User Tag List

Page 1 of 2 12 LastLast
Results 1 to 20 of 31

Thread: The Tonewood Project

  1. #1
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default The Tonewood Project

    A lot of what I'm doing is for reasons that are probably a little obscure but from previous adventures in audio electronics I'm going to assume that most readers are far less interested in the theory and mathematics than am I, so I'll try to concentrate on the practical. If you want more theory just ask, I'm happy to oblige. To see what you might be getting into, look here.

    This is “The Tonewood Project” because the end result uses a variety of Australian hardwoods which are renowned as tonewoods by Australian luthiers – see for example here, here, here, and here.

    It didn't start out that way but that's a story for another day.

    The woods are used to form tubes in combination with other materials chosen for mechanical and acoustic properties, namely boron fibre, carbon fibre and basalt fibre. These are then combined with stainless steel (currently KVA's MS2 alloy) to make the bike by an amalgam of standard metalwork and composite tube to tube construction techniques. Altogether this has been quite a journey as I had to learn the skills required for the steelwork and the woodwork and brush up on new techniques for the composites work.

    The layup of each tube changes with its intended use. I can alter the number of layers and the compositions of these layers, including which woods are used where, which composites to use where and whether to use unidirectional or bidirectional fabric. I can also alter the layup angles for all the above, giving me a large amount of control over the tube properties. I use a simple spreadsheet system to make calculations of torsional, lateral and longitudinal properties.

    The following description is representative but not necessarily typical.

    The base layer for this tube is Spotted Gum (Corymbia maculata / C citriodora), the wood on the left in the phot below, wound at 30 degrees. The other woods shown are Mountain Ash, Tasmanian Blackwood and Rose Gum, more detail abouts woods in a future post. First the required strip(s) of wood are cut from veneer leaf. The width required is given by Pi * diameter * cos(wind angle). Because wood bends almost exclusively on the compression side the diameter should be taken as the base diameter plus twice the strip thickness, so for a tube with a base diameter of 27.4mm and a leaf thickness of 0.6mm, a wind angle of 30 degrees requires a strip width of 28.6 * pi * 0.866 which is 78mm. I cut these strips with a Stanley knife and a straight-edge on a wooden board, which works OK, but courtesy of the people at Specialty Materials (the suppliers of boron fibre), I have since learnt to use a sharpened pizza cutter to get better tolerances and a straighter cut.

    The strips are sanded both sides at 120 grit (not shown). Wear gloves to protect your work and a mask to protect your lungs – these woods have very high silica contents and the dust is a respiratory hazard. In addition some of them cause asthma like symptoms in susceptible individuals. Sanding such thin long strips is an invitation to buckling and cracking, some woods being more susceptible than others. It helps to tape the more susceptible ones to a backing board.


    A Strips by Markwine, on Flickr

    Once sanded and dusted, it's time to impregnate the wood and wind the tube. To prevent splitting, it is helpful to tape the outside ends of the strip with something robust like bookbinding tape. The epoxy I use is a low viscosity high Tg impregnating epoxy from ATL composites in Australia. The low viscosity improves penetration, the high temperature gives me good working life and stands up to the 120 degree heat cycle that the later prepregs require. The drawback is that curing requires a temperature controlled oven, so I built myself one.

    The wood is epoxy impregnated on one side by pulling the strip under a heat gun and applying the epoxy to the heated part of the wood. The phots belie the actual technique which is to do this in a continuous pass under the heat gun. The rate of travel and the distance of the gun are adjusted to get the right level of heating. You can judge this by the apparent viscosity of the resin – at the right temperature it brushes on like water; too low and it's sticky, too high and it smells baaad. I tried vacuum impregnation in a chamber I made for this purpose and solvent impregnation using a commercial epoxy additive but neither works as well as this simple heating setup.


    B Heating by Markwine, on Flickr


    C Impreg by Markwine, on Flickr

    To ensure complete waterproofing of the inner surface of the tube, I add a strip of very thin (75 gsm, somewhat less than 0.1mm thick) glass fibre to this face, cut so it overlaps slightly. The glass is also wet with epoxy. Getting this to all wind together correctly is fun and games.


    D post impreg by Markwine, on Flickr

    First, cross the strip with the mandrel at the correct angle. I'll talk about mandrels another day, they turned out to be the most difficult part of the whole process.


    E ready to wind by Markwine, on Flickr

    Wind up one end and clamp it – the clamps shown here don't work very well, I'm developing new clamps that allow repositioning. Note this phot was taken on a different day with a different mandrel: I'm using my wife's SLR and trying not to get epoxy all over it. This and my woeful photography skills mean some shots didn't come out.


    F winding by Markwine, on Flickr

    Continue winding down to the other end and clamp that. The major problem that occurs here is keeping the wind cylindrical – if part of it goes conical it throws everything out and you'll get overlaps and / or gaps. If all goes well and the sides are aligned, start applying the pressure wrap – I use a length of silicone tubing pulled taut as I wrap. I chose silicone because it has excellent temperature resistance and good elasticity. It also allows me to observe the wind as I wrap so that I can check the edge alignment.


    G Wound, start wrap by Markwine, on Flickr

    Once fully wrapped, it's time to bake the epoxy. I've found that too high an initial temperature can cause problems, so I start with a couple of hours at 40 degrees before raising the oven to 60 to complete the cure.


    H wrapped by Markwine, on Flickr

    Once cured, unwrap the tube, sand off any epoxy ridges (which form between the wraps of tubing as epoxy is squeezed out) and check the tube is cylindrical.

    To do the second layer, both the outside of the tube and the inside of the second strip are epoxy impregnated. Note the orientation of the second strip, which will wind in the opposite sense to the first. The rest of the process is as per the first strip.


    I Second Wind by Markwine, on Flickr

    That's enough for now, more later.

  2. #2
    Join Date
    Apr 2009
    Posts
    2,028
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    very awesome

    reminds me of unidirectionality

  3. #3
    Join Date
    Dec 2011
    Location
    Fort Collins, CO
    Posts
    351
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Wow, looking forward to seeing more of this project

  4. #4
    Join Date
    Apr 2011
    Location
    Boulder County, Colorado
    Posts
    489
    Post Thanks / Like
    Mentioned
    7 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Holy cow that is a lot of work to make some tubes. I'm sure it will be beautiful when finished. Is the general idea that you can sort of dial in a particular resonance (for lack of a better description) for each tube in order to obtain a specific ride quality?

  5. #5
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Quote Originally Posted by Honus View Post
    Holy cow that is a lot of work to make some tubes.
    Yeah at this stage it takes almost two weeks to make each tube. I have to get quicker.

    Quote Originally Posted by Honus View Post
    I'm sure it will be beautiful when finished. Is the general idea that you can sort of dial in a particular resonance (for lack of a better description) for each tube in order to obtain a specific ride quality?
    I can't answer that question briefly, so I'll address it in a later post.

  6. #6
    Join Date
    Apr 2009
    Posts
    2,028
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    is the wood completely dry,as in do you let it set its own environment if you get my drift, I know nothing about wood except what i have read recently but its fascinating

  7. #7
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    I get the wood to around 10% moisture, which in my environment isn't hard; today it was 39 degrees C (102 F ) with 9% RH, that's a pretty effective drying oven. One thing I've been at great pains to do is to isolate the wood from the environment so these moisture levels remain constant. Debonding due to differential moisture absorbtion is the principle cause of failure of wood / epoxy bonds - the wood expands when it gets wet, the epoxy doesn't so the two part ways.

    Since some of the woods (esp spotted gum) have high levels of extractives which are reported to create bonding problems, I have tried stripping these eg by heating the wood in a 50% ethanol solution (If you remember I'm a winemaker, extracting tannins from plant based materials is what I do).

    This stripping reduces the moisture levels but also reduces the toughness values. Since the excellent toughness is part of the reason for using these woods, that seems counterproductive. So far I haven't seen any evidence of de-bonding or delamination and I've subjected these to some abuse - my favourite demonstration is to smash one of my wooden tubes against a titanium tube of equivalent size and weight. The titanium loses. A carbon tube would fare even worse.

  8. #8
    Join Date
    Apr 2011
    Location
    Boulder County, Colorado
    Posts
    489
    Post Thanks / Like
    Mentioned
    7 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    No internal damage/microfractures? That's pretty impressive. Have you done any sort of simple load deflection test compared to a metal tube of equal size/weight? I'd be really curious to see the results. I had a long conversation at a show with the guy that makes the Renovo wood bikes and while they're not my cup of tea I was certainly very intrigued by them.

  9. #9
    Join Date
    Nov 2012
    Location
    Zurich, Switzerland
    Posts
    22
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Fucking awesome! I'm looking forward to see more.

