Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Simple to build isolation transformer that consumes less power than it gives out  (Read 361071 times)

wattsup

  • Hero Member
  • *****
  • Posts: 2606
    • Spin Conveyance Theory - For a New Perspective...
@JouleSeeker

I also just ordered two isolation transformers (IT) that have two primaries and two secondaries. The company that makes them is in Montreal.

http://www.marcustransformer.com
Model: MO350B
Voltage: Primary 120/240 Secondary 120/240

I could order them locally. To get a single to single IT, I would have to make a special order so more bucks. The one I chose is 350VA. Hope that is not to high.lololololol

When you do your connections, you should identify each wire with a letter then do tests and measure outputs but log the connections and results and this will give you a base for when connecting both ITs together.

Also consider using the two primaries in one IT as the primary and secondary and just forget about the other secondary for now. It worked very well with my toroid transformer. Seems like the two primaries are closer to the core.

Additional to the outputs I am getting, I am also getting output on the secondaries and will work out a way to measure them all at once (if possible) to get a complete output level that will be more them the 97% I am getting now. That's with these regular toroids.

It is a real vacation for me when just using the mains since no mosfets to blow up. lol

wattsup


JouleSeeker

  • Hero Member
  • *****
  • Posts: 685
Good ideas, thanks. 

Meanwhile I've gone ahead with the straightforward 2-trafo build initially suggested by Jack N.
Results this evening, sorry the vid is hurried and imperfect:
http://www.youtube.com/watch?v=Q1Y4J4VQ2JI&feature=youtu.be

Here's the text with my vid tonight:
At ou.com, inventor "Jack Noskills" has open-source presented a clever little circuit, shown at the start of this vid.  Jack presented a few variations, one without the cap C, and that is the one I replicated here.

I used two 1:1 isolation transformers as recommended by Jack, and connected them as shown in his schematic.  My load is one 40-watt bulb, which glows dimly in my light-box (previously described) and I monitor the output lux.  I also record the input and output voltages, and especially the input power and output power (using Kill-a-Watt meters) as I vary the input voltage.

 Here's what I observed:  the efficiency = Pout/Pin improves as I lower the input voltage with this system, while the light output decreases.  It's interesting that a single trafo running on the mains gives me an efficiency of about 84-85%.  With this circuit, I get about that overall efficiency at 90 V input, but as I lower the voltage the efficiency ratio appears to increase, taking the Pout/Pin ratio as displayed on the watt-meters. 

At 70V in, Vout = 47.9V; Pin = 9.6W and Pout = 11.4 so the ratio is 119% (already a surprise...).

At 67V in, Vout stays the same notably, at 47.9V.   Pin drops to 8.9 W while Pout INcreases to 11.8W so the ratio is 133!  strange IMHO.

BEFORE we get all excited, I must note that I have another way of checking on Pout -- this is the light output of the incandescent bulb.  At 70V input, the lux meter reads 38 lux.  At 67V input, supposedly the output power goes up some (although Vout stays the same), yet the lux meter reads 31 lux; down.

 I don't know how to explain all this, but in the spirit of open-source sharing, I share my latest results with Jack's circuit (sans C, but not sans souci).

In any case -- fun!  thanks, Jack N.

Happy experimenting!

T-1000

  • Hero Member
  • *****
  • Posts: 1738
Results this evening, sorry the vid is hurried and imperfect:
http://www.youtube.com/watch?v=Q1Y4J4VQ2JI&feature=youtu.be

Very good!

Now we need to take it into physics laboratories and find out what really is cause of effect  8)

Many people failed to understand where exactly we have power amplification and in result we get complex OU circuits while it is enough to have very simple one..

Cheers!

Lynxsteam

  • Full Member
  • ***
  • Posts: 172
Joule Seeker - Your experimental results are encouraging and surprising. 
Congratulations Jack - at this point we can't rule this out.  And that is a big step in the right direction!

I am continuing my avenue with your circuit as well.  My thought was that a Bedini motor is so close already to self running when tuned, that your circuit could possibly get it over the "hump".  I know it is farfetched, but that is what I am trying.

This morning I was making some headway.  I now have your circuit fed by the flyback off the collector and outputting through a bridge rectifier.  I did something and blew my last 3055 so I have to go get some more.  I had the output up to 11 volts DC and need to get to 13 volts.  I have changed the drive coil, changed distance, resistance to 33-58 ohms, deleted the isolated secondary.

JouleSeeker

  • Hero Member
  • *****
  • Posts: 685
    Sounds like you're making progress, Lynx -- and that's great!  I'd admire your tenacity and clever ideas. 

    Now don't put too much weight on my result from last night...  I warned that while the output power indicated by the Kill-A-Watt P3 meter was high, yet the bulb was DIMMING as I lowered the input (and output) voltage.

   I followed up today with a number of tests.  The most telling was putting just the 40W bulb on the output from the Variac, with just a P3 meter in between.  Plot below shows results. 

Interpretation:   The response of light-lux versus watts in is nice and linear from about 70 V (on the P3 meter) on up to mains voltage, using the P3 Kill-a-watt meter to measure Pin. Below about 70V, the P3 watt-meter is NOT reliable for measuring power.   Looks good for V > 70V on the P3 meter.

I have another watt-meter, by WANF, and it won't even give a power reading when the voltage is that low. 

THIS does not mean that Jack's circuit won't show ou with some work or tuning, but yesterday evening the output power to the second P3 meter was sitting at about 50V when I was getting "interesting" results -- and the METER IS NOT TO BE TRUSTED AT SUCH LOW OPERATING VOLTAGE; as my results today demonstrate, and as the dimming bulb last night was warning me.

It's good to have independent ways to monitor the power, such as light-output and a watt-meter. 




baroutologos

  • Hero Member
  • *****
  • Posts: 918
hello Gyula,

You probably are right by the LC meter and inacurracy to measure this type of inductances. Perhaps has to do with its probe measuring wave form or just aything.
i would like to try another small experiment to clarify this further .. but i am in vocations.. so it will wait a bit :)
Carry on.   :)

Hi baroutologos,

... Regarding your measurements on transformers, it is indeed strange and I also found big differences in measured no load input current and the calculated input current which came from measured transformer parameters. 
 
 The problem may come partly from the digital LC meter: it does not use 50 or 60 Hz test frequency for L measurements, my own LC meter (Maxwell DMM MX-25 304 old type) uses about 200 Hz in the some hundred milliHenry and Henry ranges and it is doubtful how the different mains transformer cores perform at such a "high" frequency, most cores may lose permeability to some percent but some other cores may lose even half of their '50 Hz' permeability.  This means that a 50 Hz test circuit should be used  for measuring the transformer coils.  I repeat this frequency difference does not fully explain the situation...

@ Microcontroller,
excellent. This is the measuring minimum standard that should be applied from very start.

wattsup

  • Hero Member
  • *****
  • Posts: 2606
    • Spin Conveyance Theory - For a New Perspective...
@all

Well here is my last trials using a secondary configuration as well. Mind again that these transformers are step down and hence not a true isolation transformer.

MAINS: REGULAR WALL PLUG
VARIAC AT FULL: SUPERIOR ELECTRIC MODEL BP5715
INPUT A: 121.9 VAC @ 0.24A = 29.259 WATTS
OUTPUT B: 57.7 VAC @ O.5A = 28.85 WATTS
OUTPUT C: 43.3VDC @ 0.25A = 10.825 WATTS
TOTAL OUTPUT: 39.675 WATTS
HE HE HE that does not include the heat dissipation of the hand burning bulb.
This is crazy.

This will be up soon;
http://youtu.be/HGVWXMmMs1o

Enjoy.

wattsup

TheCell

  • Sr. Member
  • ****
  • Posts: 285
Hello
because I  am interested in the outcome of this experiment, I must give my 50c .
@microcontroller
You can add up output watts in this case , cause the bulbs consume reals watts (not like caps or coils).


You are not measuring real power at the input side. (The value is to high)
But this measuring error (Input watts) results in a benefit for the cop!!
Because the real input watts are a lower value the cop will be higher.


Nevertheless measure the right 'real watt' input walues.
But I think wattsup , you are on the right track.

At the input side amps and volts are not in phase. I would try 2 things:
1)Put a 'kill a watt meter' directly after the variac . It will show reals watts. If the input voltage at these meter is not too low the measuring electronic will do it's job. For example my energy power meter works in a input volt range between 100V to 240 V .as shown on the label. If your meter's voltrange is similar to this, this is the preferred input watt measuring method.


Method 2 : Use a shunt in series with the input and measure the phase difference between the voltage and the current with a scope. Multiply the volt and amp values that you read now as input values with the cosinus of this phase difference. Than you get the real power.
For the output : there is not phase difference therefor cos(Phi) = 1.


MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Wattsup:

The AC voltage that you are measuring is an RMS voltage.  It looks like the clamp ammeter is measuring RMS current.  Assuming that is true then you can't multiply the two together, you will get an exaggerated wattage calculation.  My suggestion would be to investigate this issue and make corrections if necessary.

For your DC current measurement you are at nearly the rock-bottom output from the digital clamp meter.  Hence your accuracy is very poor.  I would suggest that you just use a regular digital multimeter or an analog ammeter with the proper maximum scale to get a better DC power calculation.

