Reddit reviews Supco VG64 Vacuum Gauge, Digital Display, 0-12000 microns Range, 10% Accuracy, 1/4" Male Flare Fitting Connection

Without tools, the best you can really do is look for oiliness on the pipe. When you find a joint that feels oily (look for dark colours on the pipe, usually covered in very fine dust. Once you rub the dusty area with your fingers you'll feel the oiliness), spray/pour a small amount of washing up liquid over it and look for bubbles. If there's any refrigerant left in the system, you'll usually see bubbles appear (or over a longer period, foam). Of course if the gas has all escaped already - and we're only talking a couple of hundred grams here - then you won't see any bubbles.

Most domestic gear has no access fitting at all so even locating the leak can be difficult if there's no visual indication. You need to get pressure in there so you can leak test - this means adding a bullet piercing valve (like this).

Then you need to put something in via your bullet piercing valve to raise the system pressure. It's best to use dry nitrogen to leak test to save wasting refrigerant, but since you most likely don't have that on hand you could buy yourself some refrigerant (most likely R134a) and pressurise with that. That's a big no-no here (both disposable cylinders and dumping gas to atmosphere by charging a system with a known leak) but I'm guessing your in the US, and those sorts of laws seem remarkably lax so go for gold. It's not like you'll be ruining MY ozone layer too, right? ;)

You also need to regulate the pressure going in to the system. This is where you need gauges. Connect the yellow line to your bottle, connect the blue line loosely to your bullet piercing valve, purge from cylinder to piercing valve by opening the cylinder tap + gauges tap, and releasing some pressure via the loose fitting, then tighten the fitting and close your gauges tap. Open the bullet piercing valve. Open the gauges tap slowly and give it ~50psi of system pressure.

Then you can go for gold with your soap, or you can lash out and grab yourself some "proper" leak detection fluid (like this), or better yet an electronic leak detector (like this one).

Once you've found your leak, you want to release your nitrogen (or reclaim your refrigerant using a reclaim plant and a spare cylinder), then repair it using an oxy/acetalyne set, or since it's only tiny pipework you can get away with a MAPP gas set.

If it's a copper->copper joint, you're laughing - polish the pipework up with emery cloth, heat the pipe until it's just this side of glowing red, and feed the joint with brown tip silver solder.

If it's a copper->steel joint, then it's a bit more of a pain. You need blue tip silver solder and flux. Clean your joint with the emery cloth, give it a nice coating of flux on every surface you need solder to stick to, then heat it up until it's a fair way short of glowing red. Feed the blue tip solder in and STOP. Unlike brown tip (15% silver) you can't just keep feeding blue tip (45% silver) as it ruins the weld.

Now, since you put on a bullet piercing valve and they leak like a sieve in the long term, we need to replace that with a schrader access valve. Since it's most likely going to be in a straight through piece of pipe, you can save time and grab yourself a pre made access valve in 1/4" pipe. Cut away the hole left by the bullet piercing valve, polish the copper and cut the pipework with a ~10mm gap using a tube cutter. Then slip your access fitting assembly in there, and follow the copper->copper joint procedure.

Of course, now that we've done all that you need to change the liquid line filter drier too. I'd recommend a 1/4" solder in core drier in place of the original copper spun drier because... well, copper spun driers are terrible. Follow the pipe cutting procedure from the piercing valve instructions and the soldering instructions from the copper->copper joint instructions and that's done too. Remember - always try and mount the drier so it's outlet is LOWER than its inlet. This turns the drier into a small liquid receiver and helps ensure a good liquid seal over the capillary tube. Speaking of capillary tubes, if it was inserted straight into the original copper spun drier CUT the capillary, don't try and unsweat it. The chances of blocking it up are about 82.5634% (approximately) when you unsweat capillarys. You're much better off chopping it with a set of capillary tube cutters and ensuring a good clean capillary. The ~30mm of wasted capillary will affect performance, but almost certainly not to any sort of measurable degree.

Then give the system a good evacuation using a vacuum pump and ensure it reaches a good vacuum (sub-500 micron) with a digital vacuum gauge.

Then using a set of electronic scales, charge your freshly evacuated system (remembering to purge!) to the charge recommended by the manufacturer.

Voila! You have just fixed your chest freezer.

..... Alternatively, pay someone to do it for you and/or recycle the components and buy yourself a new one.

I made a fixture that screwed the cap lid down. It makes zero difference in the leaks.

I've purchased a nice vacuum gauge, https://www.amazon.com/gp/product/B002JP5PAE which allowed me to eliminate the arduino with BMP183 inside the chamber. I still see the same level of leak, so I feel comfortable that outgassing by the arduino or its battery isn't the cause of the slow increase in pressure.

I noticed that the hose fittings (1/4" flare) are prone to leaking. At this point, it seems like my (pipe-thread and flare) fittings connected to the vacuum chamber are the remaining source of leaks- either they aren't tightened enough, or they have leftover oxygen in the threads.

I haven't completely eliminated outgassing of acrylic as the othe rremaining cause of pressure increase. I guess the only thing I can do is continuously pump the chamber for a couple days without exposing it to atmosphere.