I'm not familiar with the spark gap usage, nor the exploding wire thing.?
As I see it, mainstream electronics does not use spark gaps because they are too variable.?
There are effects due to the small geometry that would be needed to protect against say 8kV.
For instance, pits, or worse, spikes in the electrodes caused by previous discharges.?
Also practical PCB design would try to put the spark gap on a surface, and is then prone to humidity, tracking, cathodic filaments etc etc.
I can see it works for EHV pylons, with at least a foot of gap, but even so, a Spice model would be hugely variable.?
For mainstream
- such as LTSpice would want to focus on -
there are inexpensive sintered ZnO varistors with stable characteristics.?
Actually even EHV mains "switchyards" use similar ZnO devices, about the size of a barrel, because they absorb the energy - whereas a spark gap doesn't, much.?
So, interesting though it is, where are we intending to go with spark gaps?? ?What problem are you intending to solve??
Ultimately, what is the purpose??
For me, the purpose of LTSpice is to prove the intended behaviour, and to validate the design against component variation.?
As a bonus, it normally offers great insight and clarity, allowing further refinement - especially for small discrete circuits (2~6 transistors).?
On the flip side, it can prove that a given circuit is rubbish, doesn't work as intended.
Both good reasons to love it.?
|
"the uniqueness can be measured, but not anticipated prior to the event."
The limitations of a conventional spark gap? can be reduced somewhat by using an? exploding wire between two electrodes. This allows for a degree of freedom in the operating voltage. I used this technique with a 200uF/10kV capacitor. The wire was exploded with a much smaller capacitor.
|
Re: POSH (Powershell) tool to manage "user approved" LTspice cmp files
On Tue, Jul 11, 2023 at 11:28 PM, eewiz wrote:
Hello All
?
eetech00 wrote: I am looking for someone to help test. Anyone interested?
?
I volunteer.
Proverbially, I find myself standing on the field after the rest of troop took two steps back.
But that's alright, I'll enjoy the opportunity to delve into a working powershell script.
Thank you for all of your programming and modeling efforts.
Contact eewiz@... if you feel the need to communicate directly.
?
All for now
Hello eewiz I've uploaded a .zip file containing powershell script "DoLTspiceUpdate_ps1_20230713.txt". You will need to rename this .txt file to "DoLTspiceUpdate.ps1". To run the script: 1. Open (non-admin) command prompt. 2. Change current directory to script location, then type: ?? ? powershell.exe .\DoLTspiceUpdate.ps1 <enter> Running the script with no command line args (as shown above) will show "USAGE" information. ?? ? A logfile will be created "DoLTspiceUpdate.log" containing script screen output. Comments welcome...
|
#1 rule, It is not deterministic. Every spark gap is unique at some level. There is a distribution curve that can be exploited, in terms of multiple sampling and ratiometric sampling. Both amplitude and time can be exploited. the uniqueness can be measured, but not anticipated prior to the event. and you can have more than one spark gap as well, but one current coupling them all together, i think this would have a tighter bell curve via synchronous, similarity induced averaging. the hot spark get damped by the colder ones, and the cold ones get sped up by the hot ones (think faster)
Trying to grasp the resistance in a spark gap I had to write what I know.
If anyone can refute my ideas, let me know.
The voltage and current relationship when a spark gap's gap is increased is as follows:
??? Voltage: The voltage required to initiate a spark across the gap increases
??? as the gap widens. This is because the air in the gap has to be ionized
??? before a spark can occur, and it takes more voltage to ionize the air as
??? the gap widens.
??? Current: The current that flows through the spark gap after it has initiated
??? is relatively constant, regardless of the gap width. This is because the
??? resistance of the ionized air is very low, so a large current can flow even
??? with a small voltage difference.
??? For example, spark gaps are used in spark-ignition engines to ignite the
fuel-air mixture. The spark plug in a spark-ignition engine has a gap of about
0.020 inches, and the voltage required to initiate a spark across this gap is
about 20,000 volts. However, once the spark has initiated, the current that
flows through the spark plug can be as high as 20 amperes.
??? The relationship between voltage and current in a spark gap is not linear.
As the gap widens, the voltage required to initiate a spark increases
exponentially. This means that a small increase in the gap width can result in
a large increase in the voltage required to initiate a spark.
??? The voltage and current relationship in a spark gap is also affected by the
type of gas that is present in the gap. Air is the most common gas used in
spark gaps, but other gases such as argon and helium can also be used.