MileHigh

TheCell

  • Sr. Member
  • ****
  • Posts: 285

You can multiply RMS values of U and I to get the average power.The meaning of the RMS Value, be it voltage or current is to get a value that is equal to the case the whole device would be operated with DC voltage / amps.
If you got Urms you can calculate the average power by P=(Urms ^ 2) / R
If you can't measure current , you can now calculate Irms = sqrt(P/R) . And I is now also the RMS - value, because these are DC - like values.
And it does not matter , if he is using meters , that measure rms or not, as long the device is operating with sinewave.
His device is operating at mains frequency and with sine wave.
And for this case meters can be uses that measure mean average U or I , but do internally a multiplication correction for to show the rms (root mean square) , because it's the necessity for power calculations to deal with rms values (meaning volts and amps). If the waveforms in the device were other than sine , [/size]
then he MUST use real RMS meters, for measuring the correct value. (Not necessary here)
(And U and I must relate proportional to each other by ohms law U=R*I no nonlinearities ,
or you can forget your meter with RMS measuring capabilities)
Only the input power measurements are false.

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
TheCell:

Thank you for correcting me, you can multiply Vrms x Irms to get a proper power measurement.

If I can offer a suggestion it would be to use power resistors instead of light bulbs because light bulb resistance varies with intensity/temperature.  Then connect a True-RMS voltmeter across the power resistor to calculate the output power.  The transformers may saturate which would distort the sine wave and you can see that Wattsup's sine wave is not a pure sinusoid so using a True-RMS meter would be advisable.

MileHigh

JouleSeeker

  • Hero Member
  • *****
  • Posts: 685
Thanks for sharing regarding your fascinating replication, Wattsup!  More power to you!

You might use a watt-meter (e.g. Kill-A-Watt) on the input power also, as a check. 
Also, why not run straight from the mains?

Have you tried rectifying BOTH outputs?
____

Thanks for admitting this, MileHigh --


TheCell:

Thank you for correcting me, you can multiply Vrms x Irms to get a proper power measurement.
...

MileHigh

Magluvin

  • Hero Member
  • *****
  • Posts: 5884
MH, How can you expect people to take your suggestions seriously when it is that these people have to correct you on the fact that P=IxV?

I know that you admitted the mistake. But guess what. Now these people have to overlook all that you post. The reason is, when it comes to something as simple as you saying that P cant be derived from IxV, then what are we to think when you post something more complex than P=IxV?  Now we have to question everything

I know you have knowledge and some skills. And you have repeatedly blamed lack there of on the fact that it has been 30 odd years since tech school. Just the other day you said it in the Tar Baby thread, again.  Call me a liar on that one. ;]  If you do, I will provide a list of posts that you state just that. If you like. ;]


Yet with quite a few similar stumbles, a memorable one is the led that biases on at 1v, you still assert yourself as some sort of elder of knowledge, telling people they are fools and such, just because they are very interested in a subject.

You need to reread(and rethink) your posts before hitting that button. Because if you are going to stand up and jump in as the Wizard of OZ, if you keep stumbling like that, these people will find you in that little box, hiding, pulling levers and knobs, trying to make people believe somethings that are really not.

These people are going to experiment and continue to learn about these subjects, whether you post you interruptive oppositions or not.   If we go to look at your posts here at OU, and even just skip the Tarbaby thread posts, what would we see? Its real easy. Just click on Milehigh and click show posts.  Its all there.

I see a lot of negativity in just about every thread you post. I have not yet seen a post from you that shows you trying to help someone try to "advance" on a project.
Why? Because you always oppose any ideas or claims of OU. You are never ever positive on these subjects. Most of us know where you are coming from.

MaGs







MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Magluvin.

That's just a bullshit posting by you were you are strutting around puffing your chest out and stroking your ego because you saw me make a mistake.  If I corrected every single error you made and made a big stink about it and gave a speech you would not like it one bit.  The LED example is more bullshit because if someone else made the same mistake you would have politely corrected them and them left it at that.  With me you ran around like a crazy person screaming your lungs off.

Don't start this nonsense again where you are going to "go after me."  What you do know is that I know my stuff and I am not perfect.  Big farking deal.

In my previous posting I was trying to help Wattsup and I made a mistake.  Do NOT make the conscious decision that you are are going to try to make my life miserable every time I make a mistake. 

MileHigh

ramset

  • Hero Member
  • *****
  • Posts: 8073
MH
Quote
  Do NOT make the conscious decision that you are are going to try to make my life miserable every time I make a mistake
end Quote
 
Mags don't you remember we gave that job to Wilby??
{a side note ,I believe wilby was awarded two Gold stars for his efforts in the "Rossi" thread]
Thx
Chet