Hey everyone! Since a lot of people are saying it, yes, of course the pump mechanism isn’t a source of bleed-down because it’s after the ball valve. The ball valve and hose are part of the pump and that’s what I meant when I said “pump”. Let’s all be good sports here and give each other the benefit of the doubt on such things. 😉 Another common question is why didn’t I clamp the end plates by threading rod all the way through? The reason is that the tubes are soft at this point. They are annealed from the heat and are easily distorted. They would not take the clamping forces needed for a 120psi seal. The tube flanges are much much stronger.
I worked with fire sprinkler systems and often the ball valves "let by" and down to reasonably low pressure. This must be the case or you would be seeing oil somewhere, must be returning to reservoir right?
I knew what you mean, Quinn. We have been on a European River Cruise and the guide said there are a lot of Canadians on board. She said if you need to go to the toilet can you please say Bathroom or even better Restroom. ??? I said its taken my wife 5 years to teach me to say toilet! One massive bloke (maybe a lumberjack?) used to call it the Powder room! (Oh dear!)
FYI - it's easy to make custom sized o-rings. On a flat surface, form your stock into a circle of the correct diameter, overpassing to the outside and then slice thru both simultaneously with a single edge razor blade held at a 45º angle. Then cement the two ends together with superglue and you're good to go.
@@petem6291 Yes, it works well. I've used a larger o-ring with the the same cross section, cut and super glued to the diameter needed for a hydraulic fitting. This was on a Gatwick friction welder, ran for years. We ordered the correct o-ring but never installed it.
Really enjoy your channel, only came across it in the last couple of days from your failed boiler video, and have gone back and rewatched a lot of your videos and have got me motivated to get out into the workshop again. Keep up the great stuff
Just a quick note to thank you for your videos. I understand and appreciate how much effort you are putting in to provide us with this content. Thank you, thank you thank you!!!!!!!!!!
Your combination of knowledge, experience, humor, and humility is spot on. Your narrations are second to none. It's wonderful to learn from you and with you!
If you ever want to do this again, leave your plates at 5" and drill 3/8 holes around the outside edge and clamp them together with threaded rod. You could easily get 12 rods around the outside. Also maybe neoprene would make a better gasket? I expect the ends of the tube core aren't perfectly true. Nice work and thanks for sharing!
@@Murgoh I was thinking this too. My only worry is that this would place a lot of force on the solder joints themselves, as they now become part of the clamping structure, rather than just clamping the endplates against the new fixtures. I think that silicone pipe may have been better used sliced open to make the gaskets rather than the paper jointing material.
I've built a similar contraption for pressure casting resin parts. I found the solution to eliminating bleed down was to seal the gaskets with regular old plumber's putty. It eliminated the need for elaborate clamping, as it compensates for the subtle flexure of mating surfaces.
I have a very similar pressure testing pump that I bought at an auction, about 20 years ago. Now, at last, I know how it's supposed to work and how to use it. Thanks Quinn: you're a mine of information and an inspiration.
Cheap c clamps are great. For tight uses like this some cheap c clamps with the handles removed and some flats milled in the shaft for a socket can be a life saver.
Good idea as long as a gorilla does not get on the socket. GP does not stand for general practitioner (doctor). It stands for gorilla paws - ie overtightening.
Thanks, one of the things I really like about your approach to a problem is your way of thinking outside the box, even if if fails we learn something everytime.
Quinn knows this, bot for the rest of you who don't, ME thread series exist not just to adequately fasten parts, but to have "near scale" appearance. So 5/16"-32 "looks scale" because of the high thread count. Like Egyptian cotton sheets, or something. (Also OMG the umlauts. I am now dead)
Hi Quinn, hope you are well. I loved this video, only remark I can add is, the end gaskets I would have used rubber as in an old Lorry innertube, more robust will fill any micro gaps and not so lightly cut by the copper, what a brilliant idea you had for testing those tube joints. Love you and your humour. Take care and be safe. X
Thanks to you showing your experience I attempted one. Silver solder was a pain as well but it held 200 psi last night 😊. Thanks for all you do and show
I have learned a lot by watching you all the while being entertained. I think if I would have had someone like you as an educator maybe I would have went into engineering instead of electronics/computer science. You are amazing.... thank you
I do my fair share of soldering and after watching you silver solder the tubes i felt they would be solid. Your forethought of testing them while you had access was great. Looking forward to next vid in the series👍
its not just the gasket. 4 in dia = 12.5 square inches. x 60 psi = 754 pounds. the end plates are thin and can flex as well as the clamps. A couple thousanths will exceed the gasket compression. Thanks for the interesting project.
Hi Quinn, I think that was a good way of going about it. Gave visibility of all the tube joints on both sides which is very helpful if not essential. You got a high probability that the completed boiler would take the x2 test pressure. You'll obviously need some sort of pressure gauge and stop valve near the boiler to hold the test pressure if you can't fix the pump run down. Maybe you could say a few words about the boiler test regs you will be following.
Clämps hee hee. 60? Thought they would have peaked in the 80s? I was laughing by the end of the intro having just spent a nerdy Friday night. Great work as always.
Cool. I used to do industrial boiler-cleaning in my school holidays we pressure-tested 80 year old industrial boilers. Dad said be careful - steel never forgets - a warning to not over-test.
Thanks for the great work. I really enjoying your videos. Keep up the great work. You have a very good way of explaining the what you are doing. And why great work..
Whoa - overlaying your Fusion models on the real parts is super powerful for visualization. I don’t think I’ve seen anyone do that before! Also, it’s great to see Swarfy nesting so happily 🦆
With the hose isolated (closing the valve), the part that is bleeding down should be after the shutoff valve. With all the fittings tight, I would look at either the valve (internal sealing of the petcock), the hose (failure of the rubber in the hose, in my book the most likely culprit), or possibly the gauge leaking internally.
@@Blondihacks if you're ever looking to replace it I highly recommend Parker 919. For low pressure stuff (1000-3000 psi) it's really very nice. I like it because it is chemically inert and I have used it for everything from high pressure machine tool coolant to brake lines. And you don't have to have a crimp connector you can actually get field assemblable fittings. It's a poly tube with a stainless overbraid and the -4 size almost looks like scale model hose.
I have been thinking about the need to pressure test this before assembly since I saw the last video. And right here on your channel is a great video showing how you did it. I would not have wanted to assemble this assembly without knowing it held pressure.
The oh!? ring probably wouldn't have worked but they make o-ring splice kits so never worry about not being able to find the right size(length/circumference) you can cut and glue them and it works just fine. You're videos are an inspiration keep up the good work! One more little tidbit, a drafting compass and an exacto knife in the pencil holder makes the prettiest round paper gaskets ever.
I too have a disproportionate like for facing off bandsaws cuts. Your big steam engine crosshead video reminded me of that machining the sliding surfaces. I kind of want to incorporate the "machined surface with low spots" aesthetic into something.
Nice results! I can recomend to exchange the ball valve for a needle valve on the pump. Maybe together with a checkvalve in the circuit. This will hold the pressure much better. This is how most hydraulic handpumps work for calibrationg gauges and pressure sensors.
With all 12 clamps on it it looks like some kind of sci-fi power core (maybe for a space ship’s “arc reactor” or a big mech). Regardless, I really enjoyed this video; I found it to be a really interesting exercise. Really looking forward to the next installments of this series!
Looks solid to me but if you really want to keep testing you could run threaded rod through the tubes and thread them into the bottom sealing plate. Drill holes in the top plate and use (silicone/gasket?) to seal up nuts/washer/what ever works. Should provide a more even clamping force though I would use a thicker rubber style of gasket. Can't wait to see it in action!
I think it should still need some (a lot of) clamping force right around the rim, independent on each plate, clamping the copper and aluminum at the rim. As the pressure builds, the copper dish will deflect in towards the opposite end. Rods in the middle could help control flex in the aluminum plates when pressurized, but under zero pressure could also put a lot of extra stress on the solder joints.
@@BrianFullerton yeah, I was thinking about that after I posted. Could be solved with metal or harder wood half circles inserted between the copper end plates to prevent them from being crushed. The force would then go to the bell ends (for lack of a better word, and funny) side walls. But regardless of that, the gasket is the main problem. I would use an ⅛" thick rubber gasket instead of the paper like gasket she used. 99% chance it would work with the clamps she got.
Hey Quinn, I work on diesel trucks and have come across a lot of flat round rubber seals for fuel filters or fuel water separators that would probably be the right size. If you know anyone who works on big trucks they might have some. Also instead of clamps you might be able to seal it be passing a piece of all thread all the way through and use copper oil pan drain bolt seals and nuts. Speaking of copper washers, if the copper rims are flat enough and you can get it tight enough you might not need a gasket.
Hi, what you observe is maybe a isothermic effect. When you increase pressure the temperature will also increase, this temperature will bleed out of the system overtime, this will give you a drop in pressure. Used to work with calibration of pressure sensors, and we always needed to wait a while after increasing pressure before taking measurements.
