Joule Ringer!

[Slider2732]

@PB - I've sometimes used the diode, sometimes not. Yes, forward biasing it toward the Base removes the blue, once the pot moves past about 1/4 distance (50K pot). I have no idea how you knew that ! because it doesn't completely close off the blue at lower settings, with the diode being orientated that way. The way that the LED would act like an overdriven neon bulb in Bedini HV experiments was what I liked without using it though, like a warning beacon.
In terms of efficiency, it appears the diode should be in the circuit

Speaking of appears and non exactness, I realised that the shoe box type of light box i'd made was going to be too big for my work table. A pencil is too big for the clutter on here at the moment lol.
So, a toilet roll is now to be used
My bulbs are low wattage, 1W, 2W and 2.5W. The 1W and 2W fit into the tube quite snugly, but would benefit from a piece of electrical tape. A solar panel from one of those dollar store dancing flowers is to sit in the end, attached to a round end piece to close off the light from outside.
I'll get a solar panel figure for mains driven running and call that 100%. From there, everything recorded will be a percentage.




Great circuit there Slayer and delighted to see you in this thread !

[PhysicsProf]

@ PhysicsProf

The transformer is a 110v to 9v no extra inductor just a 1M resistor.

I see -- thanks, Slayer007!

[slayer007]

@ PhysicsProf

The transformer is a 110v to 9v no extra inductor just a 1M resistor.

[PhysicsProf]

1 question on subject : Do you think it would work with strings of LEDs instead of LED bulbs ? Example 30 LEDs in series to replace 1 bulb with 120V transformer.
A bunch of 30 LEDs would cost like 1 USD, instead of 10-15 USD for 1 bulb. They could be made easily everywhere.
But I saw that bulbs have HV caps and step a down transformer inside, and you said it was useful in your setup...
An alternative could be to put these 30 LEDs in //, and use a wall-USB-charger 120V-5V to feed them the special electricity from the SJR.
LEDs with or without resistor in series ??
Thanks for sharing
Jules

I've been wondering the same thing -- Are the "guts" of the LED bulb really helping us get out more Lumens per watt? If a source of heat-loss, then perhaps it would be better to try to circumvent somehow. Thanks for showing the innards of the bulb, Jules, and for phrasing the question!

Lots of good ideas here -- tuning notes by Pnutbutter and Matt (Hi, Matt!), and the xformer out of a wall wart by Slider.
Slayer --can you provide a little more detail on that "extra" inductor you show in your schematic? Nice vid; thanks.

[slayer007]

Lasersaber Type Exciter running off 6v @ 12.6 mA and 12v @ 28 mA.

Lasersaber Type Exciter - YouTube

[Jules Tresor]

LED string instead of expensive bulb ?

@ PB, thanks for all this details on the caps I understand you want to do all "by yourself". Most tinkers say like that. So I end up giving the money to charity organizations
1 question on subject : Do you think it would work with strings of LEDs instead of LED bulbs ? Example 30 LEDs in series to replace 1 bulb with 120V transformer.
A bunch of 30 LEDs would cost like 1 USD, instead of 10-15 USD for 1 bulb. They could be made easily everywhere.
But I saw that bulbs have HV caps and step a down transformer inside, and you said it was useful in your setup...
An alternative could be to put these 30 LEDs in //, and use a wall-USB-charger 120V-5V to feed them the special electricity from the SJR.
LEDs with or without resistor in series ??
Thanks for sharing
Jules

[Peanutbutter29]

I keep forgetting....

@ Jules, oh my thank you kindly for the offer! I would "feel" better getting one on my own, if I can find a supplier for bulbs that will ship. I don't really need a tranny as I can get those here and usually trannies are 50/60hz anyhoo.

Before I forget AGAIN, two things
* easy way right now, to get 220 to work similarly. 12-120v tranny @12v; so.....24-220 @24v! If you double the input voltage from us then a 24-220 would be the same; this may be an easy quick fix for now.

