AuthorTopic: Tiny House Chronicles: More Adventures in Plumbing  (Read 1041 times)

Guest

  • Guest
Tiny House Chronicles: More Adventures in Plumbing
« on: June 27, 2016, 07:19:22 AM »


youtube-Logo-4gc2reddit-logoOff the keyboard of Geoffrey Chia



Follow us on Twitter @doomstead666

Friend us on Facebook



 



Published on the Doomstead Diner on June 27, 2016



Tiny_Home   



Discuss this article at the Doomsteading Table inside the Diner



My preferred system for solar thermal water heating with passive gravity/thermosiphoning circulation was explained in these old diagrams: (for explanatory text see previous tiny house article on plumbing).



diagram1  diagram2



That arrangment offered these advantages:





  1. Hot water cylinder (HWC) is located indoors, hence retains heat better





  2. HWC positioned vertically, hence more effective thermolayering (smaller interface between cold water layer at the base and hot water layer above, hot water being extracted from the top)





  3. The "heat pipe core" type evacuated tube is preferable to the hollow type, because if one tube breaks, the entire system can continue to function well





Unfortunately I was completely unable to obtain my preferred type of indirect small hot water cylinder with large bore internal copper coil capable of thermosiphoning. This forced me to change tack.



diagram3My other alternative is the "hollow core" type evacuated tube system with HWC mounted on top of the tube array, the entire system which sits outdoors. This is often promoted as a stand alone system: cold water is poured into the inlet near the top of the cylinder and after a few hours, hot water is drained from the base diagram 3. The tubes fill with water from the cylinder and this water is heated directly by the sun (in contrast to the heat pipe core type where fluid picks up heat within a manifold which houses the tops of heat pipes which contain acetone).



diagram4Convection currents in the hollow evacuated tubes are set up as shown in diagram 4. Obviously the convection currents cease at night.



The stand alone system does not allow for continuous filling of the cylinder, unless a header tank with ball-valve is attached to the vent at the top of the HWC.



That arrangement was not suitable for my purposes, hence using some lateral thinking I am pursuing the following arrangement where cold water fills from the base of the HWC, hot water is extracted from the top and at night there is reliance on thermolayering to deliver further hot water. The vent connects to a vertical pipe of around 3 metres height, which ensures a constant pressure within the HWC of 3 metres water, above which pressure is expelled out of that vent (option to return it to the top of the header tank within the tiny house is shown in diagram 5).



Diagram5Disadvantages of the hollow core system are the converse of the heat pipe system:





  1. Hot water cylinder (HWC) is located outdoors, hence cools down faster (unless extra insulation is added around it).





  2. HWC sits horizontally, hence less effective for thermolayering





  3. If one tube breaks, water will immediately drain out of the entire system (including the header tank)





One other option would be to mount the array on the roof and use a solar activated electric pump to pump the water from header tank up to the HWC, however that adds electronic complexity, hence I am going with the ground based system at this time.



Having visited a friend at his offgrid homestead who currently uses the "hollow core" type evacuated tubes, I was advised this system can cope with overnight temperatures down to minus 20 degrees C.



In the case of the heat pipe core type, if water is passed directly through the manifold, at night this small volume of stagnant water can easily freeze and break the manifold. Hence in the heat pipe core type, in cold climates, it is necessary to use food grade antifreeze (eg propylene gylcol) as the heat exchanging fluid through the manifold, which then circulates through a copper coil in the HWC. My friend previously had electrolysis problems with the heat pipe / manifold type system, presumably because, in his case, water rather than glycol was directly passed through the manifold. Some systems use magnesium anodes to overcome this problem but another way to minimise that risk could be to use pure (undiluted) propylene glycol as the heat exchange liquid in the manifold, which has an electrical conductivity a thousand times less than that of pure water. He has not however had electrolysis problems with the hollow core type system.



Having received the stamp of approval for my latest arrangement after discussions with my plumber, we will try it out once the plumbing has been fitted in the tiny house.



