ESI Q and A Forums Home
 Search       Members   Calendar   Help   Home 
Search by username
Not logged in - Login | Register 

Saddle trees and changing back shape
 Moderated by: DrDeb  
 New Topic   Reply   Print 
AuthorPost
AdamTill
Member
 

Joined: Tue Mar 27th, 2007
Location: Calgary, Alberta Canada
Posts: 289
Status:  Offline
 Posted: Wed May 5th, 2010 06:38 pm
 Quote  Reply 
Chopped glass mat or tissue has one useful structural purpose - to build up thickness quickly on something like a boat hull. It uses a massive amount of resin per volume, and resin has no particular structural value to increase the bending strength of a tree.

The reason we use resin in a composite is because fabrics have great tensile (pull-apart) strength, but poor compressive (push together) strength. The goal is align the fibres to take the load as much as possible along the fibre length, and let the resin take up the tendancy of the fibres to buckle.

If we use chopped glass with the fibres in random directions, then only a tiny fraction of the fibres will be oriented in the direction of load bearing (45 degrees to the length of the bar in torsion/twist, along the length in bending). The fibres which are non-load bearing just suck up resin (adding weight), and occupy space (adding weight).

If all you want to do is put a non-structural finish on a tree, then by all means. The chopped glass is very easy to apply, and won't pucker around corners or require darting like cloth will. It's impossible to use in an optimal way, however. Resin, per weight, is also MUCH more expensive then fibreglass, so what chopper systems do is maximize the amount of expensive resin per weight of cheap fibreglass.

If you're looking for strength from the chopped glass, then forget it...it's a concept along the lines of bed-liner material.

As for whether you want to use cloth at all, that depends on how strong the tree is. Resin will have some small value in preventing the fibres of the wood from buckling or separating, but it won't be much (try testing a board of wood under a few bricks vs a resin-soaked board...won't be much difference). You'd need to have all your strength coming from the tree however.

Taping the seams seems like a waste, as well, if that's all you're doing. A good wood-glue bond will be stronger then the wood that it's bonding, so tape isn't adding any value beyond spreading the load over a slightly wider area.

Two layers of glass on top isn't a bad idea, since glass is better under tension then compression. When I build a wing spar, I always make the top cap (picture an i-beam) twice as thick as the bottom, since I know that most of the loads will be to bend the tips upwards. I wasn't sure if I could say the same for a saddle tree, since I think the loads swing both ways (bending the bar in both directions). I still made my top surface designed thicker, but not to the same extreme.


Edit: that's why I mentioned using unidirectional carbon and satin weave as a combo. The uniweb is for bending along the bar, and the satin would be placed at 45 degrees to the long axis for bending loads. That would result is the absolute lightest tree possible...which wasn't my goal first-off.


If the glass is load bearing, you'll need to contain the burst component, which the single-piece wrap is doing to some degree. You could also wrap the finished bar with thread (called fibreglass/kevlar/carbon tow) to accomplish the same thing.

The risk for me, for example is that the carbon top and bottom caps are stronger then my planned wood core. The most likely thing to fail isn't the cap, but the glue between the caps and the core. I plan to avoid this issue by wrapping with kevlar tow as a last step, and possibly lacing the cantle and fork to the bars using that as well. Kevlar is most appropriate here since it's much tougher and more abrasion resistant then fibreglass and carbon fibre.

In summary - resin = expensive, heavy and weak. Minimize use thereof.

PS - One thought if you're covering bars on their own is to use braided sleeve. It's like chinese finger-trap material...you can compress it to expand, and pull it to tighten. If you could find material big enough, it would be very easy to slip over a bar, wet out, then throw into a vacuum bag. Then you could tape the seams between the bars/cantle/fork with wide glass tape.




Last edited on Wed May 5th, 2010 07:06 pm by AdamTill

Don Fenaroli
Member
 

Joined: Wed May 5th, 2010
Location:  
Posts: 2
Status:  Offline
 Posted: Thu May 6th, 2010 09:38 pm
 Quote  Reply 
Adam,
I'm new to this forum, but I wanted to thank you and Dave for sharing some of your process. Quick question; What is the length and area of the bar you just molded and have you radiused the outside edges?
How about you Dave, what is the area of the bar on your new fiberglass covered tree and how much do you radius the contact edges?
cheers
don

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Fri May 7th, 2010 03:38 pm
 Quote  Reply 
Adam,
Very interesting notion.  To use the tube I would have to cover the piece before I assembled. We would have to make some changes in their 3-D models and perhaps the indexing system to accommodate the thickness.  My  question is if I were to individually glass the pieces and then assemble would it be as strong as when I glass after assembly? If I fillet the seams properly does the glass really add much?

