Low flow rate in closed loop system

[Stringpark]

Just about have my refurbished closed loop system ready to go. I've installed three 4x8 flat plate collectors on the roof and have connected with about 45' (total supply and return) of 3/4" CSST piping. The collectors are probably 15'-18' above the pump. I've tried to flush the system with city water (with plans to then temporality run in on straight water as a test), however, the flow rate through the system when flushing is very low (0.2 - 0.6 GPM, where the supply is capable of 3.8 GPM). This was all measured via a Flume flow meter. The supply pressure is ~45 psi. I'm guessing I have an air lock of sorts in the system (hopefully not cooked glycol restricting flow), but I guess my question is should the city supply at that pressure be capable of purging the system of air? Incidentally, for kicks I went ahead and pressurized the loop with water and ran the pump. The water does circulate, but at a very low rate <0.5 GPM.

Thanks.

[peakbagger]

Do you have a vent at the highest point? That will confirm the air lock plus is good place for pressure gauge to see where the the pressure drop is.

Yes 45 psi should give you plenty of flow. That is my system charge pressure

My guess is you have debris lodged somewhere in the system.

[Stringpark]

Thanks peakbagger! I do have a manual vent that is built into the sensor well.
Sensor Well.PNG
Now, as to whether it's at the very highest point, I'm not sure, but it should be within inches.
Array.PNG
Unfortunately, I wasn't thinking well when finishing up on the roof, and I applied a butyl-backed flexible flashing over the vent. So I guess the next step is to cut that off to access the vent, but it's going to be a mess.

Once I gain access to the vent and measure the pressure drop, what will that tell me? Also, if the trouble ends up being debris lodged in the system, do you have any suggestions for ways to clear it?

Thanks!

[peakbagger]

I am confused. It looks like If I was I standing on the bottom drip edge of the roof looking up that the flow goes into the upper left hand corner and then flows through the three collectors in series before exiting out the upper right hand corner? That is unusual. I do not pretend to know what the internals look like but most collectors would be oriented 90 degrees so they are vertical with a common bottom and upper header connected to all the panels so the flow is in parallel. Normally inside of the panel is just multiple smaller vertical tubes with fins hooked to common header. Hooking them in series means all the flow has to run through the three panels so friction in piping could be an issue and is not typical for SHW collectors. I have seen versions that were fed from the bottom with horizontal internal tubes but they were hooked in series and really didnt work well. Panels can only absorb 80 deg F over ambient and that best done with vertical internal tubes in parallel. The tubes may be small so there are lot of them and that means low velocity in each tube.

The trick with pressure gauges is realize you are measuring two different pressures, static and dynamic. If there is no flow in the line then all you are measuring is how high the gauge is relative to the source of pressure. Assuming you have steady water pressure in the street then three pressure gauges would be handy. Mount one on the line coming in from the city, one on the vent on the roof and one at the discharge where you are dumping the water. Now with no flow write down all three readings, this is static pressure. The difference in pressure between the roof and the supply gauge should be the elevation difference divided by 2.31 which will convert to PSI. If its not, one of your gauges is not calibrated. You can also just use one gauge and run up and down a ladder ideally with a temporary valve to hook up to at each point. Now start the flow through the system and read all three gauges, this is the dynamic pressure. Now subtract the static from the dynamic and then subtract the pressure from the gauge up on the roof with the one on the inlet and compare it to the difference between the gauge on the roof and outlet. My guess is you will see a much bigger value between the inlet and the vent on the roof. That means that it either piped incorrectly or you got a plug. To go farther than that would require more pressure taps on the inlet and outlet of each collector. They should be equal dynamic pressure drop (subtract the outlet from the inlet when flowing, if one is much higher than another you got a plug. If they are equal than its probably incorrect piping.

FYI , your piping piping looks small but hard to judge in photo. Measure the piping going in an out. 1/2" copper tubing is only good for around 3.5 psi with big pump, 3/4" Tubing is good for 7 GPM

[Stringpark]

peakbagger your assessment is correct, they are in series. This is a Stiebel Eltron SOLKIT system and the below drawing from their installation manual is nearly an exact representation of how I put everything together, with a few exceptions. 1.) I didn't install the automatic vent on the other side of the sensor well (top right on the drawing). 2.) I didn't use the flex "jumpers" with manual vent at the input of the array and between the collectors. The ones that were on the system at it's previous home had deteriorated, so I just used the same 3/4" corrugated stainless steel tubing that I used for the supply and return lines between the array and the storage tank. So in the end, I only have the one manual vent at the output of the array.
SOLKIT Schematic Diagram.PNG
When preparing the system, I had a couple of the panels torn apart for repairs (fixed pinhole leaks, broken seals, etc.), and although I didn't study the internals too intently, I do remember that there were several (8 or so) smaller diameter vertical pipes that all terminated to the same 1/2" footer pipe. I never looked to see how the header was setup, but I'm assuming that perhaps it is in two parts, to where the fluid would have to go down 4 of the vertical pipes and up the other 4 pipes to exit the collector. Not 100% sure about that. I may reach out to Stiebel Eltron out of curiosity.

