LION Dummy Run Shows Rising Trend in Radiation Count (Alan Smith)

This is a post from Alan Smith in this thread where he reports about a new test he has run.

Here are a couple of data-plots from a dummy-run using part-prepared diamonds (soaked for 7.5 days in D20, not 30. They were pre-baked for 7 days. Maybe a tiny amount of anomalous heat there, but no cream cakes. The most interesting thing is in the lower graph showing a steadily rising trend in radiation count. Thermo 0 is the ‘live’ port, thermo 1 is the control.

Hot plot – rad count is way down in the clutter along the bottom line.


Expanded ‘clutter zone’ with trend line for radiation detected using a Netto Geiger with SBM-20 tube 6 cms from the live fuel tube.


  • Bruce__H

    The green line in the lower plot showing the trend in radiation count should be viewed with care. It rises steadily only because a line with few degrees of freedom has been fitted to the wiggly light-brown line showing instantaneous counts.

    In reality, it seems to me that it would be more accurate to say that average radiation counts underwent a step change about 2/5 of the way into the session followed by a very slight decline.

    An interesting observation is that the step increase in radiation counts coincides roughly with an increase in temperature after following an era of temperature stability (the plateau in the upper traces in the first plot).

    • Alan Smith

      “An interesting observation is that the step increase in radiation counts coincides roughly with an increase in temperature after an era of temperature stability (i.e., after the plateau in the upper traces in the first plot).”

      Otherwise known as ‘lunchtime’.

  • Da Phys

    Dummy or not dummy, that is the question.

  • Bob Greenyer

    Looks like it could be dismissed as a little thermal creep related to me.

    A control test for this would be to put it next to a heater or a real dummy cell.

    • Ophelia Rump

      If that variance has any potential significance he should consider reversing which is the live port and which the control, to rule out sensor variance. That would fit your suggested control test well enough.

      • Bob Greenyer

        On temperature – I am referring to Geiger counter count trend

        • Alan Smith

          Yes, of course. However, against that is the following observation – when the count was steadily in the 30-40 cpm region, putting a 2mm thick piece of Al (for a couple of minutes at most) berween reactor and Geiger produced an immediate decline to the lab normal background of 20-25 cpm. Not proof of anything, but since the shielding wasn’t IMHO in place long enough to let the SBM-20 detector tube cool down significantly does suggest it isn’t a purely thermal effect on the Geiger.. But all these factors will be subjected to proper controls as we progress.

          • Bob Greenyer

            These are good tests

      • Alan Smith

        He should and he does.

    • Alan Smith

      Hi Bob. No shortage of little thermal creeps, that’s for sure. I didn’t claim anything for this observation, Frank chose to highlight it- originally I just slipped it into the ‘Control 3’ thread. There is also a marginal amount of XSH observed in the test port, but that might be artifact too.

      I will be happy to post the data-file when the data guardian returns from darkest Wales.

      • Anon2012_2014

        Got it. Just read it. Stay warm there. Weird weather.

  • Anon2012_2014

    Unfortunately we are looking at statistical noise. Fit a linear regression to it. I don’t think the r^2 or the T or F statistics are significant yet.

    • These tests are cheap and easy to perform, many more tests will be performed by Alan and others. For Alan, this is just one of his tests runs in preparation for his and others attempts to replicate Lion. At the end of the day, what will truly matter the most is verification performed by completely independent third parties.

  • Anon2012_2014

    @Alan Smith,

    Can you send Frank a spreadsheet with the radiation counts so we can run a regression at home on it? You sent him an old run but it doesn’t have the radiation counts channel.

    Thank you,


  • Pink Panther

    Can you post the raw data file please?

    • These tests are cheap and easy to perform and therefore many more tests will be performed by Alan and others. For Alan, this is just one of his tests runs in preparation for his and others real attempts to replicate Lion.

      At the end of the day, what would bring most value is verification performed by completely independent third parties.

      While waiting for some quality research I’m still working hard on calibrating my own ‘Friendly Robot’ system. It seems the Mentos I’m using in my active core somehow is interfering with the datalogger – see for yourself!

    • Alan Smith

      Yes- it will arrive. Post snow it seems The data is stuck in a boarding house in Swansea.

      • Anon2012_2014

        Anything more Alan above the noise. I wouldn’t expect anything because this was a dummy control run. Just want to confirm.

        • Alan Smith

          Nothing happening here except the lab is mostly packed- we are moving our lab into a much bigger space – it was supposed to happen tomorrow, but due to the snow the painters have been delayed, and so the move is now officially beginning on Monday, though the property managers are letting us move some equipment in before then. Furniture installers, electricians etc will hopefully ‘get on with it’. We hope to be back in action by the 15th at the latest.

          Here’s a link to a video of the empty unpainted space. Solid, spacious, and secure.

          • Stephen

            Looks great. Has the feel of place where history will be made.

            Good luck with the move

  • Anon2012_2014


    Please ask Alan Smith to send us a spreadsheet of the radiation counts vs time so that we can numerically analyze the above. The null hypothesis is that is just noise, and that the very slight increase noted followed by a decrease is perhaps associated with a solar flare or similar geomagnetic event, or heating of the GM tube changing its counting efficiency or its own noise.

    Thank you.

  • Monty

    i think more data is needed to decide if there is something going on there. Right now everything is quite noisy.

    • Ophelia Rump

      Isn’t the point of a dummy run to determine a normal baseline and test the rig for irregularities? With that in mind, the idea that “something is going on” should not even be entertained.

      You do a dummy run precisely so that you do not consider the result you see on the dummy run to be significant if you see it on the test run.

  • US_Citizen71

    The X axis is not labeled so I am going to assume that it is seconds. If so this test lasted for about four hours. If this is the case wouldn’t the obvious cause for the radiation count variance simply be solar or cosmic radiation changes? i.e. the sun moving through the sky

    • Alan Smith

      See Monty’s comment below. Don’t get things out of proportion just yet. It might have been my new Rolex that did the damage.

      • Rene

        Old Rolex watches used Radium for luminescence, then they switched to Tritium in the mid 1960s, but the newest, post 1990s, Rolex watches use photo-luminescent materials, either Luminova or Chromalight.

  • Alan Smith

    Update on the move into our new lab. Hopefully the landlord’s painters will do the business this weekend, I will be extremely pissed if they don’t, the pile of sheeted equipment on the lab floor (and my sense of lost time) grows by the day! Furniture fitting is booked to begin on Monday midday, a hefty new lathe and milling machine are arriving on Wednesday. I am organising the lab so we can use the space for interactive seminars if required, maybe 10-15 people at a time – I will be able to confirm numbers when the fit-out is done. Around half of the new lead-block wall came yesterday, delivered by a chap around the size of a small house. I’m so glad he didn’t hand me the crate without a warning about its weight!

  • Alan Smith

    Update on a second dummy run with 7/30 prepared diamonds. Just over 100 hours into the hot zone. The reactor was given a slow ramp up to 900C at the start, and was then cycled between 800/900C for many hours, with one run up to 1000C and then back to 800, where it has remained (unattended) for the last 40 hours approx. Going to check on it soon – no b/band in the lab for at least another 10 days so a personal visit is required. If anything interesting has/is occurring I will report later.

  • Alan Smith

    Here’s the latest state of play- preliminary testing of a (close to) fully isothermal reactor- minimal convective losses improve sensitivity- this is ‘first heat’ in this one.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies.