Mizuno reports increased excess heat

Looking for the exact fan for the calorimeter, I began also comparing published air flow rates for similar fans.

4 m/s air velocity in a 65 mm diameter pipe, at Standard Pressure seems to work out to 112.5 28.2 cfm. (more than my bathroom fan)

Edit: I caught my previous error above. The cfm of the fan seems to be right on the money for the air flow rate claimed compared to higher-end fans rated for 12V 5W.

Most fans in the projection design (side outlet) 12V 5W range seem to only be rated for 10 to 30 cfm.

So far, this one in the image below has the highest listed flow rate that I have been able to find the flow rate for.

.

Here is some more specifications from different companies..

.

One of those, maybe the 5 W, was 2 bucks each if you bought 200.

This is the one reported in the 2017 paper, however the photo is different.

Sanyo Denki Instruments San Ace B97, Model 109BM12GC2-1

https://products.sanyodenki.co…e/dc/blower/109BM12GC2-1/

The 24 vdc version is available from stock at Mouser Electronics, around $18:

https://www.mouser.com/Product…%252BC2DxXSdYKmf0SQ%3D%3D

Quote from seven_of_twenty

With liquid coolant instead of air, you should be able to get any temperature you want for the reaction by controlling the flow rate of the coolant or if necessary, even adding heat to it via an appropriate control system. See again SGVIT's example of a high temperature calorimeter using liquid coolant mass flow which could be made to fit Mizuno's reactors. This link should be the Google translate English version The particular example in this paper is for a 2kW heater-- well in the range of Mizuno's postulated 3kW device. The form factor fits too.

Agreed. Our hypothesis, supported by some experimental work, is that Ni should be kept at a high temperature to favor H spillover. Because H spillover is known to increase with temperature and metal dispersion. Interesting that Mizuno work goes in that sense as well. But, as you rightly said, this does not prevent the use of liquid calorimetry.

A few comments regarding a fan-based flow calorimeter. The temperature delta between input and output will not be linear with respect to the power dissipated internally. As the air inside the calorimeter is heated it expands and produces a back pressure against the fan. So there are two factors affecting back pressure: a static back pressure and one that is temperature rise dependent. Typical fan performance curves show a decrease in airflow as back pressure increases. It's possible, of course, to work with a non-linear power vs. temperature rise curve by first running a calibration over the desired range of power levels. However, changes in the ambient temperature or variations in fan performance will introduce errors.

The other approach is to construct a fan controller that guarantees a constant thermal mass flow. This can be done by placing a pair of diodes, one resistively heated and one not heated, in the inlet airstream. The forward voltage drop across each diode is then sent to an amplifier circuit that generates a variable DC voltage to power the fan. The temperature delta between the two diode junctions is proportional to the thermal mass of air moved by the fan, since the heated diode is cooled somewhat by the passing airflow. A plot of power vs. temperature rise then yields a nearly linear relationship. Note that DC power is supplied to the heater, so precise V and I measurements are possible. If anyone is interested I can provide circuit details and possibly PC boards.

Quote from jeff

A few comments regarding a fan-based flow calorimeter. The temperature delta between input and output will not be linear with respect to the power dissipated internally. As the air inside the calorimeter is heated it expands and produces a back pressure against the fan. So there are two factors affecting back pressure: a static back pressure and one that is temperature rise dependent. Typical fan performance curves show a decrease in airflow as back pressure increases. It's possible, of course, to work with a non-linear power vs. temperature rise curve by first running a calibration over the desired range of power levels. However, changes in the ambient temperature or variations in fan performance will introduce errors.

The other approach is to construct a fan controller that guarantees a constant thermal mass flow. This can be done by placing a pair of diodes, one resistively heated and one not heated, in the inlet airstream. The forward voltage drop across each diode is then sent to an amplifier circuit that generates a variable DC voltage to power the fan. The temperature delta between the two diode junctions is proportional to the thermal mass of air moved by the fan, since the heated diode is cooled somewhat by the passing airflow. A plot of power vs. temperature rise then yields a nearly linear relationship. Note that DC power is supplied to the heater, so precise V and I measurements are possible. If anyone is interested I can provide circuit details and possibly PC boards.

Although the temperature increase may cause a relative pressure increase inside the calorimeter, it seems to me that the Mizuno calorimeter will normally be in a state of weak vacuum relative to the atmosphere. This is because the inlet and outlets are 50 mm, while the fan is pulling air out at a fairly high rate. For a 4 m/s flow in a 65 mm pipe, the 50 mm holes will experience 6.6 m/s flow rates just to balance the flow outwards. Both 50 mm holes are restrictive in this case, however the one with the fan attached is effectively part of the fan ducting (which may have a similar effective restriction), whereas the atmospheric air inlet has only atmospheric pressure to push the air in to the box to keep it filled. That atmospheric air won’t flow in unless there is lower pressure in the box.

In my opinion it is better to measure the pressure in the outlet tube and correct for that than to try to modify the fan output to constant mass flow, which creates new variables. I would prefer that the fan spins at the same speed at all times, or at least operates at the same power at all times. However the earlier suggestion of an automotive type MAF sensor may be a good one if one can be found in an appropriate diameter.

Jed, I am curious what specific problems you ran into when trying to do liquid cooling? Sharing with the group, the things that didn't work and the "why" would be very helpful for a lot of us.

Quote from JedRothwell

As noted, he used water calorimetry but it cooled the cell down too much. It was difficult to make it cool just enough but not too much (Goldilocks effect).

Understood that. Its why i was suggesting insulation between the mesh and reactor wall. It would allow the mesh to run at the high temperature needed for the reaction to work while allowing water cooling of the reactor wall. I guess it might be difficult to work out exactly how much insulation would be needed.

PS I'm assuming the reactor wall isn't causing the effect.

No excess heat from a well conducted Mizuno replication

Quote from Curbina

You did not read Mizuno’s paper then. As magicsound pointed out, Deneum is working completely outside the low pressure range that Mizuno said is critical to achieve success.

My understanding is that they did run it at low pressure (300 Pa), except at the very end where they injected 30, then 100mbar. So they worked outside the low pressure range but only after getting a null result.

"...

Optimum pressure is between 100 and 300 Pa. It should not exceed 6,000 Pa. The reactant
will probably not load at less than 100 Pa. However, as shown in Table 1, once it has loaded,
pressure can be pumped down as low as 2.3 Pa and the reaction continues.

..."