Piped up a combi just today using all pulled bends on copper and only switching over to Polyplumb using straights. Plastic is great for getting from here to over there under the floor or generally out of sight but it does look a bit shite. For me a nice tidy copper pipe job with end feed joints is best, but I'm an old git. ATB from J and H Builders.
Agree totally. You can't get a better install than copper. Plastic pipe is the worst idea after lead and asbestos was dreamed up imo. Copper doesn't contaminate the water and even has natural anti bacterial and anti fungal properties.
In our apartment block, the bent plastic pipes are failing and leak. They are 35 years old and, over time, the stress on the bends has eventually led to tiny cracks gradually forming. This is on both hot and cold pipework.
They're trying to sell you something so you always have to question these demo rigs. They're feeding both pipe runs from a common low pressure pump so obviously the water will take the path of least resistance. If you fed those runs with mains pressure, you'd get plenty of flow through both. Yes you'd get more through the one without fittings (mainly because it doesn't have inserts) but you'd still get a good flow through the JG section.
Not only this but the system they are pushing uses crimp fittings which have a thick barb which further reduces the diameter, much worse than common thin wall plastic pipe inserts due to having to withstand the compression force during crimp. Factor the crimped ends into the equation and the system they are pushing ends up with lower flow of the 2 due to severe restriction, despite the restriction only being at some points. It is substantial. They are trying to lose the poor flow reputation that system is known for in the states. Press fit copper or soldered copper is still king, maintains a minimum of 15mm at all times and the wall thickness is about 0.2mm, times 2 = 0.4mm. Those barb type fittings are 2mm plus 2mm for the pipe wall = 4mm. double it as there are 2 sides to the pipe.. 8mm reduction. Marketing desperately needed lol..
just remember that certain pipe formations causing, for example, back pressure and shock waves are of advantage, and that gasses are compressable - water isn't. It's hard to compare the two apart from superficially.
Pull bends and use swaging tools, for fifty years plus.Laughed at on a recent course when I mentioned frictional resistance on modern pipe work. ……Physics doesn’t change boys.😎
I saw somewhere that swaging was no longer allowed as it thins the copper. My dad was a plumber for 50 years and I don't remember him ever going back to a failed swage joint... Go figure.
@@MaidanRustledomestic hot water... Who the fuck doesn't know that in the plumbing world? It's also not an Acronym, it's an initialism. An Acronym is "NATO" or "NASA", you pronounce an acronym like a word. An initialism, you say the letters. 👍🏻
Vena contracta!! Have exactly the same on my shower. Removed 5 plastic pushfit 90 deg bends from the plumbing and fitted nice curved pipe instead. What a difference. Lazy plumbing. Brilliant video!!
Very interesting. I was working in the central heating 57 years ago . Everything was done with black iron pipe. All the pipe work had to run slightly up , all the bends had to be done by heating the pipe or using the hydraulic pipe bender . For all the branches coming of the main pipe, we had to bend the pipe and cut the bend with what I would call a shoe . Then we had to melt a hole in the main pipe and place the shoe on the main pipe and weld it . Everything had to be easy flowing, and all the air had to be able to run to a high point with an air pot . I loved the work . In difficult positions, we had to use a metal mirror to see the welding work .
Brilliant demonstration, confirms what my dad always told me. I like the double wall plate elbow and loop which ensures the water is always fresh in the entire system (2.17)
To be fair for a true test the pipes need to be run separately, you could fit a tap at the base of each pipe then only turn 1 on at a time & then test the flow, otherwise some water will just take the path of least resistance. (Plus you could still turn on both at the same time if you wanted). Obviously the flow should be less in the pipe with 90s but I'd bet it's not as bad when run separately so it would be nice to see.
same issue with those cheapo isolator valves - the ones with the screwdriver slot that any sane plumber will never turn for fear of it weeping! Had an 9.5kW electric shower which had one of those in the supply. It was overheating the water, basically poor flow rate. Removed the isolator and put a swept bend on the supply and the difference was remarkable.
We do a similar experiment in 1st year chemical engineering. It demonstrates the different pressure losses across bends, fittings, valve openings, valve types etc.
The design "rule of thumb" is that every bend fitted was equivalent to the friction loss of an additional one metre of pipe, using the D'arcy Weisbach equation if I remember correctly from 40 years ago...
@kc8485 obviously with the right diameter with a suitable flow. The diameter I don't remember but prob 2 to 8 inch or 25 to 200mm. I have experience a lot of odd combinations tho... If you google tlv pressure loss, you can get to a simple online calculator to play.
In practice, you will use fittings and butt welded joints in chem eng and spec the pumps to suit. Nobody has got the money to spend trying to bend 304 sched 40s pipe at any size over the about 1" on site, it would take forever. There are specialist applications where you might use swept bends etc like steam blow down lines and other weird and wonderful stuff like jacketed pipework for pharma, but it's extremely exceptional. Source : engineer and project manager in the process pipework industry for decades.
@davelowe1977 and then you get the other end of vacuum applications and then 2 m diameter lower pressure gas networks with bends welded in section. But yeah, small diameter water pipework, bend it or push connect.
I definitely agree with the theory - a nice smooth, larger radius curve "must" surely allow better flow. But on the demo it looks like the bore of the fitting pipe is smaller than the curvy pipe...
@@SkillBuilderhalf mill throughout the entire length though Remember also the increased diameter, all be it slight here, has a multiplying effect on carrying capacity
Fantastic demonstration. I'd love to see a comparison between long-radius elbow fittings and continuous pipe as I always try to use long-radius fittings where I need a fitting but it would be good to see the difference.
DIY amateur here: I use copper compression fittings because I can extend or adjust pipework without cutting any copper pipe. Yes, hotels and big housebuilders don't need to modify their installations but a home owner sometimes does to move or install a basin, an outdoor tap or a washing machine for example. To be frank, I don't trust plastic (but I'm old-fashioned).
I'm not a plumber but I have a degree in chemical engineering. "Ells", as they're called in the chemical industry, add drag equivalent to a pipe length of 30x the diameter, so one elbow in a 15mm pipe is like adding 18" of pipe. They are ok when you have lots of pressure to play with but not in the feed to a gravity shower or the inlet piping to a shower pump.
But also... If you buy the 'standard' compression fittings for 15mm copper you will have a similar problem; however you can buy the 'full-bore' (or full bore) fittings which do not reduce the diameter at any point. I went to a job where the cistern feed went through the restricted couplings and you could 'hear' the constriction and turbulence. Changed to full-bore and it flowed more quickly, and silently. Similarly with heating runs, it's no good having 15mm flexi pipe, and then inserting a collar at the coupling which reduces to 10mm!
The two pipes looked like they were different internal diameters, additionally the pipes were different colors. So are we seeing a like for like comparison??? They should strive to SHOW that the pipe diameters were identical. ALSO, as pointed out by others, we must avoid the path of least resistance robbing dynamic flow from the other path. So SHOW that the feeds are NOT taken from a common pipe, BUT take both directly & independently from the header tank, to ensure same starting pressure (static & dynamic). BUT I agree (& know) that for flow rates: bends should be kept to a minimum; sharp bends are worse; and Elbows & Joints are Terrible.
I suspected it was better to avoid over-use of elbows, but I really thought your video would show I was wrong! I'm actually really surprised how great the difference is,
Many years ago, part of my job was to run flow-pressure calculations on pipework systems - elbows and tees were assigned an 'equivalent length' of straight pipe for pressure-loss. As another commenter has asked - yes, swept bends are in theory available, but most often you have to plumb with what you can easily get. On larger systems, 65mm / 2-1/2" would have been useful, but it just wasn't stocked by anyone in the correct spec we needed (BS1387 & BS1740 HW galv), so we had to make do with jumping from 50mm / 2" to 80mm / 3".
Use elbows if one needs a tight turn due to issues of space but much prefer to pull a bend for a better flow and its certainly much cheaper and less fittings to leak.
Very interesting Rodger. I’d like to see the same experiment but with copper pipe and push fit elbows. This would then remove the constriction and likely turbulence caused by the inserts.
@@SkillBuilder I would watch it. Please try and make it conclusive: same ID pipes, identical but independent water sources. You could also make more then two sets of test units as there many questions about this topic. Love your videos, cheers from Hungary.
I was taught that in the 1970's by my supervisor fitter while piping up filler machinery in a drinks bottling plant. We always used Stainless Steel sweeping bends.
