Having Fun With Some DIY Plumbing

[marvoni]

We're in the middle of a drought with 'day zero' estimated around May next year so I've decided to go off grid, or partially off grid in an effort to ensure I still have water at home.

I decided to use my borehole water for most of the house and just use council water for human consumption such as drinking and cooking. This has meant I've had to split the plumbing into two circuits, borehole water now supplies the 5 bathrooms and council water supplies just the cold water on the kitchen sink, the coffee machine and the soda machine which are both permanently plumbed in. There's also an option to revert back to council water for the entire house as well.

It's meant some serious DIY plumbing, although probably not plumbing as you know it, over the last month or two.

 

[marvoni]

The pressure of the borehole water system is regulated to around 3 bar with a PRV before it reaches the house. Flow rate varies, I'd guess when someone is using a shower or filling a sink it might be 15 l/min, if a toilet is flushed it might be 10l/min etc. I'd gusee max flow rate would be 25-30l/min but only for short periods.

 

[marvoni]

The pressure of the borehole water system is regulated to around 3 bar with a PRV before it reaches the house. Flow rate varies, I'd guess when someone is using a shower or filling a sink it might be 15 l/min, if a toilet is flushed it might be 10l/min etc. I'd guess max flow rate would be 25-30l/min but only for short periods.

 

[ShaunCorbs]

how are you at fab work 😀

you could try one of these and a piece of 50mm tube as a res / somewhere for the air to rise

27735 - Flexvent - Flexvent automatic air vents - Venting Range and Air/Dirt Separators - System Components for Heating, Cooling and Potable Water Installations - Products

or you would have to email these to check the flow rates are ok

Spirotech | maximising performance

 

[marvoni]

Thanks for the ideas, I do have a small fabrication workshop where we manufacture stainless steel enclosures so I can make something moderately complex if necessary. I'm not sure about fine machining of valve seats etc, my lathe is pretty large so that might be a problem.

I might be able to manufacture an electrical version with a solenoid valve operated by a small electrical float switch rather than a mechanical version.

 

[ShaunCorbs]

think you will be hard to make an auto air vent, but you could make a stainless steel tube with a 3/4 tapping one end and the other end to suite your system

 

[marvoni]

Do you think something like this might work?

 

[ShaunCorbs]

yes but would go with 3/4 connections and some baffles internally

 

[marvoni]

3/4" connections are no problem, I've got tank adaptors in stock even up to 4".

How would you suggest might be an effective internal baffle arrangement?

 

[marvoni]

Would it be reasonable to assume that any air travelling inside the water pipe would be mostly at the top above the water or do you think it would be well mixed in with the water?

 

[marvoni]

Something like this maybe?

 

[ShaunCorbs]

depends how much air if theres a bit that should work and great on the baffles

need to make the cylinder as big as possible

 

[ShaunCorbs]

its one of them the bigger it is and dia the better it will work and easier to make

 

[marvoni]

Okay, I had a scratch around the workshop and the largest size pipe I have is 67mm OD in classI copper and I also have a couple of copper end caps which is handy. I'm trying to avoid purchasing pipe because it means I'd have to buy 6 meters which probably won't be cheap so I'm keen to give it a go with this off-cut I have.

The size of the water pipe going to the house is 22mm copper meaning the tube I have is approx 10x larger in area so, unless I'm missing something obvious, the velocity of the water passing through the separator will be 1/10th of what it is in the 22mm pipe. Hopefully this will be sufficient slowing of the water velocity, along with a couple of internal baffles, for the air to separate....

 

[ShaunCorbs]

Worth a go / try, only thing you will waste is your time if it doesn't so worth a punt

 

[marvoni]

Made a start on it this afternoon but ran out of time.









There were a couple of design changes. The internal baffle plates were a bit over complicated for the tools I have at home so I extended the inlet and outlet pipes just past halfway into the main body and cut them at an angle. I'm hoping this will achieve a similar effect to the baffles. Also I was struggling for fittings so I welded the pipes and the end cap which from a design angle should be fine although I realise my welding won't win any awards. I'll spend some time cleaning them up before I install it.

 

[ShaunCorbs]

looks fine marv

 

[marvoni]

Thanks Shaun, hopefully I'll get it finished this week if I can find time.

 

[neilrj]

You have too much time, tools and talent...

 

[oz-plumber]

I'm glad to see you South Africans use silver solder - much like us Aussies.

 

[marvoni]

Lol, thanks for the comments. I used a copper welding rod, I'm too much of a cheapskate to use silver solder unless it's dissimilar metals where it's the only option.😀

The blank end cap that went on last I just used normal plumbing solder, that way I hedge my bets in case I need to remove it again (see photo in next post).

