So 2016 begins with the delivery of several smashed frames! A fairly standard mid-range Scott Pro One, which appears to have been broad swiped by a vehicle, although this is speculation. It has two fractures across a seat and chain stay, and also a rather nasty bend to the rear hanger. This should be super easy to mend.
The second frame has a more sever break but is a rather high end S-Works Venge. It was recovered by the Pro who destroyed it whilst racing. His account was that he was rear ended by another rider, the impact was apparently not that heavy but the damage is considerable. Not sure how far I’ll get with this frame. the entire rear triangle has been destroyed, the impact seems to have concertina the frame cracking the top tube and BB area as well.
The first stage is to realign the frame sets, which I’ll achieve with some 5 min structural adhesive, and considerable care and attention. Secondly missing material and the crack line will be filled with a mixture of milled carbon powder and resin. The area will then be heavily sanded to remove as much dead carbon as I can from the area that I will re-apply more carbon. It’s worth remembering that generally the failure occurs to the resin matrix, so I need to make sure I return structural integrity and cover any small cracks that may have propagated. Fortunately the Venge has a very thin black topcoat and cracks appear as white lines, I’m fairly confident that I can see the areas that require attention. This is not the case with the Scott, which has a very thick white topcoat designed to cover up unattractive carbon work and blemishes. The damage however is very different; the break to the Scott seems clean and as far as I can tell, less likely to have caused residual damage.
The damage to the Scott has been realigned and filled.
Likewise this image was taken during the process of realigning and filling the rear triangle. Interestingly whilst Specialized replaced the Pro’s frame immediately they asked him to swap out the seat post, clamp, fixtures and fittings. I’m thinking it will be fun to try to fabricate as much of these replacement parts myself, I may even attempt the seat pillar and a set of bars to marry up to the head tube area.
After the realignment the structural layers of carbon are added and sanded to marry up to the tube profile. The image below shows the tube before the area gets sanded back to receive the top layer of carbon.
After sanding I wrapped the tube with a thin layer of uni directional carbon which unless i decide to paint it, will be the final layer.
And the final layer – it’ll be polished with some wet and dry and a touch of polishing compound.
So because I found a crack on the S-Works at the point the chain stays meet the BB shell, I want to sand back and encase the whole BB area just in case. Also, I feel the finish will be a lot more attractive. I began by sanding the BB shell. The top coat and paint layer are impressively tough!
You can see the layup schedule Specialized used…
I’ve also done a lot of work on the top tube, rear and seat stays.
After sanding back the cracked area on the top tube, adding some bi-axial cloth and three layers of Uni directional carbon, the last layer I chose to add an aesthetic layer which will be repeated elsewhere. I hope the finish will make this otherwise standard Venge stand out!
Working around the rear stays there are a lot of cable holes and areas I don’t want resin to wander into. To these areas I’ve added a release wax and some filet wax to ensure I can remove any waste material easily.
I think I mentioned that in addition to the other six breaks in the frame, I found a crack on the stay joint depicted above. Just to be sure, and to achieve the same finish on either side, I’ve wrapped both sides. I’m less concerned by this crack, I suspect the drop out will be mounted high into each stay. I think it would be unlikely to completely fail. Better to be safe however…
I added layers of bi-axial / uni directional carbon and then wrapped the areas in a carbon ribbon to increase the pressure on these tricky angles.
The peel ply is added before the compression layer.
After a light sand you can see the fix. I’m please with it and confident the area has been restored.
I’m taking some time to cut the carbon as neatly as possible, I find this guillotine blade is really useful.
And the now structurally sound stays, which were once cracked in half… rebuilt! Once the rest of the frame is sound again, there’s just a decorative finishing layer to follow.
The finished seat stays.
I’ve reinforced the carbon around the bottom bracket / chain stay area. Mainly with a Bi-axial (or 0/45 degree unidirectional) cloth. I’m going to finish the area with a wrap of higher quality fine twill carbon. I’ll need to vacuum bag the entire frame for this layer however, compressing around the tricky multiple tubes is not easy.
I spent some time cleaning up the top tube decorative wrap, which I think looks really nice. I’ll also add a boarder strip, trying not to get the original paintwork covered in resin.
What it will look like.
So for the final layers to the bottom bracket area, I thought I’d use the opportunity to re visit the vacuum bagging process. It’s quite labour intensive and to date I haven’t managed to document the process as fully as I would have liked.
The process is used to squeeze the carbon cloth together whilst wicking away excess resin and thus bonding the layers into a composite material fit for purpose. In truth I usually try to complete all the layers that would go into completing an area of the frame within the window of full resin curing – with the intent that the area is compressed into one full cured layer. As I understand it the resin will continue to cure over the course of a week or so, this is the sort of time frame I usually give myself to complete.
Obviously in this case I’m restoring a broken area of composite that is several years old so I’m not hoping to achieve the above. In order to gain a good bond however, all surfaces are washed with a solvent cleaner or acetone and then lightly sanded to both be sure of the removal of grease, and to add a tooth to the surface.
To generate the pressure required to bond the carbon and wick away the excess resin, on this frame I’ve so far used several layers of tape that are stretched tightly. With oven heated higher temperature prepreg carbons, this is often achieve either within a vacuum bag, with the air drawn away by a pump, or with shrink tape that reduces by 5% when heated. A full vacuum achieves 14.7 PSI, however this is difficult to achieve, so realistically 12.9 PSI is the sort of target i’m likely to hit. I believe tape will achieve a lot more than this so whilst it looks a little DIY, tape or latex strips stretched over the surface are more than adequate. Issues arise with this method however when the surface area is not round, then of course equal pressure can not be applied to all areas. In the bike building industry silicone inserts are used to round off or filet areas.
Anyway… the first stage is to cut the bag material. The bag material I bought is sealed on to sides, effectively a long sleeve.
A bagging tape is used to seal the open sides. Once applied it is important to take time to press out and stretch away any areas of kinked bag or air bubbles. A full seal is essential.
So this is the bag… the next stage is to insert the hose that’s connected to the vacuum pump.
The hose connection. The lower portion of the connector punctures the bag is supposed to self seal. I tend to use the bagging tape however, just to ensure a full seal.
The vacuum pump I use.
Before bagging the frame it’s wise to round off any sharp corners, head tube and BB shell. The bag can easily get punctured if you’re not careful.
I tend to fill out the BB shell and head tube with foam inserts.
Once ready to bag the frame, I need to insert a breather cloth that allows the pump to draw the air from the area of the bag the wet carbon is in.
You can see that the breather cloth also wicks the excess resin.
And this is the finished carbon work with a layer or two of top coat. I’ll continue to build up the top coat to get rid of those super small blemishes.
I had to marry up the semi transparent black paint and matt finish but I faded the finish to reveal more of the carbon work. You can see this in the down tube below.
The finished frame. If you look closely you can see the carbon rebuild work around the entire rear end and bottom bracket area.
So you can see in the blow image most of the areas that needed re-building including the crack across the top tube. You can’t see the crack propagating into the bottom bracket area or the crack around the dropout.
Upon completion I’m a bit undecided as to what to do with this frame . After building the TT frame I had wanted to build a full aero bike for myself, but as space is limited I think I’ll utilise the aero origin of this frame and build it up as my main road frame for a while. An all rounder.
I managed to pick up a set of aero handle bars and a TT aero bar for just over £80. Perfect material for a little chop shop activity.
So the two things I want to do are mount the aero bar, which of course functionally offer the rider a complete different riding position, and also mount the arm rests.
I’d be nervous drilling into the current carbon wall of the handle bar, I wouldn’t say I fully understand the structural integrity of the bar layup design. When ridden the torsional forces must be considerable and I don’t think it’s obvious how the design will have countered the stresses. My intention then is to bond/wrap the bar onto the handle bars and surface mount a fixing on the top grip..
