PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-11-26, 19:59:42
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: 1 ... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 [24] 25 26
Author Topic: Partnered Output Coils  (Read 384976 times)
Hero Member
*****

Posts: 520
I think I speak for many who would appreciate if @tinman or someone in the know could provide a good synopsis of the situation, circuit diagram, build spec, video locations, etc., so we don't have to run around looking for them in order to possibly help out. If the coil in question is a toroid, we need to have the build spec, number of turns (true or best approximation) for primary and other coils and if a toroid, where on the ring are they wound.

Also, I would like to point out that several months ago I posted to this thread on the previous page of a youtube I had done showing a very simple effect in response to a discussion with @Smudge. Even though I was not around (been very busy super insulating my home) for a few months, I could not help but notice that no one even made a mention anywhere and I find that really too bad because many are simply missing the boat. hehehe

I would recommend to @tinman and anyone else to try and find a tape head as a scoping instrument.

wattsup



---------------------------
   

Group: Professor
Hero Member
*****

Posts: 1940
If the 1 ohm resistor is inductive,how would this result in the math trace dropping down below the 0 volt line?. Would not the inductance be active on both sides of the sine wave,and result in an equal effect all round,meaning that the math trace should stay either above or through the 0 volt line,so as it is an RMS value?.

As Poynt says the math trace is simply the two waveforms multiplied together.  So with 180 degree phase between them the positive peaks get multiplied by negative peaks and the resulting peaks are negative.  And the converse for the other half cycle also results in negative peaks.  The math trace is not rms, it is the actual values.  It does have an rms value but being voltage squared (which is really power) the rms value is of no consequence.  The m in rms stand for mean, it is the mean value of the voltage squared that is then square rooted to get the equivalent "DC" value of a waveform.  Since the math channel is already voltage squared it is the mean value that represents "DC" power.  With your simple sine waves you can get the mean value easily as half the peak of that DC offset math waveform, where the center line of that power waveform sits.  If one channel truly represented current and the other voltage then the average or mean power is exactly that mean value of the math trace.  That is true whatever the phase shift.  AC power comes in peaks at twice the frequency as you can see on your math trace.

It's a pity that inductive and capacitive effects can fool you into thinking your voltage and current are 180 degrees out of phase, whereas they are not.  So you are not pumping vast power backwards, but there still could be some small anomalous component there.  With much more data such as a number of measurements at different frequencies it should be possible to fit the equivalent circuit to that data and deduce exactly what's going on.

Smudge
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Tinman said:
Quote
i am using a 10 watt 1 ohm resistor as my CVR,and only a 1 watt carbon resistor as the second resistor.


FWIW

Most power resistors of the consumer equipment type are wire wound and higher ohm types will have noticeable impedance at 3MHz, so are not reliable as a CSR at these frequencies.

Add to that the fact that the cement block types absorb moisture which can act as a capacitance between the wire turns. Vitreous enamel types are not prone to moisture absorption, but may still have enough inductance to skew readings if used as a CSR at higher frequencies.

Carbon film and metal film types may be spiral cut to final value, which adds a wee bit of inductance, carbon slug types would be better.

Highly recommended is a quality CSR, which need not be expensive. I have fabricated my own at times from a chunk of carbon rod, although low inductance, it is not a high quality method.

Also, I can wind low value low inductance, CSR's from Manganin wire, when needed, usually 0.1 ohm or less. I always use a Kelvin 4 wire measurement technique when grading my resistors in the lower ohm range.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Group: Professor
Hero Member
*****

Posts: 1940
I think I speak for many who would appreciate if @tinman or someone in the know could provide a good synopsis of the situation, circuit diagram, build spec, video locations, etc., so we don't have to run around looking for them in order to possibly help out. If the coil in question is a toroid, we need to have the build spec, number of turns (true or best approximation) for primary and other coils and if a toroid, where on the ring are they wound.

