PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2018-08-15, 07:49:34
News: Registration with the OUR forum is now by invitation only.

Pages: 1 2 3 [4] 5 6 7 8 9
Author Topic: Dr. Stiffler returns with a new device: SFM  (Read 6850 times)

Hero Member
*****

Posts: 1333

I made a video about what i am doing with this Dr. Stiffler stuff, i think it will be much clearer for everybody.

It shows the used led strips and the coil used together with some input measurements and Spectrum Analyzer output
which is much less cleaner as shown earlier with the RF like setup (short leads).

Video here:  https://www.youtube.com/watch?v=Bp0ws2xe6VE

Picture is a zoomed in spectrum from the spectrum analyzer, left peak is the operating frequency peak 14.8Mhz,
to the right we see harmonics and noise up till 595Mhz.


Itsu
   
Hero Member
*****

Posts: 547
I made a video about what i am doing with this Dr. Stiffler stuff, i think it will be much clearer for everybody.

It shows the used led strips and the coil used together with some input measurements and Spectrum Analyzer output
which is much less cleaner as shown earlier with the RF like setup (short leads).

Video here:  https://www.youtube.com/watch?v=Bp0ws2xe6VE

Picture is a zoomed in spectrum from the spectrum analyzer, left peak is the operating frequency peak 14.8Mhz,
to the right we see harmonics and noise up till 595Mhz.


Itsu

Itsu,

I find it most interesting in your video at the 3:24 mark where you have lowered the frequency to 14.34 MHz, the Math channel indicates an input power of -112mw although the 3rd digit is hard to make out.  I wonder if this is due to your spectrum analyzer sniffer or is it normal for the circuit otherwise.  Interesting!

Regards,
Pm
   

Hero Member
*****

Posts: 2361


Buy me some coffee
Itsu,

I find it most interesting in your video at the 3:24 mark where you have lowered the frequency to 14.34 MHz, the Math channel indicates an input power of -112mw although the 3rd digit is hard to make out.  I wonder if this is due to your spectrum analyzer sniffer or is it normal for the circuit otherwise.  Interesting!

Regards,
Pm

Indeed.

Itsu is sweeping the frequency,and hits a few values during that 1 second between 3:24 and 3:25.
Going frame by frame,you can see the P/in value go negative,on both the math trace it self,and the math value as soon as he hits 14.34MHz. Either side of that shows a positive value.


Brad


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

Hero Member
*****

Posts: 1893
The Schumann frequency is normally 7.83Hz, but last year it went up to 36+Hz and moving around!

And today:-

https://www.disclosurenews.it/en/schumann-resonance-today-update/

Not many people know this, also where this frequency originates and was stable for 1000's years as far as we know (INCA time).

Now the frequency today is possibly 35.85Hz, that is a 400,000 sub harmonic of 14.34MHz :D

Just thinking aloud

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 547
Itsu,

I've attempted to replicate your measurements with my Dollar Tree LED array by trimming the series inductor and I matched fairly closely with your peak light output frequency of 14.8MHz but the maximum negative input power I could achieve was  -32mw at 14.00 MHz as you can see in the scope pix below.  I also have SA sniffer coil on the output lead and found that it lowered the resonant frequency slightly.

Regards,
Pm
   

Hero Member
*****

Posts: 1333

Ok guys,  i had a long day helping my son moving to another house, so i don't think i can check it out further today.

What does it mean?

Does it mean that power is being returned to the FG, so generated by the L3 coil/leds combo?

Lower in frequency means more on the capacitive side of the resonance point (current leading).

On another thought,  the shown current value/direction depends on how you clamp the current probe around the wire, it can be done on 2 ways (180°), so
it could be that at resonance i have negative values and outside resonance its positive.

Will check it again tomorrow.

Itsu
   

Hero Member
*****

Posts: 1333

I did a quick check with freshly started equipment (not warmed up), and can see that all calculated power values
are positive above resonance, and negative below resonance.

Not using the SA sniffer coil now.

So i think this is due to the strange open circuit we are using.
However, when extending the FG black lead and scope groundlead to the junction of the 2 diodes (like shown in
Dr. stiffler drawn diagram above), i have the same behavior more or less (resonance point much lower).

Anyway, will check it out tomorrow again.


