Minerva Monitor: "Patience my ass, I'm going to go build something!"

[CraigJ]

Looks like a complicated, yet fun project. For the table saw version, even with my glasses, I can't find the width of the boxes. I could guess, yet I'm sure that I'd be wrong at 12 3/16" ?

Also, as an alternative to LBL, could one use something like this;http://www.johnboos.com/Items_page.a...alse&csa=False and get very close? Perhaps Corian as the outside layer?

Cj

[Steve Manning]

Looks like a complicated, yet fun project. For the table saw version, even with my glasses, I can't find the width of the boxes. I could guess, yet I'm sure that I'd be wrong at 12 3/16" ?

Also, as an alternative to LBL, could one use something like this;http://www.johnboos.com/Items_page.a...alse&csa=False and get very close? Perhaps Corian as the outside layer?

Cj

Hey Craig ...... it wouldn't be fun if it were easy The actual finished width for the Minerva will be 15", front baffle is 2 1/2" thick and the back is 2". Jon would have the numbers on the table saw version. I imagine what your prescribing for the maple and Corian combo would be pretty solid combination.

[JonMarsh]

I bet you could use that stuff, though I have to say it comes pretty dear. :W

I'm planning to run the current file in Layout again now that I have a full license, but I'm also planning to completely re-do the model and drawing, as I've been learning about SketchUp's quirks, and why one must set the segment accuracy and other features and one must create components before moving pieces together- Steve has been helping me, as well as some good books and tutorial resources.

[CraigJ]

I bet you could use that stuff, though I have to say it comes pretty dear. :W

I'm planning to run the current file in Layout again now that I have a full license, but I'm also planning to completely re-do the model and drawing, as I've been learning about SketchUp's quirks, and why one must set the segment accuracy and other features and one must create components before moving pieces together- Steve has been helping me, as well as some good books and tutorial resources.

Thanks for the quick responses Steve and Jon. Regarding the :W , the secret project, unofficially named "Chautauqua-B" for you and my Grandmother, has become quite the challenge. Mating the woofers to the line source has been interesting. Dual opposed is only O.K., next step is 72" stacked H frames; fun times.

I will wait for the final cut sheet for the table saw version for the Minerva.

Cj

[JonMarsh]

I will wait for the final cut sheet for the table saw version for the Minerva.



Cj

A reasonable idea... keep in mind all of the cut parts are sitting out in the storage/work shed right now for the table saw version, so this is a bit further along than poorly documented vapor ware...

[Steve Manning]

Made some saw dust ....

Got some test pieces cut today, of course pictures below. These used shop maple and not the good stuff. This was more an exercise to check parts holding and files. We discovered a few quirks going from Sketchup to Aspire for the tool paths. We already have a couple of options to resolve those issues, some more test cuts to follow. Also waiting for the "good" bits for the machine, (hoping by weeks end) these parts were machined with what was on hand and had a little chipping as a result. Overall a good test run.

[sdl2112]

Nice progress Steve! The CNC opens up so many possibilities.

[dar47]

Looks like a complicated, yet fun project. For the table saw version, even with my glasses, I can't find the width of the boxes. I could guess, yet I'm sure that I'd be wrong at 12 3/16" ?

Also, as an alternative to LBL, could one use something like this;http://www.johnboos.com/Items_page.asp?s=r&SalesCode=&Prod=1&Page=6&Group= 10&hdg=Maple%20R-Board%201-1/2%22%20Thick%20-%20Reversible%20%20&nsf=True&nsfc=False&csa=False and get very close? Perhaps Corian as the outside layer?

Cj

Hey, Graig had the idea of using maple bread boards for fronts on my Modula MK2's and CNC'ed the wave guided fronts but the breadboards failed along a glue joint. I ended up redoing the cabs but this time I bought 3/4" x 1 1/2" finished maple mill work and glued my own slabs with epoxy and they never failed after that. Don't want to poopoo the idea but make sure their of high quality.

Bread board version that cracked.



Look similar but this is the new version epoxied,



Steve lots of test pieces looking good.:T

[JonMarsh]

What kind of glue were you using on the original version? And did it crack on the glue joint, or the wood? (You mentioned glue joint, but I'm just double checking).

I've been researching glue a bit more- our current plan is Urethane glue (Steve's got Titebond, I have a new bottle of Gorilla Glue) and I know it's super on end grain, out performing yellow glue there, but just matches yellow glue in some other modes. Of course, I'm the long time epoxy guy, but with this much stuff to glue, two part epoxy would be both expensive and time consuming.

Thoughts?

It could be that we'll just have to spring for a few cans of West System with the 200 series hardener- been there, done that before. Feedback appreciated, even better if backed with hard data from anywhere.

[Steve Manning]

The stuff I picked up http://www.woodcraft.com/product/143...-glue-4oz.aspx was recommenced at Woodcraft for this type of application. They recommended urea formaldehyde as being a little better http://www.highlandwoodworking.com/u...ystgallon.aspx, but it's a royal pain to use, very messy and the recommendations were not to use under 70 degrees, which my garage certainly is not this time of year, even with the heaters going. Both can be used for bent wood lamination's which put's a lot of stress on the glue joints, so I figured it would be a good option.

[sdl2112]

I 'm a novice woodworker but I would recommend something like Titebond III. It is stronger than the wood and I think the LBL layers would separate first. I tried polyurethane glue and what a mess. I found it hard to work with and without any benefit. From what I remember it was hard to judge how much to apply and it got everywhere. With the tight fitting assembly and the number parts you have I think a PVA glue is best. Done properly I've never had a glue failure. You know what they say....KISS :B

Here's an article that's been floating around some time from Fine Woodworking.

HowStrongisYourGlue_FWW.pdf

It might be best to buy a small amount of your candidate glues and try it on some scrape pieces.

[CraigJ]

Hey, Graig had the idea of using maple bread boards for fronts on my Modula MK2's and CNC'ed the wave guided fronts but the breadboards failed along a glue joint. I ended up redoing the cabs but this time I bought 3/4" x 1 1/2" finished maple mill work and glued my own slabs with epoxy and they never failed after that. Don't want to poopoo the idea but make sure their of high quality.

Bread board version that cracked.

Nothing wrong with poopooing an idea if you've already tried it and it didn't work. You actually got me to run to storage and look at my 1 1/2 maple and here is what I found; :jawdrop:






Looks like the glue failed on the end grain. Thanks for the tip and your MK2's are one of my favorite stand mounted speakers.

[JonMarsh]

Thanks for the kind words. They're one of mine, too, especially for the size.

To some extent, I think woods like Maple and LBL pose outlier characteristics, maple more so, because it is so hard, and so strong, and can be subject to a bit of warping or shift due to environmental factors. My experience so far with LBL is that it is more "stable" or easier to work with in glued up assemblies, but this will really be the project to assess that, won't it?

[knowledgebass]

I have had good luck with plastic urethane glue for the laminations on my ansonica build (DAP Weldwood - just mixes with water). I did have some issues with the wood plys splitting, but not any glue seams (goes back to the glue being stronger than wood), and not since I've finished it. It's used for bending laminations because it won't creep under load/stress. Most brands have a long open time and yes, they can be very temperature sensitive. I looked for indications of others using it in speaker building and could not find any examples. I stuck with it based on what l found recommended for other similar wood working (e.g. bent laminations). I did my crack repairs/filling with epoxy. I would consider using epoxy for the laminations but I always seem to make a mess with it! Clean up with the urethane glue I used was water. Also consider that different glues will will have different color glue lines. That poly urethane glue photo is pretty dark. Epoxy is clear and makes a good surface prep (e.g. grain/surface filling) for other finishes. All of these traits are going to matter at some point during finishing. Hello decision paralysis!

[Steve Manning]

I 'm a novice woodworker but I would recommend something like Titebond III. It is stronger than the wood and I think the LBL layers would separate first. I tried polyurethane glue and what a mess. I found it hard to work with and without any benefit. From what I remember it was hard to judge how much to apply and it got everywhere. With the tight fitting assembly and the number parts you have I think a PVA glue is best. Done properly I've never had a glue failure. You know what they say....KISS :B

Here's an article that's been floating around some time from Fine Woodworking.

HowStrongisYourGlue_FWW.pdf

It might be best to buy a small amount of your candidate glues and try it on some scrape pieces.

Hey Scott .... I typically use Titebond III for most everything, the only issue with a project like this, is it only has about a 10 minute open time, which when you stacking the layers with dowels having to go in place, that's not much time and rushing is never a good thing. When I built my Modula's I used Titebond Extended which has an open time of ~ 15 - 20 minutes as I recall ....... I ended up having this happen ......



I was not amused to say the least. I did use epoxy to fix this and it's been happy since.

This was using BB, and I've seen some others have similar issues with BB, both at the glue joints and between plys. That's one of the reasons for going with the LBL and looking at a different glue. The stuff I picked also has an open time of up to 45 minutes which also takes some of the pressure off as well.

[sdl2112]

I didn't realize that polyurethane had an open time of 45 min...that's good. Good luck and I look forward seeing the stack-up :B

[Norm]

CNC Notes

Steve appears to have most of the following well in hand. Here is an overview of my experiences in case it will help.

For a few years now on and off I have been using CNC to build speakers enclosures. Here is an example using Fostex F120A from my days enamored with full range drivers.



I have experimented with spinning tool CNC routers/mills, laser cutting, and waterjet looking for the magic process that outputs perfect parts as fast as a laser printer can spit out pages of text, still looking for that. These days I am happy if I get some good parts before they turn out the shop lights and kick me out.

I will say laser cutting and waterjet are paths I have tried but will likely not use as a primary source in the future. Let us start with what I learned their then get onto spinning cutters.

The F120A box above was laser cut. With wood of course that leaves charred edges. The shop that did those cuts also warned me, and I elected to go ahead, that their laser output a cone of light. Thus the thickness edges were not perpendicular to the length & width sheet plain. When the layers were glued up a truly ridiculous amount of sanding was required to produce the result seen above. Much of the sanding was to remove the charred non-right angle edge. However it was also before I started pinning the layers together using dowels. Without dowels during glue up the layers move which is VERY problematic. If a layer is off even 0.02” the eye easily sees this. On the side it stands proud it is quick and easy to sand down to match adjacent layers. It is the side where one layer is lower than their adjacent layer that leads to an unhappy place. There the choice is to feather the edges together so in that area one has a low spot or dip. Or sand and sand and sand then sand some more until all the high layers come down to match the low one. The thought of doing that with something as hard and expensive as your bamboo material I find terrifying.

Waterjetting provided accurate shapes. The shop thought I was crazy wanting to put plywood on the flooded work surface of a waterjet cutter. There were also a few unexplained nicks in the cuts where the jet was apparently deflected; I was not present when the parts were cut so these remained a mystery.

Now let’s talk about the manly art of cutting with razor sharp tools spinning at 15,000 rpm zipping about crashing into things.

I have traveled the road of both buying shop time on commercial CNC mills and hands on* DIY thanks to machine time at the local maker space. Nomenclature note these machines are often called CNC routers in the woodworking arena but in 2016 they all have some Z axis capability so henceforth l am using the term ‘mill’. *Actually, hands off when the CNC mill is running, that’s the point and tends to keep your ten digits attached and complete. After hitting ‘start’ just stand back and watch with the ‘stop’ button nearby.

The advantage of working with a commercial shop is they often have better machines, scheduled calibrations, with experienced programmers and operators. The critical aspect is your relationship with the shop and ability (and by ability I mean budget) to make your work worth their effort. Where I have located shops with woodworking CNC capabilities they are in the sign and cabinet industries. On three different occasions I have had my jobs greeted with enthusiasm at first only to have that fade. I believe our high-end designs look interesting then once they get into it the time involved compared to their regular production signs, cabinets, and fixtures kills the interest. In each case I provided paper and Auto-CAD DXF format plans and received a fixed price quote. In hindsight I believe they underestimated programming time and time futzing on the machine perfecting the runs. As the hours passed and their profit shrunk I was greeted with less and less enthusiasm. In two of the instances they just quit returning my calls, eventually I can take the ‘go away’ hint!

The advantage of joining a Fab Lab (http://www.fabfoundation.org/) that has CNC mill capabilities is you get access to the soft and hardware tools to do it yourself. I never refuse to return my phone calls! Challenges include climbing the software and hardware learning curves and limited time on the machine. Being 100% geek I enjoy the process of learning to import my design into the tool path program and defining all the cuts. Then transferring the design to the machine, spinning up the spindle, and making sawdust is certainly this geek’s odd idea of fun. The limiting factor is machine time. My local FabLab has a limit of 2 hours/day for a given user on the CNC mill to share the resource among members. I have seen users cut a 4x8 foot sheet of parts in that time, on the other hand I have run out of time and aborted a job on a sheet half that size. It all comes down to the complexity of your design and tool path programming decisions. One quickly learns the art of maximizing part yields and cut rate into the time allotted.

Enough rambling about the search for machine time, let’s get into the nitty gritty using Jon & Steve’s Minerva Monitor project as an example.

Above Steve wrote “I've been going through a bit of a self induced crash course on vacuum tables for CNC Routers of late….et. al….”. I 100% agree with you “this just seems like a kluge” assuming you are looking at one of the systems that pulls a vacuum thru a MDF sacrificial layer. The first shop I was “fired” from I believe contributing to that was frustration experienced as “parts coming loose during the process” off the vacuum platen that worked fine for his signs. Of course anything can be made to work if we work at it hard enough. When Wilson Audio was advertising with a campaign featuring photos of their factory and workers one photo featured their CNC mill. On it was a job specific vacuum hold down platen with the outlines of the parts to be milled in relief for tool clearance and inside each cut path a gasket and inside that holes to pull the vacuum through. That’s how you hold down parts with vacuum and make it stick.

The machine I am learning (Shopbot programmed with Vcarve Pro) can be setup with vacuum hold down, but FabLab uses the more DIY friendly method of clamping and screwing the work to a sacrificial bed. Of course this has its own set of tradeoffs. Parts can still come loose if I screw up the clamping or tabs and when using up-cut tools the center of the material wants to lift up reducing Z-axis depth accuracy. One needs a plan on how not to hit your hold down screws (says the guy who recently wrote off an end mill on a steel wood screw). The collision with the hold down screw occurred because I eyeballed where to place it on the sheet. Lesson learned, those now go in the tool path programming. I like to drill the project’s holes in the first pass so I have taken to starting with the material clamped down along the edges then run a pass drilling the alignment and hold down screw locations. Then cutting stops while I get busy with the electric screw driver driving wood screws into the programmed locations. Given the glue up layer alignment issue mentioned above I am somewhat maniacal about my alignment dowel holes and do not use them for hold down during cutting. Some of the time spent drilling extra holes can be taken back, while the dowel holes need to go all the way through the material the hold down screw holes I can drill just deep enough to mark the location letting the power screwdriver complete the hole when I screw the material down.

The advantage of screwing down the material to the sacrificial bed is it stays put during cutting. The disadvantage is the number of screws required can get out of hand. My last speaker project using CNC milling was a mini-monitor built up from layers of birch ply forming the sides and back of the box, thus many parts are ‘C’ shaped. By design this allows tight nesting of the parts helping to maximize the part/scrap ratio. The array covered a 4x4’ (122x122cm) sheet and as I programmed tool paths I realized by the time everything was cut even with tabs between the parts and scrap the pattern was such a lacework things could move in the XY directions due to cutter deflection. So I had the bright (it seemed at the time) idea of deleting all the tabs then adding 4 (!) screws to every C shaped part. The lure of this idea is the parts are 100% ready for glue up when they come off the mill. The reality is so very many screws I must have spent 10 minutes screwing it down and that many again freeing my completed parts.

Z-axis accuracy also plays into all this greatly. Factors affecting how accurate cut depths can be maintained include the mill’s repeatability and calibration accuracy, delta of the materials thickness, bit and material deflection during cutting, work bed flatness; and accuracy of calibrating the Z-axis during job setup. When it is all spot on the onion skin effect mentioned above to maintain vacuum clamping and tricks like stopping a V-cut leaving just the veneer so the box can fold and glue together can be obtained. The rig I presently use alas is at the other extreme, it cannot seem to maintain Z-axis repeatability to 0.2”/5mm. One part of a job may fail to cut through leaving unintended onion skin, another area and tabs disappear from my parts and are cut into the sacrificial bed. On the plus side this machine maintains wonderful accuracy in the X & Y axes. When glued up using ¼” dowels in the interior locating holes the C shaped parts aligned so well 10 minutes sanding and it looked like the assembly was carved out of a 10 inch thick piece of plywood.

So, given all that when Jon and Steve announce they have +$1k of bamboo plywood about to go on a brand new just installed CNC mill red lights and alarms go off in my head. Here are my suggestions specific to this design.

1. This vertically layered relatively simple design is a good one for an initial use of CNC milling. Fewer larger parts will speed cutting. When I see designs adding complications like wave patterns inside the box or details like ergo’s above I think hours added on the mill. These mills are like Jon on his Yamaha, you can go faster in a straight line compared to twisting left and right. Also the plan to revert to manually cutting the driver holes & rebates nicely sidesteps the Z-axis calibration issue.

2. We see Steve already following this next advice. Test cuts are your friend. I am guessing this is the first time this shop will be cutting bamboo plywood (or as I call it “epoxy phenolic with some grass thrown in”). The chip load tables and formula will get you started on spindle RPM and feed rates but they yield a range to test within. For example the from Asia to the big box store sort-of birch plywood I used for my last project yielded a cleaner cut with less inner layer tear out at faster feed rates, quite counter intuitive. Setting aside how anxious one is to produce the project parts and running controlled tests pays off in cut quality and allowing as fast a feed rate as possible. Once that feed rate and RPM setting is chosen given the investment in materials I would be sorely tempted to get a couple of sheets of less expensive material and practice cut examples of all components as Steve is. This to verify the design with prototype pieces, test hold down/tabbing scheme and to check the run times against the tool path software estimates. In my Shopbot + Vcarve system the estimated run times are always low, sometimes by 100%!

3. Regarding the glue discussion I have used Titebond II for all my laminated cabinets and have never suffered a failure. I pour some into a dish and use a disposable foam brush to apply a thin coat across the entire mating area. With a little practice this yields minimal squeeze out compared to previous method of applying from original bottle and spreading. Of course bamboo plywood is so full of the phenolic like materiel binding it together I would be reaching out to both the material’s manufacturer and the adhesive companies tech support for advice based on the chemistries involved. Steve on your Modula's I that suffered the crack I am curious if those had a long dowel or something like all-thread rod running embedded in the length of the stack-up? I am wondering if that crack was due to adhesive failure or differential expansion of materials within the item.

[Steve Manning]

Thanks for the write up Norm ..... a lot of information to digest.

I can certainly see why water jet and laser did not work well for you, in particular the laser. I ran an electron beam welding facility for 6 years and your right, the "beam" is not like you see in the movies, it's tapered, think optics and lens, etc. Once you add any kind of thickness to things your in trouble if you want parallel edges.

As I noted earlier, we discovered a couple of minor issues with our test cuts that has some rework underway with drawings. The machine cuts what you tell it to cut, if the programming is off so will the part. If we had to go with screws or clamps that would certainly have to be taken into account, bits running into screw heads would get expensive considering the price of the specialty bits that are coming to cut the bamboo. Better to spend the time up front with the coding than with sand paper later.

There was no issues with parts moving during the cutting, though more test pieces are going to be run and if needed tabs will be added for holding parts in place if it's deemed necessary. The shop owner does not like screws and feels that between skinning and tabs there should be no problems. At this point I don't see a reason to argue his opinion, but will if things look iffy.

I have also spoken with some other board members that have cnc'd bamboo and have discussed their lessons learned with the shop owner. Types of bits, feed speeds etc, We will take it slow and go from there.

Though to some the Minerva cabinet design seems over the top, it is actually a rather simple design in it's concept. Your certainly right about getting too fancy and getting in over your head. Part of the reason for doing this version of the project, is as a proof of principle with cnc and bamboo. Considering that I did my Modula project http://www.htguide.com/forum/showthr...ghlight=magico with a jig saw and a hand router, we were looking at a more efficient, accurate and stable way of building a similar cabinet design.

To answer your question on the Modula cracking, no bolts or dowels were used to hold the lamination's. Though I do expect that differential expansion of materials did come into play some with the baffle being glued in place. What it does say though, is that BB moves, as does any wood, though I was not expecting to see as much as it did. This is one of the main reasons for going with the lbl, it is far better in the stability department.

[dar47]

Lots of good brain storming going on here. :T

For the glue Steve mentioned the urea formaldehyde which I first encountered in the kitchen cabinet business back in the 80's, used in a combined 3 way pressure with micro wave to glue door panels from random length material. It was heat activated and you could literately smack a piece on steel right out of the press and it would not fail on a glue joint unless it was maple, Oak, Alder anything semi porous no problem, Maple it didn't absorb into the grain as much. This is like the Bamboo. Notice the material used in the glue strength publication, oak everything sticks to that and it just shear glue strength. I think for this application you don't need ultimate glue strength, you need surface tension grab strength and that is what epoxy does so well on many kinds of surfaces.

Now that wonder glue from the 70's Weldbond, haha. My dad loved that stuff and I got a jug left over from cleaning the parents house out. Coolest thin my mom did with that is mix with sand and a little cement and you have a great long lasting patch. She patched her front concrete steps and it did last.8O It has great tack if I remember and could work well in this application especially if there is any concerns with shifting layers with not completely tight dowels holding the layers as your clamping them. If you have time and some extra pieces you do a small longitudinal study to see? Work time is not as long as it tacks faster.

With the investment and if I was sure the layers had no sliding issues with tight dowels, the epoxy gives a controlled work time with the added benefit of the oozing epoxy is worked into the end grain both inside and out. It will feel any irregularities but the lbl is very uniform and stable as other have stated. This is a great sealer base for your finish too. If you have 2 sets of hands and plastic down no problem. After Ben got started in it with the Ardent's, you couldn't pull it out of his hands it went on every thing, haha. Like anything working time, temp and some experience with the glue helps, a little like finishes I guess.

[dar47]

Urethane glues, not a fan it's for outside high moisture areas, it expands and can foam which I would think is not good for this compressed application.

http://www.adhesives.org/adhesives-sealants/adhesives-sealants-overview/adhesive-technologies/chemically-curing/two-component-%282-c%29/urethane-adhesives

The DAP Weldwood which is different is a urea formaldehyde based glue, the white stuff I have (Weldbond) a PVA based.

http://www.thewoodwhisperer.com/videos/a-sticky-situation/

Weldbond says it can bond most anything, so bamboo might be a good application.

http://www.weldbond.com/cabinetmaking

If your going to use Titebound, I would think 1 has the longest setup time, 2 is stronger and 3 is for outside not needed and expensive.

[CraigJ]

Dar47,

The knowledge base here quite often amazes me. I've use Gorilla Glue to build a dozen+ kite boards in laminating 3 m baltic birch. Urethane is not fast to apply to pieces and the amount of expansion is ? and it is messy. How much water to apply...is the wood porous enough.....

Has anyone contacted the MFG of LBL and asked what adhesive to use for this application?

If West Systems works, from experience, I would think that rolling it would be the fastest and easiest method.

Cj

[dar47]

No kidding, I have learn a lot as we have a great mix of experience with wood guys and genius sparks and wire guys.;x(

Ya, I have use the gorilla too but didn't like it as it seems to have never ending oozing and foam like appearance at the joints when cured. In panel lay up like this it's expansion would worry me. In a product like PL premium (paste like) great for foam and carpet lining of cabinets, instant grab as it's a construction adhesive. Epoxy can be rolled, I like to pour then use a small foam brush and then strike the surface with a plastic scraper to level. I have never had a bad result with West Systems so there is that comfort level just not the expense level. 8O

As with the Ardent's when you get through all the planning, hauling the sheets, CNC and have a pile of expensive LBL in front of you, you just don't want to have a break down in your glue up. Ben and I got nothing conclusive on the glue to use so we went for "I know this will work". I have some lbl left over for some small fronts and I may try the Weldbond on it, but that's in the summer.

[JonMarsh]

A lot of interesting points here... I should make some test panels or pieces to lay up, and do some test gluing- I have a big bottle of Gorilla glue sitting in front of me. (Well, not right now, I'm in Starbucks, and it's actually sitting in front of the HDTV in the mancave).

The points about LBL ply being an bamboo based phenolic material are right on, of course; that's why I like it so much, why it's so dimensionally stable, and easy to work with using epoxy glues- one piece, the Nascent front panel, I've glued up with Titebond II, but everything else, and I mean everything, has been one epoxy or another, the majority of it with West systems with (I believe from memory) 212 system hardener, the long working time version.

I have some scrap pieces I can do some gluing test articles- probably not this weekend, as I've got to get the car in for service (where I normally have it done) and will have to hang out at a near by Starbucks while it's being done most of Saturday. But who knows, maybe I can get to it on Sunday, though there's a lot of stuff on the agenda. The time in Starbucks should give me the opportunity to work on Ron's crossover with the 12MU as long as I get all the files together on my MacBook Pro tonight.

I'll also see about installing SketchUp and spending some time updating the drawings, and doing the video training material I bought from Taunton publishing.

[Steve Manning]

I was also thinking of performing some test pieces from the drop out when the parts are machined. Let it setup overnight and do a little destructive testing. :twisted:

I've attached the write up from Titebonds website on the Polyurethane Glue for those interested. Scott I was off on the work time, it's 30 minutes (which is still good), clamp time is 45 minutes, which is also good.


Titebond Polyurethane Glue

Titebond Polyurethane Glue is a breakthrough in adhesive technology. It is the only polyurethane glue to combine a long 30-minute working time with a short 45-minute clamp time*. It is a versatile, professional-strength glue specifically formulated for multi-purpose applications.

In addition to its superior wood-to-wood performance, Titebond Polyurethane Glue is ideal for metals, ceramics, most plastics, HPL, Corian®, stone and other porous/non-porous materials. It is ready-to-use, offers excellent sandability and is unaffected by finishes.

Conforms to ASTM D-4236

Product Features

100% waterproof (Passes ANSI Type I & II water-resistance testing)
Bonds virtually everything
Epoxy-like strength - No mixing
Short clamp & fast cure
Excellent sandability
100% solids
Solvent free

Physical Properties
Type Polyurethane Calculated VOC 0 g/L
State Liquid Weight/gallon 9.55 lbs.
Color Brown Flashpoint > 200°F.
Dried Film Yellow Freeze/thaw stability Stable
Solids 100% Viscosity 8,500 cps
Storage life 12 months in tightly closed containers at 75°F.

Bond Strength ASTM D-905 (On Hard Maple)
Temperature Strength Wood Failure
Room Temperature 3,500+ psi 60%
150°F. Overnight 3,000 psi 50%

Application Guidelines

Application Temperature Above 50°F.

Assembly Time After Glue Application 25-30 minutes (70°F./50%RH)

Minimum required spread Approximately 6 mils or 250 square feet per gallon

Required clamping pressure Enough to bring joints tightly together (generally, 30-80 psi for HPL, 100-150 psi for softwoods, 125-175 psi for medium woods and 175-250 psi for hardwoods).

Method of Application Easily spread with a roller, spreader or brush. *NOTE: This product cures by reaction with moisture. Changes in humidity, temperature and moisture content of the glued surfaces will affect open time and minimum clamp time. Maximum open time can be determined by excessive foam development in the adhesive. Minimum clamp time can be determined when squeeze out of adhesive becomes firm.

Cleanup Mineral spirits while glue is wet. Scrape or sand off dried excess.
Limitations
Titebond Polyurethane Glue is not for structural applications or for use below the waterline. For ease of application, the glue, temperature and materials to be bonded should be above 50°F. Lower temperatures will cause the glue to thicken.

[sdl2112]

I recently glued a stack of 1 inch LBL for my top using Titebond III for the adhesive. I had excess scrap when I cut them to size, so I did a quick test this evening. I wanted to see if my glue line, the LBL glue line or the bamboo itself would give first. Take my results with a grain of salt...it took me like 10 minutes. I performed force tests in two orientations. In both cases the bamboo itself failed. I believe the bamboo is very strong bending parallel to the grain...not so much perpendicular to it. That is why the three cross layer bamboo ply is so strong. Stacking may not be so much.

My bamboo stack top


Test set up with work table clamp to apply bending force


First test piece break


First test piece detail


Second test piece rotated 90 deg to test different grain orientation

[Steve Manning]

Nice Scott .... How long did you let it sit between glue up and testing?

[sdl2112]

Steve, these were glued back in October. This is what I cut off the tops when I was cutting to size.

[Steve Manning]

Steve, these were glued back in October. This is what I cut off the tops when I was cutting to size.

I would say that's plenty of cure time

[ergo]

I'm using this project also as a chance to learn CAD software (Rhino 5 mainly).

Just toying with different ideas. One would be to only make the front, back and shelf as a glue sandwich and top and bottom as solid pieces (thickness of 2 boards).
This could potentially save quite nr. of boards... but not sure how easy it would be to assemble or will it be structurally as good.

[Steve Manning]

I'm using this project also as a chance to learn CAD software (Rhino 5 mainly).

Just toying with different ideas. One would be to only make the front, back and shelf as a glue sandwich and top and bottom as solid pieces (thickness of 2 boards).
This could potentially save quite nr. of boards... but not sure how easy it would be to assemble or will it be structurally as good.

Interesting idea Ergo ..... I can certainly see where that would save on wood. Though I bet that would be rather interesting to try and glue up, especially keeping things aligned. Can't really say about structurally, though anywhere you have extra joints, there could be a potential failure point. Might be worth trying the idea with mdf or something else cheap to validate the concept.

[JonMarsh]

Ergo has a future in real product engineering at a reasonable cost... :W :B :T

we'll have to settle for being the high priced Yahoo's... simple and fast.

[JonMarsh]

How I wasted/ worked my butt off this Sunday...

well, this has been a very productive Sunday, on NECESSARY, not FUN productive things.

A big part has been following up on necessary chores and also getting things put together or re-arranged for the next stage of work.

That included assembling a stereo microscope with USB camera for upcoming assembly and test work... (mine is using an eBay sourced 17" Macbook Pro, good price)



and freeing up a desk/work table for the scope and other assembly tools and hot air rework station by consolidating the 2010 Mac Pro running WIN10 with my 2013 Mac Pro on one desk.



Oh, you didn't get the memo that we're going to ultra miniature SMD for crossovers from now one? :rofl:

Those with sharp eyes may notice the Audioegine A2's on the desk; what they can't see is the S8 subwoofer sitting under the desk, or the careful tweaking of level, phase, and crossover frequency I did to tune up this setup- best desktop audio I've ever head, but then I've never built my own... at this price, I couldn't pay myself to do that. MacPro 2010 drives the A2's input with USB, Mac Pro 2013 uses the headphone output into stere mini plug. Yes, two computers, one audio system to rule them all.

Most of you may not realize that the Windows Desktop on the left is on an Apple 27" Cinema display, but the Mac desktop on the right is on a 5K2 Dell display requiring dual display port connections to supply the necessary bandwidth. Bass-ackwards, huh? :W

Oh, and yes, that is a bamboo desk, actually one of those motorized sit/stand models. I love it... good tools will always serve you well!

[JonMarsh]

Researching Urethan glue on some forums...

OK, this is some of the feedback I found when searching on this topic and looking at wood working forums... (yeah, what do these guys know about building speakers, but maybe they know something about glue?)

Because of Titebond's general reputation in the industry, I expect good things from them. But these forum posts do raise questions in my mind, and make me feel fairly comfortable about this 36 oz bottle of Gorilla glue I have sitting in front of me...

This just goes to show that testing under different conditions will end with different results.

When I was testing glues on IPE, I tried both tightbonds 1 and 2, elmers carpenters wood glue, Weldwood plastic resin glue, and Gorilla glue. Not Tightbond polyurethane, but I would suspect that Tightbond would blend their polyurethane to be at least as good as Gorilla glue. They have a pretty substantial name to uphold.

With each test piece except Gorilla glue, I was able to break the test samples apart on the glue line, by holding one half in a vise and rapping the other half with a hammer. And by holding one half in the vise and using a pipe wrench to twist them. I was not able to split the gorilla glue on the glue line. All I ended up with was a bunch of little pieces of IPE glued to other little pieces of IPE. I also could not get the plastic resin glue to break on the glue line by twisting.

I'm sold on Gorilla glue, and will remain so until I experience failure. Wet or dry, I dunno about that, but I will continue to use it wet, because that's what the directions say, and that's the way I performed my test.

Come March of next year, my tables will have endured one year of Michigan weather. Could get pretty harsh here in a few months. (hope not) I will add a report to my site at that time how things held up.

To follow up on my original test of Titebond Polyurethane I have some more recent information and tests that you guys and gals may be interested in. After numerous failures while performing comparitive tests with TP glue I decided to call Franklin about the problem. The gentleman I spoke to asked me numerous questions about how I was using the product. I seemed to have passed his tests because he could not explain the reasons for the failures I had experienced. He offered to send me another container of TP and I accepted. When the glue arrived, I repeated the same tests as before. That was a T joint using 3/4" ply. I carefully dampened both pieces where the jint would be and applied the product to both pieces and clamped being careful of not getting too much squeeze out. After curing for 24 hours I put the test piece in my vice and lightly hit the vertical piece with the palm of my hand. It popped like a carrot right down the middle of the joint leaving both poeces undamaged. I swear that paint may be stronger. In view of my tests, I would recommend that this product be avoided.

[dar47]

Okay I'm going to glue some Weldbond tomorrow and see how that does on maple scraps since it's a wonder plastic glue without all the mess, haha.

[Steve Manning]

Jon,

I think we need to change the thread tittle from ..... Minerva Monitor: "Patience my ass, I'm going to go build something!" to Minerva Monitor: "Patience my ass, I'm going to go glue something!"

[JonMarsh]

Yeah, but maybe that should be "Patience my ass, I'm going to go test glue something!"

A colleague of mine that I helped hire, who is Danish (but I don't hold it against him... ;-) and I have a favorite motto which drives some of our colleagues in Europe and the Far east a little over the edge...

"In God We Trust; all others we verify"

Of course, there are variations of that, such as

"In God We Trust; All others pay cash"

But we don't need to go there...

OTOH, when I do get out to the shed sometime in the near future, I will use the Gorilla glue as the test adhesive. BTW, tried to buy Titegond Polyurethane glue locally, no joy- and going on Amazon, they had the curious item in the product detail that that it was discontinued by the manufacturer- might have just been that particular container size or something...

Until we do our own testing, I'd say TP Urethane glue is off the table for now. I'll order one bottle from Amazon and do some tests here. If you feel like also following up on that and sharing your thoughts, it would be fine and appreciated.

[Steve Manning]

Yeah, but maybe that should be "Patience my ass, I'm going to go test glue something!"

A colleague of mine that I helped hire, who is Danish (but I don't hold it against him... ;-) and I have a favorite motto which drives some of our colleagues in Europe and the Far east a little over the edge...



Of course, there are variations of that, such as



But we don't need to go there...

OTOH, when I do get out to the shed sometime in the near future, I will use the Gorilla glue as the test adhesive. BTW, tried to buy Titegond Polyurethane glue locally, no joy- and going on Amazon, they had the curious item in the product detail that that it was discontinued by the manufacturer- might have just been that particular container size or something...

Until we do our own testing, I'd say TP Urethane glue is off the table for now. I'll order one bottle from Amazon and do some tests here. If you feel like also following up on that and sharing your thoughts, it would be fine and appreciated.

Since I have already picked up two 12 oz bottles of the Tite Bond Polyurethane Liquid Glue for the project and have some test maple plywood test pieces in the garage, I'll do a glue up and see how things turn out. Pictures and comments to follow.

[TEK]

I don't quite qet this. What's all the fuss about? All of you have glued a lot off stuff before - without much issues.
Is it the lamination or hard wood that causes the challanges?

I'm using this for regular wood and BB.

http://m.casco.no/produktinfos.php?id=968&bereich=17&produktgruppe=132

I have tested it on some scrap pieces of BB. No breakage along the glue line, just everywhere else :-)

[Steve Manning]

I don't quite qet this. What's all the fuss about? All of you have glued a lot off stuff before - without much issues.
Is it the lamination or hard wood that causes the challanges?

I'm using this for regular wood and BB.

http://m.casco.no/produktinfos.php?id=968&bereich=17&produktgruppe=132

I have tested it on some scrap pieces of BB. No breakage along the glue line, just everywhere else :-)

I think your right TEK ...... there is more than one way to skin a cat ...... oops mine just took off running.

[TEK]

Not mine :-)

[Steve Manning]

Not mine :-)

[ATTACH=CONFIG]25085[/ATTACH]

Cute .... We have four of the fur balls around here.

[Steve Manning]

Glue Time

Well since this has become a hot side topic ....... Weather was nice today, mid 60's and sunny, so I opened up the garage door and played. I did two different tests with the Titebond Polyurethane Liquid Glue.

First test was going over kill on the glue. Over kill, in this case was nothing more than applying glue to both surfaces of the parts, instructions recommend just one surface.

These pictures show squeeze out right after clamping when glue is wet.

[Steve Manning]

Waited an hour for glue to set. Instructions recommend 45 minutes, which I suspect is at 70 degrees.

First two pictures are of final squeeze out.






5 seconds with a putty knife .... this stuff cleans up just as advertised.



A few additional minutes of scrapping and squeeze out is gone. This is certainly the easiest glue I've used to remove excess.

[Steve Manning]

Second test ..... I only applied glue to one surface and then clamped.

Again, initial squeeze out, less than the first test but still all the way around the parts.





After an hour of drying .....



After cleaning off the squeeze out ....






Of note, this was done with maple plywood, not bamboo. These were from the initial test pieces that were cut earlier. Glue applied with a foam roller.

Stress tests will follow in a day or so.

[JonMarsh]

Man, it's almost a shame to do stress tests on parts that look that nice... I'm just going to have some 3-4" wide LBL boards that I'll cut to smaller pieces, and do some different glue tests.

Just ordered a new anti-static setup for assembly bench for those ultra compact SMD crossovers we'll be working on... :rofl: :B