Efficiency of (flat) thermal collectors?
Paul Ciszek
collectors? i.e., what fraction of the sunlight hitting them is
typically translated into heat inside the house?
--
Please reply to: | "Any sufficiently advanced incompetence is
pciszek at panix dot com | indistinguishable from malice."
Autoreply is disabled |
Morris Dovey
> What is a typical efficiency for non-concentrating solar thermal
> collectors? i.e., what fraction of the sunlight hitting them is
> typically translated into heat inside the house?
WAG:
For air-heating collectors the number should be somewhere between 60%
and 80% - but I suspect that it may be closer to 40-50% for a "typical"
panels.
Water heating panels should be somewhat lower due to the need to
incorporate a powered control system and pump.
--
Morris Dovey
DeSoto Solar
DeSoto, Iowa USA
http://www.iedu.com/DeSoto/
separatus
SRCC rates a SunEarth panel at 68%-71% at high noon on average.
However, remember this is the highest you will ever get, and only for
moments per day. Flat collectors have a bell curve, making almost
nothing at 8:30AM for example. In fact, for a flat plate some 80% of
the daily heat collection is between 10:00AM and 2:00PM. That is one
of the reasons evacuated tubes make more heat... an evacuated tube
essentially goes to it's rated maximum efficiency and stays there all
day.
Michael Eckhard
TrendSetter Solar Products, Inc.
Morris Dovey
> On Jan 2, 9:46 pm, Morris Dovey <mrdo...@iedu.com> wrote:
>> Paul Ciszek wrote:
>>> What is a typical efficiency for non-concentrating solar thermal
>>> collectors? i.e., what fraction of the sunlight hitting them is
>>> typically translated into heat inside the house?
>>
>> WAG:
>>
>> For air-heating collectors the number should be somewhere between 60%
>> and 80% - but I suspect that it may be closer to 40-50% for a "typical"
>> panels.
>>
>> Water heating panels should be somewhat lower due to the need to
>> incorporate a powered control system and pump.
>
> However, remember this is the highest you will ever get, and only for
> moments per day.
For that particular panel - which, for anyone who didn't pull up the
SunEarth specs, appears to be a (year-round) rooftop DHW panel. Other
designs for other applications (such as space heating) can and do
improve on that range.
> Flat collectors have a bell curve, making almost
> nothing at 8:30AM for example. In fact, for a flat plate some 80% of
> the daily heat collection is between 10:00AM and 2:00PM.
An over-generalization. It's worth pointing out that not all "flat
collectors" are/have "flat plates", and that the 10am - 2pm interval is
dependent on a number of unstipulated factors.
> That is one
> of the reasons evacuated tubes make more heat... an evacuated tube
> essentially goes to it's rated maximum efficiency and stays there all
> day.
Hmm - interesting assertion.
separatus
> For that particular panel - which, for anyone who didn't pull up the
> SunEarth specs, appears to be a (year-round) rooftop DHW panel. Other
> designs for other applications (such as space heating) can and do
> improve on that range.
Yes, that particular panel which is considered a very good panel. I am
not aware of any flat-plate collector at this time with significantly
higher output. If you are aware of a 4x10 collector with a rated
output 10% higher than the EC40 collector from SunEarth, I would love
to see the make and model here.
> > Flat collectors have a bell curve, making almost
> > nothing at 8:30AM for example. In fact, for a flat plate some 80% of
> > the daily heat collection is between 10:00AM and 2:00PM.
>
> An over-generalization. It's worth pointing out that not all "flat
> collectors" are/have "flat plates", and that the 10am - 2pm interval is
> dependent on a number of unstipulated factors.
You are right I did not run an FChart for this application. However, I
am curious as to whether you actually intend to dispute that flat
plate collectors have a heating curve that follows a bell curve?
> Hmm - interesting assertion.
Not really. With the same absorber area and the same maximum
conversion efficiency, an evacuated tube will trounce any flat plate
collector on the market. Just because the SRCC rating sheets don't
account for it, their performance projections (annualized) show it
off.
Morris Dovey
> On Jan 6, 7:42 am, Morris Dovey <mrdo...@iedu.com> wrote:
>> For that particular panel - which, for anyone who didn't pull up the
>> SunEarth specs, appears to be a (year-round) rooftop DHW panel. Other
>> designs for other applications (such as space heating) can and do
>> improve on that range.
>
> Yes, that particular panel which is considered a very good panel. I am
> not aware of any flat-plate collector at this time with significantly
> higher output. If you are aware of a 4x10 collector with a rated
> output 10% higher than the EC40 collector from SunEarth, I would love
> to see the make and model here.
I'm not aware of such a panel with a 4x10 form factor.
>>> Flat collectors have a bell curve, making almost
>>> nothing at 8:30AM for example. In fact, for a flat plate some 80% of
>>> the daily heat collection is between 10:00AM and 2:00PM.
>> An over-generalization. It's worth pointing out that not all "flat
>> collectors" are/have "flat plates", and that the 10am - 2pm interval is
>> dependent on a number of unstipulated factors.