I spent a few hours with the bars attached to my current road bike to ascertain a comfortable riding position. And of course this can be altered with the fore and aft position of the saddle.
First job is to strip back the top coat and paint work.
A couple of handy Garmin/computer fixings offer the opportunity to test the position. Also a TT seat post usually has two distinct flip positions to really increase saddle position from normal capacity.
The bonded bars have a really nice aesthetic. I’ve used a 5 min structural adhesive to create the first fix, I’ll subsequently do an all over bar lay up of carbon to both secure the fix and also add some additional rigidity.
This is the kind of finish I aiming for, nicely presented by Livi…
Also, whilst mending the frame, I’ve been trying to purchase the missing frame fixtures and fittings.
More to follow…
I’ve recently had to fashion a new mech hanger to adapt a frame to receive a SRAM red group set. Poor images but…
.a Time Trial Frame.
This is the first commission I’ve received that I’ve agreed to take on. I’ve had a few request in recent years but apart from the commitment over such a long period, I’ve also been reticent to expose another persons safety to an as yet unqualified structural integrity. As I’ve mentioned before one of the key qualities to building for me is that it remains a practical exercise of trial and error, it’s quite a different thing however to exposes someone else to this method.
So the deal here was that the commission was taken on with the commissioner (as it happens an engineer) involved. He would specify the geometry and as much as he could witness the build for himself. I’ll test ride the frame and he can make an assessment before he rides it with intent.
It was a tough build, harder than I expected, with a number of key issues to negotiate. I was interested in frame stiffness and aerodynamics. Weight was always going to be difficult to keep down – it is a large frame with over-sized tubing and I would inevitably want to over engineer the tubes and joints. So I would concentrate on fit, aerodynamics and finish and let the weight take it’s own path. In the end I aimed for a comparative weight to the Argon 18 E-116. By no means a super light frame but at least then my effort would be comparably to a legitimate machine. With the ISP and fork installed this was achieved.
The finished frame commission.
The nervous moments after a year of development and just before a test ride.
I wanted the finish to be exposed layers of carbon without the, as I call it, decorative top layer.
The bottom bracket area… as you can see i still have to attach the front hanger that I’ve just finished making and route cable through to the front mech.
We went for a 1 1/8 head tube to keep the front end profile down. I couldn’t find another TT specific alternative available to order. It’s not clear here but the head tube is profiled to a point to help push the airflow sideways.
A complete donor bike has been bought and these are amongst the other components the beautiful Reynolds wheels that will be attached. The fork we’re looking to install is a Deda TT fork.
The build story…
So with the help of Brian at Impact Cycle http://www.impactct.co.uk/ and Geoff at TrailNet http://www.trailnet.org.uk/ two people I more or less met out riding, this morning we had a bike fit and drew up the geometry for the TT project. Next step is to research and build a curing oven to allow us to use some Prepreg carbons, and also to do some R&D on shrink tapes that compress at high temperatures. None of this I’ve attempted before so it will be a learning curve.
This is the geometry that was generated. It’s based on a Kestrel frame that Geoff has happily ridden for years.
Black epoxy filler, should help with those pesky blemishes – next to the shrink tape i’ll be experimenting with.
This is the chosen rear end.
Glued and mounted the gear stop on the selected TT rear end. It comes in at around 350g as it is. I’ll lose more of that total with a reduction in the mono stay length. The previous rear ends I’ve used were more like 200g.
The seat stays are pretty aero and the chain stays are quite chunky so I predicted it would serve me well. It has a conventional brake mount which we were all happy with, we didn’t want to add too many complications although I suspect it would be pretty simple to mount the brake under the BB area.
I began construction on the seat tube, the tubes have bulkheads running through in an attempt to increase rigidity against torsional forces… at least that’s the theory, at the very least it will increase the stiffness of the foam to allow me to create the really narrow airfoil shape.
