Sight Glass Leaks ????

Help Support and Requests
1

OK, first post.

I do not have an industrial sized system. All metal pieces are small enough to be hand carried. My CLS sight glass leaks. I know that a DIN wrench is the proper tool to tighten these. I don’t own a proper DIN wrench, but I use something very similar, called a jaw-style wrench (specifically Hopkins 10631). It looks like a cross between an oil filter tool, channel lock pliers and a pipe wrench.

Jaw%20wrench

I understand what has to be done to repair the leak. I am mostly interested in hearing more about the mechanics of the repair. Round metal tanks are hard to grip. Every time I attempt to tighten my sight glass, it is a wrestling match. What process and aids do you use to properly seal and tighten? How would you secure a round, open ended, tank?

I have considered placing the tank over something solid, like a fence post, but am worried that this will damage the interior of the spool. I am particularly concerned with the sharp edges where the flanges are secured. I don’t want to ding that edge and compromise its sealing ability. How do you hold your tank securely?

Now, to the sight glass, and its port. The sight glass is a round, screw on unit. sealing material is provided with it. This provided sealing material should theoretically be effective at securing the port, all by itself, assuming that sufficient torque has been applied.

However, trying to torque this sufficiently has proven to be quite difficult. Every other threaded connection in the CLS is sealed using PTFE tape. Has anyone used PTFE tape on the sight glass threads? Should taping these threads be even considered? Does anyone have any good suggestions, advice, comments?

1 Like
2

This is how I tighten my customers sight glasses. This works 98% of the time. Sometime the gasket needs to be adjusted but this works the majority.

Clamp it down to a spool so it’s way more stable. I step on it with the clamp to the floor so it can’t turn. I use a large pipe wrench and use myself as leverage to tighten the sightglass.

Here is my :spoon: for the day.

image
image3024×4032 4.8 MB

11 Likes
3

You don’t want to put ptfe tape on the threads of a sight glass, the threads aren’t the sealing surface.On the other hand a ptfe gasket under the glass would be a good choice. You could use a ratchet strap to secure your tank to a wooden post in order to tighten the nut on the sight glass.

3 Likes
4

Dosen’t that mark the nut up? Or is that the killa signature!? :grin::+1:

6 Likes
5

my stuff aint winning no beauty contests. Its always function over fashion for me. LOL But i guess thats my signiture. It works though.

5 Likes
6

That thing I use is reasonably gentle, unless you slip. It does have teeth.

1 Like
7

This works in a pinch, and wont mar as badly.

8

I examined the nut. This is the marring. I had to circle it otherwise you might have missed it. :joy::rofl::joy::tipping_hand_man:

A little mark like that shouldnt have an effect on anything.

image
image1537×2050 625 KB

2 Likes
9

https://www.amazon.com/Dixon-DIN-SMS-60-90SP-Adjustable-Fittings/dp/B00LT0LTLY
I like this wrench, it is a multi-tasker too!

2 Likes
10

…and just to tie these two threads together…and maybe reiterate the utility of the search function

http://future4200.com/search?q=sight+glass+leak

5 Likes
11

I appreciate your suggestion, concerning the search function. I did look first. Probably should change the title to “How do you secure a round, slippery-ass, tank, to work on it”

As an update: Probably got the sight glass tightened, but it might not have even been the problem.

I originally noticed that there was a leak in my collection vessel, while it was submerged, during a butane run. I was getting a trickle of tiny bubbles, from around the sight glass area.

I set out to resolve this, by attempting to tighten the sight glass retaining collar. It was very awkward, so I asked here, about solutions to secure the tank, to make it easier to work on.

At first, I was doing only vacuum testing, mainly because I thought that I knew what was leaking. I kept evacuating, losing the vacuum, and tightening this and that. Really quite frustrating, so I decided that it was time to pressure test.

Ran the pressure up to 100psi and discovered that the sight glass was actually sealed. That stream of tiny bubbles was coming from a pinhole leak in the weld, the pot itself was leaking. Stupid hole is so small that you can barely detect it, even when looking directly at it.

I had been tricked. When I first noticed the leak, the spool was the other direction down, so all of the bubbles rose over the sight glass, and made it look like they were coming from it.

So the moral of this tale is:

Vacuum testing will show you that a container has a leak, but there is no way to know where the leaks are located. You really have to pressure test to see where that leak is actually located.

Use a vacuum first, I feel that it is much easier (way less messy). If the container has any leaks, you will lose vacuum, but don’t tighten anything yet.

Once you know that a container is not properly sealed, it should then be pressure tested. You need to be certain where the leaks are, before you can do whatever is needed to correct the problems. Don’t make assumptions, otherwise you will waste time tightening stuff that doesn’t need it.

Now I have to find a local welder to fix that pinhole.

BTW: learned another lesson too. After you pull a vacuum, if you are going to turn off your vacuum pump, disconnect (loosen) the vacuum line.

If the valve you are using has even a small leak, the vacuum in your container will probably pull oil out of your pump, and up and into the vacuum line. So just loosen the vacuum line nut, the leaky valve will suck air, instead of oil.

I am going to always loosen that nut, from now on. A half turn of a wrench is no big deal, especially compared to pump oil sprayed all over your good stuff. And yeah, I know…there is an oil in the lines topic.

3 Likes
12

We are in Fresno ca if you need that pin hole drilled out and welded.

4 Likes
13

If possible, I always test under pressure first, and locate leaks with soapy water.

For stuff that won’t take pressure a good gauge and a tank of Helium will get you there.

14

Yeah, I agree that there is really no need to check vacuum first. No don’t need to do both

However, I am just kinda lazy (damn Indica), and dragging the compressor out of the garage, getting out buckets and filling them (for submersion testing), or even just making a mess with soapy water is usually more than I want to do. That is, until I KNOW that I have a leak.

So, my lazy process is to attach the vacuum line, pull a vacuum, close the valve, loosen the vacuum line, and go do a lot of dabs for a couple hours. If I come back and (assuming I can still see) the vacuum isn’t holding…then it’s soap time, with pressure.

Oh yeah, a question. Is it acceptable to completely immerse a container with a gauge attached? I am mainly referring to just under water, but 99% ISO with dry ice also crossed my mind.

I do not submerge my gauges, I am just curious if they would survive.

The pressure measuring mechanism would be under pressure, so I don’t think that the measurement would be affected. However, the case is not sealed, and I doubt that there would be any internal pressure. I think that the case would fill with liquid. Don’t know if this would be bad.

1 Like
15

There are leaks that will occur under pressure that will not occur under vac.

Eg: vac your system down. Remove a clamp. You should not loose vac. Try and run it?

5 Likes
16

Old automotive trick is to put a folded up rag or plastic between the wrench and the part being tightened to not mar the finish

2 Likes
17

that and even at certain psi, leaks will not be visible. I had a leak at 100psi that didn’t show until I hit at least 80psi.

1 Like
18

i find that many “new” sight glasses fail when they get cold the first time. the gaskets shrink and leaks happen at -40 or so.

2 Likes
19

I nailed a 2x4 rhen toe nailed 2more 1 on each side to make a cradle of sorts when i did my 200lb tank. Worked great. Just ratchet strap it and should work fine it did for me