[greybeard]

Better be able to balance at speed

At my age that's the biggest problem, especially going round corners.


John

[jonixllc]

At my age that's the biggest problem, especially going round corners.


John

You'd better stick to simple harmonic motion then - low amplitude please

[cncagain]

Hi Guys. I've been enjoying this thread since early last year. I am considering building a manual mill the size of an x3 or maybe a little bigger using epoxy granite. I have made test pieces for both the linear slides using the moglice substitute, and for the e.g. itself. From both experiments, I think I am going to move onto an actual build. My real problem is with the z axis. No quill. So maybe cnc is the only "easy" way to do this. I already have a xylotex setup and mach 3 on another machine.
I am thinking of at least two styles of mill for this project:
a. make a sieg x3 style machine, roughly same capabilities.
b. make a smaller frame version of a machine like the Mikron hsm 600. Seems much more rigid...
Which would be a good choice for making stuff around the size of or smaller than a storebought loaf of bread?
I had been looking at just buying a used millrite. They seem to cost more than a used bridgeport around here (almost 2x as much with shipping). Rockwells and clausings are even worse-price wise. Shipping on all these guys is almost as much as the machine! I want American made, but a bridgeport is just too big for my side of the two car garage. Hence, my e.g. plan. Has anyone put any thought into a manual machine using e.g.? I saw the bases by cncbridges. I want to do the whole darn thing using e.g.
I know the thread seems now mostly about formulation, but I plan to use this stuff. I'd have been proud to have made stuff the quality of Walters. I'd have used it too!
Any ideas?
Thanks,
Ken

[veteq]

Good grief.......50,000G

Hey John,
Can you show me how you calculated that?
Sorry really dont know.

regards,
Roy

[ckelloug]

Ranchak,

How are your experiments going?

Epikur,

Cool pdf? Is there an English version?

cncagain,

A lot of us want to use E/G to make something. Because good E/G is a lot less rigid than steel and bad E/G is similar to pine in stiffness, a lot of work has gone into trying to get the formula right. My own efforts on the formula are stalled due to a combination of day job work and getting organized enough to finish the research. I also have some packing model optimizer problems that I need to account for. The equal parts recipe from ages back is the best result I currently believe in but better results will come.

ncmeinc,

As far as I know, Philadelphia resins was just bought out.
see http://www.itwpolymercastings.com/

EricR,

That insert post looks pretty cool.

Regards all,

Cameron

[greybeard]

Hey John,
Can you show me how you calculated that?
Sorry really dont know.

regards,
Roy

I cheat.

I use a nomograph from www.centrifuge.jp/calculation/

Cheers
John

[turnpoint]

Hi there,
I just found this discussion, so please excuse me if this has been covered. I just read as many posts as I could mentally handle-- about three hours reading--learned a lot.

I have a background in composites-- mostly tooling for high performance carbon fiber parts.

I find this discussion very interesting in that I am currently constructing 2 heated laminating tables that have a poured core of epoxy with graphite chip aggregate. The graphite in the core helps to transfer heat efficiently. The more graphite chip aggregate I use the less epoxy I have to use, and the better the table transfers heat. The tables are laid up in a mold. The laminate schedule is: 1/8" tooling gelcoat surface, 3 layers of 19oz triaxial cloth, 1/2 inch of the epoxy graphite, heating elements, 1/2 inch of epoxy graphite, and then 3 layers of 19oz triaxial cloth. Once out of the mold the tooling gelcoat is machined flat (large gantry 5-axis machine). With the heated flat table it is very important to have a balanced laminate which will heat up uniformly-- otherwise it warps.

Anyway, what I was wondering is if anyone has created carbon fiber box beam structures with the epoxy granite fill. I ask because carbon fiber is stiffer and has a better coefficient of thermal expansion than steel. Also, it sounds like you already have to make a mold to create E/P parts--- so why not put down machinable epoxy and carbon fiber laminate first before doing the E/P pour? Then the E/P will act as a core in compression, while the laminate skins take the tension loads -- which they are much better suited for(simplified).

Just wondering if someone has done this before,

Also, a good source of epoxy in bulk is http://www.appliedpoleramic.com/
They have many epoxy formulations that could work well for this-- probably one of their low viscosity infusion/RTM epoxies would be best. I think their minimum order is 5 gallons-- per gallon it is about $45 for their basic resin system.

Hope this helps,
Brandon

[Eric R.]

Hi Cameron,

I have received some sample inserts (brass and stainless steel) and these are really nice inserts. Exactly what I was looking for. Next week I will make a small mold and create a sample using some of these inserts, to find the best way to place these in a secure and precise way. Actually I am currently designing a large 5-axis milling machine, mainly for wood and foam milling, and as I am crazy enough not to follow traditional paths, this is what I am planning to do:

• the machine base: E/G. For the machine base I have a large drum (200 Ltr.) of Hexion EPIKOTE 05102 with EPICURE 04998, which promises a lot of fun
• moving bridge: carbon fibre beam, using vacuum resin injection. I have a nice stock of bi-axial and tri-axial carbon fibre, to create a large diameter/thick wall beam. The challenge will be to get the inserts positioned and molded in a secure and precise way. But I will find something . For creating the beam, I will place the entire mold in a vacuum chamber, therefore not risking deformation of the mold due to bending of the walls as a result of the vacuum.
• motion systems: THK caged ball guideways and linear motors; Z-axis balanced with gas spring. That is also why I want to use a carbon fibre bridge, just to lower the moving mass, which will allow me to use somewhat "affordable" linear motors. Carbon fibre is ideal for this, also due to its superior characteristics compared to steel with regard to stiffness and limited thermal expansion.
• CNC controller: EMC2 based

Brandon,

I will be creating a carbon fibre box beam as you refer to, but without the E/G fill as I will be using the beam as a moving brigde and therefore want to reduce weight. In order to get the right mechanical characteristics, the beam will have rather thick walls (approx. 1/2 inch), which will take quite some carbon fibre, keeping in mind that I will create the beam using vacuum resin injection. Actually I am thinking of constructing the beam out of multiple smaller beams, packed together into one large beam, to increase the stiffness even further.
Regarding your suggestion for an E/G filled carbon beam: what application do you have in mind for such a beam?

Any suggestions/comments/questions are welcome!

Regards,

Eric

[turnpoint]

Hi Eric,
Sounds like the machine you are building will be top shelf- I hope you post your progress.

In regards to your response:
I was thinking that the best use of the E/G filled carbon fiber would be for all non-moving parts. The goal would be to have a stiff and stable structure. Carbon fiber for low CTE and stiffness-- E/P for dampening and mass.

I agree that the moving bits of the machine would be better as pure carbon fiber beam structures to reduce the mass-- and to get better acceleration and deceleration. My 5-axis gantry weighs a couple thousand pounds-- everyday I watch it and think there has to be a better way.

But let me know how your tests go-- I have some experience that could help out. Infusing carbon fiber can be a challenge--- the fibers pack more densely and are harder to pull resin through. With that said there are some things that will make infusing carbon fiber with epoxy more enjoyable. First is to use the lowest viscosity resin you can find-- under 400 cps. Second is to raise the temp of the resin (including fibers and mold) to at least 90 deg. F (this necessitates slower hardners-- you want at least an hour pot life at 90 deg). Third is to infuse into the carbon stack in the Z direction (don't expect to pull resin across your part more than 6 in). You want to set up supply lines and appropriate flow media on the top of the laminate stack and pull the resin down into the carbon laminate.

In regards to threaded inserts in carbon fiber--- the 1/2" wall thickness will work well for this. But laminating them in place seems like it will really test your patience. It would be simpler to drill and tap the inserts into the beams after the infusion. Best would be to have this done from a CNC shop that deals with carbon fiber--- you could machine the linear rail mounts and drill the holes for the inserts at the same time.

Brandon

[veteq]

everyday I watch it and think there has to be a better way.

Just woke up, thats a nice way to start the day.
Deja vu, LOL
I know that loop!

Regards,

Roy.

[ckelloug]

Eric,

It sounds like folks with a lot more experience than I have are advising you but my research indicates that on paper, 3M novec 4430 or novec 4432 might make your wetout go better on the carbon fiber.

see http://products3.3m.com/catalog/us/e...er/output_html

Also, no fair that you can get Epikote 05102 on your side of the pond. Hexion doesn't sell it here.

[veteq]

Just wanted to show my last LOOP in designing,
Since machining is so expensive i`m back at the point off building compleet EG, making a nice base that is vibrated in a wood mould. Then glue in the inserts that are placed on a precision jig with liquid metal, align the linear bearings and seal them with liquid metal.

The L-frame is on the way...

Regards,
Roy.

[Eric R.]

Hi Brandon,

Thank you for your advice, which is highly appreciated! I have done a lot of research on vacuum resin infusion myself, and your information is a welcome addition to it! I recently did some testing and so far the results are very satisfactory. For the final beam I am still fighting with the entire setup, as I want to apply the infusion lines (and all other stuff) on the inside of the beam. When I find some time, I will make a schematic drawing of the setup and post it. I would definitely appreciate your experienced vision on my plans. I will also see if I can setup a FEA, to simulate the flow pattern.

Hi Cameron,

Sorry you cannot buy the Hexion stuff on your side of the Atlantic. As I travel quite often to the US, I could bring some with me, but I doubt whether federal regulations will allow me to do so, as for these chemicals special approvals may be required.
Regarding the resin for the carbon fiber, I have not made any final decisions yet, but probably I will not use the Epikote 05102 as Hexion cannot provide me with sufficient data on product characteristics for the cured product in combination with any fibers. They only have info on their intended use, which is for mineral casting purposes. Probably I will select Hexion 04908 instead, which is meant for infusion purposes. It has a very low viscosity, long potlife and requires elevated temperatures for final curing. Thanks for the 3M Novic link, as that looks very promising!

Regards,

Eric

[turnpoint]

Hi Eric,
I look forward to seeing pictures of your machine build. I really think your onto something and want to know your progress.

For setting up the infusion lines inside the box beams, the easiest would be to use a foam core inside the box. First you would find the right combination of resin lines and flow media by doing small tests on a flat table with your full laminate thickness. (These tests will also give you your laminate thickness which you need to make the foam core the exact dimensions you need). Then you would use a hand router to create this pattern on the surface of the foam that will be inside the box beam. You could leave the foam in place as the weight will be fairly negligible for your application. If you don't like it in there, you could do this with polystyrene foam that you melt out with Acetone after the beam is cured. BUT, Melting with Acetone will leave you a toxic sludge that you won't know what to do with--- so better to leave the foam inside.

Hope some of that is useful,
Brandon

[Eric R.]

Hi Brandon,

When I have completed all tests and prepare for the big thing, I will make sure to document everything, including pictures, and post it.

I have also been thinking of using the approach that you propose, using a foam core, but I am currently experimenting with a removable and reusable kind of frame, to which I apply all the materials in an opposite order compared to traditional vacuum resin infusion and use the mold to achieve the exact dimensions. I have not figured out all the details yet, but I am working on it.

Thanks,

Eric

[jvalger]

I've been reading this thread with great interest. My company does carbon fiber vacuum infusion and I've had the same thought as I look around the shop at our machine tools. Carbon for stiffness, wrapped around a massive core for damping.

Eagerly anticipating positive news from your trial, and with apologies for a long post, here are lessons from the trenches of carbon fiber vacuum infusion I wish someone had told me way back when.

You will fail the first few times you try it, no matter how much you prepare. So use fiberglass, which is 10x cheaper, until you get the setup and process right. However, once you get it to work once, as long as you do it exactly the same way, it will work every time. This makes it less attractive for one-off projects than for repetitive part building. Prepregs are higher cost but much more likely to work the first time if you are only going to do a single part, and the results are comparable.

Carbon does not take side impacts at all gracefully, so either another laminate (e.g., kevlar or sheet metal) or a good thick gel coat is essential if you want the part to enjoy a long life in a machine shop environment. You don't want a tool to be dropped on its point onto a layer of infused carbon fabric.

Composites fail primarily because of interlaminate shear - layers separating. If you have a foam core, the core stretches with the adjacent carbon layer so no shear at the junction. If you have a stiff core like E/G, you have to give a lot of thought to how you'll keep the carbon layer from separating from the E/G layer as things vibrate and deform, or expand at different rates when heated. Interested in what others think about this.

If you can factor out shrinkage and creep, vinyl ester infusion resins are MUCH MUCH easier to succeed with than even low-cps expoxies. Especially if you're a newcomer to vacuum infusion. Around 120-150cps is typical, flows like water. It is often realistic to count on VE flowing laterally through the fabric without flow media. Not a huge difference in mechanical properties vs. epoxy and you aren't relying on the resin for strength anyway, the carbon is there for tension and the E/G for compression. Typical vinyl esters also have a much higher deformation temp than typical epoxies and do not require a post-cure.

There are three different ways to seal the vacuum: plastic bag/sheet, rigid mold, and silicone bag.

Plastic bags tend to push down into whatever they are on top of, specifically the fabric, thereby restricting space for resin flow. You have to either use flow media on the outside, which also requires a release ply, all of which adds to the setup time, or on the inside, which can create interlaminate shear problems. Another big downside is the effort required to seal it up. Don't underestimate this if you've never done it. Vacuum bagging is very tolerant of small leaks but vacuum infusion is not, and if anything goes wrong before the resin gels, you will likely have to toss the part and start over.

If you use a two-part rigid mold (RTM) you don't compress the spaces out of the fabric weave so resin flows much more readily. Often you can do away with flow media altogether.

A silicone bag is a middle ground. Stiffer than sheet plastic so easier on flow, cheaper and easier to build than a second mold half.

Finally, beware of heat buildup. Even a thin section can easily generate heat of 200F or more. A 1-in thick hockey puck section left in the bottom of a cup can rise to over 300F. Heat has to be released in order for the resin to cure, and the thicker the resin the harder time heat has flowing out. Test under realistic but small-scale conditions to make sure you aren't going to melt your mold or other tooling. Remember that heat buildup also dramatically affects gel time.

Jeff

[the4thseal]

you can not buy small amounts prepreg. i wish.

[turnpoint]

Jeff,
Great post--- I haven't thought about the junction of the carbon fiber and the E/G --- but I think you are right about that as a possible issue. If you could get a layer of toughened epoxy between them, that has enough elasticity, this could help prevent delamination.

Using pre-preg would be more reliable if you aren't comfortable with resin infusion--- which can be a real art. I thought I just saw a special for prepreg fabrics at Fiberglast in small quantities.

http://www.fibreglast.com/?gclid=CLH...FRZCagodfwRg4w

Brandon

[jvalger]

Here is one online source for as little as 1yd, though only 3k twill and a single choice for resin:

http://www.fibreglast.com/showproduc...brics-215.html

Or, find a shop that uses a lot of the stuff and wouldn't mind parting with some.

Or, do it yourself. Simpler than it might seem with manually-operated, diy equipment. Pass the fabric around one roller at the bottom of a resin bath to wet it, over a second, higher roller, then down between two rollers (one acting as the drive roller) to press out the excess resin. If you are careful, you can even make the V between the two main rollers double as the bath - the fabric itself prevents the resin flowing out the bottom. If you don't want to build your own I've seen these online.

Jeff

[turnpoint]

Hi Eric,
You stated: "I am currently experimenting with a removable and reusable kind of frame, to which I apply all the materials in an opposite order compared to traditional vacuum resin infusion and use the mold to achieve the exact dimensions. I have not figured out all the details yet, but I am working on it."

In response:
There is a slick way to make a reusable interior form. I have only seen this used on hollow spars, but it seems like it could be used for simple machine tubing.

You shape the interior void space using closed cell upholstery foam (squishy). You could incorporate the resin pathways into this rough foam shape too. Once you have the shape you want, you coat the foam with a film of 2 part silicone (the kind you make reusable vacuum bags out of-- ie Airtech or similar) . One warning is the foam and silicone are very hard to work and thus you might want to do this with the help of a mold.

The beauty of the silicone/upholstery foam forms is that, under vacuum, the closed cells will expand and act as an interior bladder-- compacting the laminate and reducing resin use. But once the part is cured, you can peel/stretch the form to get it out of the tube. Because it expands, a smaller interior size form can be used which makes loading the dry stack into a two piece/clam shell mold a little easier.

Brandon