The breakdown voltage of a gas is the voltage required to ionize the gas,
and it is different for different gases. This means that the voltage required
to initiate a spark in a spark gap will be different depending on the type of
gas that is present in the gap.
-- Ron Davison principle Efficient Electronic Solutions 760.505.4380
“What I cannot build, I do not understand”?Richard Feynman?
|
It's always negative?differential?resistance, whether in an arc, a Gunn diode, the input into a loaded boost converter.? If not ballasted you can make a negative resistance oscillator.?
The term ballast (bare load) is indeed borrowed from the nautical sense - where a top-heavy payload is ballasted by a fixed keel load, lower down.?
The same principle is applied in arc lamps, emitter resistors in paralleled BJTs, shunt loads on LDO's - and anything where a fixed contribution drowns out unwanted variations.?
|
There is an old GE databook on Neon lamps online that discussed adding radioactive gas to certain (apparently not all) lamps and why. Once I downloaded it I lost track of where I found it.
I worked at a small electronic fluorescent ballast company in the mid-90's and my boss contracted a consultant physicist who shared a patent and lawsuit against a magnetic ballast company with two other people.
The consultant told me a plasma is formed in arc lamps, with a negative resistance (as defined by the slope of the v-i curve) being the reason a ballast (resistance or impedance) is required to stabilize the plasma current. He said the association of power being produced by a negative resistance wasn't applicable in that application, just the dynamic v-i curve.
The peculiar choice of the term ballast (I wondered) was an analogy to a nautical ship ballast, which stabilizes the vertical orientation of the ship.
Murray -- Murray
|
|
|
Re: Warning: Multiple definitions of model ...
I got my invite awhile ago. Qspice will download it's installation files and quit because it doesn't run on wine.
|
Re: Warning: Multiple definitions of model ...
You're a better subject (victim?) than I, who received the
invitation a few weeks ago.
Have at it, Tony.
Donald
On 2023-07-12 17:50, Tony Casey wrote:
toggle quoted message
Show quoted text
Well, who'd have thought... Out of the blue, I received an
invitation to be a beta tester this evening. ?
--
Regards,
Tony
On 12/07/2023 13:34, Tony Casey
wrote:
I
can't really comment on QSPICE as I haven't tried it. It's not
clear how I can. I signed up for access to the beta test, but
heard nothing more. I haven't joined the user group, either. But
I'm not alone there - only 183 people have, so far.
Nevertheless, good luck to them. Having competition is good.
|
@Mike,?
nice retirement job, much appreciated by many.?
I did a similar thing, a charged inductor that spikes into a capacitor, to make a big "ping" at the LC frequency, it was driving an ultrasonic ceramic transducer.? Even got a patent for it and earned ?1.?
OK it was a primary-side FET and an on-time, not a spark-gap, but a similar "impulse drive" into a parallel tank circuit, a triangle envelope with a good crisp leading edge - perfect for echoes.?
Doesn't seem to be much interest in GM tubes, of cheap versions thereof.? My idea for a quantum-powered lottery number predictor, that everyone can afford, looks like it'll have to wait...?
|
It’s still in a restricted beta. Google QSPICE to find the signup site.?
toggle quoted message
Show quoted text
On Jul 12, 2023, at 2:47 PM, Robin Gangopadhya <robin@...> wrote:
?
Is the QuickSpice good for downlaod & use?
yep. LTspice was my retirement job
On Jul 12, 2023, at 1:38 PM, Richard Andrews via groups.io <richardandrews.ma@...> wrote:
?Wow, locating hydrocarbon's!
|
Recovery time is also dependent on the recombination time for the gaseous ions involved not just on the circuit time constants. Recombination times are typically in the order of ten microseconds or so.
?
Bruce
On 13/07/2023 06:41 NZST Bonkers <stevens.kg@...> wrote:
?
?
I'm not sure what you mean there, I think the arc, or the detection event in a GM tube can be considered 'low' resistance.
I'd like to re-establish credentials as to why we're discussing this on LTSpice's forum.?
1) to show interest in spark gaps, arresters, neons, GM tubes -? what simulation models exist, what's missing and might be nice to have or to develop.?
2) how they are used - I can offer some detail on GM tube drivers, real measurements,?
3) maybe a simplified model for the GM tube as a narrow pulse current sink, enough to discharge its [3pF] capacitance by 50-100V? - in a time that I can confirm by measurement?