*Also, on my caps used with these 2 versions and being multiples. Two main reasons; First I wanted people to get common radio shack parts. Second I've mentioned I don't have a lot of parts still, so my caps are limited and I HAD to use those for proper values. Radio shack, nor myself had a .023 for the SJRC or a .015-.016 for the "tuned version"; same applies with the .124 on the secondary. So YES single caps are the same here, nothing funky from the multiples.

@ slider! great job with another system! Ya' at this point the v1.2, at least, is using common power transformers found in wall adapters and DC supplies. I've used these Radio shack trannies before, so I knew they had em'. Most wall adapters (not light flat SMPS) use a tranny, rectifier and filter cap. Though I didn't want a DIY saying "go tear apart a adapter". I DID try a 12-24 @24VA supply, but didn't like it for tuning.

All my "tuned" coils run in the mode, from your video, where you are drawing Maximum current . So far....
Oh also, are you runnning the diode on the base? This may eliminate the blue flash, not sure.

Lastly, some may find interest in the radio shack Audio output transformer for 2.69.

This runs a single bulb better than the Toko, but not full bright of course. This IS L2 dominate, but you can "transfer tune" the primary for lower current in certain modes. With the bulbs we are using; positive bias has a 6-9v operation range, while a negative bias runs 9-12v. Current draw "stock" will be 5-600ma @ 6-9v (positive) and 300ma @ 12v (negative). It does warm some since it's size is small and not rated for high current, but cheap, tiny and single bulb light-able.
I'm sure a lower wattage bulb than 7.5 would draw less and light a bit better. Heck, easy then to make a single bulb flashlight, but I'm going slightly different now.
EDIT: The audio output is an E-core like the power transformers. Also, I did a quickie transfer tune and got a decent light with enough lower current for no tranny heat. I did a quick video to measure output and such. Not great but okay for the small size and cost.

Thanks, Sorry for posting again!
PB

[Slider2732]

Thanks for all that info PB !
Presumably a scope can show us when we're getting close to correct tunings and, in so doing, can train the ears and eyes ?


Here's a hopefully useful tip - wall warts
I had the idea to take apart a 120-12 regular Class 2 wall transformer this evening. Mainly, it was to see just what was inside, but also to try it out with the same circuit as in my last video. Inside this one was probably typical, input is direct, output is a simple half rectifier of 2x diodes, a cap and 2x small microhenry chokes on the DC output.
But - it works very well, with a simple transistor switch circuit and they might work well with the SJR circuits
The Goldilocks point is actually denoted by blue from the red/blue flashing LED this time. The circuit will start up at 10mA 12V and, below that, the 2W LOA LED blulb will flash along with the LED from about 5mA
The circuit enters the useful area at 110mA, similar to the much larger transformer and will again trip into over current and use up to 0.7A for little extra light gain.

Main thing - regular wall transformers work fine and could breathe new life into old abandoned wall adapters. They may also be easier to source in countries without parts stores.

Here's a pic of this one, transistor is a 2SD313:


(method to gauge light output with a light box still under construction)

[Jules Tresor]

220V parts

@PB, we will ask Ash or Andrew from Panacea to send you a set of parts, bulbs + transformers in 220V. I would be honored to pay the bill. If you agree then PM me or Ash. It's important to find the working circuit with 220V because most poor countries are using that kind of voltage rated parts !
PS: I could send that stuff myself directly to you if you dare giving me your details, I'll use DHL for faster delivery. Here in Malaysia/Thailand they use 220V like in Europe.

[Peanutbutter29]

Capacitors + Coils "In General"

I'd like to ramble some on caps and coils in general. I'm not making any assumptions to what is known already or not; So I'll try to explain the perspective completely. I'm sure many understand some, a lot or all of this already.

First, I want to look at the 2 primary modes that we "know of" for coils to run; these being Lumped mode and slow wave mode.
-Slow wave mode is what a "true tesla coil" was designed around. Corum and Corum did a wonderful paper on this a while back. As Tesla mentioned, this applies when our physical wire length and operational frequency bear a relation of "not less than" 1/6 wave; of course 1/4 was optimal. The term slow wave was applied by Corum as this relies on velocity factor in design to achieve this 1/4 length; "relative to the speed of electrical movement in the coil".
Here though, we have a situation where current and voltage vary and are NOT even across a coil length. E.G. standing waves. This is related to the reflection (high swr) caused by the Operational frequency and physical length matching, or closely matching a 1/4 or any other higher "Supportive" (odd) harmonic.