Itsu
   

Hero Member
*****

Posts: 1333

By the way,  i was in Email contact with the Dr. about his 18-1 kit and also asked him about the used CREE leds board

This is his answer:

Q) I guess i do not have the correct CREE board and not many (nobody actually) has one as there seem to be NO 13W equivalent to 60W @ 120V CREE bulbs available, only 9.5W.
A) The boards I have were purchased from 'Aldi' for a low cost of $4.95 for four bulbs and the are of the 60W equivalent size. One round bulb I have used in the videos is a 100W equivalent purchased at Home Depot
 
Q) Can you show a picture or layout  of the used bulb (is it a bulb or more a spotlight (due to the flat surface of the leds) or any other specification?
A) In one of the last three videos you can see a close up, its hex shaped with no on board electronics.

The second answer does not answer my question (bulb or spotlight like model), so i asked about that again.
The shown CREE boards are indeed hex shaped, but flat, so i don't think they can be in a bulb like lamp as it
shines only in one direction, while a bub needs to shine 360° around.

Anyway, perhaps he still answers.


Itsu
   
Hero Member
*****

Posts: 547
Itsu,

I ran a another test of the same setup minus the SA sniffer but with a non-inductive 100 ohm resistor is series with the input lead.  I then took input power measurements and compared the probe measurement to the current produced by the differential voltage across the 100 ohm resistor and used the results to produce the mean power.  CH1(yel) is the output from the SG and input to the 100 ohm resistor, CH2(blu) is the output of the 100 ohm resistor, CH4(grn) is the probe current, and Math(red) is the power in mean watts.

The results are shown in the scope pix below.  The first is the power input based on the current as measured with the current probe and is -6mw.  Keep in mind the resistor is in place for both tests.

The second pix is with the 100 ohm current measurement and shows to be 48.8mw which I believe to be the most accurate measurement.  The proof is in the fact that the p-p output voltage of the resistor shown on CH2 is less than the input voltage shown on CH1.  If we were really pumping energy back into the power supply, the output voltage would be higher than the input.

So, unless I'm missing something here, we are victims of the current probe's inability to accurately measure at these frequencies.

Regards,
Pm 
   

Hero Member
*****

Posts: 1333

PM,


good check with a csr, which i normally also do, and will do today.

As my current probe should be able to handle up to 60Mhz, the frequency should not be a problem, unless the shown
400-600 Mhz signals on the SA are messing with the probe.

In your screenshots i also notice the phase difference between the csr signal and the current probe signal, which
also could account for the negative power calculations (your blue signal is even on the trailing side of the voltage,
meaning an inductive load).

More test to come,    thanks,   itsu 

 
   
Hero Member
*****

Posts: 547
Itsu,

Here is another test using a 1 ohm csr with the same probe connections as before.  The results certainly came out different.

The first shot is using the current probe for the current measurement and the second is using the csr.  Both are still in the negative input power region.  The csr measurement is shown in mvv but is actually mw.

I've also included a pix of the harmonics of the setup at resonance using my Rigol SA with a sniffer laying close to the inductor.  These odd harmonics are not present when the LED are not conducting.  I also had a 10cm length of wire connected to the junction of the 1N4148s which adds capacitance and lowers the resonance frequency.  This wire was not attached for the scope pix.

I'm curious as to what your results will show.

Regards,
Pm

Edit: Corrected scope pix for csr
   

Hero Member
*****

Posts: 1333
Thanks PM,

but my scope is not so sophisticated that it can use double math functions, like in your case using the
math result of the CH1 - Ch2 (voltage across the csr) and then multiply that result by the CH1 voltage.

But using a 1 Ohm induction free resistor as csr in the red lead (yellow: FG red lead, blue: csr output
both compared to the FG black lead) and using CH1 - CH2 for math function i get screenhot 1.

The green trace is the current probe at the red lead (before the csr) and shows similar wave form as the math,
but different values.


Added my SA peak table


 
Itsu
« Last Edit: 2018-06-16, 17:37:57 by Itsu »
   
Hero Member
*****

Posts: 547
Thanks PM,

but my scope is not so sophisticated that it can use double math functions, like in your case using the
math result of the CH1 - Ch2 (voltage across the csr) and then multiply that result by the CH1 voltage.