Current design of the header tank is shown in diagram 6 and the plan is to elevate it above the loft floor on a heavy duty support base as in diagram 7.



Diagram6  Diagram7



The plan to thermosiphon water heated by the wood stove through the header tank remains, however I discovered that the "Hobbit" stove I ordered could not incorporate both external air intake system and the backboiler tank together. Hence because the former option was far more important, I gave up the latter. This has actually worked out favourably because I now plan to harvest heat from a copper coil wrapped around the base of the hot flue and because this will be less efficient than the backboiler tank (which sits inside the combustion chamber) there will be little to no risk of the water in the header tank overheating (the main purpose will be to raise the temperature of the water in the header tank from finger numbing coldness, perhaps 10 degrees C to a tepid temperature, perhaps 20 degrees C. The header tank will then serve as a modest thermal mass heat radiator through the night).



 



G. Chia, June 2016



Guest

  • Guest
Tiny House Chronicles: A Plumbing Polemic
« Reply #1 on: July 15, 2016, 06:10:33 AM »


youtube-Logo-4gc2reddit-logoOff the keyboard of Geoffrey Chia



Follow us on Twitter @doomstead666

Friend us on Facebook



 



Published on the Doomstead Diner on July15, 2016



Tiny_Home   



Discuss this article at the Doomsteading Table inside the Diner








The ongoing tiny house chronicles: A venturi exhaust system for the shower stall and solar hot water arrays incorporating a direct HWC



Geoffrey Chia, July 2016



DIAGRAM 1  DIAGRAM 2  DIAGRAM 3  DIAGRAM 4 HeatpipeCoreEvacTubes



DIAGRAM 5 HollowCoreEvacTubes  DIAGRAM 6 HollowCoreEvacTubes  DIAGRAM 7 Thermosiphoning HeatpipeCoreEvacTubes  DIAGRAM 8 Thermosiphoning HollowCoreEvacTubes



VENTURI EXHAUST SYSTEM FOR THE SHOWER STALL:



I could not find any "off the shelf" exhaust fan models specifically designed to be located in a shower stall which operated on 12V or 24V DC.

A simple computer-type fan can certainly do the job, at least in the short term, and there are many of suitable size eg 12cm diameter, which can run on 24V DC diagram 1. However even if rated as "moisture resistant" they may not cope with sauna type humidity and are likely to fail prematurely. Even though they are quite cheap, the idea is to seek durable items and not be forced to keep replacing them. Furthermore heat loss from this open vent will be substantial, especially at night. A removable internal insulated cover can be added, however that will be another finicky add-on.



24V DC blower type ducted fans are easily available but are probably not suitable to be used "in-line" for air extraction. If embedded in the wall, due to the longer profile than a computer fan, it will stick out awkwardly. Even if mounted exteriorly and extracting air via a duct, it will be directly exposed to the sauna type moisture and may also fail prematurely, being made to do a job it was not designed to do. One way to ensure it has no contact with such moist air is to position the blower upstream of the shower exhaust port, to create a venturi effect to suck air out of the shower stall diagram 2. This will admittedly be an experimental system however I received no practical objections to the design from the intellectual resource folks I consulted (Doone W: scientist, engineer, geologist and homesteader, Lara N: architect, designer & builder, and Lara J: mathematics and physics expert). Any failure in execution of this system will of course be the fault of the author. This system will sacrifice some efficiency for better durability and longevity. If the external "inverted U" pipe or duct is well insulated, this configuration will help prevent heat loss because warm air from within the tiny house will tend to sit static in the top bend of the inverted U at night when the external atmosphere is cool (unless there are strong external breezes).