As before the foam is shaped on a long strip of sand paper. I created the carbon strip by pressing layers of unidirectional carbon between two sheets of (flat) glass.
Seat tube progress.
You can see the bulkhead more easily in this top tube off cut. There are two bulkheads running up the seat tube from the bottom bracket. Probably unnecessarily there’s a continuous section running through the top tube and also sections in the down tube meeting the head tube and bottom bracket. I didn’t want a continuous section in the down tube as it would play havoc with internal cable routing.
The carbon arrives. After a conversation with Hana Kolarova and engineering student and sponsored rider, who happens to be building a frame as part of her studies, she passed on some of her research. It suggested aiming more specific 0/45o biaxial cloth at the head tube and BB areas which of course makes sense but seeing the data really forces it home. It’ll be interesting to experience the outcome.
I tend to use tailors chalk and a very thin slice of masking to demarcate the cutting line and either use scissors or a guillotine blade to cut down the tape line. This method stops the fabric from fraying and the tape tends to easily peel off when resin is applied.
I worked with the slotted 68mm BB shell as usual but this time I generated a sleeve rather than using the available carbon box as I’d done before. This would commit me to routing gear cable externally over the BB area. I plan to drill out the Ali shell, mainly so that when I adhered it there will be even more resistance to an over tightened bottom bracket twisting the shell. Whilst I don’t have experience of this myself, I’m told that it’s a common problem.
Wrapping and vac bagging the tubes.
On this occasion I have chosen to work with a low viscosity resin that should flow through the fiber more easily, in addition I’ve chosen a slow catalyst that should allow the vacuum time to remove more excess resin than I’d managed with the less viscous fast drying products. In addition the new resin has a post cure capacity at a higher temperature, which basically means I can knock up an oven and bake the frame to a harder finish. I was going to use pre pregs however i could find the variety of cloth required.
First working layer on the seat tube, top tube and BB shell. I’ll shape the integrated rear wheel shape into the seat tube next. It’s easier to shape when the tube has some rigidity.
The width of the mono stay fitted really well onto the seat tubes, a super strong fixing will be easy.
You can see more clearly here the variation of fibers applied. Depicted here is the 0/45o fiber.
Next was to set the frame into the jig. I really need to make a proper jig, I’m more or less happy that the material I work with can be aligned however I do spend a great deal of time checking and double checking, there are more full proof home made solutions.
After applying a few layers to the original cores I’m assessing the frame in the jig before completing the individual tubes. The next layer will be several layers of uni directional cloth around the ISP and several layers of +/- 45o biaxial cloth around the head tube and BB join areas. The rear end mounts nicely onto the seat tube. It was tricky to assess the length of the seat tube against the arc of the rear wheel aperture.
Finally the TT frame tubes have been bonded. It’s fairly neat and considering the amount of carbon used a starting weight of 1.1kg gives me a fair amount to play with when shaping the tube intersections.
I was interested to see how this Fairing Compound will work on the TT joint filet’s. They use it on plane wing construction, and as it says it’s super light. The contents of this entire 1ltr tub weighs just a few grams! l believe Guru Cycles use something similar, a molding putty on their joints before the layups so I thought this might also work. You mix just a little resin into the compound and it sets hard to a sand-able finish.
I cut paper templates that fitted the frame before cutting the full compliment of layers to be applied to the joints.
I’ve just begun constructing the oven to facilitate the curing process of the resin.
I stripped back an old storage heater, removed the bottom cage and thermostat. I bought a simple chef’s thermometer with an alarm and stuck the probe through into the oven. The walls of the oven are made from 50mm storage insulation that has no ignition point… meaning it can’t catch fire. Anyway the top temperature I’m aiming at is only 80o so it’s not a huge concern. I don’t really have room for the oven so I’m really only looking at a one off use.