It's all on OU.com where this started and you can pick it up from there.  As a synopsis Tinman made his own toroidal core using his own mixture of iron particles in an epoxy bond.  He then wound 70 turns on it which I imagine go all way round the core.  He then moulded a larger core around this wound toroid so now the coill is completely surrounded by core material.  This inner winding is the inner secondary.  He then wound two more coils of 70 turns bifilar on the large core.  These are called the outer primary and the outer secondary.  Feeding the outer primary he gets a voltage step up to the inner secondary see image here taken from OU.com.

Smudge
   
Group: Tech Wizard
Hero Member
*****

Posts: 1194
...
 It is not the capacitance that is the problem, it is the inductance of the CSR.  To put that into perspective there is a phase difference of 75 degrees between the voltage across the 1 ohm CSR and the current through it, a zero phase shift in the 1.2 ohms then a 60 degree phase for the series combination.  
...

Hi Folks,

I picked a snapshot from Brad's video to refer to the 1 Ohm 10 W resistor he used as the CSR, see attached picture. Such type of ceramic or cement bodied high wattage resistors are wire wound and can be scavenged from older television sets, CRT monitors etc  (ION nicely discussed it in his above post) and no wonder Smudge deduced from the measurements the inductive nature of such resistor which is pronounced already in the lower MHz range. (Brad wrote the voltage divider test "behaved" correctly under 400 kHz or so).

Perhaps the cheapest alternative for Brad locally would this 1 Ohm, half Watt rated resistor (I do not know whether there is minimal order fee):
http://au.mouser.com/ProductDetail/IRC-TT-Electronics/RC20GF1R0JTR/?qs=sGAEpiMZZMuDPtTs5Gda27WoUNA4qsgZx2gj2CcDFfk%3d  

Unfortunately its tolerance is 5%. From its data sheet it turns out it is a carbon composition, non inductive resistor.  There are other than 1 Ohm values too at mouser from this type.

Digikey has a wider selection in non inductive resistors, see here:
http://tinyurl.com/qfkpwwj    

 but it has this nasty policy:

Free shipping to Australia on orders of $200 AUD or more.
A shipping charge of $34 AUD will be billed on all orders less than $200 AUD.   :o
All applicable duties and taxes are the responsibility of the customer.
Incoterm: CPT


Gyula
   
Group: Guest
Around here we try to use the Aryton-Perry wound precision non-inductive Ohmite resistors. 3 Watt, tolerance 1 % and inductance less than 1 nH at 1MHz.
They aren't very expensive from Digi-Key, but evidently you'd need to order a mass quantity to get a break on the shipping to Australia.

WNDR10FE = 0.1 ohm, 1 %, 3 Watt non-inductive, etc.
   

Group: Professor
Hero Member
*****

Posts: 1940
Having identified that the CSR has inductance that seemingly screws up the measurements, it is possible to "unscrew" those measurements.  Brad's "calibration" test has all the information needed to accurately establish the value of that inductance.  Then that info can be used to "unscrew" the real measurements.  This still shows a value of negative resistance in the primary circuit at that 3MHz frequency, so the anomaly is still there.  I will produce a paper doing all this to include vector diagrams that show what's going on for those not skilled in vector math.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
A question for all.

So as i have this correct in regards to the math trace on my scope.
If the math trace is already the mean value of V x I,dose that mean if the bulk of that math trace sine wave is below the 0 volt line on the scope,that there is more power flowing out of the circuit than what is flowing in-->or a negative resistance effect as Smudge puts it?.


Brad


---------------------------
Never let your schooling get in the way of your education.
   

Group: Administrator
Hero Member
*****

Posts: 3217
It's not as complicated as it may seem...
Brad,

The math trace on your scope is made up of instantaneous (not mean) measurements and the product of those measurements of the two wave forms.

