I recall working on an arc lamp power supply, which used a spark gap for ignition. I wondered, and still do, are there any design techniques or formulas, to obtain a specific desired voltage pulse across the gap? Or is it cut and try, regarding the gap size and current? (the current break, that is) -- Rich
spark plugs
Started by ●December 21, 2023
Reply by ●December 21, 20232023-12-21
On Thursday, December 21, 2023 at 4:03:35 PM UTC-8, RichD wrote:> I recall working on an arc lamp power supply, which used > a spark gap for ignition. > > I wondered, and still do, are there any design techniques > or formulas, to obtain a specific desired voltage pulse > across the gap? Or is it cut and try, regarding the gap size > and current? (the current break, that is)Sure, there's formulas; an auto spark coil takes high current from 12V on the primary to build up field energy, then when the points open, the stored energy (less a bit for the resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor (often a PTC type) gives a lower current for the now-hot engine. Energy is conserved, a little under L times the square of the current, over two, per spark. The gap size has to be small enough that the arcing doesn't occur in the wrong place... and the conductivity of the gap determines the spark duration (low gas denslty, lower resistivity and faster spark).
Reply by ●December 21, 20232023-12-21
On 22/12/2023 11:03 am, RichD wrote:> I recall working on an arc lamp power supply, which used > a spark gap for ignition. > > I wondered, and still do, are there any design techniques > or formulas, to obtain a specific desired voltage pulse > across the gap? Or is it cut and try, regarding the gap size > and current? (the current break, that is)You can buy high voltage spark gaps - the one I bought for that job in 1972 was glass-enclosed tube with an inert gas filling. At the tine I'd bought a copy of a 1920's text book on the conduction pf electricity through gases, for which there was already a well established theory - search on Paschen's Law. https://en.wikipedia.org/wiki/Paschen%27s_law My first spark gaps was made in the workshop, and used two steel ball-bearings to form the gap, mounted in a silica glass tube, The arc lamp was exposed to the UV light from the spark, and that help start the lamp. The glass in the store-bought spark gaps wasn't UV transparent and the light they emitted didn't start the lamp nearly as often. -- Bill Sloman, Sydney
Reply by ●December 24, 20232023-12-24
On Thu, 21 Dec 2023 19:07:11 -0800 (PST), whit3rd <whit3rd@gmail.com> wrote:>On Thursday, December 21, 2023 at 4:03:35?PM UTC-8, RichD wrote: >> I recall working on an arc lamp power supply, which used >> a spark gap for ignition. >> >> I wondered, and still do, are there any design techniques >> or formulas, to obtain a specific desired voltage pulse >> across the gap? Or is it cut and try, regarding the gap size >> and current? (the current break, that is) > >Sure, there's formulas; an auto spark coil takes high current from 12V on the primary >to build up field energy, then when the points open, the stored energy (less a bit for the >resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor >(often a PTC type) gives a lower current for the now-hot engine. > >Energy is conserved, a little under L times the square of the current, over two, per spark. >The gap size has to be small enough that the arcing doesn't occur in the >wrong place... and the conductivity of the gap determines the spark duration (low >gas denslty, lower resistivity and faster spark).What a terribly unscientific explanation. I'm aghast, quite frankly.
Reply by ●December 24, 20232023-12-24
On Sunday, December 24, 2023 at 9:52:02 AM UTC-8, Cursitor Doom wrote:> On Thu, 21 Dec 2023 19:07:11 -0800 (PST), whit3rd <whi...@gmail.com> > wrote: > >On Thursday, December 21, 2023 at 4:03:35?PM UTC-8, RichD wrote: > >> I recall working on an arc lamp power supply, which used > >> a spark gap for ignition. > >> > >> I wondered, and still do, are there any design techniques > >> or formulas, to obtain a specific desired voltage pulse > >> across the gap? Or is it cut and try, regarding the gap size > >> and current? (the current break, that is) > > > >Sure, there's formulas; an auto spark coil takes high current from 12V on the primary > >to build up field energy, then when the points open, the stored energy (less a bit for the > >resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor > >(often a PTC type) gives a lower current for the now-hot engine. > > > >Energy is conserved, a little under L times the square of the current, over two, per spark. > >The gap size has to be small enough that the arcing doesn't occur in the > >wrong place... and the conductivity of the gap determines the spark duration (low > >gas denslty, lower resistivity and faster spark). > What a terribly unscientific explanation. I'm aghast, quite frankly.If you want science, look up Georges Charpak's work (Nobel prize 1992); the physics of a spark is nonlinear, nontrivial, and depends on... EVERYTHING. There's no simple general formula. Maybe if you define a geometry, and a gas pressure, and a temperature profile, and an energy budget, you can design a detector gas to equal the magic mixes Charpak was known for.
