Investigating "anomalies" in Bifilar coils

[TinMan]

Brad,

LtSpice has a model for Tlines that has an infinite number of lumped L/C segments and no resistance so it is lossless.  They also include a Tline with loss as well but, one can model a transmission line by creating your own L/C lumped segments that will give reasonable accuracy to known real transmission lines. 

However with the MEI device, if energy is entering from outside the Tline from say the aether for example, then that would never be visible in the simulation which is where I believe you were coming from and is correct.

Pm

Indeed

Sim's were never meant to simulate the unknown.


Brad

[TinMan]

OK,so im looking at getting serious with this now.

So what do you guys think about using this copper shielding tape ?

Looks like the go to me--2x rolls ?

http://www.ebay.com.au/itm/EMI-Shielding-Conductive-Adhesive-Copper-Foil-Tape-1-2-inch-x-55-yds-12mm-x-50m-/131961038332?hash=item1eb97da5fc:g:tZ8AAOSwNRdX9Qxz

Or this 5mm tape

http://www.ebay.com.au/itm/New-Arrival-5mm-x30M-Copper-Foil-Conductive-Tape-EMI-Shielding-Adhesive-Barrier-/151182809429?hash=item2333329955:g:1VsAAOSwzJ5Xdjtl


Brad

[Grumage]

OK,so im looking at getting serious with this now.

So what do you guys think about using this copper shielding tape ?

Looks like the go to me--2x rolls ?

http://www.ebay.com.au/itm/EMI-Shielding-Conductive-Adhesive-Copper-Foil-Tape-1-2-inch-x-55-yds-12mm-x-50m-/131961038332?hash=item1eb97da5fc:g:tZ8AAOSwNRdX9Qxz

Or this 5mm tape

http://www.ebay.com.au/itm/New-Arrival-5mm-x30M-Copper-Foil-Conductive-Tape-EMI-Shielding-Adhesive-Barrier-/151182809429?hash=item2333329955:g:1VsAAOSwzJ5Xdjtl


Brad

Yes that's the stuff!! Try looking at Slug repellent tape, might be cheaper but it's the same....

[TinMan]

Well,looks like i am ordering a couple of roll's.
The 5mm stuff should do to start with.

Anyway,who is going to have a go at working out where the !apparent! extra energy is coming from?.

Here we have a situation where we have both displacement current,and conduction current,both creating a magnetic field,that i would think is 90* to each other,and also opposite in direction in regards to the two wires.

Maybe Smudge is the man for this?


Brad

[Chet]

Partzman did mention computer ribbon tape?

[TinMan]

Partzman did mention computer ribbon tape?

Never have looked into it Chet,but im ordering 2x 50 meter rolls tomorrow.
Maybe the capacitance value will be too high?-1uF lol.

Now,a frequency down around 10KHz would be nice,as i have the higher powered wave form generator sitting there waiting,and it runs of a 12 volt battery--total isolation 


Brad

[poynt99]

Again, if you would please make this suggestion exact and explicit on a schematic, that would be much appreciated. I think I know what you mean but please, to avoid any confusion, show it on a schematic. Thanks...

Sorry.

See diagram below.

[partzman]

Here is a test result that needs confirmation by others using an isolation transformer with their particular coil.  In this case the tests indicate that the COP is improved with DUT isolation from the generator ground with my bench setup.

The schematic is redundantly included for ease of reference.

The first pix shows the results with CH2 connected across R2 with the second pix being a snapshot of CH1 for use in cosine calcs.  Note the phase between C2 and Ch1 and also the Math channel Pin.

The third and fourth pix show the same as above respectively except the inverted CH2 is now across R3 on the input side.  Again note the phase between C2 and CH1 plus the Math calculated Pin.  The COP is greater with the input current measurement as compared to the output current measurement. 

Again the 4-wire resistor measurements: R1 = 49.96 ohms, R2 = .994 ohms, and R3 = .997 ohms.

Pm

[ion]

How would a step up ratio look for the input " voltage transformer "?

With known turns ratio into L1 ( being open ended ) would this not " up the ante " ?

I'm ordering some more " Copper Slug repellent tape " today hoping to try and join in.

Cheers Graham.

Dear Graham

Altering the turns ratio would be good for optimizing the impedance match from the generator to the DUT. IOW if the input to the DUT requires a higher voltage (than the SG can deliver) but at lower current  then the iso transformer can be made as a step up, with higher turns count on the secondary.

Regards

P.S. 18 to 20  deg BTDC and running fine.

