Click here to Find a Contractor in your area.
In fairness to all, we don't discuss pricing on the Wall. Thanks for your cooperation.
Need to contact us? Visit

Drain back with evacuated tubes

michaelmichael Posts: 302
I thought I read some time back here on the wall someone was using evacuated tubes for Drain back. Anyone?

I know the cons of doing that?

What positives?

all comments




  • Kevin_in_DenverKevin_in_Denver Posts: 588
    edited February 2010
    e-tubes w/ drainback

    First off, only  "heat pipe" style e-tubes can be used.

    One issue but not a deal-killer:

    1.  Sunny day power failure.  When the power comes back on, the dry stagnating panels will get hit with water that will flash to steam.  Some water hammer will be heard, but SunMaxx, Apricus, and Sunda survive this event without problems.  The header is brazed copper and very robust, the thermal shock of going from 420F to 70F in seconds is OK.  The compression fittings at each end also  survive.  Customers may be concerned about the noise, so some controllers can be set up to prevent pump activation whenever the collector sensor is above 180F.  That setting will throw away  heat under certain conditions.

    The manufacturers are mostly silent on this application, so I'm doing some of my own testing.  I've talked to an e-tube installer in NM who said "no reason to ever do anything but drainback" and he has a dozen or so systems installed.  It's worth looking at because it's cheaper, more efficient, and solves overheating issues without more equipment.

    As you may know, most of the e-tube issues from 10+ years ago have gone away.  Reliability, hail resistance, and cost are now acceptable.  Snow is still an issue, with a different answer for different applications and local weather.

    Here's an older thread:
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • FortunatFortunat Posts: 103
    consider header geometry

    If you are going to try to do drainback with a heat pipe type evacuated tube collector (Apricus, Sunmaxx, thermomax, etc), consider carefully the local geometry of the header and how well it will drain. Unlike the smooth headers in a harp style flat plate collector, most the ET headers have local high and low spots that don't necessarily drain as you might hope.

    We have had some experience with collectors installed in a drain back where the conventional 1/4 inch per ft or more manifold slope was not sufficient to prevent local pooling of water in the header and subsequent system failure due to freeze. Antifreezing the system would have solved the issue, but this was a direct system with several hundred gallons of storage so the cost would be prohibitive even to get to 20%.

    Before you try it, get your hands on a manifold, fill it with water, tilt it and watch it drain.

    Good luck,

  • Kevin_in_DenverKevin_in_Denver Posts: 588
    edited February 2010
    other e-tube drainback advantages

    Good points, Fortunat.  The Viessmann/Thermomax header won't work.

    The others I mentioned are a 3/4" (22mm) tube with 1/2" vertical internal cross-pipes that block some water from leaving, but not enough to cause a problem.

    If you only have one panel, you probably don't need to tilt it, because it's draining from both ends.  That keeps things looking better.

    Another advantage is in the typical attic piping.  A 4x10 flat plate collector can be difficult to plumb because the lower header winds up way out by the eave.  An e-tube would have both pipe penetrations much closer to the peak of the roof, making access much easier.  It also easier to achieve the proper slope on the supply and return from that point over to the pipe chase.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • michaelmichael Posts: 302
    Drain back with evacuated tubes


    How about the stagnation heat on the insulation in the collector header?

    The other problem I have heard is when there is stagnation the tubes tend to lose their vacuum?

    These are things I heard.

    Thanks for the tips. Does anyone on the wall actually have one installed on their home?

  • Kevin_in_DenverKevin_in_Denver Posts: 588
    edited February 2010
    header insulation

    The SRCC test includes dry stagnation, and this will weed out problems.  At COSEIA last week, Tim Delaney showed us a cutaway of the Sunda header.    The header pipe is supported by a ceramic spacer, then glass wool, then foam.  That collector had overheated, and some of the foam was brownish, but only where the ceramic spacer directly touched the foam.  So yes, there could be trouble in a design where the copper header touches foam.

    That reminds me of an instance in 1982 where a flat plate collector actually caught fire.

    Prior to the Apricus design, which is a double glass wall, most e-tubes had a glass to metal seal.  Daily dry stagnation over a long period would cause that seal to fail due to slight differences in thermal expansion.  Sunda thinks they've solved this with a more robust seal that stays much cooler.  Apricus and Sunmaxx don't have any glass bonded to metal.  Sunda is sticking with their design because a single glass wall lets 6-10% more energy inside the tube than a double wall.  The double wall tube is also much cheaper to replace after glass breakage.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • michaelmichael Posts: 302

    That's great info.

    Much thanks
  • Correction: Apricus Header Draining

    Upon further inspection, the Apricus header needs to be sloped in order to drain properly.   Here's a drawing of it, it isn't just a straight tube with cross pipes like Sunda and SunMaxx.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • More e-tube drainback testing

    I have one Apricus AP-22 collector that I'm putting through its paces.