In order to get the average (or mean) of the math trace (since we can't seem to figure out how to get the scope to do it) you have to eye-ball it yourself. In other words you approximate it based on how much area of the wave form you see above the 0V ref line, compared to how much you estimate to be below it.

For example, if the math trace was a perfect sine wave and it was sitting right on the 0V ref line, the average power would be 0W. In other words, the average of a perfect sine wave is what ever value is sitting right down the middle of the wave form. Another example, if the perfect sine wave math trace was going between the 0V line and the +1VV line, then the average power would be +0.5W.


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
Brad,

The math trace on your scope is made up of instantaneous (not mean) measurements and the product of those measurements of the two wave forms.

In order to get the average (or mean) of the math trace (since we can't seem to figure out how to get the scope to do it) you have to eye-ball it yourself. In other words you approximate it based on how much area of the wave form you see above the 0V ref line, compared to how much you estimate to be below it.

For example, if the math trace was a perfect sine wave and it was sitting right on the 0V ref line, the average power would be 0W. In other words, the average of a perfect sine wave is what ever value is sitting right down the middle of the wave form. Another example, if the perfect sine wave math trace was going between the 0V line and the +1VV line, then the average power would be +0.5W.

Ok,i now have a 1 ohm non inductive carbon CVR.
In the scope shot below(blue trace across the 1 ohm CVR),would my average power now be close to -50mW ?.

Edit-sorry,the math trace is 500mV per devision,so would that make the average power around -250mW ?.


---------------------------
Never let your schooling get in the way of your education.
   

Group: Administrator
Hero Member
*****

Posts: 3217
It's not as complicated as it may seem...
If the math scale is set to 500mVV/div, then -250mW would be a good estimate.


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
If the math scale is set to 500mVV/div, then -250mW would be a good estimate.

OK,now that is sorted,we can move onto the next part.
I did the same test with a higher value carbon resistor,and this time the voltage values on the scope are near spot on to what the resistor values should be this time. So that has answered the question about the large 10 watt resistor i was using as the CVR being inductive-as you guys worked out.

But rather than jump straight into !why ! we still have this negative power situation,i would like to look a bit more as to why the current is leading the voltage. My thoughts are that we are now high enough in frequency that the inductors capacitance value is at work here,and if not,then maybe this capacitance that Smudge was talking about with the resistor it self.

Now ,i know there is some sort of real power in the primary,as i can still dissipate around 10mW's across the 100 ohm resistor that is across the inner secondary. It's not much,but it is there.

The other thing i tried was to put a variable capacitor from an old radio tuner across the primary coil. By increasing the capacitance,i can increase the current and decrease the voltage. If i reduce the capacitance,i can increase the voltage and decrease the current--but either way,the math trace stays put. The only reason for this test was to see what effect increasing and decreasing the primaries coil capacitance had on the current and voltage input.


---------------------------
Never let your schooling get in the way of your education.
   

Group: Professor
Hero Member
*****

Posts: 1940
OK here is the analysis using Brad's two measurements, one with a pair of resistors and the other with one resistor and the transformer primary.  The negative resistance is still there.  Sorry about the file name, no offence meant to Brad  :-* .  MS Word automatically abbreviated the title.

Brad, your math trace is voltage squared, so the division was 500 milli-(volts squared).  With your 1 ohm current sensor the volts squared relates directly to watts.  And yes your trace had a mean value of minus 250mW.

Smudge
   

Group: Professor
Hero Member
*****

Posts: 1940
OK,now that is sorted,we can move onto the next part.
I did the same test with a higher value carbon resistor,and this time the voltage values on the scope are near spot on to what the resistor values should be this time. So that has answered the question about the large 10 watt resistor i was using as the CVR being inductive-as you guys worked out.

But rather than jump straight into !why ! we still have this negative power situation,i would like to look a bit more as to why the current is leading the voltage. My thoughts are that we are now high enough in frequency that the inductors capacitance value is at work here,and if not,then maybe this capacitance that Smudge was talking about with the resistor it self.

