Glycol heater system problems

[spareparts]

Hi, new to solar and have an issue with my glycol system. We bought our first home and a previous owner had installed a solar heating system.

There's three panels on the roof and two radiator type devices in two separate rooms in the house. Each has a fan and kicks on when it reaches temp I'm new to solar so if you need me to find specific info just let me know what will help out.

Problem I keep running in to is what I think is vapor lock. I have to have the system purged of air/gasses and filled up with about one to two gallons of glycol every month when the pump gurgles and won't supply hot glycol to the radiators. I can't find any leaks anywhere that would account for this missing volume.

Pressure starts out at 30psi when the unit is filled on a sunny day and steadily drops as the month goes by. Dial reads 10psi now, circulated enough fluid to turn on one radiator unit for about 5 mins and then stops...pump still running but no more heat.

Is it possible that the panels are overrated for this system, generating too much heat and boiling off the glycol? Should I cover one? Or cover, say the bottom or top third of all three?

Eventually we'll install another radiator or two but first need to have some double or triple pane windows installed and a few other projects around the house.

[peakbagger]

Any chance you have an air elimination valve on the top of the system?. They are known to leak.

[spareparts]

Any chance you have an air elimination valve on the top of the system?. They are known to leak.

There is a manual valve up top. I'll spray it with soap bubbles tomorrow and see if I detect any leaks. I haven't found any liquid glycol leaking out of it though, I have a flat roof and have checked top to bottom several times recently looking for leaks.

The nearest solar technician is 45 mins away, at his request I opened the valve at one point and steam was was coming out of it.

[inetdog]

There has to be some sort of pressure relief valve in the system, possibly draining in a way that hides what is coming out. That along with an inadequate expansion tank (possibly fluid filled by now instead of air filled) can lead to overpressure when the glycol expands causing a loss of glycol, then when the system cools even a small inward air leak can fill up the remaining space from the lost glycol. Instead of glycol returning from the expansion tank to fill the volume, air comes in and is not eliminated when the system heats again, causing the boiling glycol you saw when you tripped the air elimination valve on a hot day.

Bottom line suggestion, you have two problems:
1. Waterlogged (glycol filled) expansion tank and
2. A leak, possibly at a relief valve, that drains glycol where you cannot see it.

Draining the system and pressurizing it with air is one way to try to locate that leak.

[spareparts]

A new expansion tank was installed just a few months ago, last November I believe.

I'll look around on the roof and under the crawl space some more and see if I can locate a pressure relief valve. The only thing I've noticed so far is what I think is the air bleed valve, which looks like a copper elbow pointing up, with a plug threaded into it anand a bolt screwed in to the plug (best way I can describe it).

[peakbagger]

Might be worth taking an air gauge and checking the charge pressure on the expansion tank. Could be set too low which effectively means its not doing what its supposed to leading to very high system pressure when the system is hot. No matter what glycol doesn't disappear into thin air. If you are adding fluid then fluid is leaving the system. Odds are its only happening when system is heated up and at maximum pressure. If you knew what the pressure rating of the collector was, you could fil the system with air and do a careful leak check of the entire system.

Odds are the piping you describe with the plug and bolt is where an automatic air purge was installed. They make charging the system easier but are somewhat notorious for leaking over time. Knock on wood mine is still up on my roof with 20 years on it and the system still hold pressure.

If was a potable hot water system with heat exchanger I would be worried about the heat exchanger leaking into the potable system. The other place to look is if there is backup heating system on the same loop to the radiators. Boiler systems tend to be run at low pressure, usually 15 psi so a leak in a higher pressure SHW system would end up in the boiler loop.

[spareparts]

Went up on the roof this morning. I live in the mountains and we have a bit of snow melt, so looking for leaks up there this time of year is deceptive. Also complicating it is the flat roof design.

But I did find it! There's a pressure relief valve located down lower on the pipes, half encapsulated by plastic pipe and insulation. With the system at low pressure right now I can't see any leaking, but there's a definite PG slick running away from it that I mistook for snowmelt, and when I stuck my finger in the valve, it came out covered in PG. Valve looks ancient.

How can I figure out what valve to replace this with? Looks like any standard brass valve found on a water heater. Stamped with two numbers, the first is hard to read as the data tab has been kinked a couple of times: 84373 (I think) and 75. Any other info printed on the tag has long ago faded away. I'm guessing the first number to be the part number and the second is the PSI rating?

The panels themselves are Grumman model 332A that measure 4x8 feet.

[J.P.M.]

Went up on the roof this morning. I live in the mountains and we have a bit of snow melt, so looking for leaks up there this time of year is deceptive. Also complicating it is the flat roof design.

But I did find it! There's a pressure relief valve located down lower on the pipes, half encapsulated by plastic pipe and insulation. With the system at low pressure right now I can't see any leaking, but there's a definite PG slick running away from it that I mistook for snowmelt, and when I stuck my finger in the valve, it came out covered in PG. Valve looks ancient.

How can I figure out what valve to replace this with? Looks like any standard brass valve found on a water heater. Stamped with two numbers, the first is hard to read as the data tab has been kinked a couple of times: 84373 (I think) and 75. Any other info printed on the tag has long ago faded away. I'm guessing the first number to be the part number and the second is the PSI rating?

The panels themselves are Grumman model 332A that measure 4x8 feet.

All "domestic" pressure relief valves ("PRV's"), in N. America anyway, have a "test lever" that must be accessible so as to be able to test the valve and ensure it's functioning and with flow ingress and egress not blocked or fouled by crud/sediment/scaling/etc. If the device you describe has such a lever, it's most likely a pressure relieving device, although they are usually located at the high point in any line they are protecting. If it has no such lever, it probably isn't a PRV (but it of course may still be leaking).

The test lever mechanism on a PRV and parts of the PRV associated with the test lever are notorious leakers which is why (and somewhat ironically I'd add), installers and plumbers often imply or downright suggest that the PRV not be tested, as doing so will most likely result in an early leaker.

If there is no test lever, it's original designed purpose was not as a PRV. If it's at the low point in a loop, it's most likely purposed as a drain.

There also may be more than one PRV on a system, usually at least one for each separate fluid circuit, sometimes more than one per fluid circuit, depending on duty.

IF it is the PRV for that loop, to replace it and have the best chance of being safe, find the pressure and temp. rating of each component in the loop (fluid circuit) that the PRV is attached to. Then, find a PRV at big box hardware with an ASME stamp on it and with a pressure rating stamped on it that's BELOW the lowest pressure rating of any of the components in the fluid circuit and below the lowest temp. rating of any of the components in the fluid circuit. Having a higher pressure rating on a PRV than the lowest rated device means that the other device will do the honorable thing, protect the PRV, and blow first.

I'd also suggest you consider replacing any air bleed devices at the top of the system, and any others in the system after checking to see if any air bleed valves up there are functioning as they do not last forever. And, FWIW, adding a redundant (?) PRV at those points at the same elevation as any air bleed valves.

Another FWIW, I spent about half an engineering career designing heat exchangers for power plants, chemical plants and refinery applications. Most of the HX's I've ever seen for domestic solar applications are pretty poorly designed. If you have a HX in the system, good or bad, any HX that's been around as long as a Grumman collector is most likely fouled and you're paying a large performance penalty for it. That device may be the source of elevated system temps. due to lower low caused by excessive HX fouling.

Over time, and as others have noted/written about, I suspect that the system overheated at some point(s) and maybe/likely/still does so as a result of reduced flow, probably or at least partially due to fouling (dirt in the system), or as you suspect, a vapor lock or any number of reasons. If due to fouling/blockage, that fouling restricted/reduced the flowrate (which often tends to be system wide, mostly at restrictions, sharp edges and changes in flow direction), will cause the collectors to overheat and so raise the coolant (glycol) temp. above its boiling point. Then, either the system fails as a leak or a PRV event (blowoff) occurs. That may be an ongoing situation with reduced flow from the fouling causing high(er) collector temps. and fluid boil off and/or overpressurization from the higher collector and thus fluid temps.

Overall, without being there to inspect the system, it's difficult to troubleshoot.

Overall, you have an old system that needs a good technician to go over it.

None of us is as smart as all of us, but we're all operating here under the disadvantage of not being there.

[spareparts]

Thanks for taking the time for such a detailed explanation.

It is indeed a PRV, has the little test lever.

There's no heat exchanger in the system, unless the fan powered radiators in the house count.

If I do have blockage, what's the procedure to clean out the system? Cause if this is the case I may stop here snd cut my losses. I just can't sink much more money into this old system. A few minor repairs are no problem, but as it is so far it will take me two more winters of it running efficiently to recoup what I've put into it versus what it saves on the gas bill.

[inetdog]

The "radiator" with fans is indeed a heat exchanger. It is just a liquid to air heat exchanger rather than the liquid to liquid heat exchanger used in domestic hot water systems.

[J.P.M.]

Thanks for taking the time for such a detailed explanation.

It is indeed a PRV, has the little test lever.