    After a half hour of dry stagnation under full sun, when I turn on the pump I don't get any boiling or water hammer with 100F inlet water. This is a pressurized drainback system at 60 psi, but I was surprised. I'll try it at hotter inlet temps and with 2 collectors.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • FortunatFortunat Posts: 103
    Apricus in drainback


    That cutaway is a great illustration of what I'm talking about. Our experience with these collectors in drainback here in the northeast is not good.

    It takes a lot of collector pitch to ensure that the last little pocket of water drains out of the manifold sufficiently. 1/4 in per ft won't do it (i have a couple of destroyed manifolds to prove it).

    Apricus still carries drainback as an option in their literature, so it must be doable, and it sounds like you are doing a bunch of research so I'm sure you'll have it covered...

    by the way, from another thread...what 'cracked' when you shocked the system thermally? a heat pipe? or part of the header?

  • Kevin_in_DenverKevin_in_Denver Posts: 588
    edited April 2010
    Apricus Thermal Shock

    Test Conditions:

    Full sun.

    Solar Panel is an Apricus evacuated tube AP-22

    The near-panel piping is corrugated Stainless Steel, the

    rest of the solar supply & return is PEX.  (We are also testing the limits of  PEX in a pressurized drainback collector loop.)

    After 2 hrs of stagnation, the copper header measured 375F,

    but the actual temperature of the absorber could be much higher.

    Bottom of tank temperature was 80F

    The pump was turned on to test for system boiling and thermal

    shock.  Once the pumping started after

    about a minute, the boiling sounds were minor.

    The maximum return temperature measured was 160F.

    There was no spike in the collector loop pressure.

    Flowrate: 1.5 gpm

    After the test, I noticed that tube #3 was shattered just

    below the midpoint.

    Even though the cold collector fluid doesn’t circulate

    directly through the tube, the heat pipe can conduct so much heat away from the

    absorber that is causes localized cooling that stresses the glass of the

    evacuated tube, which may break.  The Apricus design apparently puts the heat pipe too close to the glass absorber.   SunMaxx had an older design

    like that , but their current design has the heat pipe on axis.   Stay tuned.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • hot rodhot rod Posts: 6,711
    stagnation temperatures

    here is a formula Siggy worked up to show potential stagnation temperature in a typical flat plat collector. Plug in the Y-intercept and slope from the collector you are using to get an idea of what you may be dealing with.

    I have heard of tube cracking in the sunny, hot Arizona conditions also. Lots of temperature swing at work there.

    Bob "hot rod" Rohr
    trainer for Caleffi NA
    The magic is in hydronics, and hydronics is in me
  • FortunatFortunat Posts: 103
    Stagnation temperatures


    That equation seems to make sense. The other place to get stagnation temperature is either from the collector spec sheet or from a testing lab. SPF (the european equivalent of SRCC) tests stagnation temperature as part of their test protocol and lists it on the test report.

    Their database is available online just like SRCC and I tend to find the SPF numbers more accurate in lots of applications.

    A typical test report is attached in case anyone wants to see one and is too lazy to search for their website. Comparing the published tested stagnation temperature to the theoretical temperature derived by Sigenthaller is left as an exercise for the reader...

    Great solar day today....gotta get out of the office to go enjoy it!

  • hot rodhot rod Posts: 6,711
    nice data

    I notice the formula they use produces a curved line. I like that they show Gross and Apertue area outputs. That's where the evac tube number can get skewed.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    The magic is in hydronics, and hydronics is in me
  • More Stagnation Testing AP-22

    I'm using this thread as my lab notebook.

    After a full day of dry stagnation, the collector sensor reads 385F. (70F ambient). For ease of attachment, I use a bronze 3/4" Ford-style compression fitting with a silicone seal. It works well on the 22mm Apricus header. The Ford fitting surface temp maxed out at 200F.

    From there, I use a Falcon corrugated stainless water heater connector with a silicone O-ring. The thermal conductivity of the SS is poor enough to drop the pipe surface temperature to ambient in just 3 inches from the threads of the Ford fitting.

    These look just like the stainless piping that many manufacturers are pushing for retrofit installations. The trouble is, that pre-insulated stuff runs $20/ft., and I think flexible collector loop lines are tough to keep straight enough to ensure constant slope and drainback.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • Pex forced to fail

    As I mentioned earlier, I'm also testing PEX in the pressurized drainback loop.

    I was able to get the PEX to fail by closing a valve on the collector loop return line, and turning the pump on. The AP-22 generated enough steam to blow the supply line. The return line did not blow because it has a 36" Falcon fitting and 36" of galvanized pipe before joining to PEX. The drainback collector loop was also pre-pressurized with compressed air to 100psi.

