Author Topic: Primary coil designs and configurations  (Read 519 times)

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Primary coil designs and configurations
« on: July 26, 2018, 12:51:42 AM »
After a lot of years having fun with spark gap tesla coils I moved and abandoned my hobby for a long time.
A recent purchase of a SSTC kit made me tinker again, just on a much smaller level.

I always tried to get a better understanding of how this HV HF stuff really works and interacts.
Most solid state systems use either a flat or cylindrical primary coil.
If I check other devices that operate in similar frequency ranges I usually see litz wire for the primary and often the secondary coils as well.
And it makes sense to minimise losses.
If my old school understanding of core free transformers is still correct than we would want to minimise unwanted effects because they only waste energy.
Solid state coil designs seem to focus mainly on the driver part and how to best couple the two coils.
The actual coil desing seems to be more a thing of size and placement...

But what if we try to combine old school HF designs with our primary coil designs?
I tried to find some info that compared different coil designs in an otherwise unchanged system but did not get much useful things at all.
Basket coils, special winding patterns and so on come to mind here...
Not to mention bifilar coils.
Sure, most of these designs require much more turns than what we use to have but why exactly do we try to use only a hand full of turns?
As far as I undesrstand the problem we need the right mix of magnetic field strength, size of the field, magnetic coupling and q-factor.
For some reason I am starting to think that neither standard flat nor cylindrical coils can be the perfect type here.
Same for using standard wire or a coil on an etched circuit board.

Does anyone here experiment with different coil designs and performance checks?
Or at least with sizes, placement and distances?
Maybe even measurements of the magnetic field strenght and shape?
Why not use HF coil designs in a fun toy that operates at similar frequencies?

Offline Mads Barnkob

  • Administrator
  • High Voltage Senior
  • *****
  • Posts: 692
  • Karma: +4/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Primary coil designs and configurations
« Reply #1 on: July 26, 2018, 09:54:39 AM »
Hi and welcome to HVF.

I think that the run-of-the-mill 50k impedance DRSSTC is simply made as it is with copper tubing, regular enameled wire etc. because it is easy, cheap and well proven. The flat primary low coupling is simply enough to work great at high current impulses. There are losses, but they are minimal or irrelevant in our quest to throw out lightning, I love the saying in Tesla coils that "Everything below 10A is leakage current" and that pretty much describes much of the philosophy around it.

There is however no doubt that the system is far from optimal, depending what you want to achieve, if you like playing MIDI a DRSSTC is hard to beat in performance, but if your quest is for longest arc to secondary ratio, the QCW topology really shows that the energy transfer control is much more important than then coil design/coupling is. It still needs thoughtful design to lower losses, so here you actually see much more RF design practises used, but it is the voltage envelope that its driven with that is the main contributor to its performance.

There is a couple of things that really gets in the way of doing these experiments. Money and space.
Most people do not have the necessary expensive equipment to make some real good measurements that can actually be compared and the space for many different components and coils.
The whole research work was intensely done during the SGTC years and when OLTCs came, but with the DRSSTC it seems to rely on all the practices from back then. There is just less people taking on amateur research on their own, as you can just google for an answer now. 

I would love to make a large experimental setup, but I am limited in equipment and space, so instead I made some more theoretical design tables, but still within best practice limits: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/secondary-coil/

One thing I want to do at a point is complete my IGBT test bench and do a wide range of maximum resonant switching frequency, peak current and gate voltage on different IGBT modules, not quite the secondary circuit, but at least its some research :)
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline Uspring

  • High Voltage Experimenter
  • **
  • Posts: 58
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #2 on: July 26, 2018, 03:54:06 PM »
Quote
Sure, most of these designs require much more turns than what we use to have but why exactly do we try to use only a hand full of turns?