Reply by ●December 24, 20232023-12-24
On 12/21/2023 8:07 PM, whit3rd wrote:> Sure, there's formulas;The problem is quantifying the environment. The medium between the electrodes plays a role -- as do the shapes of the electrodes, etc.> an auto spark coil takes high current from 12V on the primary > to build up field energy, then when the points open, the stored energy (less a bit for the > resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor > (often a PTC type) gives a lower current for the now-hot engine. > > Energy is conserved, a little under L times the square of the current, over two, per spark. > The gap size has to be small enough that the arcing doesn't occur in the > wrong place... and the conductivity of the gap determines the spark duration (low > gas denslty, lower resistivity and faster spark).We had a VC approach us shortly out of school (70's) asking us to design an "electronic ignition" -- no rotor, etc. Back then, 1MHz processors were the norm. And, only 1802's were low enough power to run in an automotive environment. We knocked together a prototype in short order (the code is trivial -- as is the hardware, if you don't have to plan for production!) and demonstrated how we could adjust the advance and dwell based on RPM and "general performance criteria". The VC had hoped we could enhance the value by compensating for plug degradation, indicating needs for service, etc. You can't (cheaply) *look* at the plug voltages but you can look at "characteristics" of each firing. And, if you ignore absolutes and, instead, do relative comparisons (this plug to that, this plug NOW to this plug a month ago, etc.) you can make some crude inferences. But, there are just too many variations to account for! No two plugs are produced alike. No two coils (we had a coil per plug). Wires get dirty and rerouted. etc. None of us had a desire to make a career out of this -- or ANYTHING automotive -- so we bowed out when he wanted to present to the automakers. The only way to do ANYTHING is in huge quantities -- which means, you do FEW things! (We'd rather have lots of varied challenges than just a few) Amusingly, I now find myself reviewing those (ancient) notes as I look to design an ECU for "one last *big* ICE" (for myself). I think it would be cool to be able to dial in operating constraints, dynamically: performance, fuel economy, emissions quality, etc. [This sparked from a discussion with a colleague over designing similar controls for race cars... let the controls "learn" the characteristics of the track so the driver doesn't have to tell the controls when he's entering a straightaway, deep curve, etc.]
Reply by ●December 24, 20232023-12-24
On Sun, 24 Dec 2023 16:54:06 -0700, Don Y <blockedofcourse@foo.invalid> wrote:>On 12/21/2023 8:07 PM, whit3rd wrote: >> Sure, there's formulas; > >The problem is quantifying the environment. The medium between >the electrodes plays a role -- as do the shapes of the electrodes, >etc. > >> an auto spark coil takes high current from 12V on the primary >> to build up field energy, then when the points open, the stored energy (less a bit for the >> resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor >> (often a PTC type) gives a lower current for the now-hot engine. >> >> Energy is conserved, a little under L times the square of the current, over two, per spark. >> The gap size has to be small enough that the arcing doesn't occur in the >> wrong place... and the conductivity of the gap determines the spark duration (low >> gas denslty, lower resistivity and faster spark). > >We had a VC approach us shortly out of school (70's) asking >us to design an "electronic ignition" -- no rotor, etc. >Back then, 1MHz processors were the norm. And, only >1802's were low enough power to run in an automotive >environment. > >We knocked together a prototype in short order (the code >is trivial -- as is the hardware, if you don't have to >plan for production!) and demonstrated how we could adjust >the advance and dwell based on RPM and "general performance >criteria". > >The VC had hoped we could enhance the value by compensating >for plug degradation, indicating needs for service, etc. > >You can't (cheaply) *look* at the plug voltages but you can >look at "characteristics" of each firing. And, if you ignore >absolutes and, instead, do relative comparisons (this plug >to that, this plug NOW to this plug a month ago, etc.) you >can make some crude inferences. > >But, there are just too many variations to account for! >No two plugs are produced alike. No two coils (we had a coil >per plug). Wires get dirty and rerouted. etc. > >None of us had a desire to make a career out of this -- or >ANYTHING automotive -- so we bowed out when he wanted to >present to the automakers. The only way to do ANYTHING is >in huge quantities -- which means, you do FEW things! >(We'd rather have lots of varied challenges than just a few) > >Amusingly, I now find myself reviewing those (ancient) notes as I >look to design an ECU for "one last *big* ICE" (for myself). >I think it would be cool to be able to dial in operating >constraints, dynamically: performance, fuel economy, emissions >quality, etc. > >[This sparked from a discussion with a colleague over designing >similar controls for race cars... let the controls "learn" >the characteristics of the track so the driver doesn't >have to tell the controls when he's entering a straightaway, >deep curve, etc.]Indeed. Could easily be done with a Raspberry Pi or an Arduino even that matter I would imagine.
