|
Post by alanmh on Mar 19, 2012 7:59:45 GMT -5
Hi New to this forum and Iam thinking about how to design a SHW system. A key consideration for me is overproduction in hot summers if I go away for say 2 weeks. Should I worry about this?
Can a panel produce enough heat to boil the water in my tank? or will it just naturally level off at a safe temperature?
Many thanks
|
|
|
Post by cye on Mar 19, 2012 18:32:16 GMT -5
hi alan, welcome to the forum.
solar hot water panels certainly can boil your tank and you should consider some form of heat dump mechanism to deal with any excess heat.
commercial installers don't always build this into their designs, and many installers will simply oversize the tank to reduce the risk of this happening. oversizing the hot water tank is far from ideal and to be avoided.
Many differential temperature controllers will have a setting whereby the solar pump is switched off when the tank water becomes too hot. Whilst a useful feature as a failsafe, this method should generally not be relied upon because, when the pump has stopped, your panel(s) will boil and that's not good for the panels or plumbing either!
whereabouts in the world do you live? if you live in any area where there are no water shortages, then you can use a very simple heat dump mechanism called the 'run-off' method. when the tank gets too hot, a solenoid valve opens in the domestic hot water pipework to run off some hot water down a drain. equivalent to turning the hot tap on and letting it run down the sink. this allows the pump to continue to run and prevents both the panels and tank from boiling.
if the run-off method is not an option, then a common heat dump solution is, again using a solenoid valve, to divert the hot solar fluid in the solar loop away from the coil in the tank to a radiator of some sort. in this way the pump continues to run & prevents the panels boiling, and the heat is dumped into the air.
yet another variation on this theme is one whereby, when 'dumping temperature' at the tank base is reached, the central heating pump (but not the heating) is activated, thereby the central heating coil in the tank cools the tank. this assumes the heating will not be running when the panels are getting a grilling!
sure, give us some further info on your proposed setup and we can advise further.
best regards
cye
|
|
|
Post by alanmh on Mar 20, 2012 6:37:31 GMT -5
Thanks for the reply Cye I am located in Berkshire in SE England. We have E/W facing roofs, and a south facing gable wall. I am currently weighing up using ither two panels, one on each roof or a vertically mounted panel on the gable wall. None are shaded, so I think its a bit moot and down to when hot water would be needed. I am still undecided on flat panel v tubes, but am edging towards tubes as they seem to have a better winter time performance. I plan to make my own controller and have been considering a drain back system to combat frost and overheating, but am still concerned how this would work in practice given the potential high temperaturs on the panel/ manifold. The reason for the drainback is that I want to introduce the system in two stages, the first being direct heating of the hot water (no antifreeze) and then introduce a secondary coiled mains pressure DMH tank at a later date. I am looking into designing the controller around a jeenode jeelabs.org/tag/jeenode/ as this can also be tied into my HAH www.dbzoo.com/livebox/hah and report to Pachube www.pachube.comUsing this design, a solenoid dump would be feasible. we dont have a water meter and I guess this would be rare in practice. Not sure about the radiator dump as it would heat the house at a time it was already quite warm, I guess I could bury it? Sounds like a major job? I will share my desing when done.
|
|
|
Post by cye on Mar 20, 2012 15:23:55 GMT -5
Hi Alan, all sounds like an interesting project.
Here's some things possibly worth considering when planning your design:
[1] Very few houses with E/W orientations have roofs facing dead east and dead west. Generally, one roof will face slightly North. If it were me, if angled mounting brackets on the South facing gable are not practicable, I would double up the panels on the roof facing slightly south, rather than bother wasting panels on the roof facing slightly north. You'll collect 80% of the energy on an east or west facing roof compared with a south facing, but this starts to degrade substantially once there is any northerly element.
In my case, having main roofs facing not very south, I used our garage roof. Easy to get access to too.
[2] Drainbacks are nice for frost protection but are rare here in UK/Ireland (though there's one a hundred yards from me, an american system). Not having a closed plumbing loop, drainbacks need a very high head pump, whereas in a closed loop system, the water being pushed upwards is balanced somewhat by the weight of water being pushed down the other leg. Therefore, with closed loop systems, the pump only needs to overcome the resistance of flow caused by the bore of the pipe, rather than to have to contend with gravity too. Simpler, less expensive, and more energy efficient pumps can therefore be used with closed loop. The closed loop pump can be a common jogger rather than a fell runner!
