemble more of a 4mm thick bar. With the shock installed, I let out all of the air and used a strap to hold the frame in full compression to ensure I had clearance prior to welding.
cross-section of the part:


Overall the frame went together better than I anticipated. It is likely overly heavy (I am hesitant to weigh it) and probably has flaws in geometry and/or welds, but it was a fun process and has been very fun to ride (I finished it about a month ago and have put many big rides on it since then).

Hopefully this post is inspiring to someone out there - or helpful in some way. I learned quite a bit throughout the process and likely will be improving upon this in the near future. Also, if anyone has further questions, suggestions for improvements, or sees anything that I did dumb or completely wrong please let me know.

Cheers.

 

[Cord]

How much material did you leave in the bores of the bearing houses? Did you then ream them to size, or interpolate mill on the CNC? If you did mill them, what kind of accuracy can you consistently hit? I presume you are sizing the bores to something like N7 for a press fit?

 

[a_j_p]

Nominally I undersized the ID by ~0.033mm, i.e. for a 24mm bearing OD I put the bore nominally at 23.967mm. I believe this would be considered more of a P7 or S7 rather than N7; however I don't have plug gauges in this range so there is likely some discrepancy here. I landed on -0.033mm somewhat experimentally, a good range where my machine consistently provided an acceptable result. I finished all of the bores on the CNC with a 3/8" tool I dedicated solely for finishing bores until project completion. It took a few test parts to dial in F/S, cutter comp, and finishing strategy but once established I just had to make sure I followed the same process for each part.

Likely this would have been much easier if I just purchased the correct reamers instead, and I probably also could have used 304 tubing instead of 17-4 barstock (if I selected my bearing size with stock tubing in mind). But I opted to keep my options open; 17-4 barstock was cheaper than 304 tubing, and a single diameter barstock was easy to adapt to most any bearing size. Also, using the CNC did let me easily control height/diameter of the lip in the housing (see photo below) and create that feature in the same setup as the bore finish.

 

[Cord]

Cheers! And in that picture of the model can I see a bit of the lip missing? Is this to aid bearing removal later? Details details!!!

 

[a_j_p]

Yes I removed a portion of the lip on all of the bearing interfaces to allow all of the components to slide onto/off-of their respective axles. Some of the components (i.e. the upper portion of the swing-arm that is only held together by a single bridge) may have been flexible enough to assemble without doing this, however this was the only way to get the main pivot onto the frame as well as the rocker. Essentially as long as I follow the axis of the cut-away, I can slide the component directly onto the axle and then bolt it together. I don't have great photos of actually cutting the lip from the backside of the bearing housings but it was pretty easy and I designed in enough tolerance that even a sloppy machining job would have done the trick.


The hardest part was matching the width of the welded in axle to the attachment points of the welded-together yoke - ensuring that I had enough clearance to be able to get the swing-arm onto the frame, yet not being so loose that when I tightened things down the bearings would bind. I was nervous about this and actually built a little fixture to center the axle in position when tack welding - but it turns out this wasn't necessary.



Instead of making the centering jig shown above - it would have been easier to just make the axle longer than needed, and put the frame back onto the milling table and use an endmill to cut it to the desired stick-out on both sides. The reason I say this may be easier is because I actually had to do this anyway to face the axles post welding them in. Well I actually had to do this twice, once to face the axles as just mentioned, and once to shorten the main pivot axle when my welding at the yoke connection pulled the two bearing interfaces slightly tighter and I couldn't assemble to the frame. For me it would have been easier to weld an oversize axle into the frame, weld the yoke and bearing attachments together, then set the frame up on the machine to both cut to length and face the axle at once. This also would have saved time machining that centering fixture in the photo above.


I guess for additional details, below is a cross-section of the main pivot axle bearing housing + bolt as it pierces the bottom of the seat tube. This was by far the most critical as far as space-constraints go, there is not a ton of room to fit 2 bearings in a housing, + a bolt between the ST and the cranks + chain ring. I chose the width of the bearing housing based on what I was comfortable I could weld the tube to, and then placed the lip where the bearings sit against in a position that best centered the stack-up. I'm assuming there are simpler ways to do this, and while I don't have first hand knowledge to how Walt (of WaltWorks) does/did this, he did have some decent photos of some of bikes he has made that I used for inspiration posted on his website blog.


For reference, I did make all of bolts that interface with bearings. I would assume that you could just as easily design around some standard shoulder bolt from McMaster, or purchase a linkage bolt kit from another bicycle manufacturer and design around those, but to keep the profile small, weight lower, and to not have to worry about designing around existing items, I did make my own. The main pivot bolts are 17-4 M12x1.75, the minor pivot bolts are 6061 M12x1.75 and the upper swing-arm link uses 6061 M10x1.5. I loosely based the aluminum bolt design around what came stock on my old Intense Tracer.

 

[Cord]

Brilliant, thanks for all the detail. I really appreciate hearing how other people tackle jobs.

 

[Artemus147]

Awesome post! I have yet to tackle a full suspension frame and this definitely got me excited to start designing one of my own. Great work, looks like a fun bike to ride!