Reply by ●December 25, 20232023-12-25
On 25/12/2023 8:42 am, whit3rd wrote:> On Sunday, December 24, 2023 at 9:52:02 AM UTC-8, Cursitor Doom wrote: >> On Thu, 21 Dec 2023 19:07:11 -0800 (PST), whit3rd <whi...@gmail.com> >> wrote: >>> On Thursday, December 21, 2023 at 4:03:35?PM UTC-8, RichD wrote: >>>> I recall working on an arc lamp power supply, which used >>>> a spark gap for ignition. >>>> >>>> I wondered, and still do, are there any design techniques >>>> or formulas, to obtain a specific desired voltage pulse >>>> across the gap? Or is it cut and try, regarding the gap size >>>> and current? (the current break, that is) >>> >>> Sure, there's formulas; an auto spark coil takes high current from 12V on the primary >>> to build up field energy, then when the points open, the stored energy (less a bit for the >>> resistivity of the iron) blasts the spark hard. After warmup, the current limit resistor >>> (often a PTC type) gives a lower current for the now-hot engine. >>> >>> Energy is conserved, a little under L times the square of the current, over two, per spark. >>> The gap size has to be small enough that the arcing doesn't occur in the >>> wrong place... and the conductivity of the gap determines the spark duration (low >>> gas denslty, lower resistivity and faster spark).>>>> What a terribly unscientific explanation. I'm aghast, quite frankly. > > If you want science, look up Georges Charpak's work (Nobel prize 1992); > the physics of a spark is nonlinear, nontrivial, and depends on... > EVERYTHING. There's no simple general formula. Maybe if you > define a geometry, and a gas pressure, and a temperature profile, and an energy > budget, you can design a detector gas to equal the magic mixes Charpak was > known for.Cursitor Doom doesn't want science, and doesn't understand it even when his nose gets rubbed in it. You can expose a horses arse to education but you can't make it think. -- Bill Sloman, Sydney
Reply by ●December 25, 20232023-12-25
On 2023-12-22 01:03, RichD wrote:> I recall working on an arc lamp power supply, which used > a spark gap for ignition. > > I wondered, and still do, are there any design techniques > or formulas, to obtain a specific desired voltage pulse > across the gap? Or is it cut and try, regarding the gap size > and current? (the current break, that is)That's strange, an arc lamp is low voltage, high current. The ones I have seen needs the points to touch for an instant, then separate, either automatically or manually. During use, the points can be at a distance of about one centimetre. Using a a high voltage spark to initiate the thing? I don't know if that is possible, I have not seen it. -- Cheers, Carlos.
Reply by ●December 25, 20232023-12-25
On Mon, 25 Dec 2023 14:06:06 +0100, "Carlos E.R." <robin_listas@es.invalid> wrote:>On 2023-12-22 01:03, RichD wrote: >> I recall working on an arc lamp power supply, which used >> a spark gap for ignition. >> >> I wondered, and still do, are there any design techniques >> or formulas, to obtain a specific desired voltage pulse >> across the gap? Or is it cut and try, regarding the gap size >> and current? (the current break, that is) > >That's strange, an arc lamp is low voltage, high current. The ones I >have seen needs the points to touch for an instant, then separate, >either automatically or manually. During use, the points can be at a >distance of about one centimetre. > >Using a a high voltage spark to initiate the thing? I don't know if that >is possible, I have not seen it.In the older style transformer-only arc welders, the arc has to be struck manually by momentarily shorting out a low voltage/high current secondary winding; typically 50 to 80V. This can be a messy and frustrating process at times, even for a skilled tradesman. Newer style inverter welders use HF AC for a 'soft start' which is much cleaner and faster. I've not heard of a high voltage start method either.
