MIZUNO REPLICATION AND MATERIALS ONLY

Quote from Bruce__H

Second, it continues to be weird, weird, weird that output power during excess heat runs pretty much exactly matches input power.

I discussed this in the Indian presentation. It gives me the willies. Muto and Mizuno had no comment about it.

It seems to be a coincidence. There is no question that output power adjusted for losses is considerably higher than input power. The losses are easily estimated during calibrations. The estimates may not be very precise, but you can be sure there are substantial losses with such a large box. Even though it is well insulated it has to be radiating a lot of heat.

The data I showed was from years ago. I have not seen the more recent data. I do not know what an "incubator style" calorimeter means.

Quote from Bruce__H

By "incubator style calorimetry" do you mean the same insulated-box-with-flowing-air calorimetry as used by Mizuno?

When I first saw that, I looked it up and nothing popped up. Not my specialty, but will ask what that is one of these days.

Quote from Paradigmnoia

Rather than improve the calorimeter, it is lossy as can be, and has a massive thermal sink in the box itself

Mizuno's system is improved in the sense that when you compare recently collected temperature or power time-time to those from older data sets you will see that the older ones wriggled around all over the place whether it was activated mesh runs or calibrations. Now things are much tighter and more reproducible. I am sure that you are correct about the presence of systemic biases and errors but they are now at least there all the time in the same way.

Quote from Paradigmnoia

Which means that the coincidence that the excess heat values resemble the input values can be easily remedied by a layer of foam to push the losses back into the box, raising the delta T, and therefore the reported output power values will numerically increase.

(Or perhaps the subsequent calibrations will run over 100% recovery, requiring adjustments to the air mass calculations.)

I see what you mean ... I just don't see how the sort of problems you describe would differentially affect the calibration and activated-mesh runs. If measured excess heat is artefacutal here, it must have something to do with how these two types of runs are set up. That is where to look for problems. So far I don't see any.

Yeah, I don't know what an incubator style calorimeter is. But I know of an amusing example of something like that. When Richard Oriani did his first experiment, he used a seebeck calorimeter that was designed to hold a baby. He told me that and I did a double take: "hold a what?!" A baby, no kidding. It must have been a big calorimeter. How did the baby breathe? I asked. I recall he said something like, "they didn't leave it in there for long." Apparently, a doctor or medical school loaned him the instrument.

Here is a "whole room calorimeter" for an adult. It measures O2, CO2 and H2O, as well as heat, I assume. The photo shows an ominous looking black crate-like box sitting next to a large gas tank, but I don't think they stuff the patient into the box. I think there is a room behind it.

https://www.sablesys.com/produ…-room-calorimetry-system/

Quote from JedRothwell

The photo shows an ominous looking black crate-like box sitting next to a large gas tank, but I don't think they stuff the patient into the box. I think there is a room behind it.

Since this is the internet someone might think I mean that. Scroll down that SableSys.com screen and you will see there is an actual room for the patient. Not a black crate. Plus they have systems for non-whole room environments.

Quote from Bruce__H

Mizuno's system is improved in the sense that when you compare recently collected temperature or power time-time to those from older data sets you will see that the older ones wriggled around all over the place whether it was activated mesh runs or calibrations. Now things are much tighter and more reproducible. I am sure that you are correct about the presence of systemic biases and errors but they are now at least there all the time in the same way.

The Saito/Muto (whatever) recent tests hang together much better than previous work, I agree.

Quote from Bruce__H

I see what you mean ... I just don't see how the sort of problems you describe would differentially affect the calibration and activated-mesh runs. If measured excess heat is artefacutal here, it must have something to do with how these two types of runs are set up. That is where to look for problems. So far I don't see any.

I haven’t come across any kind of systemic error that favours ‘extra’ heat detection and that also could be introduced only to the “activated” device conditions, repeatedly, over years. There are plenty of ways to mess with the calorimeter results, but most (if not all) of them should work equivalently to calibration and active conditions.

I do suggest that the air inlet is the most error-prone part of the calorimeter, and steps should be taken to stabilize how air enters and the temperature is measured there. The inlet orifice area in particular is quite variable due to plumbing and wiring crawling through the hole.

Quote from Paradigmnoia

I do suggest that the air inlet is the most error-prone part of the calorimeter, and steps should be taken to stabilize how air enters and the temperature is measured there. The inlet orifice area in particular is quite variable due to plumbing and wiring crawling through the hole.

The temperature should not be measured there. Or, not only there. It should be measured some distance away from the box. The ambient air temperature should be the same as the air temperature at the inlet, to within 0.01 deg C.

All temperatures should be measured at multiple locations.

I just spoke with Mizuno and he asked me to contact the forum and apologize for the delay in his reply. He is quite overwhelmed with personal matters. Dr. Muto's experiment was done with the following equipment: https://www.yamato-scientific.com/product/show/dh650_2/

This is more or less a convection oven with a powerful recirculation fan. Calibrations were done to relate heater power to temperature with a dummy reactor.The dummy reactor at 150W input, gave around 90C. With the active reactor, the temperature increased to 160C, equivalent to 364W with dummy reactor. I hope this clarifies the experiment. We have several partners around the world attempting to replicate this.