  10. #10
    Join Date
    Dec 2012
    Location
    Edge of the world
    Posts
    59
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Errrrr......


    YES!
    moar!!!




    Edit: Also that Australian tonewoods gentleman lives quite close by me, which comes in handy.

  11. #11
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Once I've formed the tube I start adding new layers of material to achieve the properties I want. This post is going to concentrate on the many ways not to do this.

    For this tube I'm adding a layer of unidirectional carbon – this is a 350 gsm material and thus a lot thicker than a standard ply of around 200 gsm.


    A Carbon by Markwine, on Flickr
    The base tube is coated with epoxy and the carbon fabric is wet through then the wrapped around the tube at the appropriate angle (in this case 30 degrees). Once everything is laid up I made sure the outer side was properly wet then laid another layer in the opposite orientation. These were then overwrapped with peel ply (a material that absorbs resin but doesn't bond, so excess resin is simply peeled off the surface after the cure), compressed using the silicone wrap technique as per the wooden tube and put through the oven to cure.


    C Carbon wrapped by Markwine, on Flickr

    It turns out that the peel ply was the only thing I got right. There were so many things wrong at once that took a while to figure them all out. The first and most obvious was that the silicone tube wrap which works well on the wooden strips which are quite rigid in the long axis works extremely badly on the fabric which is more easily deformed. The end result was a tube which was completely unusable due to surface irregularities.


    B Carbon by Markwine, on Flickr

    Trying to sand the tube down to remove these irregularities made the other problem apparent – a huge level of fibre distortion, enough to ruin the properties of the tube. It was only after some follow up research that I realised that this is an inherent problem when trying to compress multiple plies of composites which are laid in a direction which gives any hoop strength. The reason is one of those things that is retrospectively obvious* – the very reason we are laying in this orientation is so that the fibres resist deformation in the hoop and this same property means that when they are compressed axially they cannot move except by distorting. The heavy fabric simply amplified this effect.

    To get around this problem, I changed techniques. Any composites which are laid with any hoop orientation are laid and cured as single plies of at most 200 gsm fabric. These plies are only approximately 0.2mm thick, limiting the possible distortion. To help with this I also changed the way I wet out the fabric to a “quasi prepreg” scheme as outlined here: DIY Carbon Fiber Prepreg. An easy solution.

    This had the added benefit of making the plies much easier to handle.

    I also scrapped the silicone tube wrap technique and tried both wrapping with PVC electrical tape and compressing with heatshrink tubing. I didn't take any phots of these but you're not missing much - they both give quite poor surfaces, not as irregular as the silicone but not nearly good enough. Because I'm alternating layers of fabrics and wood the surface of the cured fabric layers has to be very smooth to match the flat rigid surface of the wood laminae or I'll end up with patches full of pure resin.

    Time to bite the bullet and go with vacuum bagging, so I bought myself a small single stage vacuum pump and some bagging materials and set to. One steep learning curve later I'm getting far better results but that's a topic for another day.

    * I reckon there should be a single word for "that which is retrospectively obvious", probably in German, after all they have "schadenfreude" and "backpfeifengesicht".

  12. #12
    Join Date
    Aug 2009
    Location
    Germany
    Posts
    295
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Mark, if you are aiming for a cosmetic surface you won't get it without post-coating. Vacuum bagging will give you the best quality part but will not leave the surface smooth. To get a good, smooth surface you need to add a topcoat. Something like T30 from Hexion.

    If you're putting a layer of wood over the carbon then it would be best to leave the peelply finish alone.

    You don't necessarily need to buy a vacuum pump. You can use an old refrigerator compressor to make one for next to free.

    The word you're looking for is: "Nachträglichkeitsoffenbarung" or "Duh!"
    Cheers
    Kevin

    PolyTube Cycles

  13. #13
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Kevin

    Thanks for the advice, and especially for the word. My wife, who spent a year in Germany, hooted with laughter when I told her.

  14. #14
    Join Date
    Oct 2011
    Location
    Gisborne, New Zealand
    Posts
    361
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Man this looks complicated. Any particular reason you chose to wrap all that material over stainless steel tubes instead of aluminum or titanium tubing? Look forward to more pics!
    Steel Bamboo Aluminum Wood Titanium Magnesium ETC

    (Pick your poison, ride it like a stuck pig!!!)

