WatchUSeek Watch Forums banner
1 - 20 of 39 Posts

·
Registered
Joined
·
1,143 Posts
Discussion Starter · #1 · (Edited)

How to build a 10 ATM/100m Watch Water Pressure Tester for $40 (assuming you already have a bicycle pump or an air compressor)




There are a few ways you can go about doing this depending on a) what you already have and b) how elaborate you want to get.
The simplest and cheapest way to go is to just buy the following items:
1) Water Filter Housing ($33 Shipped) Here is a link: Pentek 158110 3/8" #5 Clear/Blue Water Filter Housing | Water Softeners Filters
Everything else can be purchased at Home Depot/Lowes or online at a myriad of places
2) 1pc 3/8" npt Plug ($3)
3) 1pc 3/8" male to 1/4"npt female reducer ($3)
4 ) 1pc 1/4" npt tp Air Compressor fitting or 1pc 1/4" npt to Schrader Valve (Buy the air compressor fitting if you have an air compressor and buy the other one if you are going to use a bicycle pump) - ($2)
Total Cost $40


A quick note on these parts: The female ports on the Water Filter Housing are 3/8 NPT (National Pipe Thread) There are 2 sizes of fittings that are common and you will find at Home Depot/Lowes etc. They are 3/8 NPT and 1/4 NPT. Typically 3/8 are used for Paint sprayers and other uses that require more flow and the 1/4 NPT are used for most Air Tools like Nail guns etc. If you have a compressor you could skip the reducer and buy 3/8 NPT to Air hose connector, but if you are going to a Shrader Valve (bicycle pump valve) you might not be able to find the 3/8 fitting directly to the Shrader Valve. Here is what the necessary parts look like:
Here is the Plug: It doesn't really matter which side you put it on because both ports are open to the container, but you might as well put it into the port labeled "Out"

Here is the reducer. You might be able to avoid using this part, but I bought and used it.

You are going to need one of the following 2 parts. The first one if you have a compressor and the 2nd if you plan on using a bicycle pump:



A MORE Elaborate (and bit more expensive) way:
I have a compressor that is about 25 yrs old that works just as good as the day I bought it. FYI (For you young guys) a compressor is a handy thing to have. You are not going to use it every day, but I have used mine many times over the years and I would highly recommend buying one. One like I have is only about $100 and there are MANY uses for them.


I wanted to be able to disconnect the Tester from the Compressor. If you want to be able to do that you are going to need a Shut off Valve. I also wanted to be able to monitor the pressure inside the vessel with the compressor disconnected so if you want to be able to do that you are going to need a gauge that is not attached to the bicycle pump or compressor. If you want to go this route like I did it is going to cost you about $10 extra, but I thought it was worth it. Here is what you are going to need if you go this route:
1) Water Filter Housing ($33 Shipped)
2) 2 x 3/8" male to 1/4"npt female reducer ($5)
3) 1 Shut off Valve 1/4" NPT ($3)
4) 1 Pressure Gauge 1/4" NPT ($7)
4 ) 1pc 1/4" npt tp Air Compressor fitting or 1pc 1/4" npt to Schrader Valve (Buy the air compressor fitting if you have an air compressor and buy the other one if you are going to use a bicycle pump) - ($2)
Total Cost $50


Here is what mine looks like:



Here are a couple other items you are going to need:
1) PTFE tape (free as I had it, but I did find some for only 79¢)
2) Wrench to tighten the fittings - (Free as I had this)
3) A Bicycle Pump or a Compressor - (Free as I had this)
4) Coat hanger or something else to hang the watch on -(Free as I had this)
5) Pliers to bend the coat hanger -(Free as I had this)
6) Scotch Tape -(Free as I had this)


Procedure:


Assemble the fittings to the Water Filter Housing. It is going to be really obvious what to do here. If you screwed up and got something wrong you will just have to go back and buy a different part. Put the plug or the gauge in the "OUT" port and put the fittings to attach to your compressor or bicycle pump into the "IN" port. If you decide to use a shut off valve this needs to go into the "IN" port on the housing side of your air fittings (see the pic above of my setup) Use reducers as necessary. Once you pressurize the housing you are going to know pretty much right away if it is leaking. If it is leaking you are going to have to go back and put some PTFE tape on the fittings where it is leaking and that will solve your leaks. To test for leaking mix a drop or two of dish soap (Palmolive or Dawn) with about 1 oz water. You can do this in a soda cap or small container. Then just pour the soapy water on your fittings and you will immediately know if you have a leak. If it is leaking you will see bubbles forming and if it is not leaking there will not be any bubbles forming (again this will be really obvious)


Now that you have your pressure container together you are going to need something to hang your watch on. I took a coat hanger and bent it so I could easily hang a watch on it and so it would not move around in the container. As you can see I made it so it just fits inside the water filter housing. Here is what mine looks like:





And Now on to the testing Steps:


1) Attach watch to stand
2) Fill Water Filter container about 1/3 full of water
3) Drop stand and watch into the water filter container (watch side up so it is not in the water)
4) Attach Housing top (and fittings) onto the body and tighten
5) Pressurize the container to desired pressure
6) Let sit for minimum of 30 minutes at pressure
7) Quickly release pressure from the Water Filter Housing chamber.
- This water filter housing has a pressure release valve built in. It is the little red button on top of the cover. All you have to do is push it down and it will release the pressure from inside the container. If you put a shut off valve in there like I did, you can use that to release the pressure as well.
8) Once the container is depressurized immediately dunk the watch into the water so it is fully submerged. I just rotated the housing onto its side. I made my stand so the watch will just rotate onto the side of the housing and be submerged in water. I guess you might want to wait 5 or 10 seconds before dunking the watch in the water just to be sure the crystal doesn't pop off. If it is going to pop off it will do that pretty quickly and you don't want the watch to be in the water if this happens. The reason the crystal could pop off is if the watch is leaking the pressure inside the case will be greater then the pressure outside the case and if the crystal is not fitted properly (ie it is too lose) it could pop off.


One of two things is going to happen:
a) The watch will be waterproof and in this case you will not see any bubbles escaping from the case of the watch.
b) The watch will have leaked. So what has happened in this case is that at this point inside the case is a higher pressure than outside the case and now the pressure is trying to escape from inside the watch. If you see bubbles, don't worry water will not get inside the watch unless you leave it submerged until all the pressure has escaped. Just remove the watch from the water and remove the case back and go to work on the seals.


Here are some pics of the above procedure:


1) Attach watch to stand



2) Fill Water Filter container about 1/3 to 1/2 full of water. You want the water to be just below the level of the watch.



3) Drop stand and watch into the water filter container (watch side up so it is not in the water)



4) Attach Housing top (and fittings) onto the body and tighten



5) Pressurize the container to desired pressure


6) Here is what I used to pressurize it.


7) Let sit for minimum of 30 minutes at pressure
- Nothing really to show here. Hopefully you don't lose any pressure. If you do, you can give it a few pumps or attach back to the compressor. If you are losing pressure you probably should figure out why and fix the problem for the next time.


8) Quickly release pressure from the Water Filter Housing chamber. You might consider only releasing some of the pressure to make sure the crystal does not pop off before dunking the watch in the water! It is possible that if the watch leaks and if the crystal is not fitting correctly the higher pressure inside the watch (after leaking and depressurizing the chamber) could cause the crystal to pop off. This is the reason they have Helium release valves in high end deeply rated watches. The same principle can happen. Helium enters the watch at really deep distances and then when you come up quickly the inside of the watch will be at a much higher pressure than the outside and the crystal can pop off.



9) Once the container is depressurized immediately dunk the watch into the water so it is fully submerged. Here I have just set the housing on its side. The watch will then sit on the side of the housing sitting in the water.



Here is a close up of what that looks like:



If there are no leaks, it will look like the above pic (no bubbles coming from the case) In a way it is a little disappointing as nothing is happening. LOL


You should be happy to see no bubbles though as this means your watch is safe!


A NOTE about the Pressure and how Safe is this?
You might have noticed that on the picture of the Water Filter Housing it says it has a maximum pressure of 125psi. This is only 125/14.7= 8.5 ATM or 85m. I have a Mechanical Engineering background. Knowing that everything is over designed I looked up "Safety Factor" (this is the term that is used when talking about how much something is "over designed") Here is what is on wiki about this:


"Appropriate design factors are based on several considerations, such as the accuracy of predictions on the imposed loads, strength, wear estimates, and the environmental effects to which the product will be exposed in service; the consequences of engineering failure; and the cost of over-engineering the component to achieve that factor of safety. For example, components whose failure could result in substantial financial loss, serious injury, or death may use a safety factor of four or higher (often ten). Non-critical components generally might have a design factor of two. Risk analysis, failure mode and effects analysis, and other tools are commonly used. Design factors for specific applications are often mandated by law, policy, or industry standards.


Buildings commonly use a factor of safety of 2.0 for each structural member. The value for buildings is relatively low because the loads are well understood and most structures are redundant. Pressure vessels use 3.5 to 4.0"


What does all this mean?
It means that this Water Filter Housing is likely "Over Designed" by a factor of at least 3 so 3 x 125 psi = 450psi. I am not saying that you should routinely take this thing up to 450psi, but I think it is quite safe to go to 150psi (only 1.2 x what it is rated)


Everyone has to decide for themselves what they are comfortable with, but I am comfortable taking this up to 150psi. Why then, do you ask, do my pictures show approximately 132psi or about 9ATM? Well my compressor has a regulator on it that prevents it from going over 150. When I have the regulator maxed out it cuts out just shy of 150psi and after dumping some air into the Water Filter Housing all I have left is about 132 or 135psi. I could buy a different regulator, but I have decided that 9ATM is by far good enough for me.


Here is a pic of the housing thickness next to a dime. It is pretty dang thick and a whole lot more "Safe looking" than a Nalgene bottle!



Ok, here is my disclaimer......
If you blow yourself up or wreck your watch it is your own fault etc etc. Don't come crying to me....




I hope this was helpful!
 

·
Registered
Joined
·
967 Posts
Great job. Here is mine. I built it few years ago, after I saw it online. Mine uses tire valve stem on one end so I can use a bicycle pump to set pressure. The other end is quick release. Works great.



 

·
Registered
Joined
·
1,143 Posts
Discussion Starter · #4 ·
Looks like you used virtually the same water filter housing. I knew about this idea for a long time before building it. The main piece that I didn't now about was the water filter housing. I guess it could be anything, but it has to be strong and also has to be clear so you can see through it. The built in ports and the threaded top make this thing ideal though...
 

·
Registered
Joined
·
475 Posts
If i'm reading it correctly, the methodology is a bit off from standard procedures.

Standard "bubble" test is;

Partially fill container
Insert watch but do not immerse
Pressurize and wait five minutes
Immerse watch (if you see a steady steam of bubbles, remove immediately)
Slowly release pressure and check for a consistent release of bubbles.

If bubbles are present then it's a fail
If no bubbles are present then it's a pass

Your approach would only work if there was a very slow leak which introduced a delay in pressure equalization in the watch. In the event of a more serious leak, it would allow ingress. I also cannot recommend either methods for compression cases like Vostoks as they rely on water pressure to be water resistant.
 

·
Registered
Joined
·
967 Posts
You are right Lokifish. If the watch does leak a lot, and it can depressurize between the time of releasing the pressure and submerging - then you are in trouble. That is why I installed quick release valve. I didn't like the built in one as it seemed somewhat slow. Also, when I test my watches I don't start with 120psi. First try is 20psi then I move up. Never had a problem and it did catch few leaks. Better than nothing - and much cheaper than commercial alternatives. And if it does leak a lot and it floods the watch, service it again and do another try. I'm not testing a customer watch. This is a hobby and I thinker with it.
 

·
Registered
Joined
·
1,143 Posts
Discussion Starter · #7 ·
If i'm reading it correctly, the methodology is a bit off from standard procedures.

Standard "bubble" test is;

Partially fill container
Insert watch but do not immerse
Pressurize and wait five minutes
Immerse watch (if you see a steady steam of bubbles, remove immediately)
Slowly release pressure and check for a consistent release of bubbles.

If bubbles are present then it's a fail
If no bubbles are present then it's a pass

Your approach would only work if there was a very slow leak which introduced a delay in pressure equalization in the watch. In the event of a more serious leak, it would allow ingress. I also cannot recommend either methods for compression cases like Vostoks as they rely on water pressure to be water resistant.
ok,
maybe I should modify my procedure. I built this thing before testing anything. I have only done a few now and they have all been my own.

I built the thing before becoming an expert on the procedure so it is certainly possible that my procedure could be improved.

Here is the thing...
If you are truly worried of a massive leak, then I'm not so sure your "standard bubble test" is going to work either. The principle that makes this test work would not cause bubbles to come out of the inside of the case the way you do it. At least the way I understand.....

Lets say there is a massive leak. Pretty much no seal at all and almost an immediate equalization of pressure inside and outside the watch. once you dunk the watch in your procedure there would be no difference between the pressure inside the watch and outside it and hence no bubbles. The difference in pressure between inside and outside the case is what is going to cause a steady stream of bubbles.

Maybe I revise my procedure to release some pressure, then dunk the watch and slowly release some more pressure. This should keep the inside of the watch dry even in the case of a massive leak AND detect a leak in all types of leaks.

If the pressure inside the watch is greater than in the container and the watch is dunked and it is leaking (even massively) then the air is trying to get out of the inside of the watch and the pressure inside the case is going to keep the movement and everything else in there dry, right?
 

·
Registered
Joined
·
133 Posts
This is mine:


This is slightly more expensive because of the dial indicator. It would be better if the dial indicator were better.
 

·
Registered
Joined
·
967 Posts
Looks nice, but how reliable is it when testing some of the divers with supper stiff case designs (watches designed for 1000m for example)? My understanding is that watches like those don't flex enough to show accurately in a dry tester.
 

·
Registered
Joined
·
21 Posts
Looks nice, but how reliable is it when testing some of the divers with supper stiff case designs (watches designed for 1000m for example)? My understanding is that watches like those don't flex enough to show accurately in a dry tester.
I would think it would be at least as effective as testing a 1000m watch with a 10 bar of air pressure
 

·
Registered
Joined
·
133 Posts
Looks nice, but how reliable is it when testing some of the divers with supper stiff case designs (watches designed for 1000m for example)? My understanding is that watches like those don't flex enough to show accurately in a dry tester.
I have tried with a Seiko SKX007, which is a sort of stiff. With 3,5 bar pressure it was shrinked around 25 µm, which is a significant amount. If it does not change for example within 15 min, and still it recovers after pressure is released, there is no massive leaking. The bubbling in the same case should not be significant either.

Also I have tried with Dievas Vortex, which is rated for 500 m. I don't remember how much it shrinking with 5 bar pressure, but also significantly.

Better dial indicators can show measures with a couple µm resolution...
 

·
Registered
Joined
·
21 Posts
Care to elaborate mechanics of it?
the mechanics of it are that instead of testing something at 125 or 130% or what have you of a rated depth, both systems would be testing it at a fraction of rated depth. So you could easily claim "well it tested good when i had it" and still have a failure around 20% rating and perform dismally 'in the field'


edit: more plainly what I am getting at; each system is good at what it does within the limitations of the system, so to say that "what if XYZ that is wildly outside of system limitations occurs" isn't really a valid comparison or critique
 

·
Registered
Joined
·
967 Posts
The problem with dry testing, as I understand it, is that it relies on crystal deflection. When the case and crystal are build like a tank, it takes a lot of pressure difference before crystal deflects, hence 10 bars of pressure might not be enough. In this case a watch that leaks like a sieve and one that is watertight behave exactly the same at 5 or 10 bars. In order to test some of these watches you might need 30 or 50 bars. I don't have abilities at home to test at those pressures. If I'm wrong in my assumptions, please educate me.
 

·
Registered
Joined
·
7,966 Posts
The problem with dry testing, as I understand it, is that it relies on crystal deflection. When the case and crystal are build like a tank, it takes a lot of pressure difference before crystal deflects, hence 10 bars of pressure might not be enough. In this case a watch that leaks like a sieve and one that is watertight behave exactly the same at 5 or 10 bars. In order to test some of these watches you might need 30 or 50 bars. I don't have abilities at home to test at those pressures. If I'm wrong in my assumptions, please educate me.
It relies on deflection, and yes some watches can be so rigid that my dry testing machine will reject the vacuum test, but it rarely rejects the +10 bar pressure test for that reason. That usually happens when a watch is rated for extreme depths, like 1000 m or more. Here's an example of how I handle a situation where the case fails a vacuum test, and I need to confirm that it is still sealed. This is my dry testing machine that uses vacuum and pressure for testing water resistance - set to the dive watch program, which is the most demanding setting,. It starts with a -0.7 bar (gauge obviously) vacuum, and then a +10 bar pressure. The watch in question is a Panerai 049, rated to 300 m:



During the vacuum test I can see that the test is going to fail based on how the numbers are going - note that the level of deflection being measured here is very small as this machine is extremely sensitive:



Watch failed the vacuum test, and passed the pressure test:



For watches that fail either test, I can perform another test on them using a program called "leak finder." This pressurizes the watch for en extended period of time, and closely monitors the deflection:



The machine is checking the deflection to an even higher degree than in the initial test, to determine if it's safe for a water test - this one is:



I then submerge the watch in a glass of water, and watch for bubbles like you would in a cheaper wet testing scenario like the OP is simulating with his home made device:



No leaks:



But to be sure, I'll use the heater for my high pressure wet testing set-up to perform a condensation test:



Drop of water applied once the watch is warmed up:



Inspected with a loupe under strong lighting, and no condensation, so the watch is fine:



I know there is this idea you are suggesting that the results of a dry test will be a mystery, but if you work on watches daily as I do, there aren't going to be any surprises about the water resistance of a watch you are working on. Based on the condition of the watch, knowing how the testing process actually works, and loads of experience you usually know going into the test how the watch is likely to perform. So in the scenario you presented, I won't be left scratching my head deciding if the watch is too rigid or is leaking - that will generally be pretty obvious to someone with experience.

And of course I do have testing equipment that will go far beyond 10 bar - my wet testing equipment can test to 125 bar.

But the dry testers are much more than the example shown above where a dial indicator is resting on a crystal. In addition to being much more sensitive than that is, these machines don't just measure overall deflection, but the rate of the deflection change, and what the pattern of change is are also looked at in the software. Parameters are set for all sorts of case information, such as crystal types, diameter ranges, etc. I have programed custom tests on this machine for situations that are not covered by the pre-programmed tests.

Hope this helps.

Cheers, Al
 

·
Registered
Joined
·
967 Posts
Awesome explanation. Thanks Al.

I guess with dry testing, you have to have extremely sensitive depth gauge (is that proper name?). I was not questioning purpose built device to do dry testing, but home made one. As you pointed out, it has to be very sensitive, and looks like other software controlled parameters come into play as well.
 

·
Registered
Joined
·
133 Posts
Awesome explanation. Thanks Al.

I guess with dry testing, you have to have extremely sensitive depth gauge (is that proper name?). I was not questioning purpose built device to do dry testing, but home made one. As you pointed out, it has to be very sensitive, and looks like other software controlled parameters come into play as well.
No, not really. Like I wroted earlier, a poor dial indicator works usually just fine, at least the watches that I have tried (rated up to 500 m). Better one would be good if you need something special data such as leaking speed etc. which is not possible in a bubble tester anyway.
 

·
Registered
Joined
·
21 Posts
647 Dial Comparator

something like this would be pretty dang sensitive, with the indicating lines at .00005". I have no experience with dry testing of watches, so i don't know how much deflection they are looking at, but if you are going to home brew one and don't mind spending a couple hundred $$ on a quality instrument it could be a good way to go. can't imagine it would take much to measure in multiple positions instead of just on the crystal for deflection

traditional style dial indicator with same graduations but more overall travel. half a tenth is 1.27 microns, just for reference for the metric folks (most of the world) edit: so, a $300 dial indicator linked below would have about 38 micron total measurable range in either the positive or negative direction (case shrinking or expanding).

http://www.starrett.com/metrology/product-detail/metrology/metrology-products/precision-measuring-tools/indicators-gages/Dial-Indicators/25-109J


 

·
Registered
Joined
·
7,966 Posts
647 Dial Comparator

something like this would be pretty dang sensitive, with the indicating lines at .00005". I have no experience with dry testing of watches, so i don't know how much deflection they are looking at, but if you are going to home brew one and don't mind spending a couple hundred $$ on a quality instrument it could be a good way to go. can't imagine it would take much to measure in multiple positions instead of just on the crystal for deflection

traditional style dial indicator with same graduations but more overall travel. half a tenth is 1.27 microns, just for reference for the metric folks (most of the world) edit: so, a $300 dial indicator linked below would have about 38 micron total measurable range in either the positive or negative direction (case shrinking or expanding).

25-109J Dial Indicator

The Witschi above measures to 100ths of a micron...
 

·
Registered
Joined
·
21 Posts
what would be considered an "average" amount of movement for a watch that could be used and pass? obviously not talking about a very stiff case. how many 100th of a micron of movement?
 
1 - 20 of 39 Posts
Top