But using a 1 Ohm induction free resistor as csr in the red lead (yellow: FG red lead, blue: csr output
both compared to the FG black lead) and using CH1 - CH2 for math function i get screenhot 1.

The green trace is the current probe at the red lead (before the csr) and shows similar wave form as the math,
but different values.

 
Itsu

Itsu,

What you can do with your TDS scope is capture the differential waveforms, then use the Math channel to calculate CH1-CH2 and save the result in Ref1.  Then use the Math channel to calculate Ref1 times CH1 and the result is the input power.  You can skip dividing Ref1 by the csr because it is 1 and I could have done the same.  All the Math measurements should be done using the "mean" values.

Regards,
Pm
   
Hero Member
*****

Posts: 547

Added my SA peak table

Itsu

OK, thanks.  There are some differences in the harmonics.  Going shopping today to see if I can find any 60w Cree lamps. 

As a side note, I've tried lower frequencies with larger inductors and from 1-2MHz I can not see any negative input.  So, I have an idea to build an ATL (asymmetrical transmission line) to operate in the 13MHz range and see what happens with higher power levels.

Regards,
Pm
   

Hero Member
*****

Posts: 1333
I followed your REF solution, but find a problem with the CH1 and CH2 values.

Your CH1 (yellow) signal (1 Ohm csr screenshot above) shows less amplitude (20.42V) compared to your CH2 (blue)
signal (20.56V), while mine show the other way around (CH1 yellow = 16.64V, CH2 blue = 16.47V), see my 1 Ohm csr screenshot

The result CH1 - CH2 = negative (-35.39mV mean).

When continueing (using CH1 x REF), the final result is shown in the screenshot and shows a input power of 279mW (mVV).
(the green trace is the current probe)

Why is your CH1 amplitude smaller then CH2?
Are you sure the first math should also be in mean?  (instead of rms).
Good luck with the shopping,  perhaps your local ALDI still has some.


Thanks,  itsu
   
Hero Member
*****

Posts: 547
I followed your REF solution, but find a problem with the CH1 and CH2 values.

Your CH1 (yellow) signal (1 Ohm csr screenshot above) shows less amplitude (20.42V) compared to your CH2 (blue)
signal (20.56V), while mine show the other way around (CH1 yellow = 16.64V, CH2 blue = 16.47V), see my 1 Ohm csr screenshot

The result CH1 - CH2 = negative (-35.39mV mean).

When continueing (using CH1 x REF), the final result is shown in the screenshot and shows a input power of 279mW (mVV).
(the green trace is the current probe)

Why is your CH1 amplitude smaller then CH2?
Are you sure the first math should also be in mean?  (instead of rms).
Good luck with the shopping,  perhaps your local ALDI still has some.


Thanks,  itsu

Itsu,

That is a good question and I was going to test for this but went shopping.  We went to Aldi's and about a half dozen other stores and found no Cree lamps but I see they are available on Amazon.

Anyway, the difference is in the channel gain variations.  I've attached scope pix taken one after the other to show what I mean.  The first scope screen is the same as previous where CH1 is the SG side of the csr and CH2 is the output side of the csr.  The second shot is with the probe connections reversed along with the Math.  The reason CH2 is larger in voltage than CH1 in my previous post could be due to one or both of the following reasons.   First, the gain of CH2 is slightly greater than CH1 and second, the input truly is negative raising the output side of the csr above the input side.  In this case, it is a combination of both.

Viewing the first scope shot of the CH1-CH2 input, we see the p-p of CH1 at 20.45v and the p-p of CH2 at 20.64 for a difference of .19v p-p.  Viewing the second scope shot with the probes reversed or CH2-CH1, we see the p-p voltages are equal.  From this we have determined that CH2 has a slightly larger gain than CH1.  To determine what the voltage differential is, we can apply dv = (d(CH1-CH2)-d(CH2-CH1))/2 = (.19-0)/2 = .095v p-p.

Now, we can use dv to recal our CH2 while assuming CH1 is accurate.  So, with CH1-CH2 we now have CH1 = 20.45 and CH2 = 20.64-.095 = 20.545 all p-p.  Likewise with CH2-CH1 we have 20.26-.095 = 20.165 and CH1 = 20.26 again all p-p.  Now we can compare the rms differentials in these voltages to see how they would be used in the final power calcs.