There's no heat exchanger in the system, unless the fan powered radiators in the house count.

If I do have blockage, what's the procedure to clean out the system? Cause if this is the case I may stop here snd cut my losses. I just can't sink much more money into this old system. A few minor repairs are no problem, but as it is so far it will take me two more winters of it running efficiently to recoup what I've put into it versus what it saves on the gas bill.

If any of the thanks is meant for my brain spoor, you're welcome.

Actually, you do have heat exchangers in the system - the fan powered radiators exchange heat between the collector coolant that passes through the devices and the air that passes through the devices./them. Such devices are heat exchangers. The term "radiator" is a somewhat inaccurate historical accident.


Cleaning of any closed system, from power plant boilers the size of your house or larger, to solar thermal systems, down to your coffee maker (a type of boiler) can be attempted with varying degrees of success.

As a practical matter, unless it's possible to dismantle and inspect, clean and the reinspect and finally reassemble every piece and see (inspect) every bit of the internals and surface, it's impossible to know if a system is "clean", with that term having somewhat different definitions depending on service. I'd try chemical cleaning after having the system serviced and see how or if it changes the performance.

To remove/reduce scale: Flush with solutions of WEAK acids (5% concentrations maybe) such as formic, citric, oxalic, acetic (vinegar). Circulate for 2 hours or so and flush/rinse. Check the Ph of the rinse and repeat until neutral (Ph ~ 7.0).

To remove/reduce organic growth/oil/sludge: A 3% or less strength of caustic solution. Fill system and circulate for a couple of hours. Use as warm a solution as possible/practical. Rinse and check Ph of the effluent and flush as before.

For more info, check the net under "cleaning of heat exchangers"

In both or any cleaning case cases, after cleaning, fill the system, pressurize and check for leaks.

When checking for leaks, unless at VERY low pressures for something like a bubble/soap suds test, ALWAYS use liquids and NOT air or gasses for the test fluid.

Pneumatic pressure testing without extensive precautions is dangerous and can be fatal.

Like when a bomb explodes.

A failed liquid test gets things wet.
A failed pneumatic test under something exceeding operating pressure gets people dead.

If it was me, I'd see what it cost to have the system inspected and components serviced/replaced as necessary. Then, I'd flush the system with vinegar and check for leaks with soap suds and VERY low pressure (like 1-2 PSIG) and then check performance.

Beyond that, I'd be careful about more extensive (mechanical) cleaning (disassembly/cleaning/reassembly). First challenge is finding someone who will do it and who also knows what they're doing and is willing to do a good job for a price. Second, once disturbed, the system may never work the same again. And, you just may be increasing the probability of chasing an increasing number of leaks forever.

For that matter, the system may be old enough that chemical cleaning that's extensive enough to restore some performance may put strains on an old system that may be severe enough to cause more damage (sort of killing it in the attempt to save it).

Without being there, I can't offer much other than to say you may have to face the reality that the system may have reached the end of its service life. Or, a mild cleaning may allow it to continue in service for a while longer.

[spareparts]

Thanks again for your help. I've really learned alot in just the short amount of time that I joined this forum. We'll sleep on it for a couple of days and figure out what we want to do.

[LucMan]

My first question would be what does the glycol in the system look like? Is it brown or black? Indicating that the glycol was subjected to repeated stagnating temps with out being replaced when needed. If this is the case then it may not be worthwhile to go any further with additional repairs. Glycol degrades over time and use requiring replacement. When it cooks in the collectors it creates a nasty concoction that coats all the surfaces in the system especially the collectors until they become restricted, usually terminal for the collectors.
If it looks fresh, check for PH and freeze protection, replace the blowoff refill and purge the system and you should be good to go. The Gruman collectors are 1970's vintage that are quite tough and can easily last 30+ years. The most common issue is that the glazing on the panels degrades over time. That being said you know where you stand at this point in time with the age of your system.
Adding additional "radiators" is not advisable as the panels only provide a limited amount of BTU's to the system, so save your money on that project.

[J.P.M.]

Thanks again for your help. I've really learned alot in just the short amount of time that I joined this forum. We'll sleep on it for a couple of days and figure out what we want to do.

The greater the amount of money the decision involves, the greater the probability education about the subject will reduce the amount of the bill. What do you want to accomplish ? How do you plan to get there ?

Good luck.

[spareparts]

My first question would be what does the glycol in the system look like? Is it brown or black?

It is a dark purple color. I drained some in to a cup, about an inch of fluid, I dropped a penny in there, I can see it in the bottom but the fluid is dark enough I can't make out the details.