Troubleshooting Edgebander Glueline Failure

Cabinetmakers discuss the issues that might cause edgebanding adhesive joints to fail in service. April 20, 2008

Is there anyone else having trouble with Jowat clear edgebander adhesive?

Forum Responses
(Adhesive Forum)
From Jeff Pitcher, forum technical advisor:
I would look a lot more closely at your equipment before jumping to the conclusion that it's a glue issue. The Jowat unfilled glue works very well and it's highly unlikely that it would be causing any problems.

From contributor L:
I'd second Jeff's response. Check your application roller temp with an infrared thermometer ($100+-.) Run some clear edgebanding so you can see the actual glue line. Check your pressure section to be sure all the rollers are making contact. Is your pressure beam and feed track holding the parts very tightly so they donít move? Panel edges at a true 90 degrees? If you are sawing with a slider, the operator can let the outside corner of a large panel droop, causing a non-square edge over part of the cut.

From the original questioner:
Thanks for your replies. We've had this machine for three years. We've had techs come out and check the machine, our techniques, our panel saw, glue temp, trueness of our cuts, etc. We are not new to this and have exhausted all of the usual suspects.

From contributor F:
You don't mention the edge material, the substrate, machine name and model, feeds or speeds, etc. There may be more comment forthcoming with more specific info. Start with the thought that there is no bad glue, nor anything else for that matter, and we look at the whole megilla.

From the original questioner:
To say that there is "no bad glue" is a little naive. That's like saying there are no bad machines or there is no bad material.

Edge material over the three years has included single layer tape .5 mm., paper back, fleece back, 1 mm, 3 mm, 7/8", 1", EdgeIt, EdgeCo, and one other I can't remember. We've used PVC, we've done the clear tape to check the spread. We've done this on veneer core, MDF core, particleboard core, and the MDF-veneer core combo.

One of our early suspects was the ambient temperature of the shop and thus the overnight temp of parts left, especially during winter nights, so we got a heater and made sure the parts were at least at spring time (70-80s) before we banded. Glue temp has been adjusted to allow open time to compensate for the possibility that we were losing too much heat at some point. We've heated the guide rail before starting so it would suck any heat out of the part before the glue was applied, although the glue rep said that wasn't necessary for us in our situation.

All of the parts are hand fed into the machine at the speed recommended in the tech manual and the glue pot temp has been checked repeatedly both at the feed roller and at the heating irons inside the pot.

We have even run extra material and left it sitting around various places in our shop, and on our parts racks, some for two years now, and what's amazing is that after running the parts, checking for "flappers" and being satisfied that the edgebanding is sufficiently secured, and actually having it checked by the tech from our service company, now, after all this time, from time to time lying on the floor will be a piece of the edgebanding, or we'll notice that an edge has turned loose.

In the beginning, especially, we didn't care what the problem was - we were willing to relook our technique in all phases of the parts processing and rethink the order and even the machinery we use.

We even bought a Holtzer panel saw (1265) in order to get consistent cuts. We learned that even MDF core plywood will give you serious banana cuts and have learned to use dust cuts to eliminate those. We learned that on dust cuts, the blade will torque or twist if you don't cut off enough material and have adjusted our optimizer to allow for those. Everything and more that you have suggested was cured or eliminated as suspects during most of the first year.

We are a small shop but for us, two to three million in cabinets is a lot of liability standing out there waiting to fall apart. Believe me, we wished that it was something as simple as an out of square cut or the glue was too cold or the parts were too cold or dusty on the edge.

From Jeff Pitcher, forum technical advisor:
There's one way to rule out the glue. Is all the edgebanding failing? If so, it could be the glue. If not, the glue has no way of deciding that it might stick only part of the time.

From contributor F:
Have you tried any other glue? In particular, a "natural" product? As I recall, Jowat has two different clear products for the edge bander - one is filled and one is not. Jeff can check me on this point. Jowat also has a clear hotmelt with a similar number that is not for edgebanders, but packed in a similar bag. If you have straight clean fresh cuts, let's not wake the sleeping slider.

From the original questioner:
Thanks for your response. Regarding the glue with filler and the one without, we did try both during our various experiments, and I believe that we had more problems with the one with the filler than the one without. But honestly, it's been two years or so since we did all that and I could be wrong.