It's a matter of the power supply to the coils. For a SGTC you need enough voltage to fire the spark gap, so often NSTs, MOTs ore even pole pigs are utilised. The size of the MMC is then determined by the requirement to charge the MMC back up to full voltage for each mains half cycle. The primary inductance then follows from the match of primary and secondary resonance frequencies.

For a DRSSTC, the power supply is a bridge delivering 100s of Volts and Amps. Again the match of frequencies ties you down to a primary L*C product. The impedance of the primary, related to L/C, can freely be chosen, but it has an impact on how much current the tank will draw for a given bridge voltage. Ideally the current draw will be just as large as the bridge safely can handle, i.e. max power input.

For both SGTCs and DRSSTCs this amounts to a handful of primary turns.



Offline flyrod

  • High Voltage Enthusiast
  • *
  • Posts: 10
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #3 on: July 26, 2018, 10:28:15 PM »
Most solid state systems use either a flat or cylindrical primary coil.
If I check other devices that operate in similar frequency ranges I usually see litz wire for the primary and often the secondary coils as well.

There is some discussion of primary coil shapes here:

http://www.loneoceans.com/labs/qcw2/

Copper tubing works well for the primary because it can be spread out.  HF current only travels on the surface, and the tube is hollow.  Tubing is also used frequently in induction heating where the skin effect make it a good conductor, and it can also be cooled with some kind of fluid flowing through it.  If you wanted the same coil to be more compact then litz wire would help with that, but at the high currents involved I think cooling could still be a problem. 

I have never seen a bifilar or trifilar secondary, but it shouldn't be too hard to try.  A trifilar secondary could be made that's not much longer and would have lower resistance and lower inter-winding capacitance.  I wind toroid inductors trifilar and they seem to work better than similar inductors wound with a single heavy wire.

Offline Teravolt

  • High Voltage Enthusiast
  • *
  • Posts: 34
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #4 on: July 26, 2018, 11:24:00 PM »
I am glad some one has had a simular intrest. my gole is to make a better tesla coil. some of the electronics I want to develope do things like direct drive for mosfets and IGBT's to speed up there performance and get rid of gate drive transformers. Another is to get away from the universal driver and develope another feedback circuit from scratch. lastly I want to try change the coil its self by changing the dimensions like going to a coil with 1.5:1 and adding parallel resonant circuits to the primary or adding physical capacitors to the output terminal. I am using a sweep generator and o-scope to see the afects of each change. My gole is to make a tesla with a QCW topoligy in the 350 to 500kHz range

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #5 on: July 28, 2018, 05:26:54 AM »
Thanks for the replies guys!

At least I am not totally alone on the quest of understanding and improving :)

Back in my glory days of big coild, thanks to unlimeted power supply at my work place, I exerimented a bit.
Of course only the "lame inventer way" by trial and error LOL
Anyway, my biggest success or improvement was replacing the primary copper tubing with a thick strand of litz wire.
Only had a bout 15m that was salvaged from some wrecked MRI machine but it was about as thick as my thumb and had about 600 single strands in it!
One thing I noticed here was a "ringing" of the wires.
Once the system was in resonance the single strands produced a humming sound.
I first noticed it at far too low frequencies to operate the coil as a high pitched sound.
But a friend later confirmed with some fancy audio equippment that in resonance the coil was literally screaming.
Within the audible range it would have been enough for hearing damage he said.
Sadly I was desperate for money one and sold the entire setup :(

Now, many years later I again started to mess with magnetic fields.
Just out of curiosity at first...
A demagnitiser, some misued holding magnets, attempts to bend laser light...
Some coil systems showed quite different power usage despite similar electric properties.
Here I learned what and how certain materials channel or even focus a magnetic field.
Of course I had to turn my thoughs into something to test them with, so a quite powerful and self resonant induction heater was built.
My own ferrite mixes, coil designs and certainly enough mosfet smoke to make a very satisfying cloud of failure.
But the main thing I learned here was how additional coils not only allow for a far more stable configuration of the half bridge but also to limit and the power that was sucked into the system.
Finding the right balance between core materials (suited for the desired frequency and saturation), inductivity AND placement (in terms of max wire lengths) was a true pain in the $%&'*.
Too little inductivity or too much core saturation and I was down another 10 bucks on mosfets.
Too much inductivity and a bit of luck and performance was just crap.
Not so much luck and the feedback from the coils triggered both mosfets at the same time,creating more digital smoke.

Pretty much the same techniques apart from not having a dedicated feedback loop are used for simple SSTCs.
Please correct me if I am now thinking totally in the wrong direction:
A tesla coil is mostly tuned through feedback from the secondary.
But if the secondary resonance frequency would be close enough to that of a proper ZVS primary configuration it should give quite impressive results.
Well, at least as long as the difference between the two are not getting to big.
On the other hand a primary LC circuit that already matches the secondary should perfom even better in a standard PLL system with feedback from the secondary.
My thought here is: Why force a frequency ont the primary circuit if we can try our best to actually match it first?

Now in terms of coil designs:
If I would be able to use my scope to properly tune my little chinese toy I could start with some simple tests.
First with the original setup by just changing the position of the etched primary, up or down...
This should, at least for my setup and configuration show where the "sweetspot" in terms of distance from the dirst secondary turn is.
A coil that is not flat should produce a much longer field between the magnetic poles, preferably a cylindrical coil first.
The field has different properties but how exactly would I make sure the comparison is correct?
Is the same inductivity enough and I just just judge by how long and powerful the arcs are?
How do I factor in fun things like magnetic coupling or stray inductance? Can I measure it by simple means or do I need a simulator?
I have some leave coming up in a few weeks and would like to be prepared so I provide some meaningful data that hopefully helps and inspires other people here.

Last but not least another remnance from my induction heating days.
I am almost certain I got this really, really wrong from all what I read over the years and what RF or HF transmission technolgy tought us...
With my heater experiments I realised that the right ferrite and right shape can make a huge difference in where and how focussed the magnetic flux is.
A flat coil has both poles basically in the same physical location, while a cylindrical coil or otherwise non-flat coil has the poles further apart.
Now please correct my if I am wrong for the first conundrum:
We only utilise the the top part of our magnetic field from the primary, while the bottom parts does no work in empty space.
With the right core material it would be no problem to provide a short magnetic pathway with far less losses than air for the unused bottom half of our field.
In return the other half would get a significant gain in flux density and through this an even further reach.
As we are not operating in a true and double resonant fashion this increase should reduce the power requirements or if same increase the output.
But even the proper DR systems should benefit big time if my thoughts would be correct.
Just for the fun of theory and assuming above field modifications would show the desired effect:
Better magnetic coupling and a much stronger field would allow to use longer secondary coils.
To increase performance here the same design as in old school RF transmission coils or our primary coils could be used.
Even spacing between the secondary turns instead of as tight as possible!
My theory is that above changes would result in far less losses in the secondary coil through the reduction is stray inductance.
Each turn of the wire could then have its own magnetic field around it without being affect that bad by the adjoining turns.
As a result the voltage can be even higher and of course we have less problems with arcs going to the gorund istead of going full distance.

A lot of ideas to check and so little time :(
 

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #6 on: August 01, 2018, 02:38:44 PM »
I did some more reading through the old Tesla patents and at least learned it is complicated to translate patent wording into normal sentences LOL
Anyway, after comparing notes I realised that Tesla not just based his coils on the pure electrical side of things.
In todays times we base our secondary coils usually around a 1:5 ratio for the size or for the SSTC's closer to 1:3.
However, I noticed that the coils Tesla used all differ in their dimensions.
Considering that it makes no real sense to have a bunch of different sizes he must have had a reason.
Some articles he published in regards to his wireless energy patents mentioned the secret.
All his coils are physically matching the frequency they operated on!
The secondary usually at 1/4 of the wavelength while the addtional coils were at 3/4 of the wavelenght.
He clearly pointed out how important this ratio is.
Similar story for the relation between primary and secondary coils.
Again 1/4 to 3/4 for their diameters in respect to the desired frequency.
With that in mind it also become clear why his Wydencliff experiments required such big coils - he aimed at frequncies suited to utilise the ionosphere as a "reflector".
Around 9 hours well spent as these over 100 year old relations were new in this form to me.