Now ,i know there is some sort of real power in the primary,as i can still dissipate around 10mW's across the 100 ohm resistor that is across the inner secondary. It's not much,but it is there.

The other thing i tried was to put a variable capacitor from an old radio tuner across the primary coil. By increasing the capacitance,i can increase the current and decrease the voltage. If i reduce the capacitance,i can increase the voltage and decrease the current--but either way,the math trace stays put. The only reason for this test was to see what effect increasing and decreasing the primaries coil capacitance had on the current and voltage input.

When you say the math trace stayed put I assume you mean its apparent "DC offset" stays the same.  This is to be expected since the value of capacitance will not change the real power no matter in what direction that power flows.  And it is that "DC offset" that is the mean value of real power.

And the reason your primary is looking capacitive is because it is buried in an artificial high K dielectric so it will exhibit capacitance.  At your 3MHz frequency its inductive reactance that shunts the capacitance is so high that it has no effect.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
OK here is the analysis using Brad's two measurements, one with a pair of resistors and the other with one resistor and the transformer primary.  The negative resistance is still there.  Sorry about the file name, no offence meant to Brad  :-* .  MS Word automatically abbreviated the title.

Brad, your math trace is voltage squared, so the division was 500 milli-(volts squared).  With your 1 ohm current sensor the volts squared relates directly to watts.  And yes your trace had a mean value of minus 250mW.

Smudge

Great paper Smudge.
It would be good if those more versed in the measurements you presented could also take a look,and confirm your findings. It is not that i think you are wrong Smudge,it is more that the more we have that say it is correct(as that sort of math is way over my head-well some of it anyway),the further we can take this at a more serious level.

I think the oppertunity here is to large to dismiss only because something is believed to be imposiable -->that being a negative resistance. It would be great if we as a group could be the first to show a real negative resistance.

As i am still keeping those that are interested at OU.com in the loop,it would be good if you could post that paper there as well. I would also like to see if MH dismisses your findings a quickly as he dose mine ;)

Some further info on the core construction,in case it has something to do with the effect being seen here.

The inner core(made from cast iron filings)was pressed to an area pressure of 2 tons,and then vacuum bagged-->so the inner core is very dense. When i cast the outer core around the !now wound! inner core,it was not pressed,as i did not want to damage the windings around the inner core. The outer core was only vacuum bagged,and so would be less dense that that of the inner core. This is probably why the inner core/coil combo has a higher inductance value ?.

Brad


---------------------------
Never let your schooling get in the way of your education.
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Smudge is commended by all I'm sure for the nice analysis in that paper, and Brad for all his hard work and testing.

Now what should be the next, step any ideas?

I don't have a two ton press, but would not mind winding something to test.

Partzman's replication of the Arie Degeus  patent is also interesting and worthy of further study.

Good job guys.  O0


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
Smudge is commended by all I'm sure for the nice analysis in that paper, and Brad for all his hard work and testing.

Now what should be the next, step any ideas?

I don't have a two ton press, but would not mind winding something to test.

Partzman's replication of the Arie Degeus  patent is also interesting and worthy of further study.

Good job guys.  O0

Ion
You will not need a press,as my original core works much the same way.
I used a standard ferrite toroid core,and cast that in the steel putty. It is not quite as good,but i believe that that is because the steel putty is not very permeable . My next core will be using the ferrite toroid as the inner core(as i believe that will be much more permeable than the pressed cast iron filings) ,and the outer core will be the cast iron filings.

My belief at the moment is that the effect is due to the two different permabilities of the two cores,and we are getting some sort of magnetic re verb from the inner core. Although this effect (the negative resistance) was not my intention with this configuration of transformer,i think it is worth having a closer look at-->it's just too valuable to ignore.