David Genadek

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Fri May 7th, 2010 04:14 pm
 Quote  Reply 
Don,

My new line of fiberglass covered trees is not a new shape it's just that they are  covered with fiberglass rather than Rawhide so currently the shapes are exactly the same as I'm using on my other lines.  My bars are currently 22 inches long put on my next generation of trees I will have three different sizes of bars to accommodate different seat leanghths.

I define the bearing surface starting in the front from where the thoracic side of the trapezius ties into the back and extend backward to the last rib and whenever possible I would like to use the anticlinal vertebrae as the backward limit. I try to keep the top of the bar about 2 inches away from the center of the horse and the bottom limit of the bar would be just above where the ribs pop out from under the Longissamus Dorsi muscle. In my mind that's all the bearing surface there is that can be utilized without interfering with the motion of the horse. From this perspective the narrowing of the waste of the saddle and the expansion of the pads on the front and the back of the saddle is counterproductive and serves no real purpose. My bars are also concave on the bottom which does seem to freak many people out. However, I have never seen concave shape on a  healthy back so the current practice of making convex bar bottoms seems to be driven by the sad fact that many horses backs are pathological.

I don't get too worked up over the radius as it seems to me if the designer has done his work and properly defined the shape of the back as a composite of the overall range of motion in the horse with the focus of the bearing surface being in the middle of the saddle  the radius then is not needed to allow for the motion as the motion has  been built into the shape of the bar.

David Genadek

AdamTill
Member
 

Joined: Tue Mar 27th, 2007
Location: Calgary, Alberta Canada
Posts: 289
Status:  Offline
 Posted: Fri May 7th, 2010 05:11 pm
 Quote  Reply 
Hi Don,

Welcome to the forum. The plan view (vertical shadow) area of each bar is 96sq", which means the true area of the bar is probably a shade over 100sq" (200sq" total for both, obviously). I used my 2D CAD package to design this bar rather then the 3D version (which I keep meaning to learn how to use), so the actual area is more difficult to calculate directly. The bar is 22" long.

As Dave mentioned, most of the need for radiusing goes away with the bar designed the way it is. I'm not designing this to fit entirely behind the shoulder like most, so there isn't the same need to radius the front edge in particular. Also, the radiusing of most trees gets "undone" by a lot of the skirt blocking that saddlemakers use, so there's actually not quite so much in the final product as most folks thingk anyway.

The crown profile was discussed and shown earlier in the thread as well, if you're curious.

Dave - to your quesion about pre-glassing components, I guess I don't really know what's going to be strong enough. The tricky part about bonding pre-cured components is that the bond between the glassed parts is entirely mechanical, rather then the chemical bond before the resin cures. It's also why it's a bad idea to do layups in stages - if you let the inner layers cure, then the bond between inner and outer becomes mechanical only.

Now, I assume that you screw the joints together anyway, but I guess I can't really say whether the mechanical bond between parts would be good enough. The glass tape will spread the bearing loads over a wider area, but as for how much is enough...might be some tests required is all I can really say. How many good trees generally fail at the joint between the fork and bars? I would have thought the bigger site for failure would be at the stirrup cutouts.

I'm in this boat as well, obviously, and I'm still batting around ideas as to how I plan to deal with it. I'm making life further difficult by planning on leaving the fork in bare glassed wood, finshed bright like a very nice saddle Andy posted about on Leatherworker a few weeks ago. The cantle will be done like the bars, only with the outer layer done with cherry laminate over carbon skins over a softwood core. I like the concept of kelvar lacing the fork and cantle to the bars, but due to the brightwork, I'll have to keep the lacing below the level of the jockeys. More thought required...stay tuned I guess.

Cheers,
Adam

AdamTill
Member
 

Joined: Tue Mar 27th, 2007
Location: Calgary, Alberta Canada
Posts: 289
Status:  Offline
 Posted: Fri May 7th, 2010 05:30 pm
 Quote  Reply 
Another thought would be to use kevlar tape instead of fibreglass for just the seaming. It's about 10 times the cost of fibreglass, but 7x the strength and MUCH tougher (far less likely to shear like fibreglass would).

You wouldn't use much of the tape per saddle, so the cost/benefit might be there.


David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Fri May 7th, 2010 08:02 pm
 Quote  Reply 
Adam,

That is very interesting but it makes sense.  They talk about the crystalline structure of the resin as it cures and how you don't want to disrupt the formation of that structure. It would make sense that if the structures are in place and you put more on top that it would not be able to lik in the same way it would if it had  cured altogether.  I have been told that that really wouldn't matter and I'm actually using a thinner epoxy to presoak the parts and that is thicker epoxy to let the glass. It sounds like I would be better off to lay the tree up give it a soaking with the thinner epoxy then lay on the glass and wet the glass with the epoxy and slap it in the vacuum bag.

I am using some screws to hold parts together. I'm actually using a gel epoxy where the surfaces meet so it will do some filling where necessary and then I screw it in place.  The boat people I work with strongly object to using any attachments and tell me that that will cause failure in the long run. Considering thousands of saddles were made simply screwing a couple hunks of wood together and then slapping some canvas on them I have to guess that any of this is probably a huge amount of overkill.  The place trees will generally break is where the stirrup weathers one over and I also see breaks in the front it seems to me they're generally caused because of the attachment of the horn.

I'm really anxious to see what you end up with so hurry up and get it done!

David Genadek

AdamTill
Member
 

Joined: Tue Mar 27th, 2007
Location: Calgary, Alberta Canada
Posts: 289
Status:  Offline
 Posted: Fri May 7th, 2010 09:22 pm
 Quote  Reply 
I have a vaquero horsemanship clinic coming up in mid June, and I've been told that showing up there in my dressage tack would be...unadvisable :) The clinic organizer was nice enough to mention that I could borrow the Harwood-made Wade that I used on the last day of Josh Nichol's clinic, but I'd like to have my own done by then.

It'll be a bit of a push, but I picked up a length of clear basswood, my cherry veneer, and a lovely piece of 0.75" solid cherry at lunch today. Next week I should be able to get a bunch done.

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Fri May 7th, 2010 09:32 pm
 Quote  Reply 
Hey Deb used to do clinics with a red English saddle.

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Wed May 12th, 2010 05:08 pm
 Quote  Reply 
Adam,

     I have some more vacuum bagging questions . The reservoir you're calling a vacuum reservoir. What is the difference between a vacuum reservoir in what they call a resin trap?  If they make something called a resin trap that means that the vacuum is actually pulling excess resin and that gets me to wondering if I could vacuum bag my Rawhide trees and let the vacuum just pull water out of the Rawhide?

     The material you call kite Dacron won't stick to the resin?

David Genadek

Last edited on Wed May 12th, 2010 05:09 pm by David Genadek

ozgaitedhorses
Member
 

Joined: Mon Apr 30th, 2007
Location: Australia
Posts: 55
Status:  Offline
 Posted: Fri May 14th, 2010 04:01 am
 Quote  Reply 
Hi David!
My experience is with vacuum pumps in chemistry labs – so not sure how your or Adam's setup differ from that.
The reservoir and the trap have two completely different functions. If your pump doesn't run constantly, but is triggered by a vacuum pressure sensor, having a reservoir gives a bit of a buffer that prevents the pump from being constantly turned on and off, on and off, on and off. It's also quite handy when you need several vacuum lines that can be used independently, to run them off the reservoir.
The purpose of trap on the other hand is to prevent solvents from getting into the pump. For that, the trap would typically be cooled with something like liquid nitrogen, dry ice/acetone or methanol. This would be a conventional lab cold trap: http://en.wikipedia.org/wiki/File:Cold_Trap.PNG.
And yes, you could suck the water out of your rawhide, but I don't think it will work in a bag setup. You need as large an 'evaporating surface' as possible, or you'd be waiting forever and a day – and you don't get that when the bag is sucked against the surface. And obviously you'd have to cool your trap with something better than plain ice, or your pump won't last very long. You are pretty much talking about freeze drying (although not all the way, or your rawhide will be too brittle), and that takes a lot of energy and, depending on the pump, can take quite a while....
Cheers,
Manu

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Sun May 16th, 2010 03:53 pm
 Quote  Reply 
Well it looks like I still have an awful lot to learn on this project.

When I first got the notion of vacuum bagging the Rawhide to keep it tight to the trees and went out and bought on old dairy vacuum pump. My idea was that I would have a vent on one end of the bag vacuum sucking from the other to create air movement so everything could dry. That pump broke the first time I used it so my experimentation came to an end.

I think with my new pump resins are getting into the pump and that's what I'm getting a smoky exhaust. The manufacturer suggests changing the oil after every use. Oh learning curves can suck.

David Genadek

David Genadek
Member


Joined: Sun May 13th, 2007
Location: Spring Valley, Minnesota USA
Posts: 425
Status:  Offline
 Posted: Tue May 25th, 2010 02:32 am
 Quote  Reply 
    Ok I just did another go with the vacuum bagging and I'm starting to get a decent result on this complex shape.  I took my tube and spaced 6 Ts in it. On  each T I took breather fabric and made strips that I taped directly to the tee. I ended up with a tube going along out side parimeter on the inside of the bag with tails of breather  material.  I first tried a one yard long bag and it just didn't have enough material to make it over all the contours. I quick did another  2 yards long and that gave me enough material to get it done with minimul of wrinkles. I used some clamps to hold the breather material where I wanted it. A few hundred more and I'll have it down. 

David Genadek

AdamTill
Member
 

Joined: Tue Mar 27th, 2007
Location: Calgary, Alberta Canada
Posts: 289
Status:  Offline
 Posted: Tue May 25th, 2010 02:38 pm
 Quote  Reply 
David Genadek wrote: Adam,

     I have some more vacuum bagging questions . The reservoir you're calling a vacuum reservoir. What is the difference between a vacuum reservoir in what they call a resin trap?  If they make something called a resin trap that means that the vacuum is actually pulling excess resin and that gets me to wondering if I could vacuum bag my Rawhide trees and let the vacuum just pull water out of the Rawhide?

     The material you call kite Dacron won't stick to the resin?

David Genadek


Hi David,

Glad your recent attempt turned out better! It's really just practice, as with all things worth learning.

Appologies for not replying earlier - hadn't noticed these questions.

First the easy one - no, the dacron doesn't stick. A few folks I know use it as a release film. Here's a decent thread on release agents:

http://www.rcuniverse.com/forum/m_2901684/anchors_2901684/mpage_1/key_/anchor/tm.htm#2901684

Second, Manu's licked most of what I would have said, but I'll add that we're actually using slightly different vacuum techniques from the look of it. My resevoir would double as a resin trap for the most part, but I tend to only associate traps with resin infusion techniques... ie, when the vacuum is used to draw resin through a dry part.

I use the vacuum not to draw resin (which leads to a slightly wetter, and thus heavier/weaker part), but simply as a large clamp. The parts are put in wet, and then left to dry.

Here's a description:

http://www.rcgroups.com/forums/showthread.php?t=1179997

As for drawing the moisture out, while I'm sure it might be possible, I wouldn't know where to look to accomplish that. I suspect that a heat box would be far more useful then a vacuum bag, from that perspective, though you'd risk warping the part.

On my project, I did get a fair amount more done this weekend.

First I rough cut all the core blocks out of bass using my bandsaw, and did a bit of rough shaping on the sander. I left them quite thick since I didn't know quite how I wanted the shape to come out - I suspect I'll be sanding most of them away afterwards to get a much thinner part, especially at the ends.



Next I did the final shaping of the bottom plate, and glued one of the cross beams in place. Then I took it out to the barn to confirm the fit one more time, which ended up passing with flying colours.

Finally, I started the process of final shaping and gluing...which took longer then expected. Most of the time was taken up waiting for the epoxy and microfibre slurry to dry, so that things didn't shift when I was placing them down. Should be able to finish that process in the next couple of days, then I get to bond with my sanding tools for a few hours to coax the final shape out of the lot. In an ideal world, I'd be ready to bag the top surface by this coming weekend.


Val
Member


Joined: Thu Apr 5th, 2007
Location: Near Philly, Pennsylvania USA
Posts: 116
Status:  Offline
 Posted: Tue May 25th, 2010 02:52 pm
 Quote  Reply 
This is fascinating! Thanks for taking the time to share such detailed descriptions and photos. 


 Current time is 10:37 am
Page:  First Page Previous Page  1  2  3  4  5  6  7  8  ...  Next Page Last Page  




Powered by WowBB 1.7 - Copyright © 2003-2006 Aycan Gulez