G'Day UFO and all others
I really like to thank you all especially those that helped me to try and debug my machine

Now with Torpex and Larry's input
I made up a single Mosfet circuit and used this to see if the Regulator and then the Oscillator were working (so I would only Smoke 1 Mosfet)
The regulator does work the 555Oscillator does Oscillate but I think it is not sending the proper signal or is not strong enough as it seems to change the frequency along with the duty cycle.

@ Larry
I was originally using Bob's Oscillator and when I was having trouble(Thats me all over) You suggested on page 28 post#111 that you modified this oscillator but as UFO suggested we should keep it simple and use the 555 Oscillator so I continued with it
The others all seem to have success the Bob's updated LM339 Oscillator page35 post #1026 Here
My Motors got me to Tap into Radiant Energy

or perhaps Larry I should build the one you have posted on page35 post #1029 the one also showed on page 28 post #111 it is here
My Motors got me to Tap into Radiant Energy

@ Torpex I believe you are using 2 double LM? in your circuit

Please I appreciate Your Input

@ John Stone

I recieved today in my mail I think from GlobalSource I thought that it may be of Interest to you Its above my head but this is what you and others have been discussing I
Fairchild Semiconductor's Power Clip Asymmetric Dual MOSFET Helps Designers Achieve Highest Power Density and Efficiency in Power Supply Designs


Kindest regards


Still working Hard at it

Hi Ian
The main difference between my circuit and Bob's is mine has the complimentary second output.
The FET you linked to is a surface mount and is only rated at 25V instead of 400V and it has 2 FETs built into one device. If you are blowing them at 400V I don't think these will work.
Did you determine what the output of your oscillator was?

Warm Regards
Larry

thank you JohnStone
1. i have nothing but seriously basic skills in these fields...i could probably manage to burn myself with my soldering iron if i tried ;-)
so i hope you will draw the opto driver some time and that i can find the FOD3180 or a suitable replacement.
The FOD 3180 are not available where i am. does anyone have some other numbers by which to identify suitable opto-couplers when ordering?
2. thanks i'll openup some psu's and see what i can find.
3. my first guess was underestimation of completion time required and second guess was isp trouble/interferance. very glad to see he's back

G'Day Larry
The fets I know nothing about I just thought they may be interesting
Thi output from the Regulator is 13.4v into the Oscillator it Oscillates Duty cycle 0.4% to 99.?% but the output varys in volts like 4.?v

as I said I made a single Mosfet circuit to test without the coil it works as far as the regulator but not the Oscillator so I connected the 6 Mosfet Circuit and it works with the regulator but not the Oscillator.

I will try the original Bob/s Oscillator that I made first and see how it works today I was just checking it to see that I wired it OK
Kind Regards

Still working at it

Again and again: we need to seprate different levels of maturity in our setups. DO NOT MIX THEM!

Level A: Low side driver with N-FET
If you are not experienced with electronics please build your setup with low side driver as posted already and get it running with HER.

Level B: Low & high side driver
The addition of a high side driver is an improvement but no essential asset for basic research.
For low voltages a direct driven P-FET is suitable if available for you - as posted by Ufo.

Level C: Galvanically separated driving

• makes P-FETs obsolete
• it is a very rugged and variable building block
• requires more expertise

C1: optical isolation
Please consider that for an opto separated driving we need more than an opt only:

• opto separation for drive signal (opto coupler)
• driver after opto (FOD3180 or similar drivers contain both items)
• galvanically separated small DC/DC PSU for the driver

Please consider that those members pioneering the galvanically separated driving do not have final sugestions as reliable instructionables yet. This is currently the direct way but requires those special components not being available all over the world. I am so sorry of that.

C2: Inductive driving
There is an alternative for driving -> via small transformers. This is a simpler but most tricky approach. I have no experience with it up to now.
I will try it for you after having success with my opto approach.
Others with experience and willing to pioneer this matter are welcome.
In order to understand the implications see page 30 or But PLEASE do not understand these expert hints as instructable for easy replication. Simplicity has its price! You do not get it for free!
Google search. "fet pulse transformer"


Please find your level of experience and perform it. Ufo tries to get all levels mentioned above with him. Wait for his next disclosure. Focus on him as guide!
@Ufo: Glad you are back and well! Festina lente! (Make haste slowly) . It is not easy to wait but do not haste! Thanks for your effort and contributions. They are very much appreciated.
rgds John

__________________
Experts spend hours a day in order to question their doing while others stopped thinking feeling they were professionals.

Don't check your oscillator output with a DMM... only use a scope. The DMM gives you the average (RMS) output and not the amplitude of the signal feeding the FETs base. Changing the duty cycle of the oscillator will change the RMS output of the oscillator, but should not change the amplitude of the signal. You need at least 4V signal amplitude to turn on the FET and your oscillator should be putting out between 10V and 12V amplitude at any frequency or duty cycle. How are you coming up with 0.4% duty cycle? 5% is really hard to see on a scope?

Exactly what are you saying here? You are getting regulator output (13.4V), but you aren't getting any oscillator output after you attach the FETS? Do you have a coil attached to the FETs?

If your oscillator works before you attached the FETs and not after you attach the FETs, it most likely isn't your oscillator, but your FET circuit. You could switch oscillators all day and not fix anything. After you attach the FETs, what dose your oscillator output look like on a scope?
If not oscillating, is it high or low? This will give you a clue as to what the FETs are doing to your oscillator.

Let me know the answers to all questions I asked here... I can't help if I don't know.

Regards
Larry

To all guys owning no oscilloscope but a PC:
Poor Man's Oscilloscope
Soundcard Scope
41Hz Audio - PC soundcard as a scope / measurement instrument

Googel search:

These progs can be great help for basic watching what is goning on at your oscillators.

In order to protect your PC please use a voltage devider 1:10 between your circuit and the sound input of your PC.

• The sound input is limited to about +- 2.5V AC!
• 20Hz ... 20KHz sine shapeed signals.
• Square about 10Hz .... 5 KHz
• NO DC measurement!
• Never apply HV signals to the PC!

Especially the prog from Zeitnitz is widely used and conatins a signal generator as well.

This is a suggestion for you low funded enthusiasts in order to enable you to overcome serious limitations. We want you to succeed as well!
I did not use these progs myself and can not help but please share your experience in order to help others.

__________________
Experts spend hours a day in order to question their doing while others stopped thinking feeling they were professionals.

Hello everyone!

Thanks JohnStone that reminds me I have a computer to setup for an oscilloscope/workstation. Great guides, that will help me get that computer going while I am waiting for my part orders.

Could you add an extra diode immediately after the mosfets as another option to protect the driver? I found some low loss 600V 11.5A and 1A rectifier diodes. I want to try to use them for for mosfet protection as well as output filtering, what do you think?

Really good specs on the rectifiers that make them look appealing but I'm a noob to this circuitry stuff, so, please be kind.
IDV02S60C 600V 11.5A
CSD01060A 600V 1A

Good to hear back from you, Ufo!

Best wishes,
matt

Thanks John
This should really help those that do not have a scope. I checked out the links and all require paid registrations of from $19 to $49.

You can get a hand held digital scope on Ebay for $79

Here is a free software scope (BIP Oscilloscope). Just basics, but will work for this project and free.
Downloads - BIP Electronics Lab Oscilloscope - 3.0

This can be used to wire up your sound card to work.



Think I will put it on my laptop so I will have a more portable scope.


Regards
Larry

Hi Matt,
where do you intend to add a diode in specific? Where cathode, where Anode?
Where does teh FET live? bottom ? top?
rgds John

@Larry. Thanks for your addon regarding PC scopes. I checekd the sites not thoroughly enough.

__________________
Experts spend hours a day in order to question their doing while others stopped thinking feeling they were professionals.

I'm sorry John,
My mistake, I was thinking of the circuit as using a P channel mosfet to drive the coil. I have N Mosfets.

Like I said, I'm a circuitry noob.
I didn't mean to waste time.

Best wishes,
matt

Hi All
I set up a photobucket type image host on my website. It is for use by people on this thread only.

