Bedini Solid State Oscillators

[kcarring]

My First Multifilar

After many, many months of experimenting with my 26" alloy rim no iron setup, and a bought PCB/Coil from rick (basic SSG), 42 cycles later I decided, ok, I will take a shot at a solid State. Especially after seeing soem videos of a fellow who goes by 49er, he really has some copper singing that fellow.

I have since taken his advice, made som modifications, including getting the protection diodes to work, and reducing my base resistors to 22 ohms each, with an additional 68 ohm resistor between my pot and trigger start. This allows for very low hertz operation, when desired.

Next on the list is a momentary switch (with a resistor?) between B & C, for a start switch. My circuit works, but doesn't like to start, easily. Been thinking about winding very fine wire at the end of the coil as a pickup to (even dimly) light an LED as an oscillation indicator.

For me, it's not about FE, at all, it's about the fact these things desulphate. I have a Yurt in the mountains and the eventual goal is to have this desulphate batteries, as they do get "left sitting a lot". This is the start anyway, my battery bank is 20 T105's and 6 Rolls. 1800 watts. Still on 12v.

As I said, I have made some improvements; noted above since this vid. It seems to have no problem charging a 27 series RV battery 100 a/H - and you know, at low amps, maybe 500 mA. So considering there isn't a charger on the market (except maybe at r-charge!) that does that... that's rather unique in itself. And offers flexibility for the "small solar panel".

One of my main goals is figure out how you can use caps to create the stability necessary to run one of these things directly off of solar.

Critique welcome, even harsh critique - I can take it.

Bedini SS SSG 2T - First Attempt - YouTube

[blackchisel97]

It looks good If you're after good desulfator: get some "serious" wire (21 or thicker), MJL 21194 and make aircore trifilar (to start with), using 100' of each strand twisted together. Make voltage divider to bias transistor so it will self oscillate (20k/10k resistors).Try both config. First,with one end of trigger wire to neg. - take a note of your charging rate. Than try inverted trigger, with wire connected to the positive rail of your input. Current input will go up but notice your charging rate. In this configuration you'll need very low input potential to start oscillator. It will charge in cloudy day. You'll need good size cap on the input (25-35V rated) to couple impedance between PVC and coil.
The reason that your circuit hesitates to start maybe too high impedance of your coil. Too much or not enough wire may cause a problem.

Vtech

[SkyWatcher]

Hi folks, Hi kcarring, thanks for sharing the video.
I'm using the joule thief circuit with the parallel resistor/capacitor/diode network off the transistor base.
This turns on the transistors much better and prevents the heat normally seen with the joule thief at higher voltage inputs.
Also, the joule thief circuit really kicks out the voltage for desulphating and charging.
Though if your going to use bedini type pulser, you could use the variant with a resistor from positive to base, then the trigger wire from emitter to resistor to base, equal resistor values seem to work best.
This should start oscillating with no problems.
And then the standard bedini, with the 20kohm from collector to base and 10kohm from base to emitter.
Hope that helps for now, otherwise, i can post circuits if needed.
peace love light
tyson

[kcarring]

It looks good If you're after good desulfator: get some "serious" wire (21 or thicker), MJL 21194 and make aircore trifilar (to start with), using 100' of each strand twisted together. Make voltage divider to bias transistor so it will self oscillate (20k/10k resistors).Try both config. First,with one end of trigger wire to neg. - take a note of your charging rate. Than try inverted trigger, with wire connected to the positive rail of your input. Current input will go up but notice your charging rate. In this configuration you'll need very low input potential to start oscillator. It will charge in cloudy day. You'll need good size cap on the input (25-35V rated) to couple impedance between PVC and coil. The reason that your circuit hesitates to start maybe too high impedance of your coil. Too much or not enough wire may cause a problem.
Vtech

@blackchisel
Thank-you very much for the input. My existing build is trifilar with 18 gauge on the runs, 26 on the trigger, and the momentary switch definitely did the trick for starting it. you say "In this configuration you'll need very low input potential to start oscillator. It will charge in cloudy day." - that may be key to getting the solid state to startup unattended without the need for any overly elaborate PICAXE diagnostics, (and consequential programmed "triggering") etc. Simple is good.

This is my setup here.