-Lumped Mode Current is by far the most common way we see and apply coils. Also, it's what a great deal of TC coils still end up with. Again, Corum did a good job of clarifying this. Whenever we are not looking at "reflective" situations (eg- physical length and Fo less than 1/6) then STANDARD formulas apply for voltage and current across a coil; Which is considered to be even (for current) and linear (for voltage).

We are, of course, running in Lumped Mode as with most coils today. However, as I showed with my first Model T video and sparking; we can create a situation where it IS possible to manipulate current placement to some degree, with reinforcement or optimally "Tuning".

So, we know we are starting and using lumped mode; so no standing and even current. Next, as I'm sure you've all said to yourself; with me talking caps changing things. What about the capacitance that's already in the coil? Yes we have a distributed self-capacitance in all coils; and yes this is exacerbated with coupling and cores. However, even with all this; the distributed capacitance plays a small role.
The reason here, is because we are running an operational frequency that's largely lower than either primary or secondary would "self-resonate at". This can NEVER be brought to such "self resonating" point as the load will always lower frequency of the circuit. You might wonder then, well the load changed the self resonating point right? Actually NO, the "load" is a tertiary variable and we can only then be interacting with the Impedance of that load; it cannot physically "be" part of the L2. So, we cannot alter this most technically.
Now, as frequency is a large portion lower than each corresponding Self-resonate; the distributed capacitance plays a small role in the apparent movement of electricity. E.G it is highly inductive, saw tooth and current leading (essentially XL not Z). This is of course looking from each coils "perspective" non-inclusive of the load or transistors effect on this phase angle, E.G- so any coupled coil or single coil here.

An important note here then. Anytime we are NOT including said capacitance into an oscillation; the coil has NO apparent preferred frequency! To be a bit more common here; if XL and XC ratios are small fractions; the coil in question is considered XL not Z.

So, if your with me now and don't think I'm nuts; here's where we introduce "Non-distributed" capacitance, in the form of capacitors. Tesla and many others mentioned there's always a relationship with Inductance and Capacitance size relative to frequency AND intended operation.
Okay say, we start with the primary and add a capacitor; what are we doing. Well, if it's too small then you DO nothing as it will still behave as negligible in the circuit. If we raise then to a point where we begin including the capacitor into oscillation at Xhz; we now will have current begin circulating IN this cap FROM the L1 primary coil. This is the first point where you can begin to say we have a "tank" circuit. Now, that we have a tank (what I simply called bound) circuit we DO notice a larger affinity for a particular frequency, though still at the edge of this. As capacitor size is raised, current circulation in the cap will rise and can even become larger than the DC input value as the Tank frequency and oscillation frequency become close.
Essentially then, this is why I refer to the caps "reinforcing" an oscillation. As long as your in a wide range of "active" tanks or binding; the coil and cap ARE actively supporting this oscillation at Xhz. This supporting varying in both degree and function at various levels.

I refer again to the model T sparking video in the first parts. I showed 3 capacitors only on the primary. to save time, I chose one too large, one too small and one just right (lol); for a SINGLE cap. I first show output (spark) with no caps at 12v@600ma. You see HIGH DISSIPATION in the switch, this is why a stock SJR heats, but you get to see it here. Also, the spark is tiny and thin. Now with the smallest "bind-able" cap we see more "apparent" output and Less dissipation. Then the largest cap we have even less dissipation and a bit longer spark. I will note, if you go larger than max shown output will DIMINISH. However, there is a middle size I show last that gives the same low dissipation and a Longer spark than either small or large cap. This would be a "transfer tuning" sweet spot for the primary, relative to a "freewheeling secondary". E.g. L2 is XL not Z.