Link:
Larry's free picture storage] -

You are free to use it if you wish. You will have to register to use it.

Regards
Larry

Hi Larry,
That's a good idea to get a central image repository going.

Maybe we should also start a wiki so we can keep talks about specifics organized?

I like the idea of IRC too for those looking to stay in touch real-time, with the chat kept logged on a page of the wiki so it stays somewhat forum like. All in due time, just some ideas to see how well everyone would participate.

Knowledge is power.

Best wishes,
matt

Hello all. Been lurking full time for 3 days, starting at page 1, finally up to page 40. Like most of you, spent two of those days hitting the refresh button on page 38-39 every hour or so
Even prior to that I was scoping out scopes. As an old electronic tech/engineer/licensed electrician/ne're do well I feel quite naked without one. I would like to add to this portion of the discussion.
I run Linux, most Linux folks are probably aware there are at least 4 or 5 totally FREE software scopes for Linux. The one I have is xoscope and it appears quite adequate (judging by watching the noise level, as I have never connected a signal, kinda scares me, since I have only the one computer. Although the circuit Larry presents would be just fine for say, checking right at the oscillator output.) I should think there are freeware Windows scopes out there as well, and of course, you could always d/l a Linux Live CD from any of the major distributions and run the Linux versions without even having to load Linux on your HD. You'd have to reinstall the scope on every boot, but not a big deal.
A couple of circuits that would provide a layer of isolation for your computer:
I especially like this idea, because all you gotta do is buy a $1 usb sound card from ebay, voila! if anything happens, buy a new $1 soundcard, computer sound card unscathed (you don't really need the link, that's the whole idea):
usb sound card
Dang! lost the other one. But the $1 USB soundcard idea would win, the other was $5 worth of parts built around a PIC, too much effort for roughly the same device. Of course, you must use a software scope with it.
Finally, I have found Aliexpress to even beat ebay on scopes. It seems to be sort of a Craig's list for China, only new items though. A sales front for them to expose themselves with. It's like DealExtreme on steriods.
$25 USB scope (good to 3khz, probably enough for this stage, but pretty limited)
Nice selection of across-the-board scopes
On the subject of wire for the coil, perhaps the most costly single item of this project (if you don't already have it in the garage), I found that about 200 feet, or 1 pound of 18awg magnet wire is just under 3 ohms. Meaning that two strands twisted up Litz style and connected in parallel would be just under 1.5 ohms, UFOPolitics target number. Here is a link to the cheapest deal I have found so far:
1 Lb. 18awg Enamelled copper wire $16 (buy two, free shipping)
The purchasers of that deal complain because it looks like the wire was wound on the spool by hand, but that shouldn't matter to us, twisting it bifilar & all.
Lastly, I would like to add that I am very impressed by the camaraderie of the entire community here, the excellent work done by so many, and of course, last but by far not least, the superb work & skill of ufopolitics himself. And by that I mean not only his skill in conveying these concepts, not to mention rediscovering them in the first place, but his social skills as well. The deft manner in which he has dealt with the know-it-alls early on has contributed greatly to the ease with which it can all be absorbed. Bravo Sir!
And thanks to all you hard-working replicators too. You play an essential role in convincing us all!

sputnik
P.S. sorry for the length of post

Hi again @All
Finaly I've got my setup with a CLF going after accidental signal generator damage...
Inductive Collapse Lighting a CFL
In this short clip I am lighting a CFL by pulsing a pancake coil and catching the BEMF from inductive collapse of the coil's magnetic field.
There is a secondary winding tuned with a capacitor to form a LC tank circuit to resonate at lower harmonic frequency of the primary (250kHz&50kHz).
On a scope across the tank circuit I see a pretty clean sine wave.
Two batteries supply 24V to the pancake coil through a MOSFET.

I am eagerly waiting for the animations Ufo.
I am into 3D CGI too need some help with rendering? I use NewTek's LightWave 3D

kEhYo

Hello kEhYo77,

Nice set there!!

I use the old MAYA 3D and I love it!!...but batch rendering I have to perform in After Effects, and premiere also...well for final tracks blending-editing.

I have built my own machines (Servers) with very nice VPU's Video Cards and pretty big capacity HD...One Terabyte...RAM is also fine...but you know videos take tons of space...but is ok...will see it soon.


Thanks and regards


Ufopolitics

I was wondering why the bedini/cole circuit is not used for the driver section?
Here is the schematic I am using a hall on the front end and you can replace the 470 ohm resistors with 1k pots to offset the switching from each MJL

Here you can see that if you use the h11d1 with (I actually ended up using a 2955) this you can then use far less current to fully switch. I was using as much as a 20k and still getting good switching. But the important thing is that you can use the op-amp circuit and only need the one stage.

Anyway hope this is helpful. I think it would simplify things and have a lot less chance of the loss of costly Fet's...

Les

Hi folks,

Just finished hooking up my opto/driver, as a low side switch with small incandesent, worked great switching 24v.

Then tried a hi/lo switch with the small light, and it worked great.

I will hook up my coil tomorrow, I am sure it will work great.
Thanks John Stone.

One thing, be sure to check +/- input to your dc/dc converter,
pin 1 was -vin and i hooked up +12v, good thing I bought 4.

Thug

G'Day Larry
My Regulator input 38.? volts output 13.5v Its OK
The Oscillator has only one pot I was measuring with a DMM. There was from 0.4% to 99% Duty Cycle the Hertz was varying as the Pot changed from 0 to full on 50K I did not use My Scope as I have not as yet really understood how to connect it to read the Hertz or Duty cycle with it so I have to study it a bit more.

The CFL Bulb did not light so I checked the voltage on the output after the diodes and there was something like 4.?V The voltage before the diodes should have been the same as the Battery voltage it was not, I should keep the readings but I didn't, even the Drain was reading very little I just can't remember what it was but it should have been at least the same as the battery I

I made a single Mosfet circuit the same as the 6 Mosfet circuit connected to a 36v bulb (3 x 12v in series)load with 36v into the regulator the output switching on the single mosfet circuit and the light came on fully bright.

On the main circuit I disconnected the coil and diodes, I then connected the 36v bulb as a load in place of the CFL bulb. When I ran the setup including the Oscillator the Bulbs did not light up. When I disconnected the Oscillator and ran the output of the Regulator to the input of the Mosfet circuit the 36v bulbs lit up quite bright so it is quite obvious that it's only the Oscillator that is not working

When I made up this Oscillator circuit I used the Large Diodes which I now realise was an over kill but I was told that it should still work OK here is the circuit



Perhaps I should replace the large diodes.
Please Larry what do you think?

Kindest regards

Still Hard at it

Hi to all my US friends.. Nice to Know UFO is OK -

Any suggestions as to where I can buy a couple of these motors.. Seems that there is no shipping to international customers and I am "down under"

I would appreciate any suggestions.
Cheers
AmpsVolts

Hello Ampsvolts,

Sorry about that...but do not worry, it is just a "Toy Motor", however, it is based on a Five Pole Armature or rotor, and 2 permanent magnets stators...The reason I chose that one is because it is pretty cheap...and easy to convert into an Asymmetrical Model.

Now, it will also work a Three Pole armature with two stators...for now, I would like to Keep It Simple...and not going into more clever design machines.
So please do not spend too much money, into a "sophisticated one"...

I will start with low armature poles videos and explanations...3 and 5...but after you learn the procedure, I know you will make almost anyone into this Tech...
The most important part is that they both be, the same exact motors, same rotor number of poles, same shaft diameter,same stator number...meaning, same model.


Regards


Ufopolitics

Hi All

Glad you came back UFOpolitics - worried like the rest!

Here's a link to the High-Speed 9-18VDC Motor in the UK:

RadioShack® Super Speed 9-18VDC Hobby Motor - RadioShack.com


Regards

John

Hello Kogs,

Too bad your still having trouble.

1. Just replace the 555 and see.

2. Ian, if I were you, I would buy a breadboard, to build a prototype, at least until you get become really good at building circuits, I know some people will disagree, but its a pain soldering and removing components, until you have a working circuit. just my opinion.

3. I would start with the simplist, 555 from Rymcybernetics, its posted back somewherere, but seperate your signal from your switching voltage. The way they draw it, you will need protection or you will fry 555's and 393's.

or

Do mad scientist circuit, less complicated then John's but works good to get started. You really cannot isolate duty from frequency adjust with the 555, no matter how you connect.

4. Maybe you did this, but post a schematic of your circuit, not the one you want to build, but the exact one you have built.

Also, hook up a cap, that will flash the test light in visable range, then you can see if the oscillator is working.

We'll get her going Ian

Thug

[Stop ever thing and learn to use your scope. You will save time in the long run.

The only thing that is obvious is that your FET board has a major problem. Without anything to trigger the mosfets (turn them on) you shouldn't have gotten the light to light the way you did it. If your oscillator works with out the fet board connected and not after words how could it be anything else other than the FET board?

Honestly I have been at this for a long time and I wouldn't think of working on this project without a scope. You are basically blind in one eye and can't see out of the other. There is no way to track down where the signal stops or at what state it stopped or if it even actually did stop if you can't see it and a DMM will tell you nothing helpful unless you know enough to be able to translate what you are seeing and at that it is very limited. Use your DMM to check your battery state and circuit resistance... that's all it is good for in the world of electronics.
Considering the handy cap we are working under I am convinced it is your FET board unless you have the coil hooked up also then it could have something to do with it... you have no way of knowing which is bad with both hooked up. The art of diagnoses is to separate, test each circuit for proper operation then eliminate the problem. If you have 2 circuit hooked up and they aren't working then it could be either one. Either separate them and test individually or replace both.

Sorry I can't be of better help.
Regards
Larry

Ian
I see 2 pots on your circuit board. I assume the 50K pot is for the frequency adjust, but what is the other one for?

Regards
Larry

Ok, Got two motors.. They actually had the right ones. Wow!
Looks like I would need a micro-scope if we are going to do something inside there...

And I Tried to post the schematic, didn't realize it didn't take. I guess I am all out of space here... apologies....

Les

UFO,
I want to start by saying thanks for starting this thread and sharing your knowledge. Great work going on here!
I'm lookin to build the lm339 circuit posted earlier in the thread but have a question about a component. I've attatched the drawing and dont know what this circled item is??
Can someone identify this component?


EDIT: my pic wont attatch for some reason! There is some component above the lm339 with a cap across the leads, I cant tell what this is. I'll try to get this pic posted.
Link for pic: Circuit -rev. 2 picture by bobfrench - Photobucket

Edit #2- I think I got it, is this just a terminal to feed the voltage regulator?

Hi Thug,
do you say that you have built your isolated driver circuit conforming my suggested circuit? I still had no time left to finish building. If you confirm the replication we can release it and encourage others to build it as well.
Did you make any modifications?
rgds John

__________________
Experts spend hours a day in order to question their doing while others stopped thinking feeling they were professionals.

JohnStone,

All I had time to do so far is ;

1. Built low side switch, tested with small incandescent, it lit and flashed.

2. Connected coil with low side switch, worked, and lit cfl, around 300 Hz.

3. Connected incandescent light to Lo and hi side, lit and flashed well. I have a vid ( sorry it's not great quality), which I will load for you as soon as I get home tonight.

I was only switching 24 v.

4. Hooked up coil, and puff, I lost my dc/dc converter, on the high side, at least it got very hot, I shut it down and had to leave, so I'm not sure what has happened yet.

John, I know I suck at vids, so please, tell me what you need me to do if its no good, where you would like scope shots, etc.

Thug

@thug & All

I tried the high-low side driven switch like that and had similar problems. Smoked some silicone too
In the driver's datasheet there info on suppressing negative transients when hard switching inductive loads by adding more resistance to the gate lead. Eventually I abandoned that config and I am working with low side only MOSFETs and IGBTs with low RdsON and high dv/dt
Tesla did abrupt switching with spark gap and mercury type low resistance commutators and others confirm that the BEMF effect is most powerful when the time to turn on/off the transistor approaches 0!

kEhYo