 



ADDITIONAL HOT WATER SYSTEM OPTIONS:



Roof mounted solar tube arrays utilising direct hot water cylinder (HWC) on loft floor



Designing a system in theory is one thing, but in practice we must always modify the design according to whatever components are available to us in the real world. I was offered false hope about obtaining a small indirect copper hot water cylinder by the false advertising of a UK vendor on eBay. Having now found the option of an affordable, small 50 litre simple direct HWC from an Australian vendor (diagram 3), this can be incorporated into the system and placed in the loft to increase the volume of hot water available and reduce concerns about inadequate pressure head for the taps. Unfortunately the roof mounted options require an additional small electric water pump and a small solar PV panel, however these are simple robust devices with good longevity. All these solar hot water system options remain free of dependence on microprocessors.



 



"Heat pipe core" type tubes (manifold on top of array) diagram 4



Caveats of this system: the manifold does NOT use glycol for heat exchange through internal copper coils in the HWC, water is circulated through a direct HWC. Hence it is NOT suitable for locations prone to substantial frost and subzero temperatures. However occasional ground frost should not be a problem as this should not affect the roof mounted system.



Pipes and Circulation:



All pipes are insulated except pipe 6, the overflow pipe from the header tank (which drains externally)





  1. Each morning, water is actively pumped from the rainwater tank at ground level into the header tank (this inlet pipe is not shown in the diagram for simplicity). Water from the header tank passively flows down through wide calibre (DN32) pipe 1, through the low resistance valve, filling the loft HWC to the brim and also filling pipe 5 up to the same level as the water level in the header tank.





  2. Water passively fills the electric pump which hence becomes primed. During the day when there is sunlight striking the PV panel, the pump drives water up pipe 2 and through the heated manifold into pipe 3. Note this manifold circuit will therefore automatically bleed air out by itself during initiation.





  3. Pipe 3 joins pipe 1 but cannot backfill into the header tank due to the presence of the upstream valve in pipe 1. Water in pipe 3 is hence forced into the HWC. Water circulates continuously through this HWC/manifold circuit in the day, progressively heating up the water in the HWC but this flow ceases at night when there is insufficient light to power the solar PV panel.





  4. Pipe 4 supplies the hot water taps. As water is extracted from this pipe it is replaced at the base of the HWC by cooler water from the header tank. At night there will be thermoseparation between the top hot layer of water and bottom cooler layer.





  5. If the sunlight is too intense and the pump is working too fast, causing the loft HWC to overfill and overflow via pipe 5 into the header tank, then the flow rate from the pump must be dialed back with the potentiometer. The expectation is that the potentiometer will be set for the brightest summer day and thereafter be fixed in that setting and not need attending. The loft HWC cannot overpressurise or boil over because such pressurised water will spill over from pipe 5 into the header tank and be replaced by cold water via pipe 1. Water in the header tank does not overheat due to the large volume of water here. as well as the heat being radiated out of the steel walls of this uninsulated matt black header tank.





  6. Over filling of the header tank in the morning is seen through the kitchen window as external spillage via pipe 6





 



"Hollow core" type tubes (small HWC on top of array) diagram 5



Pipes and Circulation:



All pipes are insulated except pipe 6, the overflow pipe from the header tank (which drains externally)





  1. Each morning, water is actively pumped from the rainwater tank at ground level into the header tank (this inlet pipe is not shown in the diagram for simplicity). Water from the header tank passively flows down through pipe 1, through the low resistance valve, filling the 50 litre loft HWC to the brim.





  2. Water passively fills the electric pump which hence becomes primed. During the day when there is sunlight striking the PV panel, the pump drives water up pipe 2 into the rooftop 30 litre HWC, filling it eventually to the outlet of pipe 3.





  3. Pipe 3 joins pipe 1 and although the pressure head in pipe 3 is higher than pipe 1, water cannot backfill into the header tank due to the presence of the valve in pipe 1. Water in pipe 3 is hence forced into the 50 litre loft HWC. Water circulates continuously through this loft/rooftop circuit in the day, progressively heating up the water in both HWCs but this flow ceases at night when there is insufficient light to power the solar PV panel.