Now with our coil we've added 1 tank or binding / reinforced the primary electrical motion. As stated above we still have L2 that is behaving OUT of it's ability to use distributed self capacitance. So it's still "freewheeling" at what ever frequency with NO apparent preference. However the same rules for creating it into a tank apply.
If we add a capacitor that is in a tank / bind-able range to the L2; we now have 2 reinforced oscillations instead of one. The interaction of these two, then begin to apply (mostly) to common rules of Z vs Z, coupling and power factor. Essentially the current sharing between tanks is related to the impedance, which can be variable from high coupling; and thus alters the "apparent" voltage / current phase angle from what either coil would have independently.

**This, and only this (2 tanks) creates a situation, where by "tuning" as I mention can possibly occur. Most technically you start with a Z load and 2 Xl's. With one cap we have a Z load, Z primary and XL secondary. With both caps we now have Z load, Z primary and Z secondary. Also we now have 3 frequencies of preferred / reinforced operation; whereas at the start we have 1 which is the load.

I'll refer yet again to the Model T spark video when I go to place the cap on L2. I start by using the same capacitor that was optimal for single binding (middle size) on the primary. I will note here, I only have about 3 sizes of HV caps; so I was forced to "Tune" to that cap, L1, L2 and the switch, bleh!
At any rate, upon the first run with both caps; we begin to see a tiny drop in input power and a much better spark. Now here's a HUGE key point; we also see, for the first time, purple that is NOT apparent with no or 1 cap and can never be. This is because because of above where both L1 and L2 are freewheeling (XL) or just L2. Another big note is that the purple section (power factor 1 / volts and amps aligned) is very narrow and NOT optimal. However, the results are still much improved and one could stop un-aware of something better. Another key note, is the cap will NEVER be the same with 2 tanks as with optimal tuning for "transfer" with a single L1 tank.
Since I had to use one L2 cap, I re-sized the L1. I only show the "right" one in the video, but it was tuned before hand to know. In this last setup you can note a Wide range of Purple; which should not be! Because the spark load Z is changing; the relative power factor of the L2 MUST shift to allow a purple spark across this variation. Another important note is that in this last stage; the load alters the switching speed of the primary AND the sound (NOT SWITCH SOUND) is apparent through the spark. This means that the last item in the circuit, the load, is dictating the ENTIRE circuit operation. Sure, you can easily say well that's what the SJR's do when you add bulbs, and you are correct. However, this is not a transistor nor is the switch triggered off the secondary. Also and worse, the HV and primary - are tied together. The point here, is that said load dominance should not be possible, but is ENTIRELY because of properly tuning both tanks relative to the switch and load. Lastly, I didn't really note, but the final running was 100-150ma less than the start.

I don't have that video public still and maybe I'll put an unlisted link here if you all want. Most of what I explained so far was shown well with the T, but you can see I could not explain all this in the video. I'd hoped some would catch it, but I think my diversion of "this is trivial" left most not interested.

So now, having said all this; I consider fully tuned most optimal, primary "transfer tuning" second and no tuning last. This is in terms of efficiency. As I've said there are many times where the coil cannot be tuned or the load is to small (relative to coil size) to tune. In fact, I believe there are more non-tunable cases than tunable. However, in ALL cases we can always add a L1 tank easily; even with L2 dominate.

If then, we look at only adding a primary cap and what we can "see" or "hear" in these circuits. Well you can hear, if you get into a bind-able range. Here's the kicker with L1 only as it's fairly simple; a small cap will only change sound, but a larger cap will lower current some. With a fully L2 dominate like the E-core, I noticed drops all the way to 10uF where the cap quickly fried from current circulation LOL!.
This is explained though above, where the larger size comes closer to the operational frequency and can allow larger circulation currents, to some extent, than the input. Essentially too, this allows the circuit to more easily accept a larger current at a slightly lower voltage. The downsides to this are mainly apparent with larger loads or outputs. You can have premature saturation of the core, create some loss from high circulation current and also overheat the transistor.