@all
I've decided, gone are the days of ripping stuff apart, and starting over. The problem with this, is; as I see it: you don't end up with enough models to "compare", and thus you are forced to go by memory as to one working better than another, or you look back at your own youtube videos.

So I am going to build more, each time, I'm saying. On the other hand, I do like to "use what I have", to a great extent.

Now, I've taken some laughing and flack for this: but, this is my next attempt, and the main reason, was, I have the materials and they were cheap; very cheap.

And that is, aluminum magnet wire. I have 1000' of enamel coated transformer magnet wire. It is just that: transformer wire, not some hobby craft wire I bought.

To potentially bypass a whole bunch of back 'n force, I did a bit of research, and quite frankly I'm not the type to fret over $20.. if it works, great, if it doesn't it wont be the first nor the last $20 I spent/wasted. I'll quote a few things from Val-Tech Inc.

Aluminum wound transformers are lighter in weight than copper wound equivalents. < TRUE

Copper-wound low voltage transformers are better for "high-impact" loads because copper has higher tensile strength than aluminum. < FALSE

Aluminum-wound transformers have higher losses because copper is a better conductor. < FALSE

Aluminum-wound transformers have higher hot-spot temperatures because copper is a better thermal conductor than aluminum. < FALSE

"To keep the temperature below the insulation rating, aluminum transformers are designed with aluminum conductors of larger cross-sectional area than copper. On average, this results in energy losses that are the same for aluminum as for copper. Therefore, transformers of similar design with the same temperature rise have roughly equivalent losses regardless of conductor material. "

So arguments aside I am going to try it and risk all 20 dollars. I bartered the guy down, got a deal, it's in my hand. Similar copper would have costed me a fair bit more and I wanted to try this.

The 18 ga. wire is rated for 6A, whereas similar copper would be rated for 10A. I think from my experiences it's pretty safe to say the stuff is not going to see 6A per strand. And if it does, I'm probably doing something very wrong.
Both strands together in my existing SS circuit, combined create a draw of maximum 1.5A and when it runs at 100Hz, even, it doesn't get that warm, really.

So the question in my mind is not whether to use it or not, I am using it and I'll find out how it performs. The question is, instead 100' like the copper should i consider going shorter, longer?

Now am i thinking about this correctly, or not:

For a given number of feet of wire, aluminum being less conductive, will be more resistive, so one could shorten the length but then one would have less turns, unless the core was slightly reduced in size?

The easier answer would be "go higher gauge". But that's not an option, because I have bought the wire, but it is 18 gauge, not 23 or 24.

What are your thoughts? Please bear in mind, I'm building a desulphator, not a free energy device.


I look forward to building this and several more Bedinis before straying too far. Although on this variant I may wire it as a Tesla Oscillator circuit with cap and diode mode, as shown by 49'er who is pretty darn active at desulphating batteries.

I had planned on using a helical bobbin style form, though.

Thanks for any input!

[blackchisel97]

@blackchisel
Thank-you very much for the input. My existing build is trifilar with 18 gauge on the runs, 26 on the trigger, and the momentary switch definitely did the trick for starting it. you say "In this configuration you'll need very low input potential to start oscillator. It will charge in cloudy day." - that may be key to getting the solid state to startup unattended without the need for any overly elaborate PICAXE diagnostics, (and consequential programmed "triggering") etc. Simple is good.

This is my setup here.

I had planned on using a helical bobbin style form, though.

Thanks for any input!

When I built my first trifilar I used #24 and later #18 on all three windings. This is a bit different than rotor based SSG.
As far as solar, I had one setup with DC/DC converter on the input holding steady output to the oscillator. However, this will also limit your desulfator according to preset V value on the converter, despite the sun condition.
I cannot comment on Al wire since I've never used such but your observations are valid. And yes, simple is good

Vtech

[kcarring]

@Blackchisel

Thanks The plan was to use #18 on a winds, including trigger this time.
It has been brought to my attention that my circuit, as wired is acting a single (doublethreaded basically) run wire, operating two transistors. Which is what I fealt when I was making it up... Thing is I can't visualize what change would be made, to have it be a two coil, two transistor independant SS SSG. That is because I did not follow a diagram... I just built it as a "test" of whether I understood the SSG circuit, enough to evolve into a multicoil setup! Perhaps I don't LOL Not surprising really LOL. Hey.. it oscillates and charges... but, I would like to find a diagram somewhere that outlines, my mistake, or a description of how to attach the coil windings differently... would suffice too. I've scratched my head and tried to rethink it, looking at it... and it's not coming to me... the difference...