Thanks so much for the explanation on static and dynamic pressure. I'm eager to check that out now. Will report back.

As for the piping, it is 3/4" corrugated stainless steel tubing.

Thanks!

[peakbagger]

Okay, I think you are on your own although the pressure testing may point you to which panel is plugged. The piping is very similar to mine and all I have is DC circulator pump running off a ancient 50 watt solar panel. I have 3/4" copper piping so in theory there is not a lot of friction loss in my system which is two large vertical panels in parallel. I have a float type high point vent and as far as I know it has never leaked but many folks pull them off. I have a 120 volt shurflow pump and a bucket to charge the system pressure up. BTW make sure your expansion tank is functional and the pressure is set correctly as typical expansion tanks are set at too low of pressure out of the box.

[Stringpark]

Funny you should mention the expansion tank, as I was actually wondering about that. The label on the tank says “pre-charge pressure 12 psi”, but everywhere else on the internet seems to recommend either the same pressure as the final system pressure or slightly less (ie 30lbs for 40lbs charged system pressure). I’m actually running the system now (in its low flow state), but haven’t engaged the expansion tank yet (I have a shutoff valve installed). The system has gone as high as 70 psi when the collectors are hot and with the pump running at its max setting. So I’d like to engage the expansion tank, but didn’t want to burst the bladder.

[J.P.M.]

One problem, aside from likely maldistribution of flow from the kind of funky u-tube internal flow arrangement those collectors are claimed to have is how to drain them.
FWIW, I'm also not a big fan of metallurgical bonding of copper and aluminum (copper headers/risers and aluminum fins), including the not too often talked about lowered thermal conductivity of bonded joint.

as for flow restrictions, my guess is, depending on how old and how long the system was in service without complete draining and flushing (and given the portrait orientation your using requiring the collectors be flushed "upside down" for complete drainage BTW), is that the collectors are partially plugged and the other system components may have the same problem due to age.

The vent/air bleed, BTW, ought to be at the highest point in the system. Close to the top won't get it done for system gas purging.

[Stringpark]

Thanks J.P.M.!

So I was able to get up on the roof the other day and measure the pressure on the output of each panel and they all measured the same. The static pressure drop was right in the neighborhood of where peakbagger explained that it should be by calculation. I didn't have the fittings needed to measure dynamic pressure, however, I did hook up a garden sprayer nozzle to the output the three panels (one at a time of course) and noticed that the output of the first panel shot water about twice as far as the next two (which were both about the same). I need to get some fittings, and then I'll try to re-attack this in the coming days. It looks like you guys are right though, and that the second (and perhaps third) panels have gunk in them. In a previous thread, LucMan had talked about running vinegar or Rydelyme through the system, so I'll probably give that a go. Guessing I'll just take a bucket and my drill pump up on the roof and just cycle the vinegar through each panel individually and see what happens. Anyway, thx again for the help!

[J.P.M.]

Thanks J.P.M.!

So I was able to get up on the roof the other day and measure the pressure on the output of each panel and they all measured the same. The static pressure drop was right in the neighborhood of where peakbagger explained that it should be by calculation. I didn't have the fittings needed to measure dynamic pressure, however, I did hook up a garden sprayer nozzle to the output the three panels (one at a time of course) and noticed that the output of the first panel shot water about twice as far as the next two (which were both about the same). I need to get some fittings, and then I'll try to re-attack this in the coming days. It looks like you guys are right though, and that the second (and perhaps third) panels have gunk in them. In a previous thread, LucMan had talked about running vinegar or Rydelyme through the system, so I'll probably give that a go. Guessing I'll just take a bucket and my drill pump up on the roof and just cycle the vinegar through each panel individually and see what happens. Anyway, thx again for the help!

You're most welcome.

Leave the vinegar in the system for a few hours. It takes time to work.

[peakbagger]

FYI, if the coils are corroded from the inside due to acidic stagnant antifreeze cleaning may cause a leak. If there was weak spot better to know it now than later. In some cases leaks can be patched while in others its not worth it as the panel are just plain shot.

[Stringpark]

Thx guys, so with the vinegar, is that straight, or a certain percentage of water mixed in, or what do you recommend?

[J.P.M.]

Thx guys, so with the vinegar, is that straight, or a certain percentage of water mixed in, or what do you recommend?