The last thing I suppose with caps is related to the transistor heat. I believe the answer for lower heat should be obvious from above, but it's because of current circulating between L and C rather than +/- input (through transistor). So, this heat reduction will occur to varying degrees relative to cap size / reinforcement. The reason you can still have heating with high loads can ONLY occur in two cases. If L2 is still freewheeling, then base timing and phase may not be correct; stealing power, reducing Hfe and heating. Also if L1 and L2 are both bound (tanked) but improperly, the same effects can occur. I suppose I should point out, this would be the case too if a complete resonance was attempted; but I lump that to improper here.

Okay that's long, but I think I covered the capacitor aspects. I hope this gives some a better understanding when "tuning" or "transfer tuning" coils. This just leaves the final "logic" and concept for how I tune my coils. Everything so far was easier to explain and less odd. From here on, I get a little wonky with what would be considered "normal" and I want to explain that well; if possible. All of this with caps will help understanding though for the rest, I hope.

Thank you much,
PB

[Matthew Jones]

I apologize in advance if what I am going to discuss is off topic a bit. But some of what I want share with ya is applicable since your using EI cores for this thing.

I have been for last 1.5 years looking at driving low impedance wound EI cores with Tesla Switch type setup. More recently in looking at this I have been working with real hot setup that creates a surge current state in the core releasing very large transient spikes for collection and use immediate use.

The spike generated in an EI core can be used immediately as opposed a monopole type setup in which they are primarily voltage in nature. What I have found is the spikes are 90 degrees out of view. So when you look at them on scope your are looking down the Z of the spike. It actually has a time value but it is 90 deg. out so you see a straight line. When you generate them in a closed inductor such as the EI cores the current trails the voltage by 90 degrees. This forces the spike to align with the time you have on your input. Or in other words turns it into a positive energy with both voltage and current that can then be used immediately.

Now the other thing about EI core is they are built with certain frequencies in mind based on the thickness if the iron plates. PeanutButter29 spoke of this in one of his videos. "Multiples of 60". But you can drive them to almost an infinite frequency and good a step up or down out them as long you use a multiple of the base use of the transformer.
The windings are also important. Now with transistors with trigger circuits I might be OFF because I drive most of my stuff with an IC so that I can maintain a good control, and I use FETS. I make sure my windings have the same resistance as the RDS on (MAX) resistance. Or a multiple of that if I am going for a higher winding. Impedance matching is important.
I use this calculator for figuring my wire length

The next thing I think is the most important because it can lead the greatest amount of power and be completly virtual in nature. Don Smith spoke about it in the video of the Tesla Symposium in 1996 video 1 of 2 chapter 12.

(Sorry about the blurry its a capture from the film.)

All three waveforms come from an LC circuit. The third wave form is overunity or so He claims, and the resistance in the circuit is what gives it to you but I have found it is the semi conductor used. Since my work is involved with the TS I have found this behavior from the Serial Switch in a Tesla switch. For instance an MJL 21194 will perform this way in the serial position with 24 volt motor run off the ground side at 144 milliseconds with 50% percent duty cycles. 144 on __ 144 off.
All semi conductors driving an inductor have point in which they can do this, (But the inductor is important) You have to really look for it. You also have a range of frequencies which this possible as well. Usually about 5 different sets work out this way. For instance the MJL at 144 ms 50% duty cycle also will do it at somewhere around 12 khz. then again a around 50 khz.
One thing to note is this not a continues expansion of voltage. It happens in sets. So you may switch 7 -10 times, every time progressively growing the voltage then it will collapse and start over. But the inductor see's it if its set up correctly. Use your scope across the collector and ground on the emitter.

I have design a routine to find it using a micro controller and an ADC. And it tedious to go through, but all transistors and fets have that possibility. Fets are real high frequancy.

So why tell you all this stuff?

