Burn out Event Data

Geschrieben von Ryan Hunt am 23. Oktober 2012.

For our loyal follower Ecco and anyone else out there that is interested, here is an analysis and the data for the event that happened and burned out our wire during a calibration with NiChrome wires in Hydrogen.

We saw evidence of 2 separate wire shorting events. The first one made the cell overheat (T_mica went up to almost 500) and the second one maxed out the power supply, making the voltage dip till the wire on the blue channel burned out.

Be impedance is the same thing as electrical resistance in this case because the power is DC. When the shorts happened, the resistance dropped.

For anyone looking to explore this further, you can dig into the spreadsheet directly: MS Excel File or Open Office version

< Zurück Weiter >

Add comment

Comments

0 #6 Ryan Hunt 2012-10-31 19:14

We looked into this power ripple and I can confirm that it is in the voltage supply and seems to be related to the cooling fan cycle.

At 22 volts it cycles at about 7 minutes while at 42 volts it is about 4 minutes

We have some changes to the we control the power supply that may make it operte more smoothly. Definitely an issue to get a hold of sooner rather than later.

Quote

0 #5 Ryan Hunt 2012-10-31 13:39

Great analysis, Scalare. Thank you very much. I think this very much merits a 24 hour run at 48W on one wire. So far we have been busy troubleshooting bugs, but we should be able to slip that in real soon. We are also working on making the data available much easier and in a better viewer so you could follow it live. Software is one of our limiting factors at the moment, but we are working on it.

Quote

0 #4 Scalare 2012-10-28 09:11

Quoting Ryan Hunt:

Notice the little ripple on the Glass_out temperatures? I'm guessing that may be convection oscillations.

Hello Ryan,
In my opinion It seems to be more depending on the total power ripple:

See also at total power of 46W the effect on T_glass of the 0.6W periodical small drop.

Is it stiil planned the long test at constant power mentioned in your answer to Ecco some days ago? It would be useful to avoid possible effect of slow transient when you will replicate Celani’s exp. (at 48W only on NiCr wire, I suppose). It is possible to appreciate a small increase of T_glass in the last 15 min., more evident on T_well hereafter. Could I propose to make a test at 48W only applied to a single wire for 24h?

Quote

0 #3 Jeff 2012-10-24 01:09

Quoting Ecco:

Thanks for the graphs. Unexpectedly, it's not immediately clear just by reading them that a physical failure in the cell occurred. One could even mistake this event for something related to unexpected LENR effects.

It is necessary to keep eyes wide open and crosscheck as much data and visual clues as possible to avoid any possible misinterpretation of apparently anomalous reactor behavior.

On that regard, using a transparent cell has been in retrospect a very good idea.

Good point, but basically always showing input and output power clearly should make it easier to understand if there is no (or if there is) production of energy. I understand output may have be derived, but input is more direct.

Quote

0 #2 Ryan Hunt 2012-10-23 18:58

Notice the little ripple on the Glass_out temperatures? I'm guessing that may be convection oscillations.

Quote

0 #1 Ecco 2012-10-23 18:06

Thanks for the graphs. Unexpectedly, it's not immediately clear just by reading them that a physical failure in the cell occurred. One could even mistake this event for something related to unexpected LENR effects.

It is necessary to keep eyes wide open and crosscheck as much data and visual clues as possible to avoid any possible misinterpretati on of apparently anomalous reactor behavior.

On that regard, using a transparent cell has been in retrospect a very good idea.

Quote

Refresh comments list
RSS feed for comments to this post

JComments