mods to John Allen's Solar Thermal system
John Allen
Sunday Morning about 11:30 my Solar thermal system started operation with a PV powered DC pump using no control logic. It relies on the coincidence that the PV makes electricity in approximate proportion to the heat collected by the water panel.
The below graph shows the first two days of operation. There was a spike in collector temp before it started because I the sun was heating the collector befor I got all the air purged and the glycol flowing.
There was a spike Monday afternoon at he end of the day because the little PV panel connected to the pump got shaded about 4:00 PM. When I discovered and corrected this about 5:00. there was still enough sun to power the pump a little.
http://svallens.com/templogger/graph.php?start=1174138136&end=1174397336&addresses=6700000105615828,F8000001096CA628,3600000109546328,3A000001095EF328,
I have been happily surprised that this use of PV as a control system works as well as it does. I expected to need to put it in series with the DTC to avoid pumping glycol when there was enough sun to power the pump but not heat the Glycol hotter than the storage. This may become a problem if I manage to get the storage temperature hotter. If so, one alternative solution is to provide some shade to the side of the PV so it cuts down the PV at the end of the day. Like yesterday, but done on purpose.
ja
--
John A Allen
John Allen
Today is a cold gray drizzling day with no meaningful heat in the collector, but the pump & PV combination managed to move some glycol anyhow. Curiously I recently re-connected the sensor on the bottom of the Metal Barrel and it is the only one to tmake the result obvious. This is because my glycol HX loop sits on the bottom of each barrel. So when some cold glycol hit that spot it chilled the storage water in the bottom only. That gradually cooled down the rest of the barrel.
http://svallens.com/templogger/graph.php?start=1174339409&end=1174425809&addresses=6700000105615828,F8000001096CA628,3600000109546328,3A000001095EF328
I'll evaluate the total BTU's wasted after the day is over and see how bad it is.
As a side note, yesterday morning Eric and I stretched out the copper HX coil for tap water in the MB and looking at the data it did not make a lot of difference in the heat extraction rates this morning. This despite the fact that almost all of the copper used to sit in a bunch at the bottom of the Metal Barrel and now it is fairly well distributed all over the metal barrel. Perhaps if I factor in the fact that the storage temp was kinda puny this morning I'd be more impressed. I'll wait for a day when that is the case and compare again. BTW the right hand scale for "rate of change" is in degrees F per hour. If I see that to 30 on a day when storage is 130 F I'll be satisfied that messing with the copper tubing was worth the fuss it took.
All my polyethylene tubing is also bunched up on both bottoms, and I doubt if it will help much to distribute it better.
Tony Luck
> Feel free to make suggestions.
I've been thinking about heat exchanger design. In some earlier e-mail
you had mentioned that getting heat into the storage tank from the
glycol loop was relatively easy ... the heat energy arrives slowly, and
you have all day to move it. But getting the heat back out again was
harder. There are several issues:
1) The water from the city supply is under higher pressure.
2) Need to maintain pressure and flow-rate (so that showers
upstairs work acceptably).
3) Need to move a lot of energy quickly.
At first I wondered about inducing some turbulence in the
flow (to avoid the problem of the water at the edges of the
pipes moving slowing and picking up a lot of heat, while the
flow at the center of the pipe is rapid, and insulated by the
slow moving water at the outside ... thus lots of cool water
might get through the exchanger). This could be achieved
by adding some larger diameter pipe at intervals in the
heat exchanger (e.g. step up from 1/2" to 3/4" and back
down again) which shouldn't affect flow rate.
But then I realized that we just have a buffering problem.
We don't need to be able to quickly heat an arbitrary amount
of water. We can slowly heat a larger amount ... and if the
amount is large enough to meet typical domestic needs
(fill the washing machine, take a shower, etc.) then we can
take our time about warming the next batch of water.
I think that the capacity here probably needs to be 30gal
or so.
So now I'm wondering whether there is an economically
viable way to do this ... either we need to include one or
more water tanks as part of the heat exchanger. Or build
the exchanger from large diameter pipe (e.g. 4").
-Tony
John Allen
I believe you are at the nut of it. To buffer or not to buffer. Rather than taking sides, I'd like to create a spreadsheet that factors in the cost of buffering the pressurized water vs the cost of having an adequate unpressurized storage and HX. In my house the morning shower routine is a rather large number and the consequences of running out are rather unpleasant. I speculate that i can expand the unpressurized storage with less cost than either an on demand heater or a very large buffer tank, but it is time to at least make a spreadsheet to compare. if I start it, will you and Joe help make it honest and accurate? I prefer to make it a Google shared file so you can edit it and all changes are tracked. Alternatively maybe we can just use an Excel file at sdcc-1 or homebrew. The Google spreadsheet is unfamiliar and slow.
ja
Today I talked to Trendsetter about all this. Their residential DHW solution costs about 12k installed with a ~$2,500 Tank-less Water heater. Subtarct current fed tax credit of 2k and you have 10K.
ja
John A Allen