If it were me looking for perfect circuit I would first start out finding that frequency in which the transistor would expand the voltage input into the collector. Then I would find an inductor body that could match frequency and wind it to match the transistors impedance at the power level I was going to use it. This should give your best output.
One other thing I would do that none of you have looked at (And it most likely will work, but don't hold me to it cause I am not sure first hand) is incorporate the current on the ground leg of the circuit into my winding pattern.
If you enter the inductor from opposite side with both the positive and negative you double the strength of the field in the core. This is 2x more efficient in low impedance winding's then using just one leg of the power.

I'll look at the circuit this weekend and see if its possible.

Anyway take for what its worth, I hope some one reads the post and gets something from what I telling you. This is all proven fact, and its very close to what your doing with this joule ringer.

Ya'll are doing great work by the way.
Thanks

Matt

[Peanutbutter29]

Bunch of things and some Good Info!

First, great job Limotor, Slider, and Totoalas! I have a few things to cover and hopefully can talk more on operation some.

I have some exciting news (lol, kinda)! I have in my hands a quote from Metglas for the toroid Lasersaber used. We can purchase them, with of course a few stipulations. First though, to note they are $630.00 each!?!?!? A bit high I would say as "pay off" would be a 100years or so. Now, we can get them down to 550. in 12 lots.
Now the stipulations:
First, you MUST have all your personal information on file with metglas.
Second, you must sign a User agreement that applies to DoD and ITAR.
- You CANNOT export this item to ANY country without written consent from the Directorate of Defense
- Any exportation is a Federal offense.
- You cannot re-distribute this product for a period of 3 years from purchase!

Also, there is a warning on the quote:
"- NOTE METGLAS/HITACHI METALS AMERICA POLICY PROHIBITS THE UTILIZATION OF OUR PRODUCTS IN
WEAPONS OF MASS DESTRUCTION AND MISSILES."

So, considering all this and the fact I probably got a background check for even inquiring about this; I don't think I'm going to look at these. Also, I've gotten around needing their advantage and this would probably hinder now; at the very least I hate winding too.

A couple things that stick out as "odd" to me. If company X purchased this core, they had to keep it 3 years before legally selling on Ebay.....why no winds, wind marks, or age? Also, when I approached the engineer with the specs; he replied within minutes WITH a model number and ALL exacting specs matching LS Core. I cannot deny that this makes me include the possibility that LS bought this new from Metglas for the 630.00; Again, I'm sure just oddities. Any Ebay price would be a steal for sure!


Next, since I'm on this subject. I did actually order some trannies (for the first time here), so when I get them all tested I should have some workings for the 220v euro guys; at least ideas. I'm very excited about a couple things here, but have to wait a few weeks to get everything.

Also, then a question for 220v LED's. which should I use / where should I find them? It seems the Ebay bulbs are mostly from China or Hong Kong; if you get from here, wouldn't it be better for 120v and avoid the 220 issue all together?

I could only find one 220v Dimmable LED in the size range we have now; but again from HK. It's only a spot type and not really the same as were using here. At any rate this is the one I looked at:
Dimmable E27/PAR16 9W 7W 6W 5W 3W Screw LED Light Fitting Bulb Lamp 110V/220V | eBay

I can select 220v and its' dimmable, so this is closest "circuit" wise; but not the best dispersion angle. Are there bulbs like these 7.5w dimmable in 220v?!? I'm trying to get at least one to know 220v aspects.

Hmmm, I suppose I'll end this here and make a separate for discussing about the "tuned" system more. I want to try and explain the caps here (in general), to help with "what" they are doing. I don't want people to miss my "concept" of tuning, so this will help as an additional primer to logic.

I suppose a last note that applies to all this from "the man"
With Tesla's first works and lectures with Disruptive discharge; he made note of the factors that are involved. He expressed these relating to what a coil might "naturally" want to do.
Frequency, Inductance, Capacitance, Coupling, Voltage and Current.
This was irrespective of the load at this time and before he knew "undampened" oscillation. I note this mainly as he included Voltage and Current into "tuning". This will relate later and why I choose particular ranges with a particular coil.

