Solar/evac tube questions

charlie_16

<span style="font-size:12pt">I have a evacuated tube solar hot water system.  Currently 20 tubes into Bradford-White Ecostor 115 gallon tank/HWH.  This is to grow to 70 tubes with second 80gal tank in series shortly.</span>

<span style="font-size:12pt">Questions:</span>

<span style="font-size:12pt">1.</span>       <span style="font-size:12pt">Pros/cons of higher or lower delta T on controller (Eagle 2).  </span>

<span style="font-size:12pt">2.</span>       <span style="font-size:12pt">My transfer pump could only get system up to 15psi when filling/purging… is this a problem?</span>

<span style="font-size:12pt">3.</span>       <span style="font-size:12pt">Right now my circulator pump is oversized-  it’ll be right when add’l. tubes are added.  If I let it run “wide open” then collector cools to near tank temp fairly quickly and the pump ends up cycling on/off.  I can throttle back by partially closing a shut-off valve…  which is better?  I haven’t sprung for a flow meter yet ($$$)</span>

Kevin_in_Denver_2

Solar Short Cycling Issues

I guess the biggest problem with short cycling the collector pump is that it may shorten the life of the pump. The second biggest problem is that the overall system efficiency drops slightly. It drops because in a system where the collector pump is off half the time means that the collectors are running at a higher average temperature.



A modulating or proportional control would theoretically throttle down the flowrate so that the collecter delta T wouldn't drop to less than say, 8 deg. F. That ensures the lowest possible collector temperature and therfore the highest efficiency.



An article in Solar Today in 1982 proposed an elegant solution to the short cycling problem without resorting to a fancy control scheme. The article proved that the collector sensor shouldn't be measuring the collector fluid outlet temperature anyway. System efficiency would be maximized if one could measure the collector "stagnation temperature".



The stagnation temperature is the collector temperature during zero flow. In order to measure that, they went on to propose that the sensor should be mounted in a small "sample collector" that has no fluid flow.



This idea never caught on, apparently because of added cost. However, the stagnation temperature can be closely approximated by taking the sensor off the pipe and attaching it to the absorber surface. If the small chunk of the absorber at the sensor location can be "decoupled" from the fluid temperature, all the better. For a flat plate collector that would mean clipping it out of the rest of the absorber. For mechanical support, leave two corners intact.



The stagnation temperature of a double wall heat pipe e-tube collector like Apricus could be measured inside a tube where the heat pipe has been removed. That would be the perfect way to control a large array, but on a small system it seems like a waste or energy. Dangling the sensor inside the last tube should do the trick. I'll be checking this out in the next few days.



Sorry for such a long answer, but I was studying this issue today, so I needed to write down my notes anyway.



15 psi should be fine, but you can always boost that with city pressure. If the air is out of the system , then it won't take much water, and therefore won't dilute your fluid very much.



As you probably know, throttling the flow is very hard without a flowmeter.

Karl_Northwind

variable flow

I would throttle the system back with a ball valve to keep a 15-20 deg delta t when the system is running full speed at full sun.  This is a hack way to do it, and will need some tweaking when the seasons change, but will get you thru.  I would (and have) done this on my own system, playing to get the flows right for optimal performance, but only where I closely monitor it. There is a big difference between what you do on your own house and what you do as a contractor for others.  I'm re-doing my system before we sell the house and I pass my jungle on to someone else.



the eagle 2 controller is ok, and an industry standard, but I'd consider either adding a delta T controller to your circ, or using a caleffi VS controller.  it'll save you some $ over the long run, as well as making the system operate more smoothly.  reducing the on delta T is ok when the pipe runs are short and the circulator is low power.  for the cost of your controller re-do you could buy a smaller circulator and use it till you add more tubes, and then re-use the smaller circulator for something else when you swap in the larger circulator. 