I guess I will re-think my approach on the coil design and cosider above factors in somehow.
It should not make too much difference how the coil is driven as we still aim for the same resonance Tesla used, at least that is what I hope.
So somehow I have to figure out what frequency range this cheap chinese driver with no documentation supports.
Then design a primary and secondary coil to match the desired frequency both in terms of inductivity and physical dimensions.

In an article, from 1902 I think, he also confirmed in some way what I thought about the secondary winding.
After experimenting with cone shaped coils to deal with voltage differences between the windings he went back to straight cylindrical ones.
And for the really high voltages he stated that an even spacing between the turns would eliminate most of the resistance and allow for even higher voltages.
He did not specify though how or why the reduced number of turns for the same coil size would make a higher voltage instead of a lower.
I guess I need to understand resonance more to figure this one out.
But as we already know from the stumpy SSTC's designs the megnetic coupling is important too.

To sum up my latest thoughts on a new approach to coil designs:
New turns out to get older and older.
What we "learned" from the early days of radio in fact translates all the way back to Tesla himself.
Considering the guy spoke 8 languages, was a mechanical mastermind and had endless number of patents around the world he might have known more than what we give him credit for.
Resonance needs to be considered throughout the design process and not just for the electrical parameters we see on our oscilloscopes.
And I mean in one way it really makes sense too...
Radio transmitter and receiver coils often consider the wavelength in their dimensions too.
Tesla just drove it all to a "new" level by really matching every single component as close as possible.
If you wonder why he did it try my theory after going through documents older than my grandfather:
I ignore the fact that most of us aim for big arcs and use a topload instead of the addtional coil Tesla used for his multiplying tranformer - same thing different purpose.
See the system as a pump.
But instead of water or pistons we drive electrical AND radio energy.
Such systems need to be balanced to work well as otherwise a lot of enery is wasted instead of being used.
If we look at potentials in a coil swinging at resonant frequency we get different voltages depending on where we tap the coil with a probe.
But we also see the point of the waveform will differ if we trigger in the same way.
If, in resonance we create a standing wave in our secondary coil that it would make sense to have the length of the coil match it.
Every antenna follows this basic principle and 1/4 and 3/4 forms are the most common in many areas.
If now the topload or additional coil follows the same rules we create "pump" with a 1 to four ratio like our 4 stroke engine.
The system forms a full cycle with no part working against each other.
What I call the standing waves would always pass throught the top connection with every forth swing resulting in a zero potential at this connection.
In my theory this means all the energy is used to build up the electrical and magentical potential of the system.
Where in a standard coild design these standing waves would constantly work against each other.
Even more if you factor in the same flaw in relation the primary coil.
We mostly see electrical and magnetical factors apart from insulation issues.
But we like to ignore out losses and a very common seems to be that the secondary coil heats up.
According to my therory this should not be an issue as the energy flow is in sync and won't be wasted for losses.
We already use (an undefined) spacing for our primary coils.
If I remember correctly than because with "touching" turns we get proximity effects and losses.
Again it makes sense to follow Tesla's footsteps by implementing the same for our secondary coils.
I could not find any specific info so far in the old documents but have to assume that a spacing same as the ire diameter will suffice.
From my old radio days I would like to add another thought:
A 1/4 to 3/4 ratio for the winding!
For multiplying transformer he used the additional coil connected to the top of the secondary while for his wydencliff experiments he used severl additional coils that where grounded.
I assume that a similar multiplying effect can be done by integrating an "additional" coil to the secondary.
The primary would mainly affect the tighly wound first quarter by its magnetic field while the RF field might include the entire coil if strong enough.
The electrical potentital is highest at the top of the secondary and the spacing the top 3 quraters of the coil should help to eliminate arcing over.
The much closer coupling with the primary from the first quarter also means the energy transfer between the two would be far greater and improve the voltage build up through longer swing times due to lower energy and coupling losses.
So again we would simulate a 1 to 4 pump or engine but this time within the secondary coil.

