ZPDM's Oscillator: A new?? oscillator

[ewizard]

No I haven't tried this circuit yet. Fried them on other experiments. I've got other tranny's laying around I may try but I'll probably pick up some of the TIP 42 and 31's anyway. Thanks for the info. It will probably be sometime after the Holidays before I actually get to this but I'll for sure be playing with this.

[barbosi]

To be fair, all reversed biased npn transistors did not have (in my knowledge) any bias on the base of the transistor.

Since the schematic that was posted has for the active transistor a potentiometer biasing its base, it would be interesting to see if that factor can make the functioning predictable. This is what I meant by the term “deterministic”.

If someone can take any stock transistor, make it oscillate and predict the frequency and eventually the duty factor based on the bias, that is surelly a gain.
Unfortunately, this will require ample testing using a oscilloscope. But I bet, determined people will do a throughout research, so I wish you the all the best.

[ZPDM]

@Ewizard,

I am glad to hear that, I may have fried one 2N222 but even there I am not sure if I didn't accidentally short it, if you open the 1 MOhm pot too quickly it does seem to "short" but for brief intervals at least doesn't seem to fry anything.

This circuit doesn't seem to mind not having a load but is quite sensitive to whatever load you attach. I originally looked at it with about 15 uF 450V caps which it plays nice with and with regular 35V caps, going from 1.5V to a 1.5 V batteries is no problem or 1.5V to 4x 1.5 V batteries in series isn't a problem. Going from 6V input to a 1.5V battery though it won't oscillate for me and looks to be drawing an amp or so for the little time I keep it connected. 6Volts to 6 Volts again no problem. This is all with my set-up of a 2 inch toroid with 70-80 turns. The, well its Christmas Eve, the "blessed" thing, does seem to like to oscillate.

[ZPDM]

To be fair, all reversed biased npn transistors did not have (in my knowledge) any bias on the base of the transistor.

Since the schematic that was posted has for the active transistor a potentiometer biasing its base, it would be interesting to see if that factor can make the functioning predictable. This is what I meant by the term “deterministic”.

If someone can take any stock transistor, make it oscillate and predict the frequency and eventually the duty factor based on the bias, that is surelly a gain.
Unfortunately, this will require ample testing using a oscilloscope. But I bet, determined people will do a throughout research, so I wish you the all the best.

Barbosi, understand my background and degree is as an MD so while I am educationally not a fool I am also not fool enough to figure I understand this circuit more than enough just to say this looks quite strange to me. I hope you and others with tools and experience in the area take time to replicate and investigate.

[Guruji]

Hi ZPDM
thanks for sharing. Why are you saying self oscillating?
There's a battery in this circuit.

[ZPDM]

Hi ZPDM
thanks for sharing. Why are you saying self oscillating?
There's a battery in this circuit.

Can't say I follow you there Guruji, it is the battery, the volts and amps from it into the inductor, which is turning on and off, i.e. oscillating. Of course there is a battery in the circuit. It is self oscillating in that like a Hartley oscillator, Joule thief oscillator, Slayer Exciter oscillator, Simplest LED oscillator and others it begins to oscillate on its own once all the connections have been made. This can be contrasted with say a Bedini SSG circuit where a first push of the rotor is required for the spinning magnets to generate an inductive signal to continue the oscillations or for that matter a physical pendulum which requires a push to start oscillating. It may be redundant to call any oscillator, self oscillating but that is the distinction I was driving at.

[ewizard]

I just came across some info from Professor Pappas on his web site that might be a part of the puzzle for this effect: http://papimi.gr/transistors.htm

Been busy but still going to give this a try. Part of the holdup has been my source for the transistors is having some web site problem preventing me from ordering them.

[ZPDM]

Thank you Ewizard,
To be honest since the New Year I have been away from electronics experiments entirely, been too busy enjoying the whole Dogecoin cryptocurrency thing. You know those young pups just sponsored the Jamaican Bobsled team with Dogecoin? Young kids using money (well Dogecoins but it sends people to the Olympics) to have fun. This is one of the most positive developments I've seen in a long time.

That is a great find concerning possible theory on what may be going on but I also don't want to get ahead of myself. I would just be very appreciative if someone just replicates the oscillator so I know that there wasn't something strangely idiosyncratic that I was doing. And if it does turn out to be an easily repeatable oscillator maybe people will start to have fun with it like they do with the Joule Thief circuit. Let me know what you find when you have a go at replication. -Z

[ewizard]

I finally got the transistors today another way as it seems his web site (thaishine.com) just isn't working right anymore. I'll probably get something hooked up in the next couple days here and post results with o-scope readings.

[ewizard]

Scope shots! Interesting circuit

I think this is a very interesting circuit. Initially using a power supply at 1.5 volts I was getting too much noise from the line and had not yet tuned the pots so I went to a couple NiMh batteries that together were reading about 2.6 volts (not fully charged). Initial tuning found a sweet spot on the scope that produced some strong spikes and full of harmonics? I am posting scope shots in order they were taken and successive shots were of different voltage input settings on the scope which show a lot of harmonics I think. All were done with a 10x probe. Adding a ground to the circuit (Tip42 collector) greatly increased size of the spikes. I had to go from 0.2V division to 1 volt division to get roughly the same size spikes just after connecting a ground wire (while on battery power). The 2.6 vots with the 2 x NiMh batteriess yielded 32 volts (Nice!) out as measured no load at the FWBR DC output. After getting the 1 megohm pot tuned in for the above tests I switched to a power supply and adjusted it to about 2.6 volts input. This yielded even larger spikes. It also seemed the circuit was very sensitive to the location of my hands even at several inches away. The TIP42 seemed to be getting quite warm initially after finding a sweet spot so I put a big heat sink on it and it seems to be staying cool enough now.