White vinegar you buy at the store is diluted acetic acid. It's a 4% concentration. Use it as it comes out of the jug. You can buy what's called glacial acetic acid, but you don't want to get involved with that unless you REALLY know what your doing.

How long has the system been in existence and, a separate question, how long has it been in use ?

[J.P.M.]

I am confused. It looks like If I was I standing on the bottom drip edge of the roof looking up that the flow goes into the upper left hand corner and then flows through the three collectors in series before exiting out the upper right hand corner? That is unusual. I do not pretend to know what the internals look like but most collectors would be oriented 90 degrees so they are vertical with a common bottom and upper header connected to all the panels so the flow is in parallel. Normally inside of the panel is just multiple smaller vertical tubes with fins hooked to common header. Hooking them in series means all the flow has to run through the three panels so friction in piping could be an issue and is not typical for SHW collectors. I have seen versions that were fed from the bottom with horizontal internal tubes but they were hooked in series and really didnt work well. Panels can only absorb 80 deg F over ambient and that best done with vertical internal tubes in parallel. The tubes may be small so there are lot of them and that means low velocity in each tube.

The trick with pressure gauges is realize you are measuring two different pressures, static and dynamic. If there is no flow in the line then all you are measuring is how high the gauge is relative to the source of pressure. Assuming you have steady water pressure in the street then three pressure gauges would be handy. Mount one on the line coming in from the city, one on the vent on the roof and one at the discharge where you are dumping the water. Now with no flow write down all three readings, this is static pressure. The difference in pressure between the roof and the supply gauge should be the elevation difference divided by 2.31 which will convert to PSI. If its not, one of your gauges is not calibrated. You can also just use one gauge and run up and down a ladder ideally with a temporary valve to hook up to at each point. Now start the flow through the system and read all three gauges, this is the dynamic pressure. Now subtract the static from the dynamic and then subtract the pressure from the gauge up on the roof with the one on the inlet and compare it to the difference between the gauge on the roof and outlet. My guess is you will see a much bigger value between the inlet and the vent on the roof. That means that it either piped incorrectly or you got a plug. To go farther than that would require more pressure taps on the inlet and outlet of each collector. They should be equal dynamic pressure drop (subtract the outlet from the inlet when flowing, if one is much higher than another you got a plug. If they are equal than its probably incorrect piping.

FYI , your piping piping looks small but hard to judge in photo. Measure the piping going in an out. 1/2" copper tubing is only good for around 3.5 psi with big pump, 3/4" Tubing is good for 7 GPM

PB: These panels have a funky internal arrangement. Think of a U-tube shell and tube heat exchanger in a vertical orientation with the U bends at the bottom.

For this collector system looking at the collectors without the glazing on them, you'll see 1 half length (or width for the portrait orientation) header. From there, the risers drop, make a "U" turn and head back up to the outlet (half) header which is at the same elevation (for the portrait orientation) and regardless of orientation, colinear with the inlet header. As plumbed for this application, the collector system can't be drained which, if the system stood unused for any length of time, may have contributed to any fouling that may exist in the collectors.

[Stringpark]

White vinegar you buy at the store is diluted acetic acid. It's a 4% concentration. Use it as it comes out of the jug. You can buy what's called glacial acetic acid, but you don't want to get involved with that unless you REALLY know what your doing.

How long has the system been in existence and, a separate question, how long has it been in use ?

Awesome, thx J.P.M.! From what I understand, the system was installed new in 2008. The previous owners let me have the system for the effort of de-installation, which I did earlier this year. The loop was completely dry when I de-installed it (pump was destroyed with cooked glycol in it, etc.), and if I recall correctly, the previous owners said it was not working when they bought the house a few years ago. So age of the system is 13 years, but I'd have to guess on the in use time and say maybe 10 years or so.

PB: These panels have a funky internal arrangement. Think of a U-tube shell and tube heat exchanger in a vertical orientation with the U bends at the bottom.

For this collector system looking at the collectors without the glazing on them, you'll see 1 half length (or width for the portrait orientation) header. From there, the risers drop, make a "U" turn and head back up to the outlet (half) header which is at the same elevation (for the portrait orientation) and regardless of orientation, colinear with the inlet header. As plumbed for this application, the collector system can't be drained which, if the system stood unused for any length of time, may have contributed to any fouling that may exist in the collectors.

That makes sense for how I imagined the header to be configured as well. These are the Sol 25 Plus collectors (Stiebel Eltron's first gen collectors, as I understand it). I've had the collectors apart, and what I remember is that the bottom 3 quarters (in portrait) looked like the below, with a common footer pipe. The "half" header you described is the only thing that made sense to me, although I never looked at header when I had the collectors apart (I was focused on fixing pinhole leaks on the footer).
Collector configuration.png