A new thought on the primary coil design based on the above:
The energy we have is pumped into our primary LC circuit.
We can cheat in the C department by using expensive capacitors, banks or whatever the money can buy.
Simply for the reason that we want all our power transfered as good as possible.
But why end there and then use a simple flat coil ?
At least for my small china toy a printed circuit board is still an option although far from ideal.
Proximity effects require quite a big spacing and being flat does not really help either.
But once we speak of real power the C department suddenly can become a big draw on the financial side.
So why not abandone the need for capacitors on the primary side altogether?
Cuts the losses down big time, saves money and without them we save space and can use shorter connections.
Ok, but how to make a LC tank without a capacitor...?
Again lets go back to the old radio days.
Loading coils for powerful transmitter were quite big to reduce or eliminate a very bad thing: Self resonance of the finely tuned coil.
At certain frequencies a poorly designed loading coil would otherwise cause unwanted harmonics on its own resonst frequency.
If there are coild designs to prevent this, including our standard flat pancake, then there must be the worst case too!?
There is and in the form of a bifilar coil as in Tesla's patents and not in the self cancelling version to eliminate the magnetic field.
Due to design we have an even spacing between two "layers", reducing unwanted promimity effects.
But the magnetic field strength is increased as effectively two layer are created in a single layer.
Tesla himself did not use this designs for his later high voltage experiments though.
I assume due to the problem that the required close proximity of the turns cause the isolation to fail.
And of course because the capacitance possible with this type of coil is still quite low.
Everything works better on higher frequencies ;)
I have to assume that in a high frequency system a bifilar primary coil would be able to replace the capacitors.
Would be really nice if the thing would otherwise require very expensive capacitors...
Even on a desktop sized PLL with only a half bridge a different approach for a bifilar coil should be possible.
Here we need a much close coupling and a flat coil is not always ideal here.
Cylindrical coils are prefered by many people if I take online images and videos as evidence.
At just over 300V isolation of the bare wire or arcing would not be a problem either.
Only problem I would see is the driver part for those high frequncies and the secondary coil design.
Would have to be very short at the frequencies possible unless you really use a lot more than the usual 5 to 10 turns.

My PLL system uses feedback from the primary to work at the correct frequency.
I can only assume that primary capacitor and the printed coil are at least close to the frequency of the secondary system.
As I already experimented with some some small addons to the provided topload with no harm an experimental bifilar coil should work despite the capacitor.
At least I hope the added capacitance can be neglected at the quite low frequencies.
But the much stronger magnetic field should result in a far great output from the secondary?
In terms of magnetic coupling: With a much stronger field is there such a thing as having it too high or too close?
If I stick to the 1 to 4 or 3/4 ratio as above would that be the perfect distance for the magnetic coupling too or should I go closer?


Offline Mads Barnkob

  • Administrator
  • High Voltage Senior
  • *****
  • Posts: 692
  • Karma: +4/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Primary coil designs and configurations
« Reply #7 on: August 02, 2018, 02:28:43 PM »
Massive wall of text, I will have to cut my answers up a bit and do over a few replies :)

Pretty much the same techniques apart from not having a dedicated feedback loop are used for simple SSTCs.
Please correct me if I am now thinking totally in the wrong direction:
A tesla coil is mostly tuned through feedback from the secondary.
But if the secondary resonance frequency would be close enough to that of a proper ZVS primary configuration it should give quite impressive results.
Well, at least as long as the difference between the two are not getting to big.
On the other hand a primary LC circuit that already matches the secondary should perfom even better in a standard PLL system with feedback from the secondary.
My thought here is: Why force a frequency ont the primary circuit if we can try our best to actually match it first?

