Heat transfer from roof loop to tank below expectations.

[Naptown]

I'm preparing a response to these questions and others by assembling a more detailed description of the hardware and system operation. It will be posted in a day or two. Please hold off additional speculation until I've posted that data.

But speculation is what we do best

[Art VanDelay]

"Thermal Bottlenecks"

LAW said, "One commentator indicated that the temperature drop of the fluid in the collector loop when passing through the heat exchanger should be about 20 oF; now, the dT is about 5 oF most of the time when there is a 50 oF dT between collector fluid and tank water. "

Right, that is exactly the type of poor performance I witnessed when I used a Rheem Solaraide. It's a ****ty design because it wraps a cylindrical tube that is 5/8" diameter around a steel tank. Even though the soft copper tube flattens slightly when wrapped, very little copper is in intimate contact with the steel. And the thick steel is a lousy conductor compared to the thin copper.

We don't need to know the flowrates to understand this problem - high flowrate or low flowrate, there should never be a 50F dT.

I think I know why Solaraide was specified, though: It's the only solar tank available at Home Depot and it's several hundred dollars less than a good tank like a Bradford White: http://www.pexsupply.com/Bradford-Wh...l-Water-Heater. The B-W boasts 14 sq. ft. of single wall heat exchange area, and the Rheem is about 2 sq. ft. of double wall heat exchange area. So the bad performance really isn't surprising.

While LAW is assembling more information to help convince us that the Rheem heat exchangers are the culprit, we can review this article by Siggy which describes the type of problem we're seeing:

http://www.pmmag.com/Articles/Column/BNP_GUID_9-5-2006_A_10000000000001179895

[MikeSolar]

LAW said, "One commentator indicated that the temperature drop of the fluid in the collector loop when passing through the heat exchanger should be about 20 oF; now, the dT is about 5 oF most of the time when there is a 50 oF dT between collector fluid and tank water. "

Right, that is exactly the type of poor performance I witnessed when I used a Rheem Solaraide. It's a ****ty design because it wraps a cylindrical tube that is 5/8" diameter around a steel tank. Even though the soft copper tube flattens slightly when wrapped, very little copper is in intimate contact with the steel. And the thick steel is a lousy conductor compared to the thin copper.

We don't need to know the flowrates to understand this problem - high flowrate or low flowrate, there should never be a 50F dT.

I think I know why Solaraide was specified, though: It's the only solar tank available at Home Depot and it's several hundred dollars less than a good tank like a Bradford White: http://www.pexsupply.com/Bradford-Wh...l-Water-Heater. The B-W boasts 14 sq. ft. of single wall heat exchange area, and the Rheem is about 2 sq. ft. of double wall heat exchange area. So the bad performance really isn't surprising.

While LAW is assembling more information to help convince us that the Rheem heat exchangers are the culprit, we can review this article by Siggy which describes the type of problem we're seeing:

http://www.pmmag.com/Articles/Column...00000001179895

I missed a couple of days and I thought the everything was solved. This is simple and ART is right, the current HX is garbage and should never be allowed for sale but the designer is the one at fault here. At best the 6 panels will put out perhaps 14-15kw (50Mbtu) (1kw/m2 X 6 X 3m2 each X 75% (if you are lucky). There is nothing wrong with a 30F dT across the panels and HX and both the panels and tank temp should rise in unison.

Doing it the way I noted some posts back, the entering temp on the HX will stay lower due to the lower temp in the bottom of the (as yet) un mixed tank. With a really slow Domestic pump, the panel temp will rise just above the tank top temp. Flow rate can be quite slow on the glycol side too, 0.5l/m2 X 18 m2 (or 2.2-2.4GPM). That slowish pumping will make for good heat transfer AND THE HEAT WILL BE WHERE IT IS USEFUL, which is not at the bottom of the tank. The only caveat to this is if you have an internal HX but there is no room for a Viessmann tank (biggest HX on the market 16ft2 for the 80gal and 20ft2 for the 120gal).

There, rant over. I'll shut up now.

[Naptown]

I missed a couple of days and I thought the everything was solved. This is simple and ART is right, the current HX is garbage and should never be allowed for sale but the designer is the one at fault here. At best the 6 panels will put out perhaps 14-15kw (50Mbtu) (1kw/m2 X 6 X 3m2 each X 75% (if you are lucky). There is nothing wrong with a 30F dT across the panels and HX and both the panels and tank temp should rise in unison.