---------------------------
Never let your schooling get in the way of your education.
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Ion
You will not need a press,as my original core works much the same way.
I used a standard ferrite toroid core,and cast that in the steel putty. It is not quite as good,but i believe that that is because the steel putty is not very permeable . My next core will be using the ferrite toroid as the inner core(as i believe that will be much more permeable than the pressed cast iron filings) ,and the outer core will be the cast iron filings.

My belief at the moment is that the effect is due to the two different permabilities of the two cores,and we are getting some sort of magnetic re verb from the inner core. Although this effect (the negative resistance) was not my intention with this configuration of transformer,i think it is worth having a closer look at-->it's just too valuable to ignore.

Great that I would not need a press, however I might be able to use a large vise.

More importantly, an identification of the steel putty used so that we can run some tests on this material alone when used as a core material (permability, eddy loss at HF etc.)

Next, an identification of the core material type or ferrite number of the other core that you used.

If the idea is to use ferrites instead of the steel putty,  a ferrite toroidal core can fit inside of a larger ferrite pot core, and any remaining space can be filled with powdered ferrite mixed with epoxy. This should make for a nice uniform device.

I have played with making toroidal cores from laser toner material, which contains iron powder.

Poured into a form and a short time in the microwave yields a finished product of iron powder molded and held  together with the melted toner plastic material. The iron powder is heated by the microwave energy which then melts the plastic binder. You need to precisely set the time in the microwave, by experiment based on the size of the core you are molding. You can add additional iron powder if needed. Be careful when working with and do not inhale toner powder.

Regards, ION


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Group: Renaissance Man
Hero Member
*****

Posts: 2765


Buy me a cigar


---------------------------
Nanny state ? Left at the gate !! :)
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Dear Ion.

I was toying with the idea of this material ??

Ferrosilicon.

Larger grain sized powder is used specifically to inoculate molten Iron just before pouring the mould.

http://www.ebay.co.uk/itm/250g-Iron-silicon-metal-powder-high-quality-super-fine-powder/320788599055?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D33958%26meid%3Dc11b3258a19b4d6a93587d94243fc9d5%26pid%3D100005%26rk%3D4%26rkt%3D6%26sd%3D320788596166

Cheers Grum

Generally we want small grain size to reduce eddy losses, but who knows what will work best for this transformer construction.

From wikipedia:

Quote
Toner is a powder used in laser printers and photocopiers to form the printed text and images on the paper, in general with a toner cartridge. In its early form it was a mix of carbon powder and iron oxide. Then, to improve the quality of the printout, the carbon was melt-mixed with a polymer.[clarification needed] Toner particles are melted by the heat of the fuser, and are thus bonded to the paper.

Originally, the particle size of toner averaged 14–16 micrometres[1] or greater. To improve image resolution, particle size was reduced, eventually reaching about 8–10 micrometers for 600 dots per inch resolution.


You can still purchase the older iron powder toner materials or salvage as I did from old laser cartridges.

Thanks for the ebay link.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Group: Professor
Hero Member
*****

Posts: 1940
Great paper Smudge.
It would be good if those more versed in the measurements you presented could also take a look,and confirm your findings. It is not that i think you are wrong Smudge,it is more that the more we have that say it is correct(as that sort of math is way over my head-well some of it anyway),the further we can take this at a more serious level.

I think the oppertunity here is to large to dismiss only because something is believed to be imposiable -->that being a negative resistance. It would be great if we as a group could be the first to show a real negative resistance.

As i am still keeping those that are interested at OU.com in the loop,it would be good if you could post that paper there as well. I would also like to see if MH dismisses your findings a quickly as he dose mine ;)


For those not versed in the math the vectors can be solved graphically by drawing the vectors to scale on graph paper.  I'll post the paper on OU.com as you suggested.  If I had your experiment in front of me I would be tempted to use a 7 ohm resistor in place of your CSR and use it as an output matched to the 7 ohm negative R.  I would also be tempted to then place an inductor across your "primary + 7 ohms" to resonate with the primary capacitance value at your frequency.  With virtually zero resistance (because the pos and neg 7 ohms cancel each other) you should have a high Q circuit.  I would tickle that circuit into oscillation by feeding from the SG via a large value resistor.  If that neg R is real you should get power into that 7 ohm CSR that doesn't come from the SG.