Now in terms of coil designs:
If I would be able to use my scope to properly tune my little chinese toy I could start with some simple tests.
First with the original setup by just changing the position of the etched primary, up or down...
This should, at least for my setup and configuration show where the "sweetspot" in terms of distance from the dirst secondary turn is.
A coil that is not flat should produce a much longer field between the magnetic poles, preferably a cylindrical coil first.
The field has different properties but how exactly would I make sure the comparison is correct?
Is the same inductivity enough and I just just judge by how long and powerful the arcs are?
How do I factor in fun things like magnetic coupling or stray inductance? Can I measure it by simple means or do I need a simulator?
I have some leave coming up in a few weeks and would like to be prepared so I provide some meaningful data that hopefully helps and inspires other people here.

How feedback and phasing works in a SSTC was discussed here: https://highvoltageforum.net/index.php?topic=431.0 and https://highvoltageforum.net/index.php?topic=122.msg912#msg912

SSTC and DRSSTC, where we have a resonant primary is two very different approaches. To your agenda the LC primary does also perform better.

Last but not least another remnance from my induction heating days.
I am almost certain I got this really, really wrong from all what I read over the years and what RF or HF transmission technolgy tought us...
With my heater experiments I realised that the right ferrite and right shape can make a huge difference in where and how focussed the magnetic flux is.
A flat coil has both poles basically in the same physical location, while a cylindrical coil or otherwise non-flat coil has the poles further apart.
Now please correct my if I am wrong for the first conundrum:
We only utilise the the top part of our magnetic field from the primary, while the bottom parts does no work in empty space.
With the right core material it would be no problem to provide a short magnetic pathway with far less losses than air for the unused bottom half of our field.
In return the other half would get a significant gain in flux density and through this an even further reach.
As we are not operating in a true and double resonant fashion this increase should reduce the power requirements or if same increase the output.
But even the proper DR systems should benefit big time if my thoughts would be correct.
Just for the fun of theory and assuming above field modifications would show the desired effect:
Better magnetic coupling and a much stronger field would allow to use longer secondary coils.
To increase performance here the same design as in old school RF transmission coils or our primary coils could be used.
Even spacing between the secondary turns instead of as tight as possible!
My theory is that above changes would result in far less losses in the secondary coil through the reduction is stray inductance.
Each turn of the wire could then have its own magnetic field around it without being affect that bad by the adjoining turns.
As a result the voltage can be even higher and of course we have less problems with arcs going to the gorund istead of going full distance.

A lot of ideas to check and so little time :(

You can not use ferrite in a air cored high voltage transformer due to the high voltage, it would simply flash over.

It is true that ferrite rods are used in induction heaters for concentrating the magnetic field to a specific area, they are called "impeder rods"

-----------------

There is one thing that you can simply not account for in the dynamic secondary system, that is the load impedance, it changes all the time and it thereby changes the whole secondary system. So there really isn't a reason to finely tune your circuit with a oscilloscope, unless you can do it live and fast enough to react on live changes from spark loading.
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline Mads Barnkob

  • Administrator
  • High Voltage Senior
  • *****
  • Posts: 692
  • Karma: +4/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Primary coil designs and configurations
« Reply #8 on: August 02, 2018, 02:55:51 PM »
I did some more reading through the old Tesla patents and at least learned it is complicated to translate patent wording into normal sentences LOL
Anyway, after comparing notes I realised that Tesla not just based his coils on the pure electrical side of things.
In todays times we base our secondary coils usually around a 1:5 ratio for the size or for the SSTC's closer to 1:3.
However, I noticed that the coils Tesla used all differ in their dimensions.
Considering that it makes no real sense to have a bunch of different sizes he must have had a reason.
Some articles he published in regards to his wireless energy patents mentioned the secret.
All his coils are physically matching the frequency they operated on!
The secondary usually at 1/4 of the wavelength while the addtional coils were at 3/4 of the wavelenght.
He clearly pointed out how important this ratio is.
Similar story for the relation between primary and secondary coils.
Again 1/4 to 3/4 for their diameters in respect to the desired frequency.
With that in mind it also become clear why his Wydencliff experiments required such big coils - he aimed at frequncies suited to utilise the ionosphere as a "reflector".
Around 9 hours well spent as these over 100 year old relations were new in this form to me.