It's a good demo - the elbowed route will have higher pressure loss due to friction, but supplying both from a common (presumably centrifugal?) pump means the flow thru the lower resistance route will build up until the frictional loses balance out. I'll dig out my dusty copy of Perry's if anyone's interested in the maths... To be truly objective they should be supplied from an "infinite" reservoir - a massively oversized pump, header tank or mains.
They are Hep2o fittings. If you are talking about the underfloor heating fittings used on Alupex there are only two per circuit and they are close to the flow control so that is a single point of resistance. It would be interesting to try the flow with and without those fittings but on underfloor heating the flow per circuit is quite low anyway, maybe 2 ltrs per minute.
I used five 90 degree 15mm copper pipe compression fittings and due to the sink position a flexible pipe, the flow rate is pathetic at the tap. The flexible pipe is I'm sure mainly to blame and the whole lot will sooner rather than later be replaced by a totally different pipe run. Interesting actually seeing just what a difference fittings can have on flow rate. Cheers Roger
A pair of benders requires skill and sometimes you mess up and ruin some copper, done it loads of times by over bended or kinking while pulling a bend. A pip bundle is so expensive nowadays so perhaps new plumbers can't afford to learn bending by making mistakes and just turn straight to a fitting?
This explains why I have poor flow to my kitchen sink. Was fine before it was moved during a kitchen refit. Plumber re-routed the hot and cold to allow a wall to be removed, he used push-fit 90's and plastic instead of copper as I'd asked him to. All done in about ten minutes instead of taking a little longer to 'do it properly'. Now that's all buried under the floor and I wait about a year for a bowl of hot water.
It's obviously going to be smoother flooding with a sweeping bend and the elbows are going to have a bounce effect as it goes through It's a great example of proof being the thing that most people won't agree with
In principle that’s correct, although I have my doubts about the validity of that rig. You can work out the end to end resistance on paper and then look at the pump curve to see what difference it makes to flow rate. With some pumps there may not be much difference.
As an electrician do a lot of showers and always alter the pipe work if surface , got a hand bender so do all the bends not only does it look better with out joints , less leaks
Impressive difference. Is this a poly pipe? Or rigid pipe that was bent? In the states we use "pex" water piping a lot. I thought it was mostly because it is easy to work with, but I didn't realize the potential for improving flow over rigid pipe.
I did my own house about 30 years ago in 22mm copper and didn't use a fitting. I thought at the time it's got to be better with bends. It's been great.
Do a flow test measuring time/flow rate on them separately as the water will obviously take the path of least resistance when running both in parallel same as if i run a hose pipe outside when someone is in the shower lol
I understand your point and it is interesting but if the bent pipe is the path of least resistance that is the whole point of the video. You have an equal pressure and flow at the pump so that is the same up to the point of the first elbow. I agree that the effect is compounded. I think there is much to explore here but the bottom line if that bends are better than elbows.
@@SkillBuilder I agree that the bends are better than elbows both for flow and better noise reduction and leak risk reduction but in this test the water will take path of least resistance so if your house has Elbows everywhere your mains water pressure will still go somewhere or maybe don't use elbows on your shower pipes so you at least get a better shower flow when someone flushes the toilet lol
This is exactly what was going through my head. For a true test the pipes need to be run separately, you could fit a tap at the base of each pipe then only turn 1 on at a time & then test the flow. (Obviously you could still run both at the same time if you wanted).
It's probably a bit of both. The plastic inserts on modern push fittings are really thick compared with the stainless steel inserts on the older grey push fittings from the 90s. Having larger radius elbows would probably help too. Hope the manufacturers give this a watch and challenge their design departments for their future products
@SkillBuilder I've always felt it's a bad system with the inserts I'm not clean on plastic but they are the nail in the coffin really currently building an extension I'm going to do all pipework in 22mm coper no messing about with 15!
The other problem in the UK is we don't use 18mm pipe in ordinary plumbing. It's either 15mm or big jump to 22mm. Often 15mm gets used when actually something larger is required, but the expense of 22mm stops it being used. 18mm is common all over Europe. Only gets used in commercial HVAC in UK.
The poly pipe company, demonstrated this years ago outside of most the plumbing traders, to promote their product, it showed how reduced labour time, every joint being possible leak, as I say this was used years ago, and the copper pipe people hit back saying, I always use plastic......to buy my copper pipe, and fittings
Do you know the internal diameter difference between them. It is neglgible and would not account for that difference and even if it did the white pipe wins. You can give it a dope test if you like but a win is a win.