Thanks again, PB

[Lidmotor]

LED bulbs

@All


I bought a 12v LED bulb off Ebay that was rated at 450 lumens @ 1/2 amp and compared that light output to the 110v that I have are running in the circuit. The 12v beat it as far as brightness but the amp draw was more. It was a trade off. The 12v LED is a bright white instead of a soft white also so it was a little like comparing apples to oranges. The 12v also got quite hot after awhile. BUT-----I have to admit that for pure simplicity this 5 watt 12v bulb just hooked up to a 12 volt battery, put out HUGE light at the cost of 1/2 an amp. It is a beautiful bulb that looks like the 110v bulb. It cost $8.15 shipping encluded, was sent from New Jersey, and arrived in just a few days. Just remember that it is a DC bulb so you can't gang a bunch of them together and get the performance that we are seeing with the AC setup.

New E27 5W 12V High-power Bright White LED Light Bulb | eBay


@Slider
I like your idea about the color changing led as an amp draw indicator. Great find.

Lidmotor

P.S>----I just ordered one of these:

5W Cree LED Lamp Bulb E27 White\Warm Light 85-260V Super Bright Metal Shell CE | eBay


It is the same bulb as the 12v LED that I bought (I think) only with AC guts. This one was $7.50 and ships from China. I got the warm white (2800 to 3500K) this time. The specs on this bulb look great! It works on 85V to 260V and 50 or 60 Hz. It is rated at 5 watts and put out 500 lm. The LED is a CREE. I should take a week or two to arrive in the mail.

[Slider2732]

Theory may guide, but EXPERIMENTS DECIDE

Here's something which may fit that thought process...
By rights, a flashing LED should flash, however, on these types of circuits they become very very useful indeed when they suddenly don't
Quite by accident, because it was the only LED to hand, I discovered a hopefully beneficial current sensing tweak to Lidmotor's method.

At low current settings, the LED will indeed alternately flash red and blue. But, move the pot to a lesser resistance for a brighter bulb and the LED glows just red and solidly. Move the pot even more and both internal lights shine out together as a warning of overcurrent, when the circuit trips into drawing heavily. Such a condition is not beneficial...for an approx extra 20% brightness there may be suddenly upwards of 0.8A used, compared to 0.13A.
The LED in the video below is from a Dollar Tree flashlight. Wish i'd kept one whole or at least the case, but they are squared box types that take 3xAAA, with 4X ultrabright white LED and 1x flashing red/blue.
The bi-colour flashing LED allows for a setting that is easy to achieve, giving the best light to current ratio for the circuit.

Here's the short demo vid: Bi-colour LED - for SJR type lighting - YouTube

[PhysicsProf]

Lots of progress here, I'm excited to see that!
Glad to see you trying things and getting that bulb to glow brightly, slider!

@Lidmotor -- somehow I missed this insightful comment of yours before, but I agree with you completely -- and thank you sincerely for this:

@PhysicProf
A scientific analysis of what we are doing is sort of a double edged sword. It is great to know exactly what the numbers are and the efficiency but it can be discouraging also. I like the way that you have an open mind about these projects and can apply the science as a (+) tool rather than a (-) one.
To many people just want to "run the numbers", look at the "laws or rules", and never build anything. One of the guys who commented on my last Youtube video tried to build this device on a simulator and could not get it to work. He asked me what was wrong and all I could think of was----his whole approach was wrong. He should just build the darn thing, watch it work, and then analyse it. I know that this sounds like a backwards scientific way but it works for me. Each project "build" adds to the overall knowledge base and the next project gets easier.

Not backwards at all, especially in a cutting-edge field like this! I am also an experimentalist -- "Theory may guide, but EXPERIMENTS DECIDE" = the gist of what I taught students for a few decades before early retirement...

Sometimes I discover something just by dumb accident. That is when I am having the most fun. You do need a basic understanding of the "laws and rules" to do this. Unfortunately I learned those things in a very boring classroom long long ago and many of them are fogotten. I needed a Dr. Jones.

And I need a Lidmotor, and a PeanutButter and a Slider ETC.! -- ALL of us pulling together, learning.
I'm confident we're going to "crack this nut" yet! High light-output per watt will benefit MANY, I'm certain.