Putting a Fluke multimeter across output of FWBR roughly doubled the size of the spikes on the scope pattern. My fingers were not touching any part of the circuit or the meter probes except on the rubber part. Scope pictures coming shortly. I used my 40Mhz scope rather than the digital 400Mhz scope as it is easier to photograph but it's still a decent scope for the $3 I paid for it at a garage sale

Also interesting is that nearly all the scope patterns were showing a negative output except the last one (I think that was when I hooked up the Fluke meter - maybe the leads were acting as an antenna since they are roughly 4 foot long each).

[ewizard]

Just getting started playing with this. IIRC those output spikes are bigger and stronger than any of the other battery chargers I've built (like the radiant ones from a couple members here and OU). Considering the low input I can certainly see how this would charge batteries good. I've got several to test it on and may see how it does with charging a set of 350 Farad caps in series (6 of them set up like Laserhackers power boost pack). That last scope shot is the one where I put the meter across it. Also once a battery is on it you see positive spikes above the zero line.

[ewizard]

I forgot to mention that I threw a real twist into this circuit. As I didn't have a coil wound yet like ZPDM mentioned I just grabbed what I had laying on the bench which was a toroid split in half with 2 strong Neo magnets between the halves in repelling mode. Pic's of the coil below. I'll be trying it with a more conventionally wound coil soon. I've got another toroid this size that has not been split (about 4" diameter) I'll put a hundred wraps on.

So far with a little more tuning I have found that 1.5 volts input is optimal. If I go higher in voltage the output voltage goes DOWN! With 1.4 to 1.5 volts in I get 51 to 53.6 volts out. Current measurement is interesting as it appears I'm only getting 0.7 microamps out and yet it is charging a battery up. Even more interesting was that I was still seeing 51 volts output with 0.2 microamps when I disconnected one leg of the input to the FWBR!!! That's just ONE WIRE connected to the FWBR input and still seeing 51 volts output at about a fourth of the output current when it's connected. I'll have to look into whether power supply and o-scope connection might be coming into play with that though....

I find that with the pots I'm using it's hard to fine tune and hit that best resonant spot. Probably need a 500 or even 100 Ohm pot in series to better hit the resonant spot. When I do hit that point you can hear it in the coil and the charging rate increases greatly. I look forward to trying a number of things with it.

[wrtner]

I have found that 1.5 volts input is optimal.

Did you measure the current here?

[ewizard]

Later I was trying higher voltages up to around 4 volts but current input tended to stay around 0.5 amps when I was getting resonance based on sound and fastest charging as well as the o-scope signal. Very narrow band for tuning to get any charging. Too little resistance and current will go high to around 1.5 amps (with loss of all spikes) and too low it will drop to 0.0 amps and no spikes. It's not something that would appear to be even close to any free energy from those numbers but it is odd how it charges batteries with such low current output. One other thing I did last night involved an 18650 Li-Ion battery (large Li-Ion) which had dropped to 1.6 volts. Normally if they go below around 2.8 volts they are bad and not rechareable. My Li-Ion smart charger would not even try charging it. So I put it on my circuit above and it went from 1.6 to over 3 volts in about 45 seconds. At that point I stuck it on my smart charger and it was fully charged in a couple hours and has held that charge this morning. So it appears to be recovered.

I was also charging a NiCad pack for a drill and while not especially fast it was still charging it from around 10.0 volts to 12.5 volts in about an hour. It seems odd to me that it can charge at all when output seems to only be less than one microamp. My guess is something is doing charging that can't be easily measured.

[ZPDM]

Later I was trying higher voltages up to around 4 volts but current input tended to stay around 0.5 amps when I was getting resonance based on sound and fastest charging as well as the o-scope signal. Very narrow band for tuning to get any charging. Too little resistance and current will go high to around 1.5 amps (with loss of all spikes) and too low it will drop to 0.0 amps and no spikes. It's not something that would appear to be even close to any free energy from those numbers but it is odd how it charges batteries with such low current output. One other thing I did last night involved an 18650 Li-Ion battery (large Li-Ion) which had dropped to 1.6 volts. Normally if they go below around 2.8 volts they are bad and not rechareable. My Li-Ion smart charger would not even try charging it. So I put it on my circuit above and it went from 1.6 to over 3 volts in about 45 seconds. At that point I stuck it on my smart charger and it was fully charged in a couple hours and has held that charge this morning. So it appears to be recovered.

I was also charging a NiCad pack for a drill and while not especially fast it was still charging it from around 10.0 volts to 12.5 volts in about an hour. It seems odd to me that it can charge at all when output seems to only be less than one microamp. My guess is something is doing charging that can't be easily measured.

Ewizard,

I greatly thank you for replicating this and exploring it with more detailed tools then I had available. I apologize that I have been busy with a new job, but I hadn't forgot about this circuit. I think it is likely if not quite safe to say that reversing a conventional transistor can lead with some consistency to some unusual and potentially useful oscillatory behavior. I'll try and get back to this later this spring and until then can't thank you enough for your efforts. If you get a chance though, will you post more about your set-up, i.e. core, coil geometry, windings, wire diameter. Seems clear to me now, flipping a transistor can be zany fun