Finished oven. Trialed it last night, the issue became ways of keeping the temperature down, it heats up super fast.
I cooked the frame for 6hrs at 40o and then a further 7hrs at 60 – 70o. The M3 structural adhesive is also a high temperature product so the entire frame post cured nicely. Following this I applied several thin layers of a UV resistant resin. I chose to apply this with a ribber glove. I spread a very small quantity over the frame and once dry buff back with a polishing compound and then reapply. The process removes the high gloss finish that I not so keen on.
It’s nice to do this on a hot day and cure the resin quickly. I chose the expose the fiber layups on this frame. I think it really looks nice and is by far the most blemish free frame I produced to date.
Following the Frame builds @ https://chrisaldgate.wordpress.com
Rapha designer and 2012 24hr TT Champion Ultan Coyle asked me to produce a set of units for the 2015 Transcontinental Race he’s competing in, the race from Flanders – Istanbul. http://www.transcontinental.cc/. We had about 6 weeks to turn it round. As it turned out not enough time to dwell on the finish but at the last recorded Transcontinental reportage it seems the design held firm through the descent of Col D’Assettia and a Montenegrin taxi strike.
If and when Ultan returns I may get the opportunity to better finish the bags and take some better images.
The brief included two units – a front unit to hold: iphone6 / dynamo power storage units / map cards / USB transformer / gloves / gels etc. And a rear unit to hold mechanical’s, tubes, multi-tool etc – also ideally an overnight bag to specifically hold: bivi / coat / sleeping mat / toothbrush etc. The bottom section depicted here is bolted onto the seat stays but other than the Velcro straps the top unit is only mechanically locked onto the frame via a locking slid mounted on the lid.
Specialized were happy to sponsor Ultan with the new TT Shiv model but as the bike didn’t come through in time Ultan thankfully allowed me to crack on and design the units for his Trek Speed Concept instead.
The front and rear units would be home to:
Front – iphone / dynamo battery / map cards / gloves / front light etc. Rear (bottom) – mechanical, tubes etc, (top) – Bivi / coat / sleeping mat / rear light.
We began with a few templates and sketches that were determined by the dimensions of specific contents and aerodynamic fit to the frame. A card template was cut and married to the frame so we could better imagine the final design.
The rear unit/s.
The front unit.
The pattern was made using various densities of uni-mould foam that get sanded to the required shape. The skeleton foam is denser than the grey foam and resists over-sanding. The pattern is then coated in body filler that’s sanded and then primed with several layers of pattern coat primer. Finally the pattern get several layers of release wax before the mould can be cast. I formed this unit as one and then cut it into the two separate units.
The above pattern was for the Tri bar clamp.
Next was to generate the moulds. The unit were divided using acrylic sheets into isolated sections that would make up the split moulds. In the blow image a two part mould for the front unit, the two sides of the bar bag, the other units were made as three part moulds. The first stage once the acrylic sheet is in place, was to apply a thick black gel coat painted on to the unit to form the detailed surface layer of the mould. Multiple layers of fiberglass are then added to give the mould it’s strength.
In addition to the basic units I had to come up with a way to connect and lock the rear units together to stop them from swinging around or knocking the seat post on those alpine gravel roads. The solution… as the two unit are braced onto the frame the depicted slider mechanically locks itself together.
To generate the lid of each unit, after the moulds were formed, I bagged the patterns once more with carbon laid up only around the lid area and cut the lids to size. The bottom lid slides onto the unit and is held in part mechnically and also by the locking screw at the rear. The lid to the top rear unit was hinged with this 7g barrel hinge. In 7g I managed to sqeeze a Kevlar reinforcement, carbon pivot and Teflon washes.
Homemade carbon road frame project.
Can you home fabricate a carbon road frame? I had heard that you could do it with if not minimal craft, at least with minimal means. I asked around but the response from those with experience, riders, engineers, designers etc was that you needed an autoclave – the capacity to compress and heat the composite in a unified moment.