http://img88.imageshack.us/img88/8444/scan0001gx.jpg

[SkyWatcher]

Hi folks, good to see the site is back up.
Hi kcarring, i looked at your diagram and you have shown a line connecting the two primary coils, if that is how you actually have it wired, then yes you need to separate the two coils for each coil to use one transistor.
Otherwise, the circuit looks fine.
Also, I noticed when using 24awg wire for higher voltage spikes, that it desphulates my 12v-7ah sla batteries much better, though it may be different with larger batteries.
Because I'm using two transistors also, 7 strands, 3 per transistor, 1 trigger.
What transistors have ya got there.
peace love light
tyson

Joule Thief Running Christmas Lights - YouTube
edit: here is a video i made showing my small joule thief running christmas lights on 1.5 watts full brightness, normally 2.4 watts from wall outlet and then the other larger oscillator coils for charging stuff, the red one is air-core and the other one has a ferrite core, they both use the two transistors shown with cap/resistor/diode off transistor base.
Testing this camera i just got.

[kcarring]

@Tyson

Thanks! I was pointed out the same thing by another fellow 49'er who's been helping me along. Duh. I usually work on these things at about 2AM when I have "me time" - which is far from ideal, so.. I snipped the bumpstart back to one side of the circuit, eliminating it being active on two transistors. I am also going to try changing the trigger configuration, try both ways, and see if it jumps right into self oscillation, eliminating the need for the bumpstart completely.

I find the circuit charges quite well, actually - beyond my expectations. Considering when I run it in what I call "charge mode", i.e. low Hz, it pulls about an amp and half, I run it at an input that measures 14 volts unloaded (it's actually a SMPS 13V 65 watt supply). I was rather surprised it even ran on an SMPS at all.

When I get my new gasifier up and running most likely it will run, most of the time on a battery charged up by the "wood-mains" so to speak; a battery charged off the charger / genset that's fueled by woodgas. So that could be anything really, because the source will be 60Hz/110VAC.

The transistors are 2N3055's, but I am very tempted to split one open, I have a reason to believe they are fakes. Imagine that, faking a cheap transistor. LOL But apparently, it got widespread (about the time I bought them CHEAP on ebay.com).

I'm going to leave this build as is, not "add on to it", per say other than maybe do some playing / modifying, but I'm about to build my next one with aluminum wire. I'm definitely "breaking all the rules here", but i figure whatever, I put a good long time, 1 yr. in on the SSG kit, learned a lot... time to play.

With the exception of a couple of projects, biggest one being a ceiling fan, I have no use for a rotor on my battery chargers. And it does add a fair bit of cost/time.

Next I will build a 10 strand, all strands 18 GA. aluminum wire with new, authentique 2N3055's so 9T I guess it'll be. Unless I opt to make the trigger smaller... which gets me to the next question...

If I make a 9T SSG, and I make my trigger the same gauge as the run coils, is it possible that a better impedence match will occur, and have it be more reliable as a self starter? This could come in real handy, given that I could then automate the run operations using PICAXE.

I am also eager to try the Tesla Oscillator configuration too. But, my guess is the 9T will draw enough to service large batteries in a relatively decent timeline... if it ends up drawing 3-6 amps in low HZ mode.

I am also curious about this:

Let's say you had a solar panel at quite a height and distance from your charging battery, and location "of utilization", so for example a solar panel way high up in a tree... and a battery + light in shaded "out of the sun" location, 100 ft. away.... if that were the case...

could you then opt to mount your SSG right up by the solar panel, set it to run low HZ, draw maximum, couple it to the solar cell with say an 80V 15,000 uF capacitor... then... run all your lines BEFORE rectifying to DC straight off the collectors... then rectify down at the charge battery... strange I know... the "extension" would have 3 wires, two being from the collectors coming down, and one being from the charge battery negative, back up to the positive rail.

My guess is no, because of that requirement of having a long wire, DC.
It would be akin to calling it a "Bedini Micro Inverter" so to speak. And then I think hmm. Is there a way to do this?