Doing it the way I noted some posts back, the entering temp on the HX will stay lower due to the lower temp in the bottom of the (as yet) un mixed tank. With a really slow Domestic pump, the panel temp will rise just above the tank top temp. Flow rate can be quite slow on the glycol side too, 0.5l/m2 X 18 m2 (or 2.2-2.4GPM). That slowish pumping will make for good heat transfer AND THE HEAT WILL BE WHERE IT IS USEFUL, which is not at the bottom of the tank. The only caveat to this is if you have an internal HX but there is no room for a Viessmann tank (biggest HX on the market 16ft2 for the 80gal and 20ft2 for the 120gal).

There, rant over. I'll shut up now.

Missed the Rheem part and I agree. that is technology from the early 70's Kind of like black paint on absorber plates.

Actually Vaughn has a 20ft² heat exchanger as a standard in all their tanks. It is submersed, copper and replaceable/removable.
The stone lining eliminates the corrosion problems with dissimilar metals.
And before you go off about it these tanks have a 30+ year history and we actually see average life span at about 25.

[russ]

Right, that is exactly the type of poor performance I witnessed when I used a Rheem Solaraide. It's a ****ty design because it wraps a cylindrical tube that is 5/8" diameter around a steel tank. Even though the soft copper tube flattens slightly when wrapped, very little copper is in intimate contact with the steel. And the thick steel is a lousy conductor compared to the thin copper.

Exactly right - then from time to time I see the heat exchangers for the shower/bath drain and people saying how wonderful they are. They are an even worse case for heat transfer than the Rheem unit. Not to mention a very intermittent flow.

[MikeSolar]

Missed the Rheem part and I agree. that is technology from the early 70's Kind of like black paint on absorber plates.

Actually Vaughn has a 20ft² heat exchanger as a standard in all their tanks. It is submersed, copper and replaceable/removable.
The stone lining eliminates the corrosion problems with dissimilar metals.
And before you go off about it these tanks have a 30+ year history and we actually see average life span at about 25.

Haha, I have a permanent dent on my shin when a vaughn tank fell whilst trying to get it out of the basement. Heavy buggers.

Interesting thing about the black paint........For the longest time I thought that highly selective surfaces like Bluetec and Sunselect were the best thing going but now i am thing from a long term perspective that black chrome or even a really good black paint is better. Yes it is less efficient and the emissivity is higher but the stagnation temp is lower, protecting glycol and O-rings or anything else that comes in contact with really high heat. Maybe you would need 10% more panel area for the same annual output but ....big deal.....

[LucMan]

Every one has their favorite tank mine being SuperStor.
Although the Rheem Solaraide tanks are not the most efficient they are about the average for the industry.
Check the SRCC OG 300 ratings, and you will see that they do work.
I am quite sure that the problem with this system is flowrate related not HX construction. Another highly unlikely but a possibility would be that the HX is not properly bonded to the tank.
I'm guessing undersized pump or piping, or maybe crud in the impeller or all of the above.

[MikeSolar]

Every one has their favorite tank mine being SuperStor.
Although the Rheem Solaraide tanks are not the most efficient they are about the average for the industry.
Check the SRCC OG 300 ratings, and you will see that they do work.
I am quite sure that the problem with this system is flowrate related not HX construction. Another highly unlikely but a possibility would be that the HX is not properly bonded to the tank.
I'm guessing undersized pump or piping, or maybe crud in the impeller or all of the above.

I have only had 1 contact with the HTP product and it wasn't very good. Bad welds on both the first tank and the replacement tank.

I do have a problem with having both solar storage and backup gas in the same tank unless the tank is at least 400L (100USG) and there is a plate inside to stop mixing from the top an bottom. Some European tanks are made that way, none here that I know of.

The OPs problem is indeed HX related. It is simply too small for the system. You can only increase the flowrate so much before you have too high a velocity/noise and pump power for the heat transfer. Maybe there is a bonding issue in the tank but I would not buy a tank that could potentially have that issue.

[Art VanDelay]

Check the SRCC OG 300 ratings, and you will see that they do work.

OG-300 tests are done with a better match of collection area to storage and heat exchanger size than the OP has.

Siggy's article above explains it pretty well.