Since then I’ve come to enjoy a very real and refreshing economy of knowledge one that contradicts the doubters assumptions. I don’t have a composite or engineering background, in fact I work in the arts in a world of opinion and subjectivity, so the distinction of trial and error is both necessary and refreshing. If the experience I collect is sound the frame will hold together, if not, it’s my neck! I’ve been fabricating frames for over five years now, although mindful not to over complicate things I continue to develope, explore and refine the process.
See below for build story and posts of developments on other projects. Happy to discuss so leave a comment if you wish to be in touch and please do post any of your ‘homemades’! This blog is a gesture of enthusiasm but without the work that others have previously posted I wouldn’t have known where to begin.
If you choose to use any of the information provided, be mindful of the consequences, I believe it takes a fair amount of craft to undertake a build – I’m simply reporting here that it is indeed possible.
Latest completed frame.
The moulded saddle came out well. It could be formed to a high gloss smooth finish but I find there’s more traction to stop the rider sliding back like this. Also I’ve always like the fabric look of it.
This is a test attempt for a more significant mould I’ll be making in the coming weeks. The first step is to secure your pattern to a release surface and filet the edges to ensure the mould does not gain a mechanical hold. Next is the release agent, in this case a release wax.
The next layer was a thick resin gel coat and once the gel coat had reached a tacky but not full cured state, several resin soaked 100g strand mat fiberglass. I then laminated several more layers of 300g strand mat to finish off.
By all accounts bike manufacturers use automotive clear coats to finish their frames.So far I haven’t paid much attention to my finish as I’m still too nervous following a build about the ride being sound. Usual I get side tracked on another build and can’t stop riding the new frame long enough to strip it back and invest the extra hours buffing and lacquering the finish.
This time however I thought I’d try this MaxMeyer Acrylic product.
A lower profile than the other saddles I’ve made and at 190g a more competitive weight. I’ve still used the ti rails but this time I bonded the rails directly into the carbon using small sections of the insides of an old Flite saddle. I sprayed the underside black before adding a topcoat of resin.
2014 . a frame build – Aero tubed frame.
Built bike that I rode today. You may notice the omission of the front derailleur! There wasn’t enough space to attach the ‘band on’ derailleur i was using, I’ll have to shape and rivet on a hanger. Pics to follow.
I also mounted the bottle cages using this handy tool a friend let me borrow. It rivets the bosses into the carbon. In addition I glued the rivets in with structural adhesive just in case the process of drilling and compressing the rivet had propagated any splits in the carbon.
Finally following the addition of hardware and the top coat of Herringbone carbon, tomorrow I will tap the BB shell and build the bike up for riding next week.
I shaped the area between the tyre and the BB alot closer this time. Hopefully the addition carbon will add to stiffness.
Here’s the ‘Sugru’ rubber mold that I formed around the internal cable bosses. I used one on each boss to compress the carbon around the boss. A lot of the larger companies use a similar method at various points around the frame where high compression is desired.
Frame before the top coat of carbon. You can see in these images the different orientations of carbon around the seat post (to strengthen the area to receive the carbon quill seat clamp) and around the joints.
Vacuum bagging the BB area.
Shaped joints –
Work in progress – shaping the head tube. Once these areas were wrapped like the image depicts, I then orientated uni directional (UD) carbon over the joints.
Work in progress – shaping the TT – Seat Tube. Here I was thinking of different shapes but in the end went as simple as possible and removed as much carbon (weight) as I felt comfortable with.
Third generation frame about to be wrapped.
I’m still using a similar fabrication process and geometry but I’ve included an aero seat tube, internal cable routing and an over sized down tube. I’ve bonded the frame before adding the final layers of carbon, which I can now include in the joint process.
Decisions re the frame began with limiting factors, like what seat clamp to go with. I chose the FSA carbon k-wing ISP.
