Ernest Smith

Topher Belknap wrote:

What about thermosiphoning?

90%RH air at 70°F most certainly WILL condense (both in the house and in the rock storage at 65°F! . . .

No . . .

65 F and 90% RH to start with, and no appreciable temperature drop. And that is just a fictitious way-past-worst-case scenario that in reality would be fixed with dehumidification and only lasts a matter of hours even without.

Typical indoor humidity is in the range of 50%. Controlling that is a matter of using related systems, as I prefer to not ask too much of any given system - heating is heating; cooling is cooling; humidity control is humidity control. They need to work together, but asking one system to do too many things can cause problems to come out of the woodwork. Just my own personal preference.

Thermosiphoning risks runaway heating. I think you have some other type of system in mind. I've often wondered why what I am pursuing has not been more widely used.

It would appear the answer lies in chronic oversimplification: some convenient change works "better" and won't change anything. Except it does change things, and causes an intractable problem. The kind of pebble bed storage I am talking about is something that has to be taken seriously in its finer details or it won't work correctly. All of the failures I have come across have been due to things that fundamentally were design flaws which could have been avoided with adequate control: trying to charge with cold air, uncontrolled convection, simultaneous charging/discharging, allowing attempted operation outside of acceptable parameters, mice (!), . . . it's a long list. There are a lot of things about these systems that can be screwed up. Like I said, they need to be taken seriously or not attempted.

I'll bow out now and descend into the "nerd cave" to do my studies and calculations.

As an aside, there are ways do do concentrated solar that works even in overcast weather. Put a vacuum tube with a head pipe at the focus of a parabolic, or even circular trough. Coating of the reflective surface can be even certain white coatings (titanium-based pigment is good), and the "focal" part is not that important in diffuse lighting. It requires a custom manifold for the tubes, but allows one tube per foot or something, instead of the usual couple of inches between tubes (saves lots of cost). Requires quite a bit of advanced design though, and you have to watch sunny weather performance - too much of a good thing.

A related project is a sunny day repurposing of an old TV dish antenna. Metal pot at the focus. It will char wastes. Charcoal is stable on the surface (doesn't rot), and provides huge ion exchange capacity (sort of - it's a biologically mediated ion exchange) for soils. In other words, it is effective carbon storage and boosts growing potential of the soil. But you do have to be a little careful of the raw materials, for metals and things.

Topher Belknap wrote:

Control logic doesn't DO anything. What are the OUTPUTS and what do they do.

In normal service, I don't see how condensation would happen. There is no source of moisture but the air, and there is pretty much no way to get below about 65 F.

Why do you think 65°F is some magical condensation prevention? . . .

Control equipment runs the fans, etc. If it is programmed to not turn on a fan under a certain circumstance, it won't.

For example, let's say air in the heat collector is cooler than the top of the bed. Charging is not allowed. The fan will not run until the collector warms up more. End of story.

The 65 F thing . . .

Right now my room air is 70 F. It is cooler on the floor, possibly 65 F early in the morning. Let's say I've been cooking a lot and humidity is 90%. That air goes in to the bottom of the rock bed, 65 F and 90% humidity. Then it works its way up and is warmed, its relative humidity drops, and it gets blown back into the room and its humidity comes back up. And so on. How is that air supposed to condense anything? That's the lowest temperature it ever gets to, 65 F or something. If it won't condense in the room, how is it supposed to condense in a rock pile at the same temperature?

Recharging would be the few cubic yards of air that was in the bed recirculated through a collector the opposite direction. There would be a tiny bit of heat loss from the lag (if there was one) between depletion of the bed early in the morning and the start of recharge in daylight, at most a degree or two. Vacuum collectors work in overcast, cold weather, That only adds heat, not moisture. The worst state the air is in then is whatever the return air was when heating stopped less heat lost in a (possible) delay before recharge. There would be a tiny pulse of cold air during startup, the stuff that was in the collector overnight.

Actually, control equipment can be programmed to keep heating from rocks until the water radiators are running. There does not have to be a lag unless the rock bed gets depleted.

The "magic" is that nothing happens to the air to increase its moisture content. The air moves through a pile of warm rocks in a great big heater every so often. As I type this that same room air is moving through a warm electric heater. No condensation there either. Magic.

Topher Belknap wrote:
Which brings me back to my original point. Why an air-to-rock system instead of an easier water-to-rock system? Especially when you already have a water system. Why add in another system using air?

P.P.S. The daytime system will be totally different - water recirculates between heat collectors and radiators inside. The thing about pebble bed heat storage is you cannot charge it and discharge it at the same time.

I have seen designs that did. Fortunately, I was able to talk those people out of the idea of air-to-rock storage altogether. You certainly can heat your house, and store surplus in rock at the same time, which would seem to me to be sufficient. . . .

I keep rereading this post, trying to figure out what mechanism you are thinking of for condensation.