    The steam was going either way, and found the supply PEX first.

    In general, you should never have a valve on the return line of a drainback system. All other possible failure modes that I tried didn't cause any PEX failures.
    Superinsulated Passive solar house, Buderus in floor backup heat by Mark Eatherton, 3KW grid-tied PV system, various solar thermal experiments
  • oldbarnsoldbarns Posts: 4
    drainback with PV pump, EV tubes and thermal dump sanity check please!

    In the interests of simplicity im thinking of installing a draindown system using a PV driven pump with Evacuated tubes

    I have found a great pump  [u][size=12][color=#0000ff][/size][/color][/u]   eco15pv

    Can you have a look at the attached please as to the design


    The Tubes and the PV will be just above the tank  (yet to purchase but I have been told 30 odd at 58mm would be best) (plus the PV)


    The tank is in and working and the input and output pipes are available,  the connections are open to the tank, no coils for this

    My woodburner also has a pair of connections open to the full 350l


    I have a loop in the tank from an oil boiler for backup    and 1 loop for UFH   and a further loop with mixer for mains hot water


    The water in the tank has anti corrosive treatment but no antifreeze


    This is a drainback system based on the diaphragm pump offering minimal resistance to the drainback and the inverted breather pipe above the manifold giving opportunity for venting and air ingress during drainback


    I know its simple and may not be quite as efficient as a managed system but with a pair of 3 way T valves (12vDC) managed by a stat on the tank I cant see any great issues but im looking for a second opinion please in case I have missed the point

    Is there anything in your opinion majorly wrong with this setup please?


    Regards rob
  • hot rodhot rod Posts: 6,711
    my thoughts

    make sure the tube collector you are considering do in fact drain 100% or you could freeze the headers. I have done a few with the collectors mounted sideways, if the manufacturer will allow horizontal mounting.

    That looks like a centrifical pump? which is fine if it can handled the lift and temperature.

    Remember those tubes will get smoking hot when drained.

    With that in mind water hitting an absorber in the tube at 400F plus will flash to steam.

    You could pressurize that loop to drive the boiling point up, 30- 40 psi could be considered.

    Or consider a solar controller to limit pump operation at extreme collector temperatures. There are 12 and 24VDC controllers available. That controller could offer additional features, drainback specific functions, data logging, etc.
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    The magic is in hydronics, and hydronics is in me
  • oldbarnsoldbarns Posts: 4
    Solar PV driving pump

    Thanks for your feedback, im hoping that driving the pump with a pair solar PV's that are over sized and angled to take on a larger arc of the sky than the solar tubes get the first glimpses of sun so that the water will not be hitting a hot collector in a manifold, ie the pump will be starting before the solar tubes get up to temp and stopping after they have had the best of the sun.

    Thanks for the tip on the vertical mounting of the tubes that might be a good idea, I'm proposing 2 arrays of 20 each but i could then potentially run them in Parrallel rather than in Series? is that viable?. you would be assured of clean drain down in this situation but could there be an issue with balencing flow rate?

    If i were to arrange the arrays in parallel but vertical i guess the best option is to drive water in from the bottom and out the top, with a breather on the top, like my drawing, you could be assured that both would be filled but not assured both having the same flow rate without some output adjustment, maybe with fulll bore valves on the output perhaps.  set it up, get the pump running and have the outputs of the manifolds open and watch the flows, trim the ball valves to induce a little more back pressur in one or the other until the out flows are the same.  that should balence them. I asum this needs to be consistant at the point the 2 flows rejoin. or am i worrying about something that really is not going to be an issue?
  • hot rodhot rod Posts: 6,711
    Reverse return

    Would be the simple piping method. Also no vent is needed on top, the return from the collector is in tha air bubble at the top of the tank, pump stops air rises up and breaks the siphon, very simple
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    The magic is in hydronics, and hydronics is in me
  • oldbarnsoldbarns Posts: 4
    potential issue with EV tubes horizontal

    Hi again,  just had a potential issue with this idea highlighted to me

    An EV tube relies on it running up hill for the thermal cycle that heats the bit that goes into the manifold

    see thermal cycle in image, if you lay it flat it is potentially going to overheat and not deliver heat to the manifold in the desired way.  im presuming most EV's have a liquid core?

    also I dont have air in the top of my tank, my tank has a built in header on top  so i would be returning just below the header but basically submerged, i guess i could just return the water into the header as it has one hole into the top of the tank and a pipe down to the bottom of the tank so the warm water would get into the tank via circulation but i put the high vent on top as a belt and braces safety measure incase i had a pump failure, any steam would work its way up and out rather than back down to the tank if it were to boil up

    regards rob
Sign In or Register to comment.


It looks like you're new here. If you want to get involved, click one of these buttons!