I bonded the frame with 3M Structural Adhesive. I think bought from Transair – http://transair.co.uk/.
The joints. I don’t bother filleting the joint like I did on the first frame, instead I’ll wrap the joints with tow, layering on more (lighter) carbon and then I’ll shape accordingly.
The BB Shell. Bought from Ceeway – you need to contact Peter directly and request the carbon parts you’re looking for. He’s really helpful!
Internal cable routing adapters. Also from Ceeway.
Tube off cuts. They vary in wall thickness. The head tube (top left) has a yellow Kevlar base layer. The seat mono stay (top right) has an internal bulkhead. The two tubes mid left are the ISP off cut next to my Deda fork off cut. Mid right is the top tube and down tube and at bottom the chain stays.
I’ll be using this UD carbon Tow to wrap the joints. Bought from East Coast or Easy Composite.
2013 – . a frame build
The frame rides really nicely but has a totally different geometry from the first frame, the difference in feel was quite shocking.
Road and safety testing a new build, ridden for the first time yesterday. Depending on how it feels/performs I may reinforce the BB area before adding a top coat and bottle bosses. It came in at 936 grams before wrapping and with the ISP weighs around 1350 grams now.
USP’s = 1 1/8 – 1 1/2 Head tube – Carbon BB Shell – Sloping seat stays – 215 gram rear end – Carbon dropouts – Triangular sloping top tube and an ISP.
Bike before the final layer.
built a jig for this build.
Pretty happy with the weight considering the ISP. This seems to be a bench mark weight at this size and using this process.
Carbon BB shell.
1 1/8 – 1 1/2 Head tube.
The finishing layer of UD carbon. I was at the bespoke bike fair and a friend who distributes Faggin bikes showed me a frame with a hand stitched leather finish. It got me thinking and this finishing coat is useful for hiding any inevitable pin holes in the finish. Looks beautiful in sun light!
2013 – a. frame build
Been away for a while working on these two.
2012 – a. frame build – Homemade road frame project. Hopefully useful for anyone wishing to get on and build bespoke!
The geometry – because this was my first build, I oriented towards what I understood to be a traditional set up. I established the geometry with research from three separate sources: my existing frame (riding position), the Cyfac Postural Manual http://www.cyfacblog.com/downloads/ and as many published frame geometries as I could lay my hands on.
I found that making a drawing gave me a better feel for the logic and dynamics of frame geometry. Given that I had no previous experience, any insight was useful. During the build I shortened the wheelbase and altered the seat tube angle. Clearly not all of the drawing is to scale, that BB shell is way too big.
The rear end: Deda – was bought from Ceeway http://www.framebuilding.com/. I bought the rear end with a steel BB shell and aluminum dropouts. in order to complete the frame I also bought all the bosses, head tube cups and an inexpensive Carbon fork.
The tubes – I started with a sheet of foam board from a craft supplier, roughly cut the tubes to size and refined the shape on long lengths of sand paper.
Any blemishes were filled and smoothed out.
I don’t have the clearest images of the next part of the process. The carbon is cut to wrap the tube the required number of layers in one roll. One side of the carbon is painted with resin. The core is wrapped in carbon and a layer of peel ply is added before the tube is compressed. I used electrical tape, applied with the sticky side out and them perforated with a needle which allows excess resin to drain.
Before wrapping the carbon, I inserted a metal stud for rigidity.
Before the structural layers of uni directional carbon were wrapped, the foam core and stud were removed with Acetone.
The cups were compressed and glued onto the head tube. I figured the plates pressing from each side would help align the cups.
The end of each tube was sanded to seat into its respective join. Where ever there are two different materials (the BB steel to carbon) being joined, the area was insulated with glass fiber. Nasa have release some interesting papers on composite uses, including on the causes of galvanic corrosion.
Part before assembly.
The next step was to build a jig. I was working to a 5mm margin of error, I’d read somewhere that this was common in factory frames. Having said that, with the jig I built, I found it pretty easy to be exact.