 

[marvoni]

I finished the assembly and installed it this morning. Here's the finished product (which I've been officially calling a prototype so I've got an excuse for the rough looking construction)....

 

[ShaunCorbs]

Have you tested it yet ?

 

[marvoni]

I tested it at various flow rates and it works very well (I'
d hazard nearly 100% efficient) up to around 12-13 litres per minute. At flow rates up to that point it removes all the air and ejects literally a drip or two of water each time the solenoid operates.

Between 13 and 20 litres per minute flow it progressively ejects more and more water along with the air. At 20 litres per minute it ejects about a shot-glass worth of water each time it dumps the air.

Above 20 litres per minute there's air getting carried through and at 25 litres per minute there's about a cup full of water ejected each time it dumps air.

I think what's happening is two things, firstly above 20 litres per minute the water velocity through it is too high for the air bubbles in the water to completely float and separate hence some air getting carried through.

Secondly I think the internal turbulance at higher water velocities is causing froth or foam at the top instead of clean water/air separation so when the solenoid opens it's actually dumping froth which accounts for the increasing water coming out of the dump pipe each time as the flow rates go up.

 

[ShaunCorbs]

Sounds like it needs making bigger dia and length also some internal baffles

But it's a proof of concept

 

[marvoni]

Have you tested it yet ?

Sorry, I wasn't trying to build anticipation, I only type with one finger so posts take time. 😀

As it is, the only time the flow rates to the house get high enough to cause the separator problems is when the kitchen sink hot tap is opened full bore to fill the sink. At all other times including when people shower, flush toilets etc the separator works admirably.

We've got 5 bathrooms so there may be a 'perfect storm' type scenario if there's guests staying or if 2 or 3 people happen to shower or flush toilets simultaneously and the separator doesn't cope but TBH I'm not that worried, it will just serve to make everyone in the house mindful that we're over consuming.

At some stage in the distant future when I've got nothing better to do I may just reconstruct it using the same components and general design but a larger diameter pipe. I'd probably go with a 100mm pipe which would give more than double the volume and half the velocity of the 67mm pipe I used.

 

[marvoni]

I already have an idea for Mark II. We dug out some threaded fittings from stock in my fabrication workshop and I purchased one extra end cap which I was short of.

 

[ShaunCorbs]

😀

I would get a longer nipple

 

[marvoni]

The assembly is going to be considerably longer than the first prototype.

We don't have a supplier for longer nipples so the only option would be to add another nipple and socket. After what it cost me for the extra end cap today I don't relish the idea of buying more fittings TBH. 😱

 

[ShaunCorbs]

You going to tig them together?

 

[marvoni]

I'm undecided. I might Tig them all except for the bottom cap so I can open up at a later date should the need for maintenance arise. I might also just use high viscocity /high strength threadlock or Stag paste and hemp and hand screw them all together. Any suggestions?

The PRV is set around 3 bar and the maximum pressure is limited by the capability of the borehole pump which is only capable of 4 bar so I'd imagine either method would be adequate.

 

[ShaunCorbs]

I would tig them together to give you maximum length

But looks good so far esp for a mk2

 

[marvoni]

I'm thinking about Tig welding 1 inch sockets (or maybe inch and a quarter) onto it for the entry and exit pipes. Hopefully larger diameter pipes will lower the water velocity entering in and exiting. I'm thinking that lower velocities will equate to less internal turbulance and better, cleaner, faster separation of the air. I'll also revisit the idea of internal baffles, there's quite a bit more internal space to accomodate them this time around.

I could draw the unit and baffle arrangement up in CAD and do a basic simulated model of the water flow but it might be more fun to set up a pump and flowmeter and do some real-world bench tests with a few different designs. I'll see how much time I can find to play with it....unfortunately I've got real work to do as well so playtime is limited.

 

[Chuck]

I'm undecided. I might Tig them all except for the bottom cap so I can open up at a later date should the need for maintenance arise.

Good idea. If it were my project, I'd be worried about the potential for stagnation and biofilm growth inside the separator. I would want to made sure it was convenient and easy to open fairly regularly for inspection and cleaning/maintenance.

 

[marvoni]

Here's a pic of MkII next to MkI to give a better idea of scale. The threads were screwed to half way which is roughly where they'd be using hemp and paste. The overall length would be around 30-40mm longer if I TIG weld it but I don't think it will have much effect on the efficiency TBH.

 

[ShaunCorbs]

DIdnt realise it was that big 😀

That is a beast