At 04:10 PM 6/20/2011, Stephen A. Lawrence wrote:
Aw, these reports drive me nuts. I'd not read
this one, I think. They have a pump for cooling
water, it seems. So when they start up, they
are pumping an estimated 6.47 kg per hour of
water. They assume that this flow rate remains the same. Why?
It's a constant displacement pump. That's what it does.
Sure. But does it pump the same regardless of
back pressure? Did they record the number of pump
cycles per minute more than at initiation?
Imagine that the pump has a leak in its internal
mechanism, say a leaky flap valve. If it did,
then the flow rate would vary with the back
pressure. I'm just pointing out that this is an
assumption, without reason and conditions for it being stated.
I'm banging my head against my desk. I have to stop doing that.
However, it's not a *terrible* assumption. It
merely raises its own problems. What I'd worry
about, here, is in the other direction. They
have this thing boiling away all the feed
water. That means they cannot guarantee that
the chamber cools at a constant rate, I
suspect, as there comes to be less water in the
chamber, the cooling rate will go down. Control
problem. This thing gets hotter, the cooling
declines, so it gets even hotter .... Unless
they back way off on the input power, quickly.
You have put your finger on the problem. There is no need to go any farther.
That statement wasn't complete. If there is a
discrepancy between the feed rate and the
boil-off rate, one of two things will happen:
either the reservoir will fill or empty,
accordingly, or they will adjust the input power
to increase or decrease the boiloff rate. They
have not stated that the input power is constant,
and it looks like Essen did not check for this.
Unfortunately, you took your finger off the
problem later on, where you apparently forgot
about the "constant flow rate" assumption...
it's that CD pump which at the heart of all of this.
I think I've heard of the pump rate being
changed. In some experiments the input power was
changed. The assumption that both remained the
same is still possible, if the values were chosen
for operation at close to steady state as to
chamber volume filled. That condition might last
for quite some time before it went over the edge at either end!
First, as I've said in previous email, the
total power dissipated in the device can be
estimated from the slope of the curve and the
temperature of the effluent. The power needed
to heat the output water goes up linearly with
the temperature of the output water, and the
power needed to heat the device goes up linearly with the slope of the curve.
Okay.
There was a claim of "ignition" at about 60C,
at which point the device started generating power.
Looks like that.
In fact, the graph says that can't be right.
If the device generated no power before it hit
60 degrees, the temperature would not have
gotten to 60 degrees so quickly, because (as
stated in the paper) the heater temperature
was only adequate to maintain it at 60 degrees
with the flow rate used in the experiment. So,
the heating curve, instead of looking like a
straight line, would look like a capacitor
charging curve, and it would have taken much, much longer to reach "ignition".
This is the problematic assumption: that the
input power "was only adequate to maintain it
at 60 degrees with the flow rate used." This is
derived from their statement: "If no additional
heat had been generated internally, the
temperature would not exceed the 60 °C recorded at 10:36."
That's preposterous, as you note. Had there
been no generated heat, there would have been a
continued rise in temperature, at the previous
rate of increase, declining asymptotically, as
you state. Quite simply, there is no stated basis for this claim.
Of course there's a simple basis for it (but not
a *stated* basis, admittedly, but it's an awfully simple step to get here):
The flow rate (which is *fixed* by the CD pump),
times the coolant temperature rise (scaled by
the specific heat, of course), gives the power
carried off by the coolant. They're claiming
that the heater power and power carried off by the coolant match at 60C.
They said that, yes. But they stated no reason
for that comment, and, frankly, it doesn't make
sense. By assuming it's true, you are led to
contradictions. How about simply considering that
there was some problem with that unsupported statement?
If they don't, then the paper is in error.
The paper is in error. Like, so? Lots of papers
are written with errors, happens all the time.
Essen should be asked about this. He's stated a
conclusion that is not supported by the data he
presents, the opposite. I think it was just a
slip of expression, that he meant to say
something else. This is an unpublished paper, not
subject to editing or peer review.
