Hi Dalibor, Great, good to hear.
The next step (more like parallel step) would be to setup an experiment with a numerical nixie. For reasons that I will elaborate on in a future post, I believe testing numerical nixies will be easier. The experiment was picked up by Hack-a-Day<http://hackaday.com/2013/11/24/measuring-the-lifespan-of-nixie-tubes/>, which was fortunate as a commenter (Tyler Laseter, I believe) had the idea of severely over-driving the tube under test to establish a relationship between darkening rate and current (Similar to Stocker's experiments). His suggestion was then to extrapolate to normal operating currents. As I explained to him as far as the IN-13 is concerned one cannot extrapolate to normal operation as any equations would be developed at completely different operating regimes. However, as a numerical nixie is already operated in the abnormal glow region maybe, just maybe, we can work out a relationship in a region not too far off from normal operation therefore not too inaccurate. I think the promise of getting results in a shorter time for numerical nixies makes at least attempting this worthwhile. In that case 3-4 samples of tube X will be needed, preferably no mercury or other inhibitors to prove the method. Even defective tubes with a single cathode working should work. More coming. Alex. On Tuesday, November 26, 2013 6:59:33 AM UTC, Dalibor wrote: > > Hi Alex, > > I can now access the website finally, interesting! About the Weston > book, You are right. It is too general and low level, there is no way > how to calculate actual sputtering rates using information provided > there. Also as You mentioned, they deal with low pressure mostly. But > there are some interesting information such as the experiment with one > shape of anode grid and different pressures. They found that the rate > of deposition of the metal on the glass at 20torr was much lower than > at 50torr, there is a relation between the shape (size) of the hole in > anode grid and pressure inside the tube. > > I think it is better to spend a few hours setting up the real > experiment than bothering with theoretical calculations in this case. > There is too many variables - metal composition, anode cage shape, > purity of the gas and so.. > > I am definitely interested in your experiment as it is far the closest > to our needs! > > Dalibor > > 2013/11/22 AlexTsekenis <[email protected] <javascript:>>: > > Hi Dalibor, > > > > I just got confirmation from Eastern Europe that they can access the > page. > > Can I ask who is your ISP/host in Czech Republic so I can check with our > > host. Thanks in advance. The same goes for anyone else that might be > having > > issues. Glad you have temporary access! > > > > It seems our timing is good ;-) Certainly, Weston's book is idea for > > bed-time reading.The chapter on sputtering (ch 4) is informative of the > > state of research in those days, but it falls short if you are reading > it to > > find out when the tube has reached its end of life. Here is why. > > > > The theoretical models presented (section 4.3) are too low level. The > > researchers tried to create models almost from first principles. The > > sputtering process is very complex, so they inevitably resorted in > number of > > assumptions and simplifications. As you go through the chapter you will > see > > that the results agree under some conditions, but disagree on others.I > think > > Weston mentions this somewhere. > > > > In terms of experimental work most setups were at too low pressures to > > sustain a glow discharge (fractions of a torr). Consequently the > > back-diffusion of cathode material that is present in commercial nixie > tubes > > was not present (as stated in top of page 138 and section 4.2.2) in > those > > experiments. Additionally the ion beam setups look nothing like a nixie > tube > > (mercury pool, charged collection target etc). > > > > As I described in SHINE, the most relevant experiments were those from > > Stocker (reference in SHINE page, page 140 in Weston's book). Stocker > used a > > cylindrical tube with a wire cathode and a cylindrical anode (much like > an > > IN-13 tube!). However he used a filament light source inside the tube, > > whereas we are using the emission from the glow itself as the light > source. > > We are not using internal light sources as anything carrying current or > > being at a potential will distort the fields inside the tube. Stocker > ends > > up with the empirical equation 4.18 in Weston's book or 2 in SHINE