[3] Your plan to implement in two stages is nice. Small steps have a better chance of succeeding and you get an early gain for little pain. So you will be using direct flow initially, where the water in your tank is actually circulated in the panels. A common feature of a differential temperature pump controller is 'freeze protection', whereby the pump operates intermittently when the outside temp falls close to freezing. This prevents freezing without using antifreeze. Such a feature is acceptable when used in conjunction with both evacuated tubes and well insulated pipework, as the vac tube design will radiate much less heat than a flat panel. One of our forum members. Colin Lloyd, has been using this method for a few years now.
Alternatively, you could implement an external heat exchanger very easily without changing your tank, and use antifreeze!
Personally I do not like mains pressure tanks for a variety of reasons, but mine is not a common view and I'll not bother you with my luddite thinking on this!
For domestic hot water I personally favour vac tubes. Principally because the performance is better than flats in conditions that are not optimal, such as cloudy weather. About 1/8 of the radiation energy available in peak summer is available in dead winter, so I wouldn't base your panel choice on winter hot water. We just don't have the weather for it unless one greatly oversizes the panels. A well and economically designed vac tube system will meet 100% of an average family's hot water needs from March through to September/October. Outside of that it's boiler, stove, etc.
Now I am playing with an arduino at present, a close cousin of the jeenode. If you could be persuaded to switch to an arduino, it would help me to have someone to collaborate with (learn from!). And also, all the code for the DTC is already available for the arduino, even for the DS18B20 one wire sensors!
Keep in touch and please do keep us posted on your progress.
|
|
|
Post by alanmh on Mar 23, 2012 7:54:42 GMT -5
Thanks again Cye, and your comments raise more questions.
1. Looking on Google maps, you are right, the west facing roof is very slightly more south facing (about 5.8 degrees). Unfortunately our property goes right up to the boundary on the south side, so angling brackets on the south facing gable wall are not possible. Which begs the question, what is the trade off / performance calculation between the roof at 264 degrees to a vertical south facing wall? Looking at this online calculator, which is really for PV panels, it suggests that putting them on the roof gives 1770KWhr compared to on the gable which would generate 1530KWhrs. That is a 13% reduction. Another chart I have from jhspowersolutins suggests that the difference is 15%, so I guess that is about right.
But surely there is also the consideration of time of day? The gable wall gets direct sunlight for 12 hours a day from the spring equinox to the autumn equinox (give or take a few minutes for the 6 degree offset). and in winter and the hole time the sun is up all winter. (info from http://sunposition.info). However the slightly south of west facing roof will only get direct sunlight in the afternoons. My concern here is when hot water is needed. Ie in the mornings for showers, evening baths and so on are not a problem as the water will have had plenty of time to heat through the day. But I am wondering if in the summer months, I can capitalise on the early morning sun. I guess the answer is no, as the gable will not get direct sun until after 08:30 when we have already left for work and putting a panel on the (just north of) east wall would be less efficient?
So in conclusion, I guess it would be better to try and store more water from the west panel on the roof. Do you agree?
2. I take you point about the pump, I hadnt thought of that .
3. Freeze protection - as you say the bit to heat in the winter will be small. This will also give me an incentive to change the tank for one with a second coil. I hadnt thought of using a secondary coil, but if I am going to that much trouble, I may as well change the DHW tank.
My reason for wanting mains pressure DHW, is that we currently use a pump that uses about 600Watts! and my wife can be in the shower for 20 minutes! But maybe that arguement should have a different thread ;-)
I also have an arduino board and could you point me to the code?
Many thanks
Alan
|
|
|
Post by cye on Mar 23, 2012 13:19:07 GMT -5
Hi Alan
You've certainly done your research! West roof sounds like the simplest solution, but,.... me sitting on the fence.... given your property is nearly EW, perhaps there is some merit in one panel on each roof. you could use a sensor in each panel and switch the pump using the higher of the two panel temps, plumbing both panels on the same solar loop. Being vac tubes with heat pipes ( which do not reverse thermosyphon), there would not be much radiation loss from the cooler panel.
Have the directions you have been running through your modelling been based on compass (magnetic) readings or true (google map)? 6-8degrees difference I think.
Tank sizing at 50L per person per day is the rule of thumb, and I find this delivers more than one person really needs. Another thing to consider is that you can run a small (original) tank hotter, e.g., 80c, thereby making it a bigger energy store, though note that panels collect energy less efficiently at higher temperatures temperatures.
Raising your cold water tank in the loft, even by 9" makes a hell of a difference to shower pressure and could dispense with the pump. Worth also checking whether the shower head is designed for vented or unvented tanks. High pressure showers are one of the reasons I do not favour unvented tanks! They increase HW usage unnecessarily.
I look for the arduino code links later for you.
best regards
cye
|
|
|
Post by cye on Apr 2, 2012 16:24:09 GMT -5
|
|