It seems like you are envisioning some other kind of rock heat storage. Every time I have encountered it on a discussion forum, there are complaints of mold, and usually also talk of charging and discharging at the same time. But in the technical literature it is pretty clear there can be no simultaneous charging and discharging in what I'm talking about.

Here is a physical description of it:

There is an insulated concrete cylinder, something like 10 ft tall and 6 ft wide. It's mouse-proof, vapor-proof, and filled with round rocks 1" diameter or something. Charging it with heat involves blowing hot air in the top and out the bottom. Discharging is the reverse: in the bottom, out the top. The purpose of that is to keep the air from it at a constant temperature. The lower temperature of return air works its way up from bottom to top. Outgoing air suddenly drops in temperature when the bed is depleted.

The things you've mentioned, cold air, outside air, overheating a house, heating and storing at the same time, controls not being able to prohibit operation . . . I don't know what kind of a system you are thinking of, but it's not at all what I'm talking about.

Topher Belknap wrote:
You mentioned this before; what I am wondering is how you get from this 'not allowed' condition back into an allowed condition. Once the rocks are cooler than the dewpoint of the air, how do you correct that condition?

Control logic, using temperature sensors as input, possibly others.

If the sensors do not meet criteria programmed into the PLC, interior heating is not allowed. Period. An IF THEN ELSE is not satisfied, so it does the ELSE.

Actually, I don't know how the thing ever would get below 65 F or so, and the dampest, coldest air it will ever see is room air at that temperature on a rainy day. There might possibly need to be a dehumidification for interior space (for comfort), but that would be a separate system.

But just for the sake of discussion, lets say the rocks freeze somehow. Maybe I'm gone for a week and the neighbor broke my collector tubes, then sprayed a hose into the rock bed. When I replace the tubes, there would be a recovery procedure, part of that ELSE above.

Air from the bed is recirculated through the collector until the bed is sufficiently warm. Maybe that is 150 F or something to kill mold. Maybe that will require humidity or some equivalent other sensor. Maybe that would require some sort of moisture removal. But heat the bed until it meets recovery criteria. Then the bed would be placed back in service.

In normal service, I don't see how condensation would happen. There is no source of moisture but the air, and there is pretty much no way to get below about 65 F.

Topher Belknap wrote:
How are you going to prevent condensation? I don't see what mechanism your proposed controller has to adjust the condensing point of the storage. I don't see any method for removing moisture from the air. I do know a number of people with blocked off rock heat storage systems, that developed mold. Near the end of summer, you will need to transition from not storing heat, to storing heat. The storage will be cooler than room temperature (otherwise it would baking the house in summer). If your summer/fall are similar to mine, you will be in a high humidity situation outside. What happens next?

The short answer is that operation where condensation forms will not be allowed. It will shut down, self-protectively, and some other system will be used.

In use the bed won't go below the lower 60s F, which would require very high humidity for condensation. If incoming air becomes unmanageably cool, it will use a self-protective shut-down and heating will transition to another system. This will all be handled automatically by PLCs programmed with the logic required to do that. I haven't found plug-and-play control equipment that is even remotely able to handle that. This will have to be an industrial process control type show. Pretty much all of the failures I have come across stem from lack of adequate control logic - preventable problems that came to be by oversimplification.

Heat loss during shutdown will be pretty slow due to insulation of the vessel. It will recirculate indoor air - outdoor air will be excluded.

Some condensation can be tolerated for short periods under some circumstances. It should be able to generate a killing heat, with discharge to waste, during transitional periods like spring and fall. Overheating interior spaces is super easy to avoid. If it has to operate it can dump the heat to waste by various means.

I'll have to instrument and log data from the water system to establish ground truth on a realistic energy flux from collectors. Then I'll do the detailed computer simulations to evaluate probable designs. Then I'll build it and see. If at any point the obstacles to making this work become less desirable than some other system, then something else will move ahead of it in the priority list.

. . .

P.S. I should add a little about the air flow.

Thermal charging (daytime) will be a closed loop. Air will be drawn from the bed bottom, run through heat collectors where it is heated, then put back into the bed top-down. Hot air temperature might need to be controlled by fan speed (slower movement means hotter air). So, air comes out the bottom, through heating, back in the top. At no time will it condense water.

Discharge at night will be a more or less closed loop too, with indoor air instead of running through the heat collector. Warm air goes out the top of the bed to the interior space. Interior air (mid to upper 60s F) comes in the bottom to replace it.

At no time during normal use does the bottom of the bed cool below room temperature. It is either recharging with heated air during the day or taking in room temperature air at night. There will be short lag time with no air flow (probably) but it will be too short for significant cooling.

. . .

P.P.S. The daytime system will be totally different - water recirculates between heat collectors and radiators inside. The thing about pebble bed heat storage is you cannot charge it and discharge it at the same time.