  4. Pipe 4 supplies the hot water taps. As water is extracted from this pipe it is replaced at the base of the HWC by cooler water from the header tank via pipe 1. At night there will be thermoseparation in the loft HWC between the top hot layer of water and bottom cooler layer. There is no flow down pipe 3 at night.





  5. There is normally free movement of air to and fro within pipe 5. The HWCs cannot overpressurise or boil over, because overpressurised vapour will exit pipe 5 (and will ultimately vent to the external air via pipe 6). If the sunlight is too intense and the pump is working too fast in the day, causing the rooftop HWC to overfill and overflow via pipe 5 into the header tank, then the flow rate from the pump must be dialed back with the potentiometer. The expectation is that the potentiometer will be set for the brightest, longest summer day and thereafter be fixed in that setting and not need attending. This overflow scenario is very unlikely if pipe 2 is narrow in calibre, thus limiting the inflow rate into the rooftop HWC, and pipe 3 is wide in calibre thus enhancing the outflow rate.





  6. Over filling of the header tank in the morning is seen through the kitchen window as external spillage via pipe 6.





 



NOTES: in this system, the rooftop 30 litre HWC is treated as no different from a simple manifold. Hence the water in the rooftop HWC is "dead water" being generally unavailable for use*. Total available hot water remains 50 litres from the loft HWC only. The system will NOT work properly (for reasons too complicated to get into here) if water is pumped into the rooftop HWC at the level of pipe 3 connection and is drained from the base of the rooftop HWC at the level of pipe 2 connection.



 



*It will be possible however to manually harvest the hot water from the rooftop HWC at night from pipe 7 by opening the manual tap as indicated in the diagram. This tap in pipe 7 must be kept closed in the day for the system to function (otherwise water will merely circulate between pipe 2 and the loft HWC without going up through the solar array)



 



**Alternatively the connection in diagram 6 with an additional valve in pipe 4 should work well without the need for manual input. In this system, there will be an abrupt reduction of flow rate from a hot water tap when the rooftop HWC empties and the hot water then derives from the loft HWC. The disadvantage of this arrangement is that having a extra valve in pipe 4 can further reduce the forward flow rate from the loft HWC compared with diagram 5. To minimise this, it will be necessary for pipes 1 and 4 to be as wide as possible.



 



Thermosiphoning solar tube arrays utilising direct hot water cylinder (HWC) on loft floor



Whether or not the arrangements in diagrams 7 and 8 will work is unclear, because the basic direct HWC is not designed for thermosiphoning. Proper thermosiphoning cylinders have a convex top, in the centre of which is located the hot water outlet. Furthermore the connection ports are larger.



In diagram 8, air at the top of the 30 litre HWC may be an issue, however it should be possible to manually bleed most of this air out using the pressure release port at the top. I intend to give them a try.



 



Potential issue with all these systems: If the level of water in the header tank drops below the top level of the loft HWC, there will no longer be sufficient driving pressure to expel water from the latter. This can easily be resolved by refilling the header tank, a matter of turning on a switch.



 



G. Chia, July 2016



Guest

  • Guest
Tiny House Chronicles: Off Grid Electrics
« Reply #2 on: July 22, 2016, 04:41:21 AM »


youtube-Logo-4gc2reddit-logoOff the keyboard of Geoffrey Chia



Follow us on Twitter @doomstead666

Friend us on Facebook



 



Published on the Doomstead Diner on July 22, 2016



Tiny_Home   



Discuss this article at the Doomsteading Table inside the Diner








Tiny House offgrid electrics: further insights and modifications



ELECTRICAL 20uly2016



As I mentioned previously we must alter our plans if the original items we intended to source turn out to be unavailable or unaffordable to us. My further research showed that although lithium batteries have fallen in price, the models available to me still remain significantly more expensive than lead acid batteries, watt hour for watt hour, even taking into account the greater depth of discharge and longevity of lithium. Furthermore due to the specific electronics required for lithium systems (battery management systems, specific chargers) and the still small market for them, the costs of these additional essential electronic components remain high.