Quote from Daniel_G

Dr. Muto's experiment was done with the following equipment: https://www.yamato-scientific.com/product/show/dh650_2/

Ah, yes. I should have remembered that is what they call an "incubator" in Japanese. I would call it a constant temperature chamber in English. The ones I have seen have a thermostat. You can only measure excess heat by measuring a decrease in input power to maintain the same temperature. That is not very precise. That is called the compensation heater method, I think.

Maybe this one has a constant input power mode, where the thermostat is disabled?

Quote from JedRothwell

Ah, yes. I should have remembered that is what they call an "incubator" in Japanese. I would call it a constant temperature chamber in English. The ones I have seen have a thermostat. You can only measure excess heat by measuring a decrease in input power to maintain the same temperature. That is not very precise. That is called the compensation heater method, I think.

Maybe this one has a constant input power mode, where the thermostat is disabled?

I agree this is not a precise system, but no one can argue that if the presence of the reactor inside the oven makes it reach the same temperature with less power input, it is an interesting observation, and is quite easy to reproduce as you use an off The shelf standardized product.

Again, not ideal as proof for anyone skeptic, but I think is a good compromise for practical purposes.

Quote from JedRothwell

That is called the compensation heater method, I think.

Maybe this one has a constant input power mode, where the thermostat is disabled?

The two methods are about equally precise. There is no advantage to the latter, except that it lets the sample heat up a little more. I am speculating that Muto is using the latter method because otherwise I don't how he could derive the graph shown here. It would never go above the temperature set by the thermostat.

McKubre used a compensation heater in his flow calorimeter. Fleischmann thought that was a bad idea, because it did not let the sample heat up much. Fleischmann thought heat is an important trigger. This is less of an issue with a sample inside a stainless steel reactor, which is inside the constant temperature chamber. A change in the chamber temperature will not affect the sample as much as the compensation heater affected McKubre's cathode.

Quote from Daniel_G

I just spoke with Mizuno and he asked me to contact the forum and apologize for the delay in his reply. He is quite overwhelmed with personal matters. Dr. Muto's experiment was done with the following equipment: https://www.yamato-scientific.com/product/show/dh650_2/

This is more or less a convection oven with a powerful recirculation fan. Calibrations were done to relate heater power to temperature with a dummy reactor.The dummy reactor at 150W input, gave around 90C. With the active reactor, the temperature increased to 160C, equivalent to 364W with dummy reactor. I hope this clarifies the experiment. We have several partners around the world attempting to replicate this.

Is there more information on the calibration, like the period of time for the temperature to settle, etc.?
There is something weird about the temperature to power input response in that graph. For example, 100 W input raises the temperature to 80 C, but an additional 50 W only raises the temperature another 10 C. The degrees C obtained per watt input drops almost in half from 100 W to 364 W.

Quote from Paradigmnoia

For example, 100 W input raises the temperature to 80 C, but an additional 50 W only raises the temperature another 10 C. The degrees C obtained per watt input drops almost in half from 100 W to 364 W.

My guess is that this is a crude measurement, with a large error margin. I think the temperature rise per watt of input declines as temperatures rise. The line goes up steeply at first and then gradually levels off. I suppose that is what is happening, but that trend is not clear from this wobbling line. Maybe I am wrong?

Presumably, there are heat losses increase even with this well insulated box, and they increase until it reaches a terminal temperature. I assume each blue dot is the terminal temperature for that power level. In other words, this is a gigantic isoperibolic calorimeter. If the insulation was much better, that would make it a gigantic adiabatic calorimeter. I have a feeling the temperature would rise a lot more until it got very hot. (Eventually, any adiabatic calorimeter has to begin losing heat until it reaches a terminal temperature, making it isoperibolic.)

The constant temperature chambers I have seen have heat pumps that work in either direction. So they can have compensation heating or refrigeration (heat pumping out), controlled by a thermostat. I think it would be very difficult to use one of these as a giant calorimeter. This data does not look like it came from something like that. Then again, maybe the blip at 100 W and inconsistent curve are caused by a heat pump?

Quote from JedRothwell

My guess is that this is a crude measurement, with a large error margin. I think the temperature rise per watt of input declines as temperatures rise. The line goes up steeply at first and then gradually levels off. I suppose that is what is happening, but that trend is not clear from this wobbling line. Maybe I am wrong?

Presumably, there are heat losses increase even with this well insulated box, and they increase until it reaches a terminal temperature. I assume each blue dot is the terminal temperature for that power level. In other words, this is a gigantic isoperibolic calorimeter. If the insulation was much better, that would make it a gigantic adiabatic calorimeter. I have a feeling the temperature would rise a lot more until it got very hot. (Eventually, any adiabatic calorimeter has to begin losing heat until it reaches a terminal temperature, making it isoperibolic.)