One thing I want to make clear: in the case of the banding failing to stay, this all happened long after initial application and, in fact, in most cases, after the cabinets had been installed for nearly a year. Specifically, we used three different bandings, but probably from the same manufacturer. We had unfinished cherry two ply on paper at .5 mm., unfinished maple two ply and pre-finished maple and cherry, same everything.

What I noticed on the panels where the glue turned loose was that there was even coverage, and there was no glue stuck to the banding. The glue was always stuck to the panel rather than the banding.

We have since switched to a different brand of glue, unfilled, and the machine runs cleaner, less glue sling on the top and bottom trimmers, less glue slap on the end of the panel (as opposed to the leading edge) and less tags sticking to the end trimmers.
It was suggested at one point that we were not using enough glue and should turn up the doser, which we did and of course we ended up with lots of glue everywhere in the trimmer section of the bander. This was a total waste of glue and increased the work necessary to prepare the panels for assembly, and ultimately did not make a difference in whether or not the banding stayed glue. In the consideration of glue temp, there should be sufficient open time for the banding to be applied to the hot glue, if the bander is running at the proper speed and the parts are clean.

Jeff, to answer your question, yes, all the different bandings applied during that first year had failures. That's not to say that every piece on every panel failed, however - on one job where we got a call back, you could pick at the edge of the banding on almost any panel and peel it off intact without leaving any banding on the panel. The glue would look smooth and flat and evenly applied. Very weird.

I'm not sure what the sleeping slider is, but yes, all the cuts were fresh, clean, straight, and square.

From contributor F:
I personally am not a big fan of backed veneer edge banding beyond fleece and less of that is better. I also find maple and to a lesser degree cherry, both close grained woods, will move in a heartbeat when given the chance. I am leaning toward the banding being the problem, frankly. Have you tried a primed wood product? It's tough to troubleshoot someone else's problem over the Internet, but given my understanding of the problem, I'd try finger jointed, unbacked veneer - maple, cherry, whatever - that had been sanded both sides and primed and is less than 6 months old.

From contributor A:
If the initial stick was good, then I would look at what can change to cause the glue to fail.

You said: "on one job where we got a call back, you could pick at the edge of the banding on almost any panel, and peel it off intact without leaving any banding on the panel. The glue would look smooth and flat and evenly applied"

Could this be a backed edgeband and the backing has separated from the edgeband material? As I understand it, the edgeband glue works by mechanical adhesion. The glue actually gets pressed into all the tiny cracks and crevices of the two materials (edgeband and substrate). I would have to look at the edgeband backing to the glue area of the glue bond since this seems to be where the bond failure was.

From contributor B:
What brand of glue did you switch to?

From the original questioner:
We switched to Dorus, unfilled.

From contributor B:
Have you noticed an improvement?

From contributor L:
We've run Dorus KS205 or 217 for years with great results. Feed speed 58'/min. Has excellent hold on all materials including wood 3/4" thick.

From contributor M:
I just bought a new Holzher 1310-6, line speed 12 meters per min. Holzher comes with about 8-10 pucks to start with. They supply lt with Dorus glue. I haven't heard of this type of glue. My supplier has Helmtin, Klirebert, and they can order Dorus. Has anyone had good success with this glue, or what is the best kind for this bander? We apply all types - PVC, wood veneer, plastic lam, 3mm PVC, solid wood. Can I use one glue for all these products?

From contributor L:
Check with your adhesive tech. Your slow line speed requires a different adhesive than production banders. Our bander has a hot air blower between the application roller and the first powered pressure roller to help maintain adhesive temperature. Our line speed is 18 meters.

From the original questioner:
Yes, we have noticed a 100% improvement. The Dorus is much cleaner than the previous glue we used, no flappers, and the post application sanding is much less than with the other glue. Considering all the other options we tried, in spite of what some of the responses said, the only conclusion we can come up with it that the Dorus, for our application, is the best glue for us to use with our machine. Thanks for all your input.

The comments below were added after this Forum discussion was archived as a Knowledge Base article (add your comment).

Comment from contributor R:
We have an IDM AH4000 series and we are currently using Joewat 280 series glue. We recently completed a large project with 3mm solid VG fir edgebanding and had the same problem en masse.