I will delay sourcing the batteries and solar PV panels for as long as possible because the prices seem to be constantly falling.



If lithium remains too expensive by crunch time, I wish to keep open the option of staying with good old, tried and true lead acid batteries, however one of the most essential features of a lead acid system must be a low voltage cutoff device located at the battery bank to prevent excessive discharge (>50%) and hence damage to the batteries.



My original plan was to run most of the tiny house appliances directly on 24V DC which should lose less energy over the transmission distance than a 12V system. There are a number of 24V DC appliances available and most DC fridges can run on both 12V and 24V, however the market is vastly bigger for 12V DC appliances. For example I was able to find 12V DC but not 24V DC models for the ceiling fan and kitchen rangehood (and hence would need to obtain 24V DC to 12V DC converters to run those).



LVcutoffThe nail in the coffin against me using a 24V household system was my complete inability to source a low voltage cutoff device for a 24V lead acid battery system. Nominal "24V" lead acid battery systems may actually deliver around 29V when fully charged, but when half depleted may deliver around 23V and should be automatically disconnected then to protect the batteries.1



The market for 12V DC appliances is massively larger than 24V, because 12V is the standard for the automotive industry and for RVs and boats. Hence it is easy to obtain a low voltage cutoff device for a "12V" lead acid system which will cut off around 11V or 11.5V depending on your preference.



Hence for DIY tiny house electricians using lead acid batteries, it may be best to stick with a 12V DC system, not 24V, and to use extra thick copper cables to minimise voltage losses over distance, especially the cable to the fridge. You can use a pure sine wave inverter intermittently for the few items where DC appliances are unavailable eg washing machine.



DIY builders must not do their own high voltage AC internal household wiring unless they are suicidal. Market pressures these days are forcing people to use AC appliances (even for RVs) and it must be admitted that the efficiency of AC appliances has vastly improved over the years, whether they be fridges or computers or TVs (which all seem to be LED with no CRT or even plasma displays being sold nowadays). Furthermore the market for and hence availability of AC appliances is magnitudes larger than that for DC appliances.



My main previous reasons to avoid 100% AC in the household and use DC as much as possible were:





  1. Everything being completely dependent on one single device, namely the DC to AC inverter, represents a potential "choke point" for total system failure. (The same can be said for the MPPT charger, however that particular item cannot be avoided no matter what system you choose).





  2. Excessive complexity – DC current from the batteries being inverted to AC, then going to individual appliances and being rectified to DC again. Much simpler for the DC current from the battery to directly power DC appliances which minimises potential points of failure and hence enhance reliability and durability.





  3. Efficiency losses (as heat) from inverter and rectifiers. In particular an inverter which is constantly on, even when no appliances are in use, represents a parasitic current drain.





  4. An inverter may be rated as highly efficient eg >90%, however that depends on the load. At optimal load eg a 3kW rated inverter running a 2kW load, it may well be >90% efficient, however at a low load eg running only a 30W laptop computer, it may only be 50% efficient, depending on the efficiency curve.





The new arguments to adopt 100% household AC wiring are:





  1. I understand that AC to DC rectifiers in just about all modern household appliances are extremely reliable. For example, many LED light manufacturers guarantee their AC bulbs (which incorporate rectifiers) for 10 years.





  2. I was informed that modern inverters do not need to be fully "on" constantly. They can automatically go into sleep mode when no appliances are on, with miniscule current consumption, and can be woken instantly when there is a load sensed.





  3. Modern inverters incorporate programmable low voltage cutoff devices. The commonest offgrid lead acid battery arrays are nominally rated "24V" DC and I understand that it is best to build up the battery system using numerous 2V cells rather than just a few 12V high capacity (eg 260Ah) batteries, because the former confer lower internal resistance. If, despite string protection, one of big 12V batteries fails, that entire costly battery will have to be replaced and until then, the whole system will run at much reduced capacity. If however a string of 2V cells fail, they can be removed and the whole system will run at only slightly lower capacity with the inverter reprogrammed to accept the lower 22V DC battery output and also to a lower cutoff voltage eg 21V (rather than cutoff at 23V for a 24V system).