For gaskets you might want to get a sheet of Klingersil 4400 , that material is resistant to water, steam, oil and many other liquids and even gases. It is easy to cut with knives, scissors and you can use holepunches.
I’ve had to part off odd slugs in the band saw. I did roughly the same thing, except instead of a through hole, I drilled and tapped on side. Then bolt to the angle block and can part off all the way….
Very neat project, Thanks for sharing! I would say 10 PSI would have been enough for a leak check of all your joints. The point of a hydrostatic test isn't for leak testing but to test structural integrity of the assembly. Failing to reach hydrostatic test pressures for a leak test is inconsequential.
I think the slow loss of pressure in the pump test may be caused by slow elastic flex of the tubing and other parts. I would guess that to stabilize at some point. Water can not escape just like that without leaking. Also, some plastics are known to absorb water a little, like 1% of weight in the skin depth.
Internal leaks in a pump would tend to flow back to the reservoir and not necessarily show up anywhere externally on the pump. That said, any small amount of water loss, no matter why will effect the pressure.
@@noelhenderson700 To be precise, what I meant was that since the pump is isolated with the valve from the SUT, the SUT pressure was dropping. The reasons for that are what I mentioned plus possible leaks. In this particular case also the paper gasket was dry when installed containing substantial amounts of air and is probably the main culprit. The gasket might be wet completely for a few days before put into place. Water is nearly incompressible and so are all metal parts. However, the parts suffer from dimensional changes when pressure rises and this will lower the pressure from the initial reading. Due to incompressibility, extremely small volume changes affect the pressure after closing the valve. One more thing, tap water usually contains lots of gases too affecting this. Using vacuum for removing gases from the system before letting water in and increasing the pressure might be a good idea.
I was thinking the same thing, as I've been designing an adhesive dispense system and I think the ever so slight expansion of the adhesive container is something I'll have to account for.
Quinn, conversation back to the o-ring thing. Now that you have quite excellent end plates and clamps, might you trade the paper gasket for a thick o-ring? It would fill the space in the copper dishes between the OD of the outer ring of fire tubes and the inside of the rim of the dish. Needs to be thicker than the rim of the dish so that the aluminum end plates will bear against it, but you would have a pretty wide tolerance for o-ring ID and OD. As pressure builds, the oring would press out to the rim of the dish, but should keep a seal down at 120 psi. I just checked the yellow book and it looked like there were a few options that would work. Could always glue up the size you need, but it looks like there are some stock options that would work, depending on the depth of that rim.
Was thinking the same, if i really needed to do that test at 120 psi. Alternatively, a ring cut with square cross section cut smoothly from a soft rubber mat (or a stack of them). Since a thin gasket can't compress much, any small deviations/deformations from flatness in the surfaces isn't likely to be filled up enough by the gasket. It's different when the gasket is sealing a wide area like a flange.
@@hieronymuslarsson1388 @Hieronymus Larsson Yes, understanding that this was an intermediate test, and that she got what she needed out of it. And that we are seeing this (possibly) after she is 10 steps ahead IRL, so our comments are not timely. But she is testing it at a critical point in the assembly, and if more PSIs were needed on such a shape , not sure how I would get there. My other idea was an external oring on the OD of the dish, in a groove in the ID of a piece of 4" sched 80 or 160 to get enough material to get down to 3.92". Would not need to be steel pipe, heavy PVC would do it at that pressure and temp...and be cheaper/ easier to machine. Soft solder in the 3 rivet holes... But I have some of that in my scrap pile, she probably does not. Fun problem to think about.
@@BrianFullerton Sure, working against the OD of the dish would work too, one big advantage being that you'd be at greater liberty to choose size of the cross section of the o-ring. Supposing the dish is round enough. Yeah it's one of these oddly interesting problems. Sealing is as satisfying to get right as it is frustrating when it doesn't work. It's neat when the pressure helps seal off itself.
Great videos. At the start of this when you were leak testing the pressure tester you were saying it was leaking slightly. Something I have come across when using flexible hoses is that they can expand a little which will cause a pressure drop. You can prove this by dropping the pressure a bit and may see the pressure rise again as the hose relaxes again. I found this while working calibrating pressure transducers.
Of course, hindsight being 20-15, looking at your band saw set up I thought of another way that might work. What if the center hole was threaded, and use a bolt that's just short of your second cut? Always, entertaining, and informative as usual!
Nice trick to make custom circular gaskets is to glue a sheet of rubber onto faced round stock and clamping an exacto knife in the toolholder, with that simply cut out the shape you need.
Surveying humor; very nice! Also, your references to woodworking tolerances are so close, you make me feel like I'm using a crayon for marking and a stone axe for cutting.
When you close the ball valve between the pump and gauge that takes the pump out of the leak equation… most likely your ball valve is leaking. Since those pumps use checks to initiate pressure build it’s not uncommon for the checks to get some debris or defect artifact in them hence why that ball valve is there.
When I saw the thumb nail I thought you’d made a nuclear reactor which wouldn’t have surprised me with your physics knowledge. Looks like you’re going to have a cool boiler when finished.
Yes, it is a perfectly normal thing to do on a Friday night... what else would anyone be doing at that time? Personally, I wouldn't know because I'm doing my hobbies on Friday nights too (only I don't post on YT, best not to advertise)... LOL 🤣😂
I keep a pair of 12 inch surgical tweezers by the lathe for bird's nests. They're only 7 bucks or so on Amazon if you want to try them. They're very helpful and they keep your hands further away than pliers.
If you made two plates over sized to where you could put bolt holes around them just outside the diameter of the core, then make two more and thread them to be bolts to the over sized one, but cut those the inside diameter to just miss the tubes! Cut those in half, then they would push on the bottom of the core plates and would give you all around clamping force! Have made ones like this to do press testing back in the day! You could also make short plates the could hold the cut plate from spreading about where they come together for more rigidness, just a thought!
Core testing, I wonder if we are going to get a clip from Chernobyl? That's a cool fixturing trick for the bandsaw, thank you for sharing. Have you considered bolting an adjustable stop bar vertically near the on/off switch to catch the head of the bandsaw to keep it from cutting threw the bolt? Or you could take one or both jaw inserts off the vise to gain a little bit of capacity. Great video, looking forward to part 10.
Nicely done Quinn. I am thinking that if your hydrostatic test unit can't hold pressure but does not leak any water the problem may be the shut off ball valve leaking pressure back into the main unit. I have had a number of lesser quality ball valves fail in this manner. Replacing it with a high quality ball valve solves the problem. Keep up the great work... Fred
A solution that might work is to get some industrial rubber "mat" (Not sure about the correct word for this since I'm not a native English speaker) but they are available with flat surface in different thickness. A relatively thin one cut into a circle and you will even out the pressure more from the clamps.
I would be tempted to rerun that test using something like 6mm thick rubber sheet instead of the gasket paper. Enjoyed that very much, particularly the method for holding large stock in the bandsaw.
you may have been able to seal against the flat inside edge of the top and bottom on the outside of the fire tubes but inside the wall, dish out the blocking plates so that there is a ring on the outside edge and using a thicker gasket material with the 6 clamps on each side
A can of flex seal in a pour form can be used in a mold to create any rubber design. Just a thought it may be another angle for you to attack some of your problems.
Not sure if I am seeing correctly but you closed the ball valve then looked at the bleed off of pressure on the gauge on the hose side of it? If so then likely your ball valve is leaking for a start before you get back to the pump seals and clearances. You will only need a few drops of water to drop the pressure too (assuming you have no air in the system).
11:40 Model engineers do indeed use a whole lot of odd threads, even here in the UK! Most modern designs still use BA threads, even those designed by the younger generation. For someone born in mid 70's between metric and imperial, even I find this weird! Why 😂
It's about using an appropriately sized fastener for the scale of the model and the smaller BA fasteners look better on a lot of engines, especially if the kit was designed with them in mind.
@@douglasbollinger8678 I had thought that metric would've been an alternative, M2.5 M3 M3.5 etc but just noticed .5mm increments seems to stop at M4 so yes, I understand now!
If you really want to earn some hate, use metric, but in JIS before 1967. They usually used a pitch just a hair above the european one. Things like M3*0.6 (instead of 0.5), M4*0.75 (instead of 0.7), M5*0.9 (instead of 0.8) ...
Paper gasket is used between to wide flat flanges then it would work. Maybe use some soft silicone sheet instead in your setup , this would let the thin edge of the copper bite into the silicone to seal slot better than the paper gasket. Keep up good work.
13:38 When I'm using loctite on small screws, I usually put a small blob on my bench, then dip the screws into it. But you know, spilling it everywhere also gets the job done. :-)
My 1st thought for gaskets (even before paper (which I do have a box of in my shop)) was backhoe inner tube (which is 3x thicker than automotive)... ;)