Topher Belknap wrote: Next question, what are you doing to get rid of that condition? With the fans off, the collectors are going to get hotter, but they are never going to reduce the dew point of that air. 80°F at 40% relative humidity becomes 120°F at 12% RH, but still the water will condense at 53°F. . . .

I'm hoping to avoid bed temperatures too far into the 50s F. If I can't make that work it will be time to start looking at other materials.

The idea is to keep indoor air temperatures close to 70 F. The thing should be sized to keep temperatures somewhere around mid-to upper-60s F all through the night, so return air will not cause bed cooling below that. Then when recharge starts in the morning, it will only be gaining heat.

The vessel containing this probably will be an insulated concrete. I've made concrete that floats in water before. It's amazing what can be done with it. Anyway, if interior space gets cool enough to cause condensation problems, it's time to switch that off and run a different heating system.

I'm getting to the "rubber meets the road" point with this packed bed idea. I like it conceptually, but it needs a brutally honest evaluation.

Topher Belknap wrote:

The condensation is a central issue. Inadequacies in control can cause condensation, and I'm not aware of any commercially available control equipment that can do it. What I am doing is designing PLC-controlled custom one-off equipment where condensation will not be allowed.

Could you be a little more specific on what you mean by 'not allowed'? If your storage rocks are say 50°F, your air from the solar is 80°F at 40% RH, what is the controller doing to prevent condensation? . . .

It's not really a big deal to program a PLC to only turn fans on when condensation is not a problem. Temperature (and possibly other) sensors are used as input, and the logic can be programmed to suit. IF THEN, AND, stuff like that.

So, for example, warm air charging might require that incoming air is at least 80 F and above the bed temperature, or the fans shut off after some time delay (can't have a lot of on/off switching during morning startup). Discharging might stop when the top of the bed (last to cool) gets below a certain temperature. The coolest the bed would get at the bottom would be whatever temperature the cool air from the return ducting is, should be over 60 F.

What I usually do for heat transfer problems is write up some time-stepping C++ code and simulate it every second or something. I've not done that yet. I also need to go through a winter with some data logging on my recirculating water system. It is sort of disturbing how some of the physical properties are not well established. Evidently no one really had cause to give these things much consideration before. I'm expecting it will be a fairly iterative process, might take a couple of winters for prototyping.

. . .

Off topic - is there a way to increase magnification on the reply box? I can hardly see what I'm typing . . .

Topher Belknap wrote:You are going to need tens of thousands of BTUs to heat the building over the night. . . . .

Heat collection is available. It works out to two systems: one for day, and a second to charge the overnight system. Heat collection for the overnight system needs to be 3 or 4 times the day system, if memory serves (been a while since I've gone through the calculations). Getting through the night in cloudy weather in January is the controlling design problem.

The condensation is a central issue. Inadequacies in control can cause condensation, and I'm not aware of any commercially available control equipment that can do it. What I am doing is designing PLC-controlled custom one-off equipment where condensation will not be allowed. It's another reason for vacuum tubes too. The collector I'm using now will build up to 200 C in direct sunlight easily with air in the manifold. I'll need to do some testing to establish the air temperatures I can actually get, but I don't expect they will be higher than that. My mention of 500 C is just because that's what some industrial systems do. They use concentrated solar to reach those temperatures.

But yes, bad air quality would be a deal breaker. It's something I've not done elaborate design calculations on, but it could potentially cause a shift to a different storage material.

Water certainly holds a lot of heat, but not a stable temperature. Paraffin is theoretically nice that way; phase change is absolutely constant temperature. I don't like the fire hazard though.

Thanks.

The bed sizing is controlled by the temperature it can be raised to. Some industrial applications can use 500 C, but that's probably an upper limit for rock integrity (thermal cycling) and starts to get into failsafe problems. If you lose control of heating circulation, you don't want to have passive convection run away with a 500 C bed of rocks.

The storage vessel can be made for horizontal air flow, but vertical is ideal. Hot air charging goes down, discharging goes up. The temperature of discharged air remains pretty constant as the cool "front" slowly advances up through the bed.

The usual problems encountered in basic residential systems are related to temperature mismatches - charging with air that is too cool, allowing the bed to condense water, etc. Not too hard to avoid those with basic electronics. I'm instrumenting my stuff for temperature and will be using PLCs and relays to control fans.

I'm in the process of a solar thermal heating system, using vacuum tubes mostly.

The daytime system is pretty straightforward - run water through the collector to heat it, then through a radiator inside.

The more interesting part is the overnight system. It will collect and store heat during the day, for extraction and use at night. The basic idea is to have an insulated container of rocks that you blow hot air down into during the day. Air temps of over 80 F are best - not a problem with vacuum tubes, even in freezing weather. Air flow is reversed at night. Hot air comes out the top while (cooled) indoor air comes back in the bottom.

There have been lots of prototypes built and tested for large-scale use. It turns out that those prototypes are about right for residential use, usually around 5 cubic yards of gravel.

Any experience or thoughts about that?