It would be pretty simple to check their
arithmetic on this one but it didn't occur to
me to do so; so, they could very well be wrong
about this, which would in turn throw off my
calculations (but not the overall conclusion).
They don't provide arithmetic to justify that
assumption, I'm reasonably sure about that. I can
understand why you'd like the statement to be
meaningful, because then you can make some
assumptions about the heat flow. But those
assumptions lead to unlikely conclusions. Unless
Essen and Kullen confirm the comment about
stability at 60 degrees without heat from the
reaction, the default here, from the experimental
data, is that this is very much not the case. It was an error.
It should therefore, be discounted. The
experimental evidence shows otherwise, so I
assume this was a simple error on their part.
I'll assume that they intended to say that "
the rate of increase of temperature would not have increased."
You can assume they said something other than
they said, but it's not what they said.
Like, we agree on that, right? It's not what they
said. Did they say what they intended to say? If
so, then they intended an error. I find that rather unlikely, don't you?
Okay, there is a possibility, that the reaction
has a rate that varies just right to produce the
linear result that you see, so it's already
kicked in below 60 degrees, at the point where
the heat being produced by the input power would
be approaching 60 degrees asymptotically.
Suppose they had some operation of this thing
without hydrogen in it, just the same input power
on the resistor, same water flow. If it
approached 60 degrees asymptotically, then we'd
have that temperature figure for comparison. But
they didn't say that, and that would be a huge
piece of evidence that they did not report. No,
it's far simpler, and less hard on them, to
assume a simple writing error, an error in expression.
The steam was claimed to be dry in this experiment.
Yes. That claim comes from a visual examination
of the steam valve, and assumes that there was
no flow out of the hose at this point. Again,
they are not explicit about this, but they only
talk about "visual checks of the outlet tube
and the valve letting out steam from the
chimney." If there is no flow out of the outlet
tube (the hose?} and there is live steam at the
valve, easily seen by the short gap where the
steam is invisible, the physical arrangement
would be adequate for this assumption to be at
least approximately correct. They measure steam quality:
"Between 11:00 and 12:00 oclock, control
measurements were done on how much water that
had not evaporated. The system to measure the
non-evaporated water was a certified Testo
System, Testo 650, with a probe guaranteed to
resist up to 550°C. The measurements showed
that at 11:15 1.4% of the water was
non-vaporized, at 11:30 1.3% and at 11:45 1.2% of the water was non-vaporized.
Did they measure this inside the "outlet tube,"
i.e., inserting the probe through the
thermocouple access port? I notice no gap in
the temperature recording from the thermocouple. So there is a question there.
However, this is where we rely on "authority."
I don't see any reason to suspect that Essen
and Kullen didn't know what they were doing.
What I can say is that *from the report alone,*
there are questions. Let's see what Stephen comes up with.
In that case, the output power was something
on the order of 10 kW once it started
producing steam. But the output power before
that point, which we can read from the graph,
was about a factor of eight less than that.
So, once it hit boiling, the output power must
have increased eight fold, very rapidly --
certainly more rapidly than the power had been
increasing up to that point. But then, since
the effluent temperature did not rise above
101C, the power generation must have stopped
its meteoric rise at exactly the core
temperature needed to keep the effluent at 101C, no more, no less.
There is a problem, but the effect of boiling
water isn't being considered. When you boil
water, the temperature of the water (at
atmospheric pressure) wlll not increase beyond
100 C, and the steam will be at that
temperature. Because of effects from the
cooling chamber walls being at slightly higher
than boiling (it must be, average), the
temperature will be *slightly* above 100, which
will, aside from bulk water, if present, dry
steam. Assuming the temperature measurements
are accurate, this is a confirmation of (mostly) dry steam.
GAAAH -- No! You are forgetting that there is a
CONSTANT FLOW RATE from the CD pump!
