I couldn't find this topic so I started.

Gas Calorimetry

Using a mass flow gas controller as input to the thermal exchange process should give good accuracy. A mass flow controller has better than 1% control of the mass flow (I.e. the thermal mass ) of the gas being controlled. A simple temperature measurement of the out flow gas multiplied by the flow rate of the mass flow controller should be a good measure of the heat produced by the system. Some precautions: Using ambient air is a bad idea as a source for the mass flow controller. I think dry nitrogen is a better, but more expensive solution. The issue is the amount of water vapor and CO2 in the air. (Human breath is about 7% CO2 vs 380 ppm of clean air) A good dehydration filter would mitigate the issue, but needs scheduled maintenance to verify long term utility. Stratification of the output air flow , from thermal gradients needs to be controlled with a static inline mixer on the exit stream prior to the temperature measurement. Good thermal insulation is required for the whole apparatus.

jdk

Hi JDK,

Some great suggestions. We are about to publish a video on the design of the calorimeter, please could you add your suggestions to the blog post when it is up.

I had been worried about the humidity issue and your suggestion may be relatively simple to integrate.

When you submit your suggestions, could you also provide links to suppliers of relevant equipment.

Thanks loads

Bob

There are multiple suppliers of mass flow equipment. see:

www.alicat.com/products/gas-flow/mass-flow-controllers/

They have a wide range of controlled flows from 0.5 milliliter to 3000 liters per minute. There is a digital input RS485 to control set points. They claim a 200:1 turn-down control over flow, but I would never make a design that pushed too close to the limits.

Using a 50:1 flow control range, a calorimeter could be set up for baseline output of say 10 watts with a low flow setting. The temperature difference between input and output streams could be noted. As the experimental power output increases, the flow rate could be increased to maintain a constant delta temperature. Using simply the flow controller, an experimental power of about 500 watts could be easily measured with the same delta T as the 10 watt baseline. Also, the relatively low temperature increase simplifies thermal insulation choices.
This design optimizes output power measurement at the expense of temperature control of the internal heating mechanism. During increased power production, the internal device temperature will need to rise, since the heat flow through the tube walls needs to increase. This external thermal flow is driven only by the temperature difference between the inner and outer surfaces of the tube.
Controlling flow rates will increase convective cooling of the outer tube, but convection will not be 100% effective. This lack of perfect instantaneous thermal coupling between the inner and outer regions will be compensated for by an increase of the inner chamber temperature.
Keeping the delta temperature small with flow control makes the calculations easy.

jdk

I just saw the video on the upcoming air calorimeter. I would classify it as a constant flow device with variable temperature output. The new device seems quite complicated. I'm not sure how to model it in my head yet, but..
I previously have made a mental model of a "constant temperature" model with variable flow. I will try to paint a verbal image of it since I think it could be implemented relatively simply.
The LENR reactor would stay the same: a flanged tube with electrical and gas connections from either end. I would place the reactor inside of a slightly larger tubular flow chamber with very good thermal insulation from the outside world. The gas and electrical connection would have to penetrate the insulation. The flow chamber has an entrance port, say a swagelok tube fitting and would be supplied with nitrogen from a "gas mass flow controller". These devices range in cost from $500 to about $2000. Modern controllers are highly accurate and stable- easily better than 1% over many days of operation. They are digitally programmable and have wide dynamic ranges- better than 100:1. in flow mass. Since they control mass flow, changes in atmospheric pressure have no impact. If more dynamic range is needed, controllers could be operated in parallel. Start the experiment with all the controllers at a minimum flow level. As the power output increases, increase the the flow of controller A until it is at maximum. Then start increasing the flow through controller B etc. With enough controllers, almost any level of power output could be measured.
The temperature difference between input gas flow and output gas flows would be the only measurements required. Assuming the calorimeter is not going to measure high speed events, nearly any thermal measuring tool would work. We need only to measure the static equilibrum temperature of both sensors. High accuracy rather than speed of response is most important.
Obviously I favor a variable flow instrument. The flow contoller solves a lot of problems and greatly simplifies design. All of the gas handling instrumentation issues have already been mastered by the controller manufacturer. We should take advantage of their successes.

jdk

I have been trying to add comments, but the "Add FIle" buttons do not seem to work.