The constant temperature chambers I have seen have heat pumps that work in either direction. So they can have compensation heating or refrigeration (heat pumping out), controlled by a thermostat. I think it would be very difficult to use one of these as a giant calorimeter. This data does not look like it came from something like that. Then again, maybe the blip at 100 W and inconsistent curve are caused by a heat pump?

The temperature rise per W ‘curve’ looks a bit better when the 25 C background is subtracted.

I like the idea of a common device for a test machine that should be fairly consistent in response overall (more calibration points at each power step would clear this up) so that instead of nitpicking the various measurements of ‘home-made’ one-off calorimeters built by various people, the relative heat difference between activated and calibration is obvious and undeniable, in a sensible way, in an off-the-shelf item. It is hard to tell if this oven is as reliable a measurement device as it could or might be.

Quote from JedRothwell

The constant temperature chambers I have seen have heat pumps that work in either direction. So they can have compensation heating or refrigeration (heat pumping out), controlled by a thermostat. I think it would be very difficult to use one of these as a giant calorimeter. This data does not look like it came from something like that. Then again, maybe the blip at 100 W and inconsistent curve are caused by a heat pump?

I downloaded a specification sheet that seems to show that it is just a rather high powered convection oven. A manual exhaust valve lets the blown air out for cool down, otherwise it is recirculated. Takes an hour to reach 500 C, at peak 7800 W for one model, so probably lots of bricks or something in there besides fluffy insulation.

Quote from Paradigmnoia

I like the idea of a common device for a test machine that should be fairly consistent in response overall (more calibration points at each power step would clear this up) so that instead of nitpicking the various measurements of ‘home-made’ one-off calorimeters built by various people,

There is much to be said for a commercially made device. It is likely to be consistent. The company probably tested it extensively. But was it intended for this use? Or only to ensure a stable temperature background? I don't know. The constant temperature boxes I have seen (called "incubators" in Japan) were stable to within a fraction of a degree, but they were not intended to measure heat, as far as I know. They work like constant temperature water baths.

The only disadvantage to everyone using this would be if they all use it the same way. There is some danger that they will have the same systematic error. It is good to have a variety of different calorimeter types applied. However, for the first several replications, if everyone uses this box (or something similar) it would be okay. At some point I would like to see something like a large Seebeck calorimeter applied. I don't know if there is a commercially made one this large. There is that "whole room" calorimeter that fits a person. I don't know if it would be sensitive enough. You can also use a refrigerator as a calorimeter. Or a large thermoelectric picnic cooler.

May this be a blinding flash of the obvious, but from the specs I can only conclude the Yamato oven being a constant temperature oven, that once fixed controls the temperature within 0.5 degrees of the setting, is being used as a gauge of how much energy takes to maintain the oven at a certain temperature setting. An active cell producing excess heat should reduce the power input to attain the temperature set, at least within the range of temperature in which the active cell produces heat. A control, inactive cell should not induce any change in the power levels, all other things remaining the same.

Quote from Curbina

May this be a blinding flash of the obvious, but from the specs I can only conclude the Yamato oven being a constant temperature oven, that once fixed controls the temperature within 0.5 degrees of the setting, is being used as a gauge of how much energy takes to maintain the oven at a certain temperature setting. An active cell producing excess heat should reduce the power input to attain the temperature set, at least within the range of temperature in which the active cell produces heat. A control, inactive cell should not induce any change in the power levels, all other things remaining the same.

That is exactly how I ran some of the tube furnace experiments. Every time the thermostatic heater kicks in the data logger records the on-time and the current. Although the current was notionally constant, some of the heater inputs were so brief that they beat the 'rise time' of the heater coils, so it was always measured and timed

Comparison of test and control is relatively easy then, once you establish the complete calibration for any given furnace.

Quote from JedRothwell

There is much to be said for a commercially made device. It is likely to be consistent. The company probably tested it extensively. But was it intended for this use? Or only to ensure a stable temperature background? I don't know. The constant temperature boxes I have seen (called "incubators" in Japan) were stable to within a fraction of a degree, but they were not intended to measure heat, as far as I know. They work like constant temperature water baths.

The only disadvantage to everyone using this would be if they all use it the same way. There is some danger that they will have the same systematic error. It is good to have a variety of different calorimeter types applied. However, for the first several replications, if everyone uses this box (or something similar) it would be okay. At some point I would like to see something like a large Seebeck calorimeter applied. I don't know if there is a commercially made one this large. There is that "whole room" calorimeter that fits a person. I don't know if it would be sensitive enough. You can also use a refrigerator as a calorimeter. Or a large thermoelectric picnic cooler.

Does the reactor require any added heat to activate it when placed in an oven?

I mean, the average temperature of the reactor at 300 W was shown to be about 50 C, with hot spots as high as 180 C or thereabouts. So if the oven were set to 175 C, should not the reactor fire up without the external or internal heater and start making excess heat of over 100 W?