4) perhaps evaluate the "quench" circuit mentioned above - is it faster than the tube, does it work, is it any benefit??
5) provide a locus for any future enquiries about this sort of thing.?
6) Ideally, find a poor mans Geiger tube that works well enough and doesn't cost ?300
A basic GM tube spec is here:?https://www.jameco.com/Jameco/Products/ProdDS/2210141.pdf
They recommend 10Meg anode resistance, with their 3pF tube capacitance that's an RC of 30us, which is in keeping with the quoted minimum recovery time of 90us (3xRC).?
I'm guessing the volts drop from V_operating [500V] to V_min [325V] minus a bit, so a swing of 200V, 250V, to guarantee the tube quenches.?
The magnitude and duration of the ionisation current is not specified, but its quick compared to 90us, I'll measure it if I can't find a published figure.?
|
Re: Warning: Multiple definitions of model ...
Well, who'd have thought... Out of the blue, I received an
invitation to be a beta tester this evening. ?
--
Regards,
Tony
On 12/07/2023 13:34, Tony Casey wrote:
toggle quoted message
Show quoted text
I
can't really comment on QSPICE as I haven't tried it. It's not
clear how I can. I signed up for access to the beta test, but
heard nothing more. I haven't joined the user group, either. But
I'm not alone there - only 183 people have, so far. Nevertheless,
good luck to them. Having competition is good.
|
Is the QuickSpice good for downlaod & use?
toggle quoted message
Show quoted text
yep. LTspice was my retirement job
On Jul 12, 2023, at 1:38 PM, Richard Andrews via groups.io <richardandrews.ma@...> wrote:
?Wow, locating hydrocarbon's!
|
Gap geometry affects the field strength for a given gap width.
With infinite plane parallel electrodes the voltage required to achieve the critical field strength is proportional to the electrode spacing.
Spark gaps with spherical electrodes were used to measure peak voltages by decreasing the spacing until breakdown occurs. The voltage required to achieve breakdown is a non-linear function of the sphere separation.
Bruce
toggle quoted message
Show quoted text
On 13/07/2023 02:52 NZST Jerry Lee Marcel <jerryleemarcel@...> wrote:
Le 12/07/2023 à 16:45, Richard Andrews via groups.io a écrit?:
As the gap widens, the voltage required to initiate a spark increases
exponentially.
Are you sure of this? I had a notion that ionization appeared at a given
field strength, which is more or less a direct relationship of voltage
divided by gap length.
Check this
|
yep. LTspice was my retirement job
toggle quoted message
Show quoted text
On Jul 12, 2023, at 1:38 PM, Richard Andrews via groups.io <richardandrews.ma@...> wrote:
?Wow, locating hydrocarbon's!
|
Re: POSH (Powershell) tool to manage "user approved" LTspice cmp files
hi eewiz
Thanks for the help..
Some user info-
All controlled cmp files have a fixed name of "MYstandard.xxx" where .xxx=.bjt, .dio, etc.
The script is currently hard configured:
1. All user cmp filenames = MYstandard.xxx where xxx=bjt, or dio, etc.
2. Parent symbol folder for all user MYstandard.xxx files = %AppData%\LTspice\lib\cmp
3. Parent symbol folder for all LTspice standard.xxx files = %AppData%\LTspice\lib\cmp (same as item 2 above)
4. When LTspice -sync is run (-u switch), it runs in foreground. The script waits until LTspice is exited.
??? So, click "OK" when prompted, exit LTspice, and the script will continue to completion.
5. When -u switch is specified, first LTspice is synced, and then the script compares the
??? two cmp files and will only replace the corresponding standard.xxx if difference.
Any comments or proposed changes are welcome.
I'll be uploading a test script file tomorrow.
|
Wow, locating hydrocarbon's!
|
I am thinking now I'll have to use e (elementary charge), calculate the number of atoms for the direct path of the gap, use Paschen's voltage and do some magic with them.
|
I've had very good luck modeling a spark gap as a switch in series with 20V drop in each direction.? The test case was a parallel tank circuit(about 10uF capacitance at 2.5KV and 10uH inductance) used to create a short AC pulse to excite eddy currents in rock formation strata in the interest of locating hydrocarbon bearing rock for oil exploration.? The current amplitude ramps down linearly instead of decaying linearly as it would with a series resistance.? The gap also quenches realistically.
|
Bonkers
murrayatuptown
Tony Casey