You are assuming no volume change in the water in
the cooling chamber, right? You are also assuming
that there is no change in flow rate. From the
experimental setup shown in Krivit's video, if
this was the same, they also weighed the input
water. There is no recording of the pump
operation, it's simply an assumption that flow was constant.
When I assume that testimony is true, I do not
assume that *conclusions* are true, and when a
witness states something without providing the
primary evidence for it, I then depend on
expertise, and even experts make mistakes,
especially in unpublished papers only available as what may be ephemera.
What I stated does not "forget" the claim of
constant flow rate, it depends only on one thing:
that the cooling chamber neither overfills (with
water then coming out the hose) nor dries out,
during the experimental period. Under those
conditions, the outlet temperature will be at
boiling, except that as it approaches being
completely dry, the temperature would start to
rise as the steam itself is further heated by the
reactor heat. Only a little, I think, because of
the inefficiency of heat transfer to steam as distinct from liquid water.
It would appear that they have chosen the pump
rate to be one which neither allows the cooling
chamber to empty, nor to overfill, at the given
input power, steady state. But there is nothing
that says that they couldn't have an internal
valve that shuts down water flow. (How does that
pump behave when pumping into a closed pipe?)
This would not affect a measurement using total
volume (weight) of water, but it would certainly
affect calculations from presumed flow rate from
what the pump was accomplishing at startup.
Consider this. Suppose that Essen and Kullen made
some unwarranted assumption in setting up their
measurements. Would Rossi correct them, if he
recognized it? I don't see that he would,
necessarily. My impression is that he wants to
look like a con artist, it's an act. (Jed has a
different explanation, perhaps, it's just a
personality thing, eccentric inventor. I certainly can't tell the difference!)
Get that straight, or you totally miss the
point. The situation is *entirely* *different*
from the situation in a boiler with a fixed allocation of water.
Not entirely different. Suppose you have a boiler
and you are heating it with a fixed amount of
heat, except for startup. Instead of having a
boiling chamber with a fixed amount of water, you
have less water in that chamber, but you supply
it at a fixed rate. How much do you supply? In
real designs, I have one of these things in my
basement, there is an automatic valve that fills
the chamber with water when the level falls.
However, suppose we are going to operate this
boiler as simply turned on. And we set, to obtain
the desired results, a fixed water flow rate. We
start out with the chamber not being full, so as
the heat is applied and it starts to heat up,
there being no boiling yet, the chamber fills. By
the time it's close to full, the boiler is
supplying full heat, which, let's say, is only
enough to boil that water, maybe a little more,
so the water level will slowly go down. As long
as the rate of decrease of the water level is
such that it doesn't run out of water during our test period, all will be well.
The assumption that water in = steam out is not
confirmed by the evidence we have. There could be
variation, from a number of sources.
That's without variation in the input power.
Variation in the input power, through the
controller shutting down the input heat as the
thing approaches operating temperature (which
could be a complex relationship), would be an
obvious engineering approach. I'd expect that a
given E-Cat has a fixed relationship as the
thermal resistance between the reaction chamber
(at 450 C) and water in the cooling chamber (at
100 C when heated up). What they would regulate
then would be the reactor chamber temperature,
which is their control access to the device.
They would have, for a given flow rate, a power
level to feed to the reactor heater to maintain
that rate of evaporation. (This assumes good
control of the reaction, by the way, that they
have operating characteristics for the particular E-Cat.)
What I don't know is why they are using a pump
rather than gravity feed, with the gravity feed
being such as to maintain the desired water level
in the C-Cat. It doesn't quite make sense to me,
and it complicates things. The gravity feed would
adjust the water flow, then, to maintain boiler
water level, which is how it's normally done. It
could also be done with a float valve, but ... moving part.
Since you apparently misunderstood that, there
is no point in reading the rest of this, aside
from a quick skim which indicates to me that the
misunderstanding of the flow rate does indeed persist.
Fine with me.