Therefore, CH1rms-CH2rms = (20.45*0.5*0.707)-(20.545*0.5*0.707) = 7.23-7.263 = -0.033v rms and CH2rms-CH1rms = (20.165*0.5*0.707)-(20.26*0.5*0.707) = 7.128-7.162 = -0.034 rms.  The slight difference is from rounding but is sufficiently accurate.  So, we see that now we have a higher output rms voltage on the csr than input which means we truly are receiving energy somehow from the LED array and the input energy is negative.

I think it would also be correct to say that we could average the negative input powers from the scope measurements to arrive at a corrected input power.  So Pin' = (0.4046+0.05055)/2 = -0.228w corrected mean input power.

Anyone is welcome to correct my figures or logic above so we all can learn.

Regards,
Pm   

   
   
Hero Member
*****

Posts: 1323
... .-.. .. -.. . .-.
Does anyone know if Aldi stock items based on location availability, or is it more centralised like Walmart  ?
We have one in our town and I could look in there. No Home Depot though, just Lowes.


---------------------------
ʎɐqǝ from pɹɐoqʎǝʞ a ʎnq ɹǝʌǝu
   

Hero Member
*****

Posts: 2361


Buy me some coffee
Itsu,

That is a good question and I was going to test for this but went shopping.  We went to Aldi's and about a half dozen other stores and found no Cree lamps but I see they are available on Amazon.

Anyway, the difference is in the channel gain variations.  I've attached scope pix taken one after the other to show what I mean.  The first scope screen is the same as previous where CH1 is the SG side of the csr and CH2 is the output side of the csr.  The second shot is with the probe connections reversed along with the Math.  The reason CH2 is larger in voltage than CH1 in my previous post could be due to one or both of the following reasons.   First, the gain of CH2 is slightly greater than CH1 and second, the input truly is negative raising the output side of the csr above the input side.  In this case, it is a combination of both.

Viewing the first scope shot of the CH1-CH2 input, we see the p-p of CH1 at 20.45v and the p-p of CH2 at 20.64 for a difference of .19v p-p.  Viewing the second scope shot with the probes reversed or CH2-CH1, we see the p-p voltages are equal.  From this we have determined that CH2 has a slightly larger gain than CH1.  To determine what the voltage differential is, we can apply dv = (d(CH1-CH2)-d(CH2-CH1))/2 = (.19-0)/2 = .095v p-p.

Now, we can use dv to recal our CH2 while assuming CH1 is accurate.  So, with CH1-CH2 we now have CH1 = 20.45 and CH2 = 20.64-.095 = 20.545 all p-p.  Likewise with CH2-CH1 we have 20.26-.095 = 20.165 and CH1 = 20.26 again all p-p.  Now we can compare the rms differentials in these voltages to see how they would be used in the final power calcs.

Therefore, CH1rms-CH2rms = (20.45*0.5*0.707)-(20.545*0.5*0.707) = 7.23-7.263 = -0.033v rms and CH2rms-CH1rms = (20.165*0.5*0.707)-(20.26*0.5*0.707) = 7.128-7.162 = -0.034 rms.  The slight difference is from rounding but is sufficiently accurate.  So, we see that now we have a higher output rms voltage on the csr than input which means we truly are receiving energy somehow from the LED array and the input energy is negative.

I think it would also be correct to say that we could average the negative input powers from the scope measurements to arrive at a corrected input power.  So Pin' = (0.4046+0.05055)/2 = -0.228w corrected mean input power.

Anyone is welcome to correct my figures or logic above so we all can learn.

Regards,
Pm   

 

Pm

I think if you place a DMM set at DCV across your LED array in the same ambient light as you carried out your test in,you will find the source of your slight extra power gain,as LEDs do act like solar panels.

I wonder what the scope would show if the same test was carried out in darkness ?.


Brad


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

Hero Member
*****

Posts: 1333
PM,

thanks for the calculations, they look OK to me, i will check my channels after a fresh calibration to see
how they deviate (using the same measuring point at the same time for both probes should show the same value)

But this means we are on the borderline of what reasonable accurate equipment can measure, which makes it tricky.