Smudge   
   
Group: Tech Wizard
Hero Member
*****

Posts: 1194
Hi Smudge,

I think you understand when I assure Brad that the math and the findings you presented in the pdf paper are ok.   :-*

There is a small detail about which I would ask: I indicated a small phase difference between V1 and V2 in Brad's scope shot (taken from the pdf paper, page 2), see my attachment. I mean the phase relation between V1 and V2 is not exactly 180 degree but there is a cca 18 nanosecond phase shift between them. I know this is quasi negligible and practically the discrepancy is just a few degree from the exact 180°. Yet this may influence a little the length of the i*R1 vector hence the amount of the negative resistance. I am not saying this small discrepancy can fully 'destroy' the negative resistance, just modify its value a little but it surely remains negative.

What do you think?

EDIT:  I think the small phase difference can be 'tuned' out by changing the input frequency a little up or down,  Brad may wish to check that.


Thanks,  Gyula
« Last Edit: 2015-10-31, 09:30:38 by gyula »
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4728


Buy me some coffee
Here is a little something else i have done-in case it is of any importance.
Below you will see a scope shot that has a green trace in it-->best i could do with windows paint,but you get the general idea.
In the two shots below,i am triggering from the blue channel(channel two),which is across the 1 ohm CVR.
In the 1st scope shot,the yellow trace is across the CVR and primary coil. In the second scope shot,the yellow trace is now across the inner secondary that has a 10 ohm carbon resistor across it.

So i traced the green on the first scope shot,so as it is the same as the trace across the 10 ohm/inner secondary combo-the yellow trace on the second scope shot. This way we can see what the inner secondaries power phase relationship is with that of the primaries voltage and current phases. I made sure the secondary was in phase with the primary at 100HTz,and then wound the frequency up to 5MHz.
The green trace would there for be 100mV per division.


---------------------------
Never let your schooling get in the way of your education.
   

Group: Professor
Hero Member
*****

Posts: 1940
Hi Smudge,

I think you understand when I assure Brad that the math and the findings you presented in the pdf paper are ok.   :-*

There is a small detail about which I would ask: I indicated a small phase difference between V1 and V2 in Brad's scope shot (taken from the pdf paper, page 2), see my attachment. I mean the phase relation between V1 and V2 is not exactly 180 degree but there is a cca 18 nanosecond phase shift between them. I know this is quasi negligible and practically the discrepancy is just a few degree from the exact 180°. Yet this may influence a little the length of the i*R1 vector hence the amount of the negative resistance. I am not saying this small discrepancy can fully 'destroy' the negative resistance, just modify its value a little but it surely remains negative.

What do you think?

EDIT:  I think the small phase difference can be 'tuned' out by changing the input frequency a little up or down,  Brad may wish to check that.


Thanks,  Gyula

If you compare that 18nS to the time for 1 cycle you see it represents about 24 degrees, so the phase lag is now about 204 degrees.  And if you then look at my vector diagram and swing vector V2 clockwise by 24 degrees you will see that the capacitance value is reduced slightly and the negative resistance value is actually increased.  I chose to ignore that but all the information is there to accurately calculate what is going on. 

Smudge
   

Group: Professor
Hero Member
*****

Posts: 1940
Just found this old paper on my computer, it discusses magnetic delay and shows some results from an experiment I did years ago.  Magnetic delay, or slow magnetic propagation velocity, is readily achieved but never ever considered to exist in transformers.  It certainly exists in Brad's device.  And it could explain why he is getting anomalous results.

Smudge
   
Pages: 1 ... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 [24] 25 26
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-11-26, 19:59:42