[TinMan]

Dear Graham

Altering the turns ratio would be good for optimizing the impedance match from the generator to the DUT. IOW if the input to the DUT requires a higher voltage (than the SG can deliver) but at lower current  then the iso transformer can be made as a step up, with higher turns count on the secondary.

Regards

P.S. 18 to 20  deg BTDC and running fine.

Oh

In all the excitement here ION,i forgot to reply to your email.

My apologies.

good to see it's sorted.


Brad

[Itsu]

Hi Itsu

For minimum capacitance,  each winding on its own half of the core would be good. Also a low dielectric material between the winding and the core or anything that can keep the wire from touching the core, such as balsa wood segments or other low dielectric material, IOW, keep a thin layer of air between the windings and the core.

COP of 10 is amazing, you are ready to loop. But be careful, use a small incandescent lamp limiter in feedback if you do.

BTW, the BNC to SG length not too critical, but the output of the core to the DUT should be as short as possible.

Good work!

Regards

Thanks ION,

"a low dielectric material between the winding and the core or anything that can keep the wire from touching the core" would be kind of hard to build, but i can give it a try.

My present 1:1 toroid has a relatively flat response across the first 100MHz according to my SA / TG, so will use that for now.

Unfortunatly i do not have much time today / tonight due to other obligations, hopefully tomorrow i can have a go at it.

Things are still confusing here, as i have seen negative power input when using the 1:1 toroid.

Hopefully it clears up tomorrow.


Itsu

[ion]

Thanks ION,

"a low dielectric material between the winding and the core or anything that can keep the wire from touching the core" would be kind of hard to build, but i can give it a try.

My present 1:1 toroid has a relatively flat response across the first 100MHz according to my SA / TG, so will use that for now.

Unfortunatly i do not have much time today / tonight due to other obligations, hopefully tomorrow i can have a go at it.

Things are still confusing here, as i have seen negative power input when using the 1:1 toroid.

Hopefully it clears up tomorrow.


Itsu

Regarding the iso transformer, what you have is probably fine, I wouldn't worry too much about it.

Partzman had suggested a method for low capacitance that is probably a lot easier, and I'm sure verpies did also at some time.

Regards

[ion]

Oh

In all the excitement here ION,i forgot to reply to your email.

My apologies.

good to see it's sorted.


Brad

Brad no need to apologize, I also missed a lot of posts and PM's running in and out trying to get the gas powered equipment up, lubed and running among other domestic things.

Kind Regards

[TinselKoala]

Can I get a witness? 

[TinselKoala]

I am now using a new toroidal coupling transformer. This one I found pre-wound in an old PC power supply. It is a high permeability ferrite toroid encased in a plastic shell, with the wire wound outside the shell. Separate windings, no contact with ferrite, etc. It seems to work quite well.

[Grumage]

Can I get a witness?

But..... What does it mean? 

I'm not used to these " new fangled " displays, please explain.

I've ordered some Copper tape 30 M by 6 mm wide.

If anyone has a moment to spare ( unlikely I know with present situation ) my old scope is a Fluke PM 3092. is it capable of doing any math functions for this particular experiment?

Cheers Graham.

[partzman]

Can I get a witness?

Let's see- Scope calculated Pin = 29.9mw, Pout = 1.5^2/19.1 = 118mw,  and Pin' = 3.72 x .0371 x cos(77.84) = 29.1mw.  Worst case COP = 3.94.

Edit: And all at 1.69MHz!

Pm

[Peterae]

I think we are talking about looping this the wrong way, fet/transistor oscillators should only be used to loosely couple to the device to start it, the output needs somehow to be fed back to reinforce the input in a controlled way using coils and caps if possible, think Tesla magnifying transmitter, think SM meltdown, if the gained energy on the output is indeed COP >1 then feeding this energy back to the tuned circuit either by direct connection or a 3rd coil will introduce a runaway situation if not controlled.
I dont think sticking a rectifier on the output and then using the energy to feed a fet oscillator is the way to go 

How you do this i have no idea yet.

[TinselKoala]

Let's see- Scope calculated Pin = 29.9mw, Pout = 1.5^2/19.1 = 118mw,  and Pin' = 3.72 x .0371 x cos(77.84) = 29.1mw.  Worst case COP = 3.94.

Edit: And all at 1.69MHz!

Pm

I'm afraid the "worst case" is even worse than that....

Here's the setup that gave those scopeshot measurements.

[TinselKoala]

But..... What does it mean? 

I'm not used to these " new fangled " displays, please explain.