I usually like to see more than 15 psi on a system, both to help keep gasses dissolved at higher temps, and prevent boiling (especially in an ET system) it doesn't take much boiing to shut off the fluid flow and produce much more boiling, and problems.  although the circ you have in there will probably force the steam thru thesystem anyway and make it work, although not something you should count on.



the idea of using city water to boost pressure is an interesting one, but I don't think I would go that route.  you have no way of knowing how much water you've added to the system, and if you have used a larger expansion tank than necessary (good practice) you could add a gallon or so of water to a 5 gallon system (ET systems usually use less fluid than flatplate collectors) and seriously compromise your freeze protection.  a good utility charging pump (wayne 1/2 hp utility pump from most hardware stores, sold under a number of brand names) will cost you $130 and will pressurize 50% glycol to 50 PSI. 



consumer grade VS controls have come a long way since the early 80's, and this practice of a spare panel sounds like a primitive pyrometer.  if a glazed box placed in the sun gets warm, there's a good chance that there is enough sun to collect heat.

not a bad idea, but we have better simpler ways of doing that now.



let us know what you do and how it works out.



Karl

charlie_16

thanks

Thanks, Kevin and Karl for both your responses.  (For the record, I am the homeowner, doing this on my own- this is not my day job)  As I understand it, lower collector temp. means lower heat loss since rate is dependent on delta T.  This is why more efficient?  The flip side I think, is that I'm trying to capture some 120+degree water in the top of my storage tank so the electric element doesn't kick in.  Would this happen sooner if I let the collector get hotter?  Higher delta T on controller AND throttle back flow rate?  (The bottom of the tank starts at around 85 degrees and has gone to 105 on the one sunny day since start-up.)

The VS controller means variable speed I assume which means VS pump also.  I have a Taco 0011 which is not variable but the rest of the collector (second location, 50 tubes) should be on line soon (leaves coming on trees will shade 20 tube collector) so hopefully flow rates can be balanced better then.

Is delta T on controller variable to account for heat loss coming back from collector to tank?  monitor temp as return enters the tank to be sure you're hotter than the tank?  Adding the data logging connector to Eagle 2 will put old dusty Pentium 2 computer to good use in the basement.

Again - thanks for your thoughts... this has been a VERY challenging home project- its working, but could be tweaked I'm sure. 

hot_rod

scrub those BTUs

out of the collector quickly. Tight delta T from inlet to outlet to limit loss to the ambient. Some recommend a 3-5 delta T across the collector. A variable speed controller would be a good choice and most have a Otc (option tube collector) function to help with tube arrays.



I like to see more like 20 psi or higher, depends on how tall the building or array is mounted. And a Karl mentioned keeping the boiling point up is part of the pressure result



hr. Systems in Europe tend to run closer to 60- 65 psi, especially evac tubes where summertime stagnation temperatures can get so high.

Fortunat

I'm not sure I agree...

rod,



If pumping power is free, then maintaining a 3-5 degree delta across the collector might make some sense. But in the real world, where parasitic electrical consumption matters, that seems like overkill. Without actually doing the math, it seems hard to believe that any improvement in collector efficiency (especially in tubes) from running the collector 5 degrees cooler wouldn't be more than offset by the increased pumping power.



For fixed flow systems, we tend to shoot for 15-20 degree dT (F) in full sun (which means something a fair bit lower most of the time). In variable speed pumped systems we typically set the target dT at 12 degrees or so.



As for system pressure, I don't understand the argument for higher pressure in evacuated tube systems. Why would I want to elevate the boiling point, unless I'm doing solar air conditioning or some other high temperature load.



Assuming you are running something like a 20 degree delta or less, and assuming that you are heating a water tank of some kind, your system will have turned off on max storage temp (typically hard coded into most controllers at 200 deg to avoid blowing the tank TRV), long before you reach the boiling point in the collectors even at 10 psig collector pressure.