Slider,

Here in The Netherlands the ALDI's could have different stocks depending on how "in demand" some stuff was.
What is already long gone (sold out) in one could be still plenty available in another.


Brad,

good point about the leds acting as photocells, but i think the measurements PM makes (i certainly do) are
with the leds emitting light (dimly), so i don't think they can/will act like photocells then.


Itsu
« Last Edit: 2018-06-17, 12:15:39 by Itsu »
   

Hero Member
*****

Posts: 2361


Buy me some coffee




Brad,

good point about the leds acting as photocells, but i think the measurements PM makes (i certainly do) are
with the leds emitting light (dimly), so i don't think they can/will act like photocells then.


Itsu

Ok,i must have missed something,as i thought the LEDs were being pulsed.

If the LEDs are being pulsed,then the extra energy from the ambient light could enter the system via the LEDs,while the LEDs are off.

Some scientists some time back,showed LEDs producing more light energy than the electrical power being supplied to them could have produced.
I believe it was the ambient heat that was the provider of the rest of the needed energy to produce that volume of light.


Brad


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

Posts: 547
OK, here is an up-to-date test after a considerable equipment warm up time followed by a scope re-calibrate.  The gain differential between CH1 and CH2 was several mv worst case.

The first scope pix is measured with the same csr setup as before and with the LED array uncovered on the bench so it is being hit by ambient room light.  BTW, on my current setup, there is no light output coming from the array during this test so I've included a check for Tinman's concern of the LED array acting as light generator and being the source of the negative energy which is a good point.

The second pix shows the measurements taken with a complete cover over the array and with the lights over the bench turned off.  There is a slight difference but the majority of negative input still remains.  Also the level of negative energy input seems to agree reasonably well with the previous averaged input after correction of the channel gains.

Itsu, I agree that we are really pushing the limits of these scopes and I would like to add that I have increased the sensitivity of channel CH1 and Ch2 to get the maximum accuracy.  I don't remember if the TDS series will allow this or not.

Edit: I have to amend my comment about there being no output for the array when it produces negative energy.  The condition ranges from no light to a small amount of light output during the the measurement of negative input power.

Regards,
Pm
   

Hero Member
*****

Posts: 1333


Thanks PM,

had very little time again today, will do some further tests tomorrow.

I did use my black box with solar cell inside calibrated to the used led strip on DC input voltage.
It shows a very poor input / output relationship, like 12mW output at 200mW input.

Video here:  https://www.youtube.com/watch?v=fVcKd1tApDM

Itsu
   

Hero Member
*****

Posts: 1274
Connect a supercapacitor in parallel with the power supply, and set up the condition that produces the "negative input power" on the scope. Now disconnect the power supply so that the circuit is only running on the stored energy in the supercap.

Does the supercap run down, or not? Negative input power means power flowing back to the source is greater than the power being dissipated in the circuit, yes? So the supercap should not run down.

But I have a feeling that it will, no matter how "negative" the scope says the input power is.



---------------------------
"The easiest person to fool is yourself" -- Richard Feynman
   
Hero Member
*****

Posts: 547
Itsu,

I can't seem to get any audio on your video. 

Also, NOTE: I have discovered that when I ask the Math channel to take the mean calculation of (CH1-CH2), the result is calculated from the differential of the mean volts of each channel which is not what we want.  Even after a re-cal there is some mean offset differential voltage between channels so I'm trying to figure out how to resolve this problem.  IOW, I don't think the negative input results are accurate or possibly even exist at this time.  More later-

Regards,
Pm 
   
Hero Member
*****

Posts: 547
Connect a supercapacitor in parallel with the power supply, and set up the condition that produces the "negative input power" on the scope. Now disconnect the power supply so that the circuit is only running on the stored energy in the supercap.

Does the supercap run down, or not? Negative input power means power flowing back to the source is greater than the power being dissipated in the circuit, yes? So the supercap should not run down.

But I have a feeling that it will, no matter how "negative" the scope says the input power is.

TK,

We are driving the LED arrays directly with our Rigol sig gens so we can't try your suggestion above.  We really are pushing the accuracy of our scopes at these frequencies with stray  pickup, etc, so the possibility of measurement error is pretty high.

Regards,
Pm
   
Pages: 1 2 3 [4] 5 6 7 8 9
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2018-08-15, 07:49:34