The dropouts, BB and forks were fixed using M6 and M16 stud into an MDF box.
The angles were achieve with two pre-cut MDF guides.
I used a section of Unistrut, with some metal parts I had lying around, to create a hold for the seat tube.
The parts were aligned and strapped into position ready for the adhesive.
Once I’d spot glued the frame, it was removed and check for alignment.
Now to flox the joints before commencing with the layups. I created the flox using a mixture of Epoxy and shredded carbon fiber. The flox creates a full weld around the join, much the same as a fillet brazed frame (but without the strength).
Next – shape the joint in order to receive the layups.
All of the shaping was done using 5mm strips of sand paper, in much the same as a fillet weld would be shaped.
Traditionally each joint should, as I understand it, receive a full schedule of layups. This means uni directional carbon being compressed over the joint at something like 0/33/45/90 degrees. http://www.youtube.com/watch?v=l9x2PjozPus – this Cyfac video demonstrates the process. I never really found clear advice regarding the repetition of layers required for each joint. Each area might well receive a different number and will return a different feel and stiffness. I used intuition and common sense, the BB requiring more support against lateral movement etc..
After the carbon (and a layer of Kevlar to protect against chips and cracks on the down tube) has been applied, a peel ply is added, both to help draw out the excess resin and also to enable the compression layer to be easily removed. I used electrical tape to compress the joint, it was punctured with a needle in order to allow the excess resin to drain away.
It is advised that all layers of carbon are compressed in one step, this helps avoid any de-lamination etc. ‘However’, because perfect results are a little difficult to achieve every time, I compromised to allow myself one workable initial multi-directional layer. This layer was shaped and sanded to achieve a smoother bed before adding the final set of layers in one go. I did this in one quick step, trying to complete the full layup before the first layer was fully cured.
It is important to avoid sanding the structural uni-directional carbon, the fibers should remain continuous and unbroken to allow energy to flow and disperse. The multi-directional carbon is more or less decorative and as such it can receive a light sand to smooth the finish.
This image depicts the first and worst joint I made.
Next, time to fix the bosses.
Running the front mech cable was a head scratch. I couldn’t think what to use to direct the cable and protect the frame. It occurred to me that the only time I’d seen a cable grommet that size was on the back end of a Bic Biro. So why not. I mounted a steel washer at the exit point and they both work well.
Mounting the gear cables to the down tube was pretty simple. I just molded flox around a pair of aluminum tubes and coated them both in carbon.
Same with the rear mech cable.
…and the brake cables.
In a few spills last year, along with my teeth, I wrecked a couple of Flite Titanium saddles and fancied trying to recycle the ti and builds a carbon version. This was the ‘very light’ out come:
The Ti bars are loose in their fixing and between the (chain ring) bolts and the saddle are a couple of rubber washes, both of which allow the saddle to flex. I made sure I ran a layer of Kevlar through the carbon layers just in case it tried to bite me!! I’ve done several thousand miles on this seat now and it seems both comfy and strong enough.
A few close ups of the finished article, the first of which depicts the Garmin center mount I also made…
Bottle cage 24g
I forget exactly but I think the total weight of the frame alone was around 1.4kg. If I become g sensitive I know the areas I over spec’ed.
I’m gearing up to build again with more contemporary geometry, shorten the head tube, lower and curve the top and shorten the wheel base. Try to make it more responsive and a little stiffer.
East Coast Fiberglass suppliers – http://www.ecfibreglasssupplies.co.uk/
Carbonology – http://www.carbonology.com/
Easy Composites – http://www.easycomposites.co.uk/
Ceeway – http://www.framebuilding.com/
The Project Junkie – http://theprojectjunkie.com/
Carbon Wasp –
Brano Meres – http://www.bmeres.com
Metal corrosion – corrosion.ksc.nasa.gov/ .
Bio-mechanics – Cefac