Welding a Plastic Tank

I’ve got a crack in my fresh water tank in the Safari and a toilet fitting torn half-way loose in the  Sovereign. So repairing them is the next “big effort.” The choice of methods, so far, are plastic welding or Scotch Weld 8005. The $30 hot air welding tool available from Harbor Freight is familiar to most, so no photo is necessary (IMHO). Here’s the Scotch Weld system, about $100 for the tool, nozzles, 10:1 injector, and two tubes of glue:

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7650 tools-s.jpg<br /><br /><br /><br /><br />
Views:	159<br /><br /><br /><br /><br />
Size:	183.8 KB<br /><br /><br /><br /><br />
ID:	51035

I labeled the rod as “suspect” because I can’t get it to glue or weld to the sheet material. Both items are labeled LDPE. As you can see in this photo, the rod material (which had been cut on a bandsaw to provide a flat face) pulled away from the sheet in the weld area and in both glue areas:

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7652 pull off 2-s.jpg<br /><br /><br /><br /><br />
Views:	137<br /><br /><br /><br /><br />
Size:	87.6 KB<br /><br /><br /><br /><br />
ID:	51036

So I thought about it and decided that either the rod was bad material or the fact that I clamped it down pretty tight might have pushed all the active glue solvents out from underneath the rod, preventing any gluing action. One piece of evidence that either of these might be true is that the squeezed-out glue was stuck pretty hard to the sheet.

New experiment–try to glue some pieces of sheet to itself. One piece to be clamped down tight and the other one just placed with momentary finger pressure. I also tried to weld a piece of sheet to the sheet, but once again no luck at all. The two glued pieces are TIGHT. Neither Bedfords nor I could break them off with hand pressure. I haven’t taken any extraordinary effort to see what it would take to break them off.

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7654 3 PIECES-S.jpg<br /><br /><br /><br /><br />
Views:	125<br /><br /><br /><br /><br />
Size:	108.7 KB<br /><br /><br /><br /><br />
ID:	51037

Bedfords (Joe) has had good luck welding this material, but we don’t know why our results are different. He left some of his material with me to experiment with (a $2 cookie sheet–why did I order the industrial stuff for $40?). I’ll test weld his material and report those results.

The next big task is to test the sheet material on the tanks. The cold weather here may delay that for some time…

——————— from LEWSTER —————

Plastic welding is sort of an art, but one that can be mastered in an hour or so of practice. When done correctly, it is just as strong as the material itself….just like a properly completed metal weld.

BTW, I have the HF welding unit. It is a bold knock-off of a $450 American made unit. Having used both, I can truly say that the HF unit, while not having the expansive selection of welding tips that the US unit has, does a proper job in skilled hands.

That said, you should cut some PVC pipe to practice on. Like any weld, the seam should fit as tight as physically possible. Tack the pieces together so the hold on their own and cut a shallow channel at the weld seam with the welder’s fine tip.

Next, you need a welding rod of the same material. THIS IS KEY TO SUCCESS!! You heat the seam and rod to the melting point and as you apply the rod to the seam, you will see the plastic flow from the rod and the base material will ‘juice’ or just slightly melt. This process is very necessary for a solid weld.

When the item is cool, it should not be able to be broken apart and in fact will break away in a different spot from the weld if done propery. After a few tries with the PVC (or ABS for that matter) practice material, your welds should be fluid and uniform and you are ready for prime time!!!

Again, selection of the proper welding rod is very important to the process. If you are not sure if the materials are alike, do a burn test. The rod and base material should react identically as to smell, smoke and burn characteristics if the are indeed the same.

The flow setting on the welder is equally important and should be vaied during the test and practice sessions to obtain the optimal air flow (and the associated heat setting)


Lew, the obvious sometimes escapes me! A simultaneous burn test! Wow, now I don’t have to figure out if it smells like a “candle” or is “acrid.” Thanks.

Yes, the proper rod is important, that’s why I can’t figure out why the “suspect” rod won’t weld. The sheet and rod are both supposed to be LDPE. Even a piece of the sheet won’t weld to itself. A plastics house here told me that it’s extremely difficult to get LDPE to weld. Maybe this is an  example of that.

Here’s a shot of the HF welder. You can see the router speed control in the background. I don’t have a too hot problem, so the controller is handy because it has a convenient on-off switch. The heat (per the expert mentioned above) is properly applied between the rod and the groove.

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7659 welder-s.jpg<br /><br /><br /><br /><br />
Views:	112<br /><br /><br /><br /><br />
Size:	112.0 KB<br /><br /><br /><br /><br />
ID:	51085Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7668 technique-s.jpg<br /><br /><br /><br /><br />
Views:	107<br /><br /><br /><br /><br />
Size:	56.8 KB<br /><br /><br /><br /><br />
ID:	51086

I’ve got photos of successful welding [yesterday] of the HDPE, polypropylene, and the cutting board that Bedfords left me. I used rod and sheet for the HDPE and polypropylene test, and a strip cut from the cutting board for that test. The welds are pretty ugly, but at least they demonstrate good adhesion. I don’t think the welder is hot enough–the welding is slow–but I’ve reduced the air pressure to 3-1/2 psi to get it as hot as I dare, which is getting pretty close to burning out the welder. Here’s just one example of the polypropylene weld–the others are similar.

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7670 pp success-s.jpg<br /><br /><br /><br /><br />
Views:	133<br /><br /><br /><br /><br />
Size:	63.5 KB<br /><br /><br /><br /><br />
ID:	51087


I took another stab at welding the LDPE. No luck. Looked good (see photo, not pretty, but looked like the other successful welds), but the rod peeled right out of the groove I had cut in the sheet. This second effort used a different, smaller diameter LDPE rod. My only advice is that even if the weld looks good, the only proof is an aggressive pull test.

I am ready to concede that LDPE may be near impossible to weld, at least for us amateurs. Thank goodness it seems to glue fine with the Scotch Weld. You can see that the two pieces that were previously glued on are still there. The longer piece, behind the rod, was an attemp to weld a piece of the sheet to the sheet, to eliminate any possibility that the rod was different material. It came off even easier than the rod.

Click image for larger version</p><br /><br /><br /><br />
<p>Name:	IMG_7660 LDPE-s.jpg<br /><br /><br /><br /><br />
Views:	114<br /><br /><br /><br /><br />
Size:	49.0 KB<br /><br /><br /><br /><br />
ID:	51178

Before I go glue up the black tank, I only have one remaining doubt. Telling the difference between LDPE and HDPE is not easy. Generally, all the samples I’ve handled tell me that HDPE is stiffer than LDPE and has a hard, slick surface feel, rather than the supple and oily feel of LDPE. I know this isn’t foolproof (to me, polypropylene feels and lot like LDPE). One thing I’m going to do for sure is the simultaneous smell test to make sure it’s polyethylene. If I had to pick right now, I’d still say the tank is LDPE (1972 Black Tank–the fitting looks to have been spun in).

—————- from DWIGHTDI ————

High density Polyethylene has a density of 0.95 – 0.96 grams /cc. (same as SG. Low density Polyethylene has a density of 0.916 -0.919. You can also use a melting stage (which is a little hot plate with a themometer attached. LDPE melts at 215 F. HDPE melts at 250F. Your problem with the LDPE rotocast tanks may be that they were using crosslinked LDPE which does not melt. You really can not weld that stuff.

An interesting way of accurately measuring SG on a small sample is to make up solutions of alcohol and water. Alchol has a lower density that water. What % ehtyl alchol = what density can be determine from googling it or using a hydrometer. If the part floats it is less dense less that the test solution. If it sinks it is more dense. If you use ethyl alcohol, you might celebrate by drinking the test solution.


Here is a few pics of the plastic welding process. I went down to target and got the cheapest cutting board I could find. With the band-saw cut 1/4 x 1/2 strips and then halved the strip to get a 1/4 x 1/4 size. Used the Harbor freight welding gun at around 3.5 psi. I had to turn the compressor reg down and the gun reg to get down that low. I cut a V in the tank making sure to round out were the crack ended. I clamped the to pull the seam together and used a square end drimmel cutter. It doesn’t have to look good. I went along the three edges tack welding(see pics). I left the rods sticking up and went back after it had cooled and cut them off. Finally I ran several beads down the crack until it was full. I worked on one side and then moved to another to allow for cooling. The whole process took no more than 45 minutes. Let me know if you have any questions.

Attached Thumbnails
Click image for larger version</p><br /><br />
<p>Name:	novdec 373.JPG<br /><br /><br />
Views:	48<br /><br /><br />
Size:	80.4 KB<br /><br /><br />
ID:	51228   Click image for larger version</p><br /><br />
<p>Name:	novdec 367.jpg<br /><br /><br />
Views:	48<br /><br /><br />
Size:	71.9 KB<br /><br /><br />
ID:	51232
Click image for larger version</p><br /><br />
<p>Name:	novdec 368.jpg<br /><br /><br />
Views:	50<br /><br /><br />
Size:	67.8 KB<br /><br /><br />
ID:	51233   Click image for larger version</p><br /><br />
<p>Name:	novdec 374.jpg<br /><br /><br />
Views:	45<br /><br /><br />
Size:	66.5 KB<br /><br /><br />
ID:	51234