>
> You are right I did not run an FChart for this application. However, I
> am curious as to whether you actually intend to dispute that flat
> plate collectors have a heating curve that follows a bell curve?
No again. I will assert that I've seen (and measured) considerable
variation in the shape of that curve - and, to the point here, the
/slopes/ of the rise and fall portions.
>> Hmm - interesting assertion.
>
> Not really. With the same absorber area and the same maximum
> conversion efficiency, an evacuated tube will trounce any flat plate
> collector on the market. Just because the SRCC rating sheets don't
> account for it, their performance projections (annualized) show it
> off.
It would seem that the area where we don't see eye-to-eye is in the
realm of conversion efficiency - and I suspect that may be closely
related to our interest in very different applications.
I intentionally build panels with a (seasonally) variable efficiency,
optimized to produce both a maximum of heat in winter and a minimum of
heat in summer - and don't much care about either projections (as
opposed to delivered results) or year-round heat production.
I'm not denigrating evacuated tubes for use in concentrating collectors.
I'm working on a Stirling engine (which operates off a temperature
differential) whose hot head is a pipe at the focus of a parabolic
trough - and I've been wishing that I could use an evacuated tube in the
design (non-technical considerations prevent) because the bare pipe is
so incredibly lossy. In that application the ratio of trough width to
the width of the target area is greater than 100:1, and just the
reflected /visible/ light losses are horrendous. An evacuated tube would
seem to offer significant help, if not with the reflective losses, with
at least re-radiated and convective/conductive losses.
For space heating, however, I'm very much more interested in energy than
I am in temperature - and because lossiness is a function of temperature
differential, I work very hard to design and build panels that run at
the lowest temperature I can manage in order to deliver the greatest
amount of energy.
Originally, I tried to prevent/minimize the losses - and then (largely
as a result of discussions here) began to design to make use of the loss
mechanisms to /contribute/ to heat delivery. One example is that the
absorber is highly reflective and absorbs on both surfaces - and yet,
for all its reflectivity, it looks flat black (and it's 'black' in a far
wider frequency range than just visible light!) The question that turned
out to be key was "How much energy does a photon lose (yield) when it's
reflected?"
That led me to examine all of the loss mechanisms to see how they could
be used to /improve/ efficiency. The answers have been surprising - and
gratifying. :)
[ You can see the flat panels at http://www.iedu.com/DeSoto/solar.html -
the trough at http://www.iedu.com/DeSoto/Projects/Stirling/Heat.html -
and a concept drawing of the high-temperature engine at the bottom of
http://www.iedu.com/DeSoto/Projects/Stirling/Fluidyne.html ]
azuredu
On Jan 6, 9:21 pm, Morris Dovey <mrdo...@iedu.com> wrote:
> I'm working on a Stirling engine (which operates off a temperature
> differential) whose hot head is a pipe at the focus of a parabolic
Interesting but I don't see how you can achieve this with reasonable
cost and performance.
> trough - and I've been wishing that I could use an evacuated tube in the
> design (non-technical considerations prevent) because the bare pipe is
> so incredibly lossy. In that application the ratio of trough width to
> the width of the target area is greater than 100:1, and just the
> reflected /visible/ light losses are horrendous.
I imagine. Do you know how 100:1 is difficult? You have to adjust
(align). If you use my method (http://wims.unice.fr/xiao/solar/
index.html) and build very carefully and with good material, 100:1
with sufficient accuracy is possible. However, watchout for the
gravity bending. If the trough is too long, this will put the tube
completely outfocus.
> An evacuated tube would
> seem to offer significant help, if not with the reflective losses, with
> at least re-radiated and convective/conductive losses.
Have you read my article on how to insulate tubes in such cases? This
is not an easy subject, most people designing small troughs are locked
up by this one.
My recommendation is never try to use glass-to-copper vacuum seals. At
this size, it will give you huge disappointments.
Morris Dovey
> Hi, me again.
>
> On Jan 6, 9:21 pm, Morris Dovey <mrdo...@iedu.com> wrote:
>
>> I'm working on a Stirling engine (which operates off a temperature
>> differential) whose hot head is a pipe at the focus of a parabolic
>
> Interesting but I don't see how you can achieve this with reasonable
> cost and performance.
Ok :)
>> trough - and I've been wishing that I could use an evacuated tube in the
>> design (non-technical considerations prevent) because the bare pipe is
>> so incredibly lossy. In that application the ratio of trough width to
>> the width of the target area is greater than 100:1, and just the
>> reflected /visible/ light losses are horrendous.
>
> I imagine. Do you know how 100:1 is difficult? You have to adjust
> (align). If you use my method (http://wims.unice.fr/xiao/solar/
> index.html) and build very carefully and with good material, 100:1
> with sufficient accuracy is possible. However, watchout for the
> gravity bending. If the trough is too long, this will put the tube
> completely outfocus.
I had no idea it was difficult. We built the trough and measured the
focal bright line at it widest point (3/8 inch). Then we divided the
width of the trough the width of that line. It didn't /seem/ difficult.
>> An evacuated tube would
>> seem to offer significant help, if not with the reflective losses, with
>> at least re-radiated and convective/conductive losses.
>
> Have you read my article on how to insulate tubes in such cases? This
> is not an easy subject, most people designing small troughs are locked
> up by this one.
Why be "locked up"?
> My recommendation is never try to use glass-to-copper vacuum seals. At
> this size, it will give you huge disappointments.
Ok - we weren't planning to do that, so (hopefully) we'll avoid major
disappointments.
separatus
> [ You can see the flat panels athttp://www.iedu.com/DeSoto/solar.html-
> the trough athttp://www.iedu.com/DeSoto/Projects/Stirling/Heat.html-
> and a concept drawing of the high-temperature engine at the bottom ofhttp://www.iedu.com/DeSoto/Projects/Stirling/Fluidyne.html]
I'll take a look. Always like seeing what others are up to!
azuredu
> > (align). If you use my method (http://wims.unice.fr/xiao/solar/
> > index.html) and build very carefully and with good material, 100:1
> > with sufficient accuracy is possible. However, watchout for the
> > gravity bending. If the trough is too long, this will put the tube
> > completely outfocus.
>
> I had no idea it was difficult. We built the trough and measured the
> focal bright line at it widest point (3/8 inch). Then we divided the
> width of the trough the width of that line. It didn't /seem/ difficult.
The bright line on the receiver is not a reliable indication.
I recommend that you follow the method described in my do-it-yourself
instruction documents, last chapter:
http://wims.unice.fr/xiao/solar/diy-en.pdf
Only when you don't see the light escaping that you can say that the
precision is correct.
By the way, I know how to judge a trough from its image. Yours seem to
have important deformations near the edges. This problem is typical
for constructions using ribs. The middle portion should be OK, but the
edge portions are probably all lost.
A simple method: examine the trough from the front, and watch the
image of the receiver. This is a magnified image, which should be
straight and regular from any angle. If you see it become serpentine
at some place, then there are problems.
Morris Dovey
> The bright line on the receiver is not a reliable indication.
> Only when you don't see the light escaping that you can say that the
> precision is correct.
>
> By the way, I know how to judge a trough from its image. Yours seem to
> have important deformations near the edges. This problem is typical
> for constructions using ribs. The middle portion should be OK, but the
> edge portions are probably all lost.
The project objective is to produce an inexpensive (less than US$200)
engine that performs direct conversion from radiant solar energy to
mechanical energy and delivers at least one full horsepower.
At this stage of the development process, it is sufficient that the
reflector provide the minimum necessary heating of the engine's hot
head. Since that requirement appears to have been met in the "quick and
dirty" prototype, attention has moved on to other aspects of the design.
To more directly address your concerns - since the reflector is a
continuous surface (without discontinuities or isolated singularities) a
deformation would need to significantly widen the bright line before it
would be considered a problem.
The fine-tuning will come after there's something to tune. :)
azuredu
> To more directly address your concerns - since the reflector is a
> continuous surface (without discontinuities or isolated singularities) a
> deformation would need to significantly widen the bright line before it
> would be considered a problem.
No that's not what I have seen.
The light usually goes out of the tube before it widens the bright
line.
The bright line is given by a portion of the surface that concentrates
to the tube.
Use my method and check the light leaks. I bet that you'll have
surprises.
Morris Dovey
Come spring, I'll check with an IR thermometer by measuring the
temperature of the hot head at the ribs and at the points midway between
the ribs.
If the temperatures between the ribs meet operational requirements, it
won't be considered a problem - and if they don't meet requirements,
it'll be a simple matter to add support between the ribs.
Thanks for your input.
azuredu
>
> If the temperatures between the ribs meet operational requirements, it
> won't be considered a problem - and if they don't meet requirements,
> it'll be a simple matter to add support between the ribs.
The temperature does not tell the story at all. What you have to
measure is the thermal output. In calories or kwh or whatever other
energy unit. Then compare with the input to compute the optic
efficiency.
Well the input can more or less be guessed out if the sky is clear.
But measuring thermal output is the only way to tell people about the
efficiency of your collector.
Morris Dovey
All true, but I don't particularly care to talk about efficiency. What I
care about is *demonstrating* that an inexpensive solar engine is
capable of doing a useful amount of work.
All of the heat is used right where it's being produced and discarded a
fraction of a second later - there /is/ no thermal output except as waste.
Serious optimization may follow, but that's not the best use of my time
at this point in the project.
-----
BTW, you only get partial credit (perhaps 3 out of 10 points) because
Carnot cycle efficiency is determined /only/ by the temperatures:
maximum efficiency = 1 - (Tc / Th)
where Tc and Th are the temperatures (in Kelvins) at the engine's cold
and hot heads.
Getting heat into the engine to maximize Th seems much easier discarding
it effectively to minimize Tc.