  4. Even if you run only one 24V DC appliance directly from the 24V DC battery system, if it is inadvertently left constantly on (eg shower exhaust fan), that could overdischarge and damage the lead acid batteries due to the lack of an intermediary low voltage cutoff device. This will not happen if 100% of appliances receive their power from an inverter which incorporates the low voltage cutoff protection.





Hence overall, if you are engaging a certified offgrid electrician to do your household wiring it may be better to go with 100% AC wiring in your tiny house. The system my electrician has suggested to me allows flexibility to accept either lithium or lead acid batteries in the future and it may be simpler to keep a spare inverter on the shelf which can be rapidly swapped if the active inverter fails. He advised me that inverters can usually be repaired, hence the faulty one need not be discarded. If you are building several tiny houses to establish a tiny house community, designing standardised setups allows the possibility of creating a microgrid.



CONCLUSION:



If your system is being wired by a professional offgrid electrician keen to offer you the latest and greatest, and you are too weak to resist the seduction of standard AC appliances (like the author), then you may choose a 100% AC house system which is completely dependent on the inverter and can keep a spare inverter handy.



If you are stronger than the author and better able to adhere to the KISS principle and/or are a DIY electrician who is not intent on suicide, you may prefer a 12V DC system which uses as many household 12V appliances as possible with only one or two items being dependent on an AC inverter. You will use extra thick household copper wires and incorporate a low voltage cutoff device at your 12V battery bank.



If you choose to go with lithium batteries in the first instance, it will be useful to ensure your system can also accept lead acid batteries in the future. This is because if/when industrial society crumbles, replacement high capacity lithium batteries, being uncommon, may be difficult or impossible to obtain. However lead acid batteries, being ubiquitous, should still be easily obtainable for a long time to come.



G. Chia, July 2016.



Many thanks to Lachlan O'Shea of Lockstar energy, specialist offgrid electrician



Any errors in this article are the sole responsibility of the author



 



Footnotes:



1. More precise lead acid battery management is more complex because the voltages mentioned refer to an open circuit without load after the system has "rested" for more than 24 hours. A fully charged "24V" system with an open circuit voltage of, say, 29V, when exposed to high load demand can drop its voltage to 23V, which is not necessarily a trigger for cutting off the system. However those details are beyond the scope of this article.





Offline Eddie

  • Global Moderator
  • Master Chef
  • *****
  • Posts: 17439
    • View Profile
Re: Tiny House Chronicles: More Adventures in Plumbing
« Reply #3 on: July 22, 2016, 11:20:11 AM »
The nail in the coffin against me using a 24V household system was my complete inability to source a low voltage cutoff device for a 24V lead acid battery system. Nominal "24V" lead acid battery systems may actually deliver around 29V when fully charged, but when half depleted may deliver around 23V and should be automatically disconnected then to protect the batteries.

I wondered about this, and scratched my head a little, because 24V off-grid systems are fairly common, and it didn't make sense that you couldn't source a low voltage cut-off.

Then a light bulb lit up in my brain (figuratively speaking). Most systems handle this at the INVERTER. In his obsession for a simple bulletproof design, Dr. Chia strayed out of the box that most solar power users live in. Almost everybody uses inverters now.  All but the simplest of inverters turn off themselves when their pre-set voltage limit is hit.
What makes the desert beautiful is that somewhere it hides a well.

 

Related Topics

  Subject / Started by Replies Last post
6 Replies
3443 Views
Last post April 01, 2016, 03:07:04 AM
by Guest
0 Replies
177 Views
Last post May 20, 2018, 08:22:41 PM
by Surly1
10 Replies
517 Views
Last post August 30, 2018, 03:18:20 PM
by RE