"Also, I noticed when using 24awg wire for higher voltage spikes, that it desphulates my 12v-7ah sla batteries much better, though it may be different with larger batteries. Because I'm using two transistors also, 7 strands, 3 per transistor, 1 trigger."

I have left over 6 MJL21194's that Intend for a project, and I have a nice big spool of 24AWG. So perhaps I'll put this configuration together. It's really quite fine by me to have many/several Bedini's as I have many friends off grid, where I can simultaneously experiment with desulphating various battery banks; have them assist me using the chragers, collect information from each situation. Is your trigger 24 as well?

[SkyWatcher]

Hi kcarring, yes both of those multistrand coils have a single 24awg trigger wire.

As far as the length issue from your solar panel, I would use another oscillator up there to convert it to ac using a secondary, then a step down transformer in your house or wherever to power the radiant spike charger.
Though never tried it, it could work well though, i think.
In case you've never seen these other oscillator circuits, here they are.


Uploaded with ImageShack.us


Uploaded with ImageShack.us
peace love light
tyson

[blackchisel97]

@ kcarring Second circuit posted by SkyWatcher can be also modified (in case of trifilar coil) by removing diode from the collector and connecting third winding to the AC side of bridge rectifier. Bridge should be rated at 1000V. You can charge directly off the DC side. Also, trigger wire which goes to the negative rail can be connected to the positive - "inverted trigger", to improve charging. Base resistor needs to be adjusted according to transistor type. I use only MJL21194 for J.B circuit. I also use 12V5W bulb in series with base resistor. It helps with tuning. In my setup with trifilar #18 I have resistance adjusted and I can barely see bulb glowing in total darkness. You don't want transistor to get hot since heat means waste.

V

[kcarring]

Thanks guys! Lot's of variants to try. The one diagram looks kind of what i had by mistake (ina different manner though) and that is two strands, but behaving as one, really, so I would imagine it'd have a similar effect as a single wire in it's place of twice the gauge, however; one might also think it would fragment the magnetic field in a different way, too, possibly, and thus have some sort of effect or another.

About scoping a bedini:

Reading above I notice some information about dropping a scope on a Bedini circuit. Is there a bunch of discussion on this elsewhere? I guess that'd be the best question..

but ... The spike. If the charge battery is attached while scoping the collector and ground clip on the negative rail, or, essentially, the emitter...

1. Does the battery not quench the "spike" faster than the scope can see it? And thus produce sort of a square wave (esp @ low Hz)
2. Is it safe to do, or does one need to put a 10K resistor between hot probe and collector or something precautionary.
3. If your source for drive energy is a wallwart or SMPS plugged into mains, when you place the ground clip - are you isolated from mains ground, or does care need be taken to make sure you don't accidentally short to ground; like you might, for example, dropping probes across an AC circuit, across a shunt, measuring Vdrop to find the current?

If on the other hand the charge battery is not connected:
1. Usually my circuit just doesn't start or stay running, but my bump start would probably get it to.
2. In the couple of incidents I have done this, it occurred when I placed a 12V 20 watt light bulb in place of the charging battery to get some idea of current flow. (I connected to both output of the 2T circuit, after the diodes). What I found was the lamp did not light up, but oh man... those glass neons FIRED and I mean HARD. They were very bright and almost purple unlike the yellowish tinge they gave off running my SSG kit. Not surprising really the circuit was draw 1.5A, not 100 mA LOL.
3. Another thing I found was strange was that when the above happened, only one neon would fire at a time, never both. But, that said, I could tap the power on and off to the circuit, and perhaps one time one neon would fire, or, another time; the other would fire. It wasn't like it was always one particular neon, just never both, at the same time...path of least resistance?