    Alfred Salgado

  15. #15
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Quote Originally Posted by Freddy Salgado View Post
    Man this looks complicated. Any particular reason you chose to wrap all that material over stainless steel tubes instead of aluminum or titanium tubing? Look forward to more pics!
    The tubes illustrated so far in this thread are wrapped on mandrels which are then removed, so the wood / composite tube stands alone.

    The stainless parts on the first prototype are the headtube, downtube and chainstays (and lugs). I went with steel for several reasons, including that it's important to me that the metal parts use lugged construction and my familiarity with the material. Stainless is quite a good material acoustically, there's a fellow in the UK who makes turntables entirely from stainless and they have a good reputation. The next two prototypes will have less stainless - one will have stainless chainstays only. I'll probably use Ti for the BB and head tube sleeves on that one if I build myself a plasma electrolytic oxidation machine.

    There are no advantages to aluminium in this application and it's an acoustic disaster zone.

    Ti is Ok, but mostly too difficult to work with (for me). I made a Ti bike with modified tubing as part of the development work that went into this project (I got it welded by someone else). It was good enough for me to think I was onto something interesting but I wanted to do better. I came up with the tonewood idea afterwards and have run with it.

  16. #16
    Join Date
    Jan 2014
    Posts
    7
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Hey Mark,

    I've been reading up on this project as much as I have been able to and figured it'd be time to ask a few questions.

    I come from a music background myself before getting into engineering and fabrication and have always been interested in acoustical science. You keep bringing up the acoustical qualities of the materials that you are using and then citing the mechanical properties (ie. strength, stiffness, etc) but you aren't mentioning any difference in the resonant frequency of the materials that you are using. From a structural standpoint, I understand the appeal of natural fiber composites but is there some other aspect of the resonant frequency that you are trying to tune in the frame?

    I've got a bit of a theory I've been working on/pondering for a little while about human perception of stiffness and it's relation to vibration (soundwaves or through the fingers) and thought that you might be working with those principles with this bike a little bit. Especially since you cited a little bit of work with tonearm geometry and (I'd assume) turntable construction which also happens to be another interest of mine.

    Thanks,
    Will
    Will Hilgenberg
    Santa Cruz, CA

  17. #17
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Will

    The first version of this was based on trying to alter frequencies of resonance and while it had some effect I learnt more from what didn't work than from what did.

    I'm now far more concerned with acoustic impedance and especially changes of acoustic impedance across interfaces between elements. This also has a relationship to resonance: resonance is caused by reflection at element boundaries and reflection is a function of differences in acoustic impedance. Equations supplied on request.

    The acoustic properties of a material are mostly functions of modulus and density: propagation velocity is the square root of the quotient of modulus and density ( = SQRT(specific modulus)) and acoustic impedance is the product of propagation velocity and density. They also govern weight and stiffness, hence the number of times they've been mentioned.

    The other major factor is the loss tangent of each of the materials involved and especially the relationship of loss tangent to frequency.

    I'm also interested in the mechanical input impedance of human body parts but I'm still learning about this and trying to work out how I can incorporate what I learn.

  18. #18
    Join Date
    Aug 2011
    Location
    Beechworth, VIC
    Posts
    2,524
    Post Thanks / Like
    Mentioned
    9 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Also please note that this thread is two years old, a lot has changed in that time.

  19. #19
    Join Date
    Jan 2014
    Posts
    7
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    Quote Originally Posted by Mark Kelly View Post
    Also please note that this thread is two years old, a lot has changed in that time.
    I am aware. I just thought this might be the most appropriate thread to ask the question. Unless I missed a more recent thread that you think would be more appropriate.
    Will Hilgenberg
    Santa Cruz, CA

  20. #20
    Join Date
    Jan 2014
    Posts
    7
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Re: The Tonewood Project

    As to the acoustic impedances at the boundaries, is this why you are incorporating stainless steel into the BB and chainstays and then overwrapping to blend it into the rest of the frame? I understand that there is also the structural element at play here and that some of the complex geometry would be more difficult with wood (ie. dropouts) but the boundary between the two materials is where you would be looking at the differences in the acoustical impedance correct?

    I would be interested in the equations but it would probably take me a while to understand and wrap my head around. It would be helpful though! As far as the boundary characteristics are concerned, does it share similar principles with RF tubes?

    Also, I'm aware of some of the mechanics behind the loss tangent in relation to electromechanical principles but I didn't realize it applied to acoustics too. That's pretty neat!

    Thanks for the info,
    Will
    Will Hilgenberg
    Santa Cruz, CA

Page 1 of 2 12 LastLast

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •