JB Weld alternative?

I have had some trouble finding a place where they sell JB weld and was wondering if this exhaust repair product [www.bizrate.co.uk] could fulfill the same function (i.e. fixing the nichrome/thermistor to the extruder parts.
AFAIK, a car exhaust is quite hot, so this could be OK, but I can't find much info on the product (max temperature). I asked the manufacturer, but figured I might as well ask here. Anyone has any experience with this or a similar product? Any danger of toxic fumes?

Cheers,
Joost

it looks like it would work. the only thing i'm not sure on is if it cures into a solid, or if it just stays in a gum-state. if it works on mufflers, i would imagine it can handle our temperatures.

i would suggest giving it a try, if it works then great! otherwise you may just need to replace the nozzle and/or barrel (or clean the gunk off...)

Hi People,

OK, to be honest, I don't know what you want to use it for, but my guess is that the epoxy moldable steel is a good alternative.
There are different versions available and I know a few of them are pretty heat resistant.

I used the stuff to build a special nut that I needed when I was building my car.
It held for a year and then I found the "official" nut so I removed the epoxy one.
Still looked good so I kept it as a spare.

Something like this, but then a little more specialist.

[www.diytools.co.uk]

Lykle Schepers

Some products designed for use on car exhaust systems do not cure fully until the exhaust pipe gets hot from engine use. Something to consider.

Why not use JB weld to attach the wire rope to the drive screw and to the nut?

interesting idea... i'm about to make that assembly, so i'll give it a shot!

Go to the Web site [jbweld.net] , sounds great available at autozone and walmart. I got the drill machine to run in all three axes and homes beautiful. The trouble is this, I am using a 200 step/rev and driving 1/4' 20 threaded rod, so it is 4000 steps/inch, (200/revX20rev/inch) and 4000 steps/25.4mm=157.48031 for both the X and Y axes. I enter these values in the preference under Ubuntu and I get a perfect inch of travel in the X and Y under the line subprogram and I'm thrilled I have a webcam, maybe I can post a picture. Under windows XP I enter the same number in the preference for X and Y, (157.48 and change) and when I go to the stepper program, the calibration number is still 30000 or something like that. In Ubuntu, this number scales down and 4000 pulses equals an inch. This drill machine has a limited Z movement, about 2" max. How high can the extruder be above the part before it screws it up. I was going to get some experimental data of time and temperature and use Newton's law of cooling to calculate how long till it reached the melting point, but what the heck, Dr. Higgs has some experience in this area. I can raise or lower the drill holder, (it is on a clamp) but I have limited Z movement. I inverted the limit switch to make down the home position. If I can be say maybe an inch, I could make parts sized 6"X 6"X3", which would fit most of the Reprap parts, yes? I would try out the JB weld on something other than the machined shaft, just to see how it works. JB two pieces of angle iron and then see how much torque it take to bust them apart. I am going to use wood screws to mount the boards on the wooden base and put it back on it's original cart so I can play show and tell at the DMA (Dayton microcomputer association) meeting on the 31 July 2007.

> If I can be say maybe
> an inch, I could make parts sized 6"X 6"X3", which
> would fit most of the Reprap parts, yes? I would
> try out the JB weld on something other than the
> machined shaft, just to see how it works.

A working space of 6x6x3 inches is going to be fine to get started. Hell, Tommelise is only doing 11x8x1-7/8 inches right now. You don't have to be able to print RepRap parts to do useful research.

What about clearance of the nozzle above the part, will an inch work, or will the plastic not form properly. I looking at the videos, the nozzle looks close to the working surface. Everyone (well almost everyone) in this group is dying for pieces to build their own machine and It's nice to get some more use out of that old drill machine. I think everyone wants to be parted and the only working machines, (Darwin's that is, no offense to Tommelise) are on the other side of the pond. The more working machines out there, the more creativity and permutations of experiments will be done.

Edited 1 time(s). Last edit at 07/24/2007 09:43PM by englewood.

englewood Wrote:
-------------------------------------------------------
> What about clearance of the nozzle above the part,
> will an inch work, or will the plastic not form
> properly.

Just don't park it over what you are trying to print while it is hot and a few mm should be fine, I would think. When I am extruding I'm usually no more than 0.6 mm above the last layer's surface.

> Everyone (well almost everyone) in this group is
> dying for pieces to build their own machine and
> It's nice to get some more use out of that old
> drill machine.

I'm probably not going to make any friends saying what I'm going to say, but it's time for some ground truths here.

There are a lot of keen guys out there like yourself who give the impression that they expect that Darwin parts are going to start pouring out of the three Darwin machines presently under construction in a matter of weeks and if not a few weeks no more than a few months.
I don't think that that is a realistic expectation. I also don't think that it's fair to the RepRap effort to have those sort of expectations.

Vik has sort of learned how to print with CAPA, to the point where he was able to print a Mk II parts set. Vik also learned that if you make a stepper motor mount out of CAPA that it melts down because stepper motors get quite hot.

I've sort of learned how to print with HDPE. I've been at it since February and I still have a lot to learn. The big thing that I've learned is that HDPE is difficult to print at room temperature. I've also learned that HDPE is not putty to be molded into just any shape that strikes my fancy. I have to design a part around the peculiarities of HDPE as much as the demands of the part.

I frankly don't think that the Darwin parts as presently designed are going to be printable with HDPE at room temperature. Right now, we have three Darwin machines within about a month or two of starting to print. If Vik, Zach, Ed and Adrian's use HDPE and their experience is anything like mine they've got 5-6 months to develop the know-how to design parts out of HDPE. I hope that I'm wrong but I fear that I'm not.

What we're looking at here is probably printing Darwin parts, suitably redesigned, in Darwins in the first quarter of next year. That's my best guess.

Now Zach has been killing himself trying to get all the early adopters like yourself going. He's cut a bid on having Mk II's milled out of metal and I think he is trying to get an order of CAPA filament. I may be wrong on that last bit.

The point, though, is that there are going to be a lot of steppers and controllers working out there a LONG time before Darwin plastic parts are going to be available at prices in the hundreds, instead of thousands of dollars.

In my opinion, if you are serious about getting a bootstrap Darwin going before then (1st quarter of 2008) you need to look at what Joost has been doing.

I think that you, personally, are going to be all right using as you plan to your drilling machine as a bootstrap RepRap. I really worry, though, about the many other keen RepRappers that I see becoming disappointed with the time lags because of unreasonable expectations.

And this is what's so nice about "open source" and exchanging experiences ... Eric has tried using JB weld on the drive screw between wire and nut, but it fell apart(I think twice). Apparently JB weld cannot really handle the forces/stress; soldering seems a better option (I soldered mine using a soldering iron @350 degrees centigrades, no propane burner needed)

Joost

I've read your blogs describing the endless hours of experimentation and finally success. I was hoping to build on your experience. I've learned there are two ways to learn, your own experience, painful and expensive in time, or through the experience of others or books. This lessons aren't as potent as personal trials and tribulations that you sure do remember! The coils have to be juiced all the time to maintain position, one of the evils of a stepper motor. I've also read JB weld sucks in torque, like concrete in tension, ZACH THE EXPERIMENT WITH JB WELD FOR TORQUE IS UNNECESSARY, WON'T WORK, DON'T WASTE YOUR TIME!

I too have been underwhelmed by the strength of JB weld. I read that it is good for attaching heatsinks to CPUs, a thin layer is apparently a good thermal conductor. The article said the downside was you would never get it off again.

So I used it to fasten 1W aluminium backed LEDs to copper PCB material. One LED failed so I thought I was going to have a problem removing it, but it came off easily with a flick from a penknife. I could easily scrape the remaining JB Weld off with the knife.

Since then several more have fallen off, presumably due the differential expansion between copper and aluminium.

In contrast I stuck some LEDs to an aluminium plastic laminate sheet with superglue and thay have stayed on.

---

[www.hydraraptor.blogspot.com]

>
> There are a lot of keen guys out there like
> yourself who give the impression that they expect
> that Darwin parts are going to start pouring out
> of the three Darwin machines presently under
> construction in a matter of weeks and if not a few
> weeks no more than a few months.
> I don't think that that is a realistic
> expectation. I also don't think that it's fair to
> the RepRap effort to have those sort of
> expectations.

i'm very sorry if i've been giving that impression. i realize it may be hard to distinguish from optimism and reality on the internet. we are much closer now than we were 6 months ago, and in another 3-6 months i definitely hope to see some printed parts flying around.

> I frankly don't think that the Darwin parts as
> presently designed are going to be printable with
> HDPE at room temperature. Right now, we have
> three Darwin machines within about a month or two
> of starting to print. If Vik, Zach, Ed and
> Adrian's use HDPE and their experience is anything
> like mine they've got 5-6 months to develop the
> know-how to design parts out of HDPE. I hope that
> I'm wrong but I fear that I'm not.

i agree, but the nice thing is we dont have to print exactly the darwin parts... we just need to print parts capable of being printed (if that makes sense) i think there were alot of shortcuts taken on darwin to save printing material and such.

>
> What we're looking at here is probably printing
> Darwin parts, suitably redesigned, in Darwins in
> the first quarter of next year. That's my best
> guess.

a good estimate.

>
> Now Zach has been killing himself trying to get
> all the early adopters like yourself going. He's
> cut a bid on having Mk II's milled out of metal
> and I think he is trying to get an order of CAPA
> filament. I may be wrong on that last bit.

true and true... although i enjoy the work i'm doing =) it seems the CAPA filament is months off due to supplier issues.

> In my opinion, if you are serious about getting a
> bootstrap Darwin going before then (1st quarter of
> 2008) you need to look at what Joost has been
> doing.

i'm going to experiment with stepper motors and a cross slide vice this weekend. i'll let you guys know how it goes. basically the biggest hurdle now is that there is no easy way to get a cartesian machine. the sooner we can solve this problem, the sooner we will be successful.

> I think that you, personally, are going to be all
> right using as you plan to your drilling machine
> as a bootstrap RepRap. I really worry, though,
> about the many other keen RepRappers that I see
> becoming disappointed with the time lags because
> of unreasonable expectations.

i also worry about this. we should probably be better about what we're portraying about the status of this project. it definitely is a research project, albeit one that is open and distributed. there are things that we've done well, and things that need some work. i think overall we're moving in the right direction and that it is fairly easy for someone to read through our site and get the correct impression of where we are at as a project.

> i'm very sorry if i've been giving that
> impression. i realize it may be hard to
> distinguish from optimism and reality on the
> internet. we are much closer now than we were 6
> months ago, and in another 3-6 months i definitely
> hope to see some printed parts flying around.

This isn't about you, Zach. I think you've been VERY upfront about what what the timelines are. The worry I have comes from comments I've been seeing out in our builder community.

> basically the biggest hurdle now is that there
> is no easy way to get a cartesian machine. the
> sooner we can solve this problem, the sooner we
> will be successful.

I stole the gantry design from Reyuki for Tommelise. It's pretty easy to build and there is nothing in the design that would preclude someone using steppers to drive it. Like Zaphod, however, it depends on threaded rods for positioning which means slow compared to Darwin. If you decide to use some variation on the Tommelise/Reyuki cartesian system, however, you may want to do a bit of a redesign of the x-axis (the gantry) to reduce the torque requirements. I used large wood surfaces in contact with each other, which requires significant torque. I did that because I had a lot of available torque with the gearmotors and went the quick and dirty route. There are other, more efficient ways to have done it.

Edited 1 time(s). Last edit at 07/25/2007 10:26AM by Forrest Higgs.

Ok, I REALLY have to suggest this then (for your sliding torque issues):

[www.igus.com]

I'm using a few here at work (the N-027 variety) and they're wonderful for low mass motion. They're also fairly cheap. They'll quote you on the website.

The sliding table gantry design of Tommelise is very common in all routers. I've dealt with machines made by at least three different manufacturers (one was a test unit for a carbide cutter company, the next manufacturer had 20 of two different types used for cutting out plastic parts, and the last is a wood router). The only real difference between them was at the head. Some had multiple solenoid dropped heads for 2D work others had two heads (one Z with a saw, one Z-A-B for five axis routing).

If JBWeld is supposed to be a strong as steel how come I can scrape it off with my fingernail after 24 hours?

---

[www.hydraraptor.blogspot.com]

Tungsten carbide fingernails?

Does anybody know if J.B. Weld benefits from a postcure? (Raising the temperature for around 2 hours to enhance the cure reaction.)

My understanding (which may be wrong---I'm not expert) is that most epoxies do, because the cure reaction requires heat, and levels off somewhere short of maximum strength if there isn't enough heat. (It slows down and asymptotically approaches a temperature-dependent maximum).

From what I've read, most epoxies do not fully set up unless you heat them to their maximum "glass transition temperature" and hold them there for a couple of hours.

If you don't, the incomplete setting reaction not only makes them weaker, but makes them behave strangely if they get hot in service---like when you use them for stepper motor mounts that get hot.

What happens is that when you approach the maximum temperature the epoxy was cured at, it softens up, and cures some more due to the heat available to drive the reaction. If the part is under stress or vibration, it will cure in a strange shape and with internal stresses, making it more prone to fail later.

The temperature at which the epoxy starts to behave strangely is dependent on the temperature it was cured at. (And how far the cure reaction got.)

Once it's been cured at a given temperature for long enough, it won't soften when it approaches that temperature later. (Up to the "maximum glass transition temperature" which depends on the formulation of the epoxy. What that is for J.B. Weld, I don't know.)

If you want to make it work for stepper mounts, I'd suggest heating it in an oven starting at a lowish temperature and slowly raising the temperature to something like 300 degrees, and letting it bake for a couple of hours. That might make it stronger and/or less likely to get weird when the motors get hot.

(Better yet, consult a composites expert to make sure this is right.)

JB Weld is supposed to be good up to 315C. I have had my extruder heater up to about 250C and it has changed from gray to dark brown. Is this to be expected or have I got some inferior counterfeit product?

---

[www.hydraraptor.blogspot.com]

I just used some JB Weld again. The instructions say it can be handled after 4 hours and it can be put to use after 15 hours if the room is above 50F.

I would say it was still a stiff liquid after 4 hours. After 15 it was soft and flexible. I.e. I could dent it with my fingernail and bend a piece 2mm thick. The room would have been no lower than 20C so I don't know what I am doing wrong. I used equal parts hardener and resin and stirred it well.

I then backed it in the oven at gas mark 6 (~200C) for two hours. It is now hard and brittle and slightly darker colour. I can snap the 2mm sample with a reasonable amount of force and it breaks cleanly like a biscuit. I didn't go up to 300C as I know it discolours.

So for some reason it does not cure properly at room temperature, or if it does it takes longer than 24 hours at 20C. As drcrash suggested baking seems to cure it but 2 hours of gas is quite expensive.

---

[www.hydraraptor.blogspot.com]

jbweld does seem to discolor when the heater is on, but it doesnt seem to effect it at all. the main purpose of the jbweld is to keep the heater wire attached to the barrel. as long as it does that, it should be fine.

Yes mine has gone almost completely black now and apart from a little cracking where the wires exit it is holding up fine so far.

Edit:
Actually mine is also acting as the insulation as I am using bare nichrome. That may be a bad idea because the wire has no room to expand.

Edited 1 time(s). Last edit at 08/12/2007 03:08PM by nophead.

---

[www.hydraraptor.blogspot.com]

We have developed a new epoxy material for the Fab@Home project called "FabEpoxy". We'd be glad to send out a few free samples to a couple of RepRap users to get feedback. Please feel free to contact me at bfinkenaur@kraftmark.biz

Cornell University and the Fab@Home machine was recently featured in a Wall Street Journal Online segment: [link.brightcove.com]
FabEpoxy was used to make the propellar seen in the video.

Best regards,
Bob Finkenaur