Interesting stuff, indeed, but a bit unconventional to work with and measurements often very deceiving, possibly even meaningless. Current in, current out.. I don't believe really (the analog meter): given the pulsing. Pretyt much anything I've done can be used only in a qualitative manner for comparison to another measurement, this is more: that is less. Or as I see it anyway... I find it all rather hard to be conclusive with. Worst being the battery cycling tests, I did 42 and I see the point, but then again... I don't see the point. I came to the conclusion that I learned nothing really. Without an accurate count of watt hours in, and watt hours out... the whole test was essentially meaningless... but... it was interesting LOL Definitely, at times it doesn't seem to "add up". But al in all OU or not, I really don't care anymore.... it makes a pretty good charger and has uncanny characteristic of treating batteries nicer. Gd 'nuff for me

[blackchisel97]

One of my energizers changed ownership recently and I just received an email about overall performance. (This is trifilar, one tranny device which picture has been posted by Virus, few posts back). I want it to share since it is unbiased test done by person who had no previous experience with this technology except, being familiar with the name of John Bedini -

I hooked unit up to my two year old (no usage) 12v/500cca battery that would not hold a charge for seven days which on my cen-tech battery tester showed 12.4v/ 9.64 m ohms/ 261.9cca/ 50% storage capacity, after three cycles of discharging to 11.5v & charging to 13.5v & after a couple of days resting now it shows 12.84v/ 6.34 m ohms/ 450.7cca/ 100% storage capacity, during charging the meters on the desulfator showed a constant 12.3v & 1.2a . I just started on my five year old high usage work truck battery that was not spinning my starter as fast as it was a couple months ago, it was showing 10.4 m ohms/ 246.9cca & 50% storage capacity. As best I can tell the charger/desulfator is doing what you built it to do.

I just built it but credit should go to John Bedini.

Thanks
Vtech

[kcarring]

@blackchisel

When you are desulphating a larger battery; say a truck batter, or RV battery, do you take the voltage right up to 15v or as close to, as possible, and cold boil the battery, before discharging / cycling? I have a battery that after 3 cycles has shown no progress, but I have only been taking it up to 14. I suppose if I left it 4 days or something it might go higher... but the situation (as it arose) has always been that, it is sitting at around 13.99 - 14.01 late in the evening and I've been scared to let it keep going all night... When I return to the battery in the morning it is back to 12.4. During load cycles it works alright, but it holds almost all of its load capacity at under 12V, and the majority of the power curve is around 11.5-11.6V, which is low. I've been running a 12V 20 watt lamp as a load, so about 1.5A or so, and this is a 27 series 80 A/H RV deep cycle. I am tempted to run it at 18 or 24V, instead of 14 (input) volts in order to "get it up" to a hard cold boil. I notice brown flakes in the cells, and I've been told you need to cycle to cold boil, such that those lead flakes return to the plates, keep doing it until the electrolyte "goes muddy" I'm told.

Thanks for any input!
Cheers

[blackchisel97]

@blackchisel

When you are desulphating a larger battery; say a truck batter, or RV battery, do you take the voltage right up to 15v or as close to, as possible, and cold boil the battery, before discharging / cycling? I have a battery that after 3 cycles has shown no progress, but I have only been taking it up to 14. I suppose if I left it 4 days or something it might go higher... but the situation (as it arose) has always been that, it is sitting at around 13.99 - 14.01 late in the evening and I've been scared to let it keep going all night... When I return to the battery in the morning it is back to 12.4. During load cycles it works alright, but it holds almost all of its load capacity at under 12V, and the majority of the power curve is around 11.5-11.6V, which is low. I've been running a 12V 20 watt lamp as a load, so about 1.5A or so, and this is a 27 series 80 A/H RV deep cycle. I am tempted to run it at 18 or 24V, instead of 14 (input) volts in order to "get it up" to a hard cold boil. I notice brown flakes in the cells, and I've been told you need to cycle to cold boil, such that those lead flakes return to the plates, keep doing it until the electrolyte "goes muddy" I'm told.

Thanks for any input!
Cheers

Hello kcarring, You can take deep cycle batt's much lower than regular LAB (I mean discharge), even 10.5V won't hurt. I was taking them down to 8.5 V standing. Don't be afraid to let it run overnight. I let batteries charge for couple days steady on my 6 filar (in case if they hesitate to climb). Sometimes it may take several cycles to lower battery impedance. I try to get them up to 14.5 -15V. Gentle tapping with rubber mallet on the case won't hurt either. Yes, electrolyte will turn muddy which is sign of plate cleaning but it will became clearer since lose impurities will settle on the bottom. I run my solid state from 15-17V with inverted trigger. On average battery my input current is about 3 -3.5A.

Thanks
V

[kcarring]

@blackchisel

many thanks for the quick detailed response