I've ordered some Copper tape 30 M by 6 mm wide.

If anyone has a moment to spare ( unlikely I know with present situation ) my old scope is a Fluke PM 3092. is it capable of doing any math functions for this particular experiment?

Cheers Graham.

Graham, I downloaded the manual for your scope and it looks like it has pretty good cursors and can use them to compute phase angle in pretty much the same way the Rigol does. I don't know if it has automatic measurements, I just skimmed quickly through the manual,  but with some concentration and good use of the cursors you can get all the info you need for manual calculations. 

[ion]

I think we are talking about looping this the wrong way, fet/transistor oscillators should only be used to loosely couple to the device to start it, the output needs somehow to be fed back to reinforce the input in a controlled way using coils and caps if possible, think Tesla magnifying transmitter, think SM meltdown, if the gained energy on the output is indeed COP >1 then feeding this energy back to the tuned circuit either by direct connection or a 3rd coil will introduce a runaway situation if not controlled.
I dont think sticking a rectifier on the output and then using the energy to feed a fet oscillator is the way to go 

How you do this i have no idea yet.

Peter

I agree the methods using multiple conversions AC to DC and back are too lossy and present unusual loading due to peak charging and bad power factor load of the DUT.

I have some many pages back posted ideas for direct feedback with a phase correction network, but have not gone too deeply into it. First more is needed to prove the coil configuration works as claimed with an alternative thermal test method (I have doubts).

I am working on a number of protection schemes

Attached is one possible simplified safe looping means, conceptual schematic. Other more completed versions with circuit values and more detail are being worked on.

Regards

[partzman]

I'm afraid the "worst case" is even worse than that....

Here's the setup that gave those scopeshot measurements.

TK,

Do you have a profile on your iso transformer like inductance of each winding and the primary inductance with the secondary shorted or, if you test without the iso, are the results the same?

Pm

[TinselKoala]

Is it sinking in yet?

Honestly, I feel like I've been kicked in the stomach. I feel really badly about this.

Poynt99 asked me what the capacitance of my TBF coil was, by the method of disconnecting the series connection and measuring across the two half-coils with the capacitance meter. The capacitance measured that way is 2.19 nF.

So he came back with the suggestion to substitute a capacitor for the PBT coil in my test setup. I asked him to make it explicit by posting a schematic, so we all know we are on the same page. So he did that, and specified a capacitor value of 2.2 nF, a standard value. I found one in my parts stash, a good quality poly film unit.

I turned on the scope and let it and the Interstate F43 FG warm up. Both instruments were still set up from last night, so I didn't even change any settings except FG output amplitude (which I always turn down low when it's not in use.)  I connected up the parts and hooked them up, with the new FG coupling toroid I described above. And Bingo. All I did was turn up the FG amplitude to get a good vertical height for the signals, and tweak the frequency a bit to get full waveforms on the screen.

Saved the screenshot and the data dump and posted them here. And I wanted someone else to run the calculations before I told the story. Partzman kindly scrutinized the scopeshots and ran the numbers, and got the same result I got.

What's the moral of this little story? Well, either I've got an overunity 35 cent capacitor on my bench.... or it makes it absolutely imperative for Partzman and the rest of us to try to find some _other_ indication of power levels, independently of oscilloscope measurements and 'traditional' power calculations, if this project is to survive. Of course, the scope is still necessary, obviously for tuning purposes, so that the _measurements_ indicate COP >1, but the scope traces and measurements alone cannot be used as evidence for OU any more.  Unless, of course, you are prepared to accept that my tiny poly film cap is also an OU device when connected in series with a couple of resistors.

So I think talk of self-looping is still very premature. It's good to have suggestions as to how it might be done, but until a COP>1 is actually confirmed by something other than scope measurements, there isn't much to work with. How do you test a self-looping system for an OU device if the test device isn't actually OU?

[TinselKoala]

TK,

Do you have a profile on your iso transformer like inductance of each winding and the primary inductance with the secondary shorted or, if you test without the iso, are the results the same?

Pm

The toroidal iso transformer has 25:25 turns of 0.72mm wire. Each winding measures 2.53 milliHenry, and with one winding shorted the other winding measures 18 microHenry.
I've already shut down for the day so I can't say for certain about results with and without the isolation transformer.

[Vasik041]

Something to consider 

http://www.eevblog.com/forum/chat/rigol-ds1052e-inaccurate-voltage-readings/


PS  Simple RF-Power Measurement
http://www.qsl.net/sz1a/download/build%20an%20rf%20power%20meter.pdf