@@SkillBuilder The internal pipe diameter has a great deal to do with the flow losses. Using Darcy Wisbech formula which uses cross section area of the pipe to calculate the frictional losses, even a small difference between the diameters will increase laminar friction and therefore slow down the flow. Cross sectional area of pipes changes radically even for small differences. Lets say the bore is 10mm, the CSA is 31.4mm2, for a 9mm it is 25.4mm2, a reduction of 17%. Yes introducing sharp bends, changes in bore, ect does reduce flow but not as much as the ID.
@SkillBuilder Not sure I can agree here I almost always do but the diameter internally looks quite different Obviously bent corners are better than fittings as well
@@glynnepritchard2526 The CSA is the only significant difference here. Hard bends will slow the flow but at such a small amount on this model it would be hardly noticeable. They also appear to be more than 1mm difference. The only reason in a domestic situation for using bent pipe is to save costs.
The inserts in the fittings are a restriction and the turn is more sudden. Sharing the flow is not entirely fair as the water will favor the path of least resistance. If you need to use fittings just use 22mm pipe up to the floor below the outlet then transition to 15mm, no problem with flow in most homes.
One can use 45 degree pushfit bends - that would have made an interesting comparison to 90s, as would a coil of pipe without sharper bends at all. Nice experiment but could have been so much better... also, several types of pushfit compared would have been a better test. But all of this becomes academic when one builds a well designed house with few or no bends and short runs from source to outlet. For retro fits all this may well be more important, I get that. Generally I am able to avoid using many fittings when using plastic piping, by using steady bends in pipe runs. Also, the pressure is not stated on this test, nor are back-pressures and these could well mean the test is flawed and results not of real-life scenario quality. I'd love to see a properly thought out test... you could make one for not much money. Thanks for the vid though - food for thought, certainly.
How did they get the bends so tight without using permanent bend formers? I'm wanting to use 15mm Hep20 PB Barrier Pipe for the first time (with traditional brass compression fittings & stainless inserts where fittings required). I plan to use some formers to cut down nos where practical. I might be missing something here, but based purely on material costs (excluding labour time), I can't see much cost advantage of plastic over copper & compression fittings for DIY size jobs.
I am more interested how much they deform under pressure and how they cope with water hammer, plus heat expansion. The water flow is irrelevant as the faucet will restrict the flow anyway and you only open it fully when filling up the tub. Such a diameter i saw only used in heating where the flow is minimal, in drinking water we use 1/2 inch pipes (16mm). Not for flow but to have room for scale and corrosion build up. Its also not DIY friendly as you need special tool for crimping, same as hydraulic lines. I always go for metal pipes. Only once i use plastic as the soil move too much and its acidic so a zinc coated pipe erode in 2 years. I start to see this like 8 maybe 10 years back. For heating where there is massive heat differences. In old days they use continuous welded pipes for hearing (and gas). They even adjust diameter.
I've been having this exact same thought recently and also looking into it. Mostly due to the mess and flow issues in my new build. A second floor flat with long pipe runs and either push fit or short radius copper elbows everywhere. And the boiler is fed by a branch of an equal tee rather than from one end of the straight through and the feed in the centre. Looking at this demonstration setup some thoughts do occur. 1: The bent pipe looks flattened on the bends. 2: The push fit pipe looks like it has a thicker wall so smaller internal diameter. That doesn't help flow. Though this might be an optical illusion. 3: How exactly are both pipe runs fed? If it is a single pump but the push fit run is supplied from a branch of an equal tee rather than the supply being put into the centre of one, this wouldn't help. 4: How would it compare with long radius elbows? I see a lot of tees on that pipework at the end. None of them look like swept tees. Very bad for flow as they look like they have sharp, short radius corners. A disappointment for me is swept tees (which only seem to be available in copper) have a very poor sweep. Also there are no swept equal tees (difficult in copper I understand) but in plastic both these fittings should be as easy as anything to make but they seem to not exist which is another disappointment.
The supply is fed from a single pump which enters on the middle of the branch of the tee so it goes left and right equally. The pipe is not flattened on the bends The wall thickness of the white pipe is very slightly more. The diameters are measure on the outside. Long radius elbows improve things but bends are still better.
What was the water pressure, and how does a higher pressure affect the result? It feels like on gravity fed systems (e.g. 0.5 bar) the impact would be proportionally much higher than mains fed (e.g. 3 bar).
