Operating a transformer on DC

[mbrownn]

I am not an hydraulics guy, but your approach with pressure and flow matches my understanding too.
As the pressure can alternate, so the flow will follow. Pressure can raise linearly, or sine like is we wish so, or any nonlinear form, or abruptly. For the later, the relationship between pressure and flow give the effect compared with a hammering action. I know you know all that and I mention it only if it could help other people.

My intention in my previous post was only related to the term “lag” which is an illusion that voltage and current are the same (due to the similarity of the wave form) but only delayed. We all know that in fact there is a derivative relationship and it depends on the load.
Try to drive abruptly a different current into an inductor and in the attempt to maintain the equilibrium, it will react with the CEMF known as spikes. Try to drive abruptly a different flow of a fluid into a pipe, the spikes in pressure will be noticed. Instantly, no delay, no lag. In both examples.

As for the test equipment, I may have wrongfully assuming that you lack an instrument to measure power factor so you could determine the phase (the lag). To me, this kind of instrument is useless in the study of pulse driven circuits. It is more an economic tool rather than research one. It is designed to know how much one would be charged for the power consumed. In hydraulics I'm not sure there are such instruments.



As I stated before, I'm not fonded by people taking ideas for free, take monetary advantage and then disappear without a “thank you”. No credits to contributors, no return with the progress or test results. Without a diagram, and not knowing you, simply I just asked the question without worrying on being too direct.


If you ask about moral coordinates rather than GPS ones, I believe I just answered your question few lines above.


Yes please, but only if you could do it without “freeloaders” taking advantage of you work. Block diagram will suffice, or any other form you may find appropriate.

I conclusion, my intervention was not meant to offend you but rather to make you trust your own judgment. You used the word “should” as you trusted the books more than your own observations. As you progress with your own research, you will notice that less and less people or books can help you. I look forward to read your discoveries.

Regards.

Thanks for taking the time to reply and with a good attitude too

The hydraulics analogy does work well for logic circuits (we have those in hydraulics too) and most other electrical circuits that are solid state. Yes, even the formation of transients has a hydraulic analogy in some circumstances but it all falls down when it comes to magnetism. For inductors I use the hydraulic pump/motor with a flywheel to simulate inductance and inductive kickback but there is no way to bring magnetism into it.

Unfortunately I no longer have any test equipment so for now I have to stick to talking about what I think is happening and discussing theories about what could be happening but hopefully that wont be for too long. I have never used a power factor meter as I believe them to be misleading, the best instrument in my opinion is an analog scope, call me old fashioned.

I Understand your dislike of intellectual theft and considered not being so open with what I am doing, but the truth is the info is less likely to get out if we all take that stance. Not only that it slows progress. I am fairly sure that working alone it would be extremely difficult to develop a device such as this, never mind taking it into production.

As you can see I am leaving clues everywhere, but to hide things in plane sight, I am not bothering to correct the false trails I went on. If someone wants to try and steal my ideas they are going to have to work out what was good and what turned out to be a false trail. I would not be that hard but it would be frustrating to do so.

My main concern right now is this ACDC problem and not being an expert on inductors and how to manipulate them, I have asked for help. The transformer actions in this device are very complex, so this is why I decided to simplify the model for people to work on. I can always introduce the complexities as the work develops if needed.

Now we have that out of the way, do you have any Ideas?

[barbosi]

In the movie "The Matrix", Morpheus said to Neo: I can only show you the door. You have to walk through it.

I noticed quite a few people pointing you to the right direction, as I perceived it.

One was quite straight pointing to a Tesla patent and I quote from it:

I may consider herein only the case of a circuit divided into two paths, inasmuch as any further subdivision involves merely an extension of the general principle. Selecting, then, any circuit through which is flowing an alternating current, I divide such circuit at any desired point into two branches or paths. In one of these paths I insert some device to create an electro-motive force counter to the waves or impulses of current of one sign and a similar device in the other branch which opposes the waves of opposite sign.

The others tried to give a suggestion on possible ways to apply the information.

I will just say that you have been shown the door. And here you have my idea: study patent 413,353. Comprehend it, not memorize it. With this comprehension, you could walk through the door.

Regards.

[mbrownn]

In the movie "The Matrix", Morpheus said to Neo: I can only show you the door. You have to walk through it.

I noticed quite a few people pointing you to the right direction, as I perceived it.

One was quite straight pointing to a Tesla patent and I quote from it:


The others tried to give a suggestion on possible ways to apply the information.

I will just say that you have been shown the door. And here you have my idea: study patent 413,353. Comprehend it, not memorize it. With this comprehension, you could walk through the door.

Regards.

Yes, this is a distinct possibility as the geometry of my device matches the Tesla patent.

[mbrownn]

There does seem to be several possible partial solutions to this problem.

The first solution is to pass the output of my circuit through the primary of yet another transformer thus altering the characteristics by altering the inductance. This second transformer has its secondary shorted until the signal is about to drop in voltage when it then is switched to open circuit. It can also be further modified by passing a DC signal through the secondary to cause a further drop in inductance.

This will require a high inductance transformer and so will reduce my output significantly.

The second solution is having two output coils and at the right time switching these outputs so that the negative side of the AC signal is opposed The rest of the time these coils could operate in series or parallel but working in the same direction.

This has the advantage of not unduly increasing impedance.

The third solution is feeding the inductive kickback that is powering my secondary through the output coil before returning to the primary. In tests this has shown to produce KV range transients

The fourth solution is by giving a non opposed path for the negative of the AC similar to the Tesla patent but I need to research this more.

[Ernst]

fifth solution?

If you sit on an office chair, one with wheels, (on a rainy office Monday), and you take a heavy book in both your hands, then if you move the books quickly to the front you will feel a reactive force moving your chair backward. If you move the books in the same manner to the back, the reactive force will put your chair back to where you started.
Thus you will not get very far.
Unless of course you move the books very quickly to the front and very slowly to the back.
The slow movement will of course also produce a reactive force, but probably not enough to overcome the friction.

Now to answer your question:
I think that by choosing a suitable waveform you might be able to produce a similar effect in electrical/magnetic terms.

Ernst.

[mbrownn]

Now to answer your question:
I think that by choosing a suitable waveform you might be able to produce a similar effect in electrical/magnetic terms.

Ernst.

In the mechanical world the thing that is commonly overlooked and is the greatest form of loss is friction. In the electrical world I consider resistance to be the same as friction so I try to avoid it as much as I can.

A coil can be low in resistance but high in impedance which restricts current as well, but by using transformers we can manipulate the impedance making it low in one direction and high in the other. We can also switch from low impedance to high impedance at will by inserting a DC signal in a secondary. I think this is what you are saying.

Impedance in a coil or transformer is not such a bad thing as not all the energy is lost to heat and is effectively stored. This energy can be released when we want it and all we have to do is give it a path so that it will go in the right direction.

Yes, this is all part of the possible solutions and thanks for the tip

[Ernst]

We can also switch from low impedance to high impedance at will by inserting a DC signal in a secondary. I think this is what you are saying.

Mmmmnnnnnot exactly.

What I am trying to say is that the output voltage is related to the change in magnetic field, which is in turn related to the input voltage rise/fall rate.
If you let the input voltage go up very slowly, the output voltage will be low, then if you create a sudden drop the reversed output voltage will be high. Thus the output voltage direction can be manipulated. If the transformer were loss-less, the average output voltage would still be 0, but since there are always losses, you may be able to completely eliminate one direction, leaving you with pulsed DC.
Combining this with switching impedance will probably work even better.

Ernst.

[mbrownn]

Just voicing my thoughts here, please give me some feedback

As the input is square wave, caused by mechanical switching, the changes in voltage are abrupt.
This leads to the AC output I get which is not a sine wave but more like a saw tooth wave.

Currently the losses are significantly higher than a normal transformer but it is likely that this will be negated by the other functions of the device. The reason for this is the transformer, motoring and generation functions all occur simultaneously on the same iron core. This means that we have only one set of iron losses and not three as is normally the case. Once the motoring and generator functions have been introduced I have not been able to measure the AC output as it is mixed with the DC and so becomes self cancelling.

With all the air gaps and large amount of iron the losses of a universal motor are large but as this has already saturated the iron we cannot use the same efficiencies for the other functions as all the iron losses were calculated in the motoring function. In truth each function will contribute in part to the total iron loss; Therefore each function will be more efficient that if it alone was tested on the core.

By magnetically preventing half the AC wave we can get the remaining half AC wave without cancellation but what are the effects of cancelling half of this wave?

Cancelling with voltage means no flow of current therefore no losses and no gains, If we cancel with magnetism in effect we have added saturation to the core which can be recovered, less the iron losses, Its a matter of timing.

Back to your post

Mmmmnnnnnot exactly.

What I am trying to say is that the output voltage is related to the change in magnetic field, which is in turn related to the input voltage rise/fall rate.
If you let the input voltage go up very slowly, the output voltage will be low, then if you create a sudden drop the reversed output voltage will be high. Thus the output voltage direction can be manipulated. If the transformer were loss-less, the average output voltage would still be 0, but since there are always losses, you may be able to completely eliminate one direction, leaving you with pulsed DC.
Combining this with switching impedance will probably work even better.

Ernst.

I agree with what you are saying, and what is confusing the issue here is that I think all of the above interactions are taking place in this machine, so It may be a case of trial and error, testing different wiring schemes, until the best one is found.

It was once said to me that there are "no transformer actions in a motor" This is laughable as almost all motors have huge transformer actions.
An induction motor uses transformer actions to power the rotor and as a result of this the efficiency is boosted to almost double that of a similar commutated motor. Universal motors mostly have compensation coils to reduce arcing on the commutator. Regardless of the method of compensation there are huge transformer actions taking place.

It is believing the simplified motor models we are taught, that ignore transformer actions, that prevents us from improving our designs.

[mbrownn]

An additional point.

Most motors are set up to run in repulsion which negates a lot of the transformer action. This device is run in attraction thus keeping the flux in the core, causing improved transformer actions.

[Hrothgar]

a picture is worth a thuosand words

Currently the losses are significantly higher than a normal transformer but it is likely that this will be negated by the other functions of the device. The reason for this is the transformer, motoring and generation functions all occur simultaneously on the same iron core. This means that we have only one set of iron losses and not three as is normally the case. Once the motoring and generator functions have been introduced I have not been able to measure the AC output as it is mixed with the DC and so becomes self cancelling.

What kind of core is it, silicon steel, ferrite, met glass, coat hanger...? Shape and size might be helpful as well.

[mbrownn]

The core is a Bosch or Delco Remy generator, so the outer case is mild steel the coil shoes seem to be soft iron and the armature is laminated. There maybe a few coat hangers in there too

Inside the generator there is the coil on the armature and four coils on soft iron shoes that are not arranged symmetrically. A trifler coil is wound around the outside of the case.

It is likely that some residual magnetism remains in the case although I don't think it is a requirement.

All these parts interact magnetically and there are known transformer interactions going on within the case. the trifler coil seems to add flux but I haven't established if this is sequential, at the same time or both. I suspect both.

Obviously the different materials have different response times but I suspect that this isn't a significant factor. Frequency could be as high as 18 to 20kHz so I could be wrong.

[Hrothgar]

Lots of coils lots of orientation

The only thing I can recommend at this point is to map how your fields are supposed to align at every few degrees of rotation and see if the orientation and direction of the field would allow for a collapse into another coil in a backward fashion. The outside of the coil has a flow in the opposite direction of the core (obvious but we sometimes forget) so make sure that the outside fields aren't working against you.

[mbrownn]

It seams like the answer has been staring me in the face all along.

Imagine a toroid with a four coils wound on it each being 90 degrees from the other. If the coil at 12 o'clock is pulsed with DC an AC signal can be detected at the other three coils. Now we want to see positive coming out of the coil at 9 o'clock so we put a diode on that coil to allow that and we get a half wave rectified output at that coil. By mirroring this with the coil at 3 o'clock we are able to get the other half of the wave. Now we can join these outputs to get a DC ripple.

The thing that has been concerning me, is I have a generated output in the coil at 9 o'clock as well and the AC signal even though blocked by the diode will still see a voltage opposition to the DC output, and so may cause a voltage and current drop.

The solution is to move the 6 o'clock coil and place it at the side of the 3 o'clock coil, again with a diode but on the opposite terminal of the coil. This diode is then connected to our 9 o'clock coil but in in series. Now we have a generated voltage and a half wave AC voltage cancelling each other out but also a half wave AC voltage pushing in the direction of the DC. The net result is the opposing AC is cancelled in the 9 o'clock coil and is shifted with its current to the other 3 o'clock coil. Hope this makes sense

As long as the phase shift from the primary is matched by all three secondary coils, I think this should work.

[Hrothgar]

Cheers!

does that mean I helped?

[mbrownn]

Yes

Sometimes when you have a problem to solve, bouncing a few ideas around helps you get over the mental block. Every one of the replies I have had, has caused me to look at a different angle of the problem. The specific help you gave me was "map ... your fields ... every few degrees of rotation...allow for ... another coil"

I was hooked up on "a coat hanger" trying to figure out the problem, but I had several coils that I hadn’t fully figured out. The answer was there all along.