Once the system is turned off and in stagnation, I'd just as soon the heat transfer fluid boil off and get pressed down to the expansion vessel sooner rather than later. It's going to happen eventually (at 100 psia, the boiling point is still just 300 degrees), so my feeling is why not let it happen sooner rather than later and limit the temperature that the glycol is exposed to. This issue is fairly well explored in a series of technical papers by Hausner and Fink from around 1999-2002. (for example: www.aee-intec.at/0uploads/dateien119.pdf)



Rod: I'm interested to know what benefits you think you might see from higher pressures?



~Fortunat

www.revisionenergy.com

Kevin_in_Denver_2

High Temps in the Collector Loop

Required reading on this topic from Siggy attached

Fortunat

expansion tank sizing and steamback

Kevin,



This is a great simplified expansion tank sizing exercise; thanks for posting it!



One additional component to consider, which is especially important in evacuated tube collector systems is the 'steam power' of the system during stagnation. Depending on collector type and plumbing configuration, it is not only the collector volume, but also some fraction of the near collector piping which will be filled with steam during stagnation. This is obviously important not only for expansion tank sizing, but also for considering the location of various system components that might be steam/temperature sensitive (any air removal devices like Spirovents, and expansion tanks).



(see the attachment, especially page 7)



In our own testing we've found that a single Apricus AP30 collector fills about 3-4 feet of 3/4 inch copper line with steam on either side of the collector near the end of phase 3. It eventually settles back to about 2 feet per side. Additional testing shows a more or less linear relationship with collector area. This assumes plumbing with good emptying behavior and moderate system pressures, roughly 1000W/m2, etc.



So, Sigenthaller's basic rule may appear conservative in most cases, but it might not always be. It makes sense to understand the dynamics of the system pretty well if you want to avoid puddles of glycol and service calls after stagnation events (whether they are related to overtemperature or to power failures or whatever).



We did our initial testing with a sight glass so we could observe the steam level as it moved in the system, but you can also calculate it fairly accurately if you watch the system pressure during stagnation (assuming you know the expansion tank volume, precharge etc).



One final thing: note the paper's discussion of collector and plumbing 'emptying behavior'. This is pretty important, as is expansion tank location! in order for the collectors to empty quickly when stagnation starts, the expansion tank must be on the collector side of the check valve so that fluid can leave the collector in both directions. Given the check valve locations on most commercial pump stations, this puts the expansion tank on the pump discharge, which I know excites a lot of hydronic professionals who have been taught to always put the expansion tank on the pump inlet.

It bothered me too at first,. But when I consider not having to get up on a morning after a summer lightning storm/power outage and drive around to 15 customer's houses to recharge their SHW systems, I am willing to ignore the purity of maintaining the 'point of no pressure change' at the pump inlet (especially since system pressure and its location on the tank outlet makes pump cavitation pretty much a non issue).



sorry for the long post....this is an issue I've spent a lot of time thinking about (can you tell?)



With Sunny Regards,



~Fortunat

www.revisionenergy.com

Kevin_in_Denver_2

First Resort, Drainback, last resort, Butler Sun Solution

"But when I consider not having to get up on a morning after a summer

lightning storm/power outage and drive around to 15 customer's houses

to recharge their SHW systems"



Nobody likes that, especially the customers.  That's why I started concentrating on drainback in 1982.  



We're wandering way off topic now, but I enjoy getting these real world results out for discussion. 



As I mentioned, I've been doing some severe thermal shock testing of an AP-22.  I managed to crack a tube after hitting a dry stagnating panel with 80F water.  The collector sensor was reading 375F.  My conclusion is that Apricus is OK for drainback, but your must use a controller that refuses to turn the collector pump on if the collector sensor is over 210F.



But if you must use antifreeze, you should consider  Butler sun solution's   Passive Integrated Over Temperature / Over Pressure System:   http://www.siliconsolar.com/passive-solar-heat-dump-system-p-501442.html



Barry Butler is a gifted inventor, and here is the double solution of handling the thermal expansion of the  phase change, and passively dumping the heat on any system.   No 3 way valves, special controls, running the collector pump at night, etc.  It uses  well-proven automotive technology, the radiator cap.  The price is less than that huge expansion tank, too.