Not an expert but do the push in fittings have a reduced internal bore and therefore reduced flow? Could be an optical illusion but the hose with the push in fittings does look smaller.
Part of my job is to unload a liquid fertiliser ship, the tanker trailers all have different attachments, some are straight pipes, others have 1 or 2 or 3 90 degree bends. The ones that have the 90 degree bends do fill slower than the straight pipe trailers. As soon as you were about to flick that switch i knew what was going to happen but i didnt think it would take that long a delay for both to be running lol
@@jonb3311 its the ships pump we use, so only 1 pump. The ship pumps off into a 6 way manifold. The lorrys were on hoses 3 and 4 (in the middle) so getting the best flow possible. There have been times a lorry with a trailer thats got 90 degree bends on it has turned up 5 minutes before a straight pipe trailer, and the straight pipe trailer leaves 2 minutes before the one that was there first. The more 90 degree bends you got, the more the flow slows.
@@SkillBuilder I just wonder if it would be the same if they were independently piped and pumped, being that the water here wants to go the path of least resistance. Someone with more than my O level physics might know.
@@SkillBuilder we have 4 bar at stopcock in our property - was tested by water supply company a couple months ago, so I assumed it's common (could be wrong of course). Would be nice to see you do the experiment though!
@@AlexGnok you could have 10bar! 3 bar is plenty it’s flow rate after that. Most unvented can prob do 60 litres per that 4 good showers 3 really good but you need the flow rate to keep up. So bigger pipes. Could but 80 bar through a straw. But imagine a shower from a pressure washer
i like the idea of pushfit but the bore is really reduced by the pipe and further by the fittings. I would like to see a similar comparison of copper bent soldered vs brass compressions. Not sure I trust having bent plastic PEX or PB anyway.
@@SkillBuilder Use it in straights but I feel like bending any kind of plastic and then applying constant high pressure to it is asking for trouble in future.
It is not under stress when it is bent. It stays there quite happily and will easily take twice its working pressure and way more heat that a domestic hot water or heating system. I think if you saw the tests you would have more confidence in it. Perhaps we will do those tests and show you how the pipe can be blown up like a balloon and then return to its orginal shape.
Sure, it would, but on these plastic pipes every fitting needs a reinforcement insert which reduces the bore, which is a huge part of the issue. You could build a similar demo with nine couplers in a vertical pipe and see a similar result.
I always use copper pipe with end feed soldered fittings. If there is any doubt about flow rate, in France where I live, you can use a huge variety of different diameters of pipe moving up in 2mm steps. Those pipes even fit inside one another, alleviating the need for reducers too. Brilliant system and never a problem. I won’t touch crappy plastic pipes, crimped fittings etc. etc. because I know they won’t stand the test of time.
As a diyer, I always use pulled bends where possible but I have found the modern copper wall is very thin and even with annealing I have trouble pulling bends using a spring. mind you a set of dodgy knees doesn't help.
@@Monkeh616 Hi. Yes I know about the proper pipe benders with formers, For many years I was a pipe fitter mainly on steel pipe but also did copper. Retired now and been helping out my son moving radiators, so the price of a bender for the amount of time I would have needed one is not worthwhile. The only time it would have been handy was I needed some small offsets in the pipes to the rad, to get it passed a beam, other than that elbows have been ok. Just looking online and I see you can get them from around £30, so if I was doing full time it would be a no brainer.
@@allanthomas-wc8ff I think I paid about 25 quid for one a decade ago, worth it even for a small job just to make life easy (and satisfying), IMO at least. Make your son buy it! He'll use it again.
As a British consumer, this is just one of the many challenges we face. Tradesmen whom have little or no understanding of why or what they are doing, it’s an artefact of our appalling education system for trades. It shouldn’t be a shock that plumbers whom are force taught Shakespeare or the names of the wives of Henry VIII don’t understand the basics of fluid dynamics.
@@SkillBuilder I don't doubt the effectiveness of bent pipe, but in this scenario the water will take the path of least resistance impacting the flow of the other even more, so it's not really a true demonstration. A twin outlet pump would have been better, but it gets the point across.
@@ninjainabowlerhatyeh, I agree with that. It's a double whammy because what you gain in one, you lose in the other. A flow rate over time would give an accurate measurement.