page > > which you one can use. But the equation only gives the sputtering rate, > not > > the loss of brightness. To put the final nail in the coffin, none of the > > above experiments or theories incorporates anti-darkening measures such > as a > > mesh anode like the IN-13 uses! > > > > We decided to measure brightness directly as at the end of the day > that's > > what causes one to throw the tube away. Doing so automatically accounts > for > > whatever processes might be taking place at an atomic level and any > measures > > taken by the manufacturer to increase the lifetime. Using this method I > > cannot do a calculation and tell you how long a tube you made will last > if > > you run it at 5 or 10mA. But I can measure it and then tell you with > great > > confidence :-) hope this clarifies the distinction. > > > > As long as chapter 4 (and other literature on sputtering) is concerned, > my > > personal feeling is that technology declined before the leaders at the > time > > had time to develop a unified model. What do you think? > > > > Alex > > > > > > > > On Friday, November 22, 2013 5:25:29 PM UTC, Dalibor wrote: > >> > >> Hi Alex, > >> > >> I think it will actually be something with settings of the firewall on > >> your server. The server doesnt respond to ping, from my IP and even > >> not from my server that is in a big czech telehouse. > >> > >> however, I checked your site from the proxy and there is plenty good > >> information! > >> > >> I was just reading about sputtering two days ago, (Weston, 1968), > >> there were mentioned some experiments for measuring the deposit > >> thickness, have You already read it? > >> > >> Bye! > >> > >> Dalibor > >> > >> 2013/11/22 AlexTsekenis <[email protected]>: > >> > Hi Dalibor, > >> > > >> > Thanks. I'm surprised you cannot access the site, no issues here from > >> > multiple computers. Maybe try from the home page > >> > http://www.saltechips.com/ > >> > and clearing your cache. Please let us know if you are still having > >> > issues, > >> > there is a wealth of information on the site including equations, > plots > >> > etc. > >> > > >> > We have data for the first 7k minutes (6810 to be exact), but I would > >> > like > >> > to collect more before drawing any conclusions. The plan is to do > curve > >> > fitting on the collected data points, thus suppressing noise and > other > >> > artefacts. If you follow our theoretical working you will see that > the > >> > light > >> > emitted is falling exponentially, so we expect the data points to fit > >> > such a > >> > curve: > >> > > http://www.ibm.com/developerworks/rational/library/mar06/cantor/fig7.gif > >> > The > >> > sampling rate of the data published is currently slow (1 point per > >> > week). > >> > This Sunday is time for the second measurement. I will need to judge > >> > whether > >> > to increase or decrease the sample rate then. > >> > > >> > Yes, you are right on current affecting sputtering rate. Also the gas > >> > pressure and composition as well as the electrode material may I add. > >> > This > >> > is mentioned in the description and an empirical equation is given. > >> > However > >> > the method we are using is independant from these factors - we are > just > >> > interested in the rate that the glass becomes darker. So the > experiment > >> > can > >> > be repeated for different currents and tube constructions to optimise > >> > the > >> > tube and operating point. > >> > > >> > We are running it at about 4mA with a constant current sink. This > makes > >> > the > >> > glow reach about 3/4 of the tube's length. In other words the tube is > in > >> > the > >> > normal glow region and we are 1/4 away from abnormal glow. In the > normal > >> > glow region increasing the current increases the area of cathode > being > >> > covered by the glow. These conditions were selected to best replicate > >> > typical usage of this particular tube, thus giving the most accurate > >> > lifespan figure possible. > >> > > >> > How is website access now? > >> > > >> > Alex > >> > > >> > > >> > On Friday, November 22, 2013 2:16:00 PM UTC, Dalibor wrote: > >> >> > >> >> Hi Alex, > >> >> > >> >> this is pretty interesting! I cant access the website, it is > probably > >> >> unavailable now, but the stream is clear! Do You have any graph how > >> >> the absorbance rise during that those 7000 minutes? > >> >> > >> >> Are You running it on specs current or on higher to make the process > >> >> faster? The sputtering is not directly proportional to current, so > it > >> >> would be nice to have data for currents from datasheet.. > >> >> > >> >> Thanks! > >> >> > >> >> Dalibor > >> >> > >> >> 2013/11/22 AlexTsekenis <[email protected]>: > >> >> > It is my great pleasure to announce that the SHINE (SALTechips > IN-13 > >> >> > Endurance) experiment is now online. > >> >> > > >> >> > For the first time since the Cold War era we are able to > rigorously > >> >> > measure > >> >> > the end of life of a nixie tube due to darkening of its glass > >> >> > envelope. > >> >> > We are super nixited as we can now accurately answer the frequent > >> >> > question > >> >> > of the lifespan of a given nixie. > >> >> > > >> >> > The experimental method relies upon a theoretical basis and custom > >> >> > precision > >> >> > instrumentation developed by SALTechips specifically for this > >> >> > purpose. > >> >> > In summary, we are monitoring the optical absorbance demonstrated > by > >> >> > the > >> >> > envelope of the tube as cathode material is being deposited by > >> >> > sputtering. > >> >> > The absorbance increases with time, thereby causing the tube to > >> >> > become > >> >> > dimmer. At some time a brightness threshold is reached which we > treat > >> >> > as > >> >> > the > >> >> > end of life of the tube. > >> >> > The method and experimental setup are described in detail here: > >> >> > http://www.saltechips.com/lab/ > >> >> > This time we are measuring the IN-13 bargraph tube that is of > >> >> > immediate > >> >> > interest to us, although the tools and techniques can be used on > most > >> >> > glow > >> >> > discharge tubes. > >> >> > > >> >> > We anticipate the experiment that is currently in progress to take > a > >> >> > while.Data is plotted live as it is collected and we also have a > live > >> >> > stream > >> >> > of the instrumentation screens available 24/7(follow link above or > >> >> > directly > >> >> > here: http://www.ustream.tv/channel/from-the-salte-lab). These > are > >> >> > the > >> >> > same > >> >> > screens we use to record data! Nearly 7000 minutes into the > >> >> > experiment > >> >> > and > >> >> > we are already collecting valuable data. > >> >> > > >> >> > The future of this method is bright as it is applied to measure > and > >> >> > compare > >> >> > other nixie tubes! > >> >> > > >> >> > > >> >> > Alex > >> >> > > >> >> > > >> >> > > >> >> > -- > >> >> > You received this message because you are subscribed to the Google > >> >> > Groups > >> >> > "neonixie-l" group. > >> >> > To unsubscribe from this group and stop receiving emails from it, > >> >> > send > >> >> > an > >> >> > email to [email protected]. > >> >> > To post to this group, send an email to [email protected]. > > >> >> > To view this discussion on the web, visit > >> >> > > >> >> > > >> >> > > https://groups.google.com/d/msgid/neonixie-l/b16aa111-40c9-49f9-b6c5-f0a5b4a788a0%40googlegroups.com. > > > >> >> > For more options, visit https://groups.google.com/groups/opt_out. > >> >> > >> >> > >> >> > >> >> -- > >> >> Dalibor Farny > >> >> http://dalibor.farny.cz > >> > > >> > -- > >> > You received this message because you are subscribed to the Google > >> > Groups > >> > "neonixie-l" group. > >> > To unsubscribe from this group and stop receiving emails from it, > send > >> > an > >> > email to [email protected]. > >> > To post to this group, send an email to [email protected]. > >> > To view this discussion on the web, visit > >> > > >> > > https://groups.google.com/d/msgid/neonixie-l/732ecf75-a44a-4ffb-b877-9f3bfad6072b%40googlegroups.com. > > > >> > > >> > For more options, visit https://groups.google.com/groups/opt_out. > >> > >> > >> > >> -- > >> Dalibor Farny > >> http://dalibor.farny.cz > > > > -- > > You received this message because you are subscribed to the Google > Groups > > "neonixie-l" group. > > To unsubscribe from this group and stop receiving emails from it, send > an > > email to [email protected] <javascript:>. > > To post to this group, send an email to > > [email protected]<javascript:>. > > > To view this discussion on the web, visit > > > https://groups.google.com/d/msgid/neonixie-l/5b1ece2d-1b69-47a7-8980-66787cda3dcd%40googlegroups.com. > > > > > > For more options, visit https://groups.google.com/groups/opt_out. > > > > -- > Dalibor Farny > http://dalibor.farny.cz > -- You received this message because you are subscribed to the Google Groups "neonixie-l" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/0851e5ea-4726-496a-b563-572a871f4076%40googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