A single ended measurement system with 3 or 4 orders of differential interaction is somewhat difficult to monitor since all the temperatures interact. A change in the external ambient temperature changes all the other temperatures in the system. The present LENR replication experiment is experiencing discussions resulting from the system complexity.

A simpler system is presented by a fixed temperature/variable flow arrangement. The flow rate can be very accurately determined and temperatures can be very accurately determined. In a calorimetric system determined by only the flow rate and the temperature difference between the entrance and exit flow locations, a nearly differential measurement is presented. The system has a flow through time, determined by the volume of the tube and the flow rates. Assuming ambient thermal stability with a longer time constant than the flow through time constant, the fixed temperature/variable flow system becomes a differential measurement system. The zero subtraction issue (baselne stability) is minimized and shorter term thermal drifts are removed from the interpretation. I think it is a much better measurement plan.

flowmeters:


www.alicat.com/products/gas-flow/


jdk

This design eliminates the need to precisely know air mass or its heat capacity.
Incoming air is first heated to a stable temperature, then it is heated again by approximately 1 degree. The exact temperature rise and wattage required to achieve it are measured. This gives a temp rise per watt at the current air speed. It automatically factors out issues about air pressure, air makeup, exact flow rate, and humidity.

Then the air flows through the test chamber. The resulting temperature rise is measured. The rest is simple math.
In a perfect world that math would be linear, but at different wattage levels imperfect insulation or other factors put a curve in the calibration. The system can automatically step a heater (located in the chamber) through a series of wattage levels. At each wattage a new calibration point is established. The resulting calibration curve is used to reduce real world errors.


The main things we need to watch out for are uneven temperature and air speed as it flows over the sensors, and insulation losses. We are using several tricks to minimize each of those issues.

I found this in my searches
extrardinary description of Langmuirs early work with hydrogen. A very important read.

www.gifnet.org/articles/Langmuir%20&%20Atomic%20Hydrogen.pdf

He describes the thermal conductivities of various hydrogen mixes with various gases. He also talks about atomic hydrogen. He also talks about the influence of tiny amounts of water in the reactor cell chemistry.

Read the whole thing. Important observations are buried through out the entire text. Langmuir was trying to improve incandecsent light bulbs, but it seems he talks a great deal about phenomena relevant to LENR.

jdk

I agree that airflow calorimetry is a good idea. However, to achieve 1C or better repeatability it is necessary to maintain a fixed thermal mass flow per unit time. A fan connected to a DC supply will not suffice. What is needed is a closed loop system that monitors thermal mass flow and provides an error correction feedback path.

Consider a pair of diodes suspended in the input air stream. One diode is heated by a thermally bonded resistor such that it maintains about a 5C temperature differential wrt the other diode for a given rate of airflow. The two diodes' respective junction voltages drive an opamp and a buffer amp, the latter powering the DC fan. Any change in airflow thermal mass will register as an error signal and adjust the fan speed. I have simulated the above described circuit in SPICE, and it appears to work beautifully. I'll build it as soon as the parts arrive.

Another good thing: the electronics for this thing cost about $20.00 and are easily obtained from any electronics parts supplier.

Jeff

This evening I completed testing of a closed loop controller for the airflow calorimeter. It consists of two diodes suspended in the airstream: one heated by a resistor and one unheated such that, with the nominal airflow, there is a ~2C temperature difference. Each diode is driven by a 100 uA current source, and the two diodes feed a difference amplifier with a gain of ~500. The difference amplifier drives a buffer amplifier, which in turn drives a small DC fan. Once calibrated the system ensures a fixed thermal mass airflow sufficient to maintain the preset temperature difference between the two diodes. The linearity of the system is proportional to its open loop gain. Over a 15-30C ambient temperature range the controller maintains a fixed thermal mass flow to within +/-0.3%. Now I can proceed to calibrate the on/out temperature differential vs. the heater power applied to the cell. Then it's on to loading the oxidized NiCu wire and looking for excess heat.

Excellent Jeff!

Please do consider using Celani's V2 protocol as it removes sooo many variables and means you can use a lot lower power in calibration and active runs to reach the higher wire temperatures needed for triggerring.

We will be looking to re-rganise the site and could have it so that other experimenters such as yourself could have your own blog thread for people to follow and assist you on with commentary.

Also it would be possible for you to join the live feeds.

B