Tesla's idea was about the wavefronts peak to be at the terminal instead of on the middle of the secondary coil. I can not find sitations right now, but I think it has been disregarded by many coilers that there is no real gain in 1/4 wave coils.

Quote
Quote from: Nikola Tesla
said in his Patent #645,576 (System of Transmission of Electrical Energy):

"The length of the thin-wire coil in each transformer should be approximately one-quarter of the wave length of the electric disturbance in the circuit, this estimate being based on the velocity of propagation of the disturbance through the coil itself and the circuit with which it is designed to be used...By such an adjustment or proportioning of the length of wire in the secondary coil or coils the points of highest potential are made to coincide with the elevated terminals D D' and it should be understood that whatever length be given to the wires this condition should be complied with in order to attain the best results."

EDIT: I found the calculations!
Quote
We can thus easily understand the benefits that we could, theoretically, have by tuning a Tesla coil following the quarter-wave antenna model, i.e. choosing the length of the secondary wire such as it is one quarter of the wavelength at resonant frequency. This would indeed ensure there is an antinode of voltage at the top of the coil, which is what we want. However, it has been experimentally demonstrated that the currents at the base and at the top of the coil are actually almost in phase (source). This shows the quarter-wave antenna model doesn't apply for the Tesla coil and that one can consider its capacitance and inductance to be lumped.

I guess I will re-think my approach on the coil design and cosider above factors in somehow.
It should not make too much difference how the coil is driven as we still aim for the same resonance Tesla used, at least that is what I hope.
So somehow I have to figure out what frequency range this cheap chinese driver with no documentation supports.
Then design a primary and secondary coil to match the desired frequency both in terms of inductivity and physical dimensions.

In an article, from 1902 I think, he also confirmed in some way what I thought about the secondary winding.
After experimenting with cone shaped coils to deal with voltage differences between the windings he went back to straight cylindrical ones.
And for the really high voltages he stated that an even spacing between the turns would eliminate most of the resistance and allow for even higher voltages.
He did not specify though how or why the reduced number of turns for the same coil size would make a higher voltage instead of a lower.
I guess I need to understand resonance more to figure this one out.
But as we already know from the stumpy SSTC's designs the megnetic coupling is important too.

Cone shaped coils are a pain in the ass to wind, that is properly the reason why he dropped them. If I remember correct they does also just behave like a cylindrical as that is what the cone's average value turns out to be identical with.

Spacing between turns for higher voltage has to do with driving the coil harder, for a modern take on this, look up Kizmo's BiggerDR that has a space wound secondary coil, it is simply to get a higher voltage-per-turn product without flashover between windings.

The number of windings is not the driving factor of voltage multiplication, normal transformer ratio theory does not work for resonant voltage rise. It also fixes another problem, which is actually the most dominating loss in a secondary coil, the proximity losses between turns. I wrote about that under the title "How to calculate the impedance and Q factor of the secondary coil system [1] [2]" here: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/secondary-coil/

I will have to read and reply to the rest later on, out of time for today :)

« Last Edit: August 02, 2018, 02:58:16 PM by Mads Barnkob »
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline Teravolt

  • High Voltage Enthusiast
  • *
  • Posts: 34
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #9 on: August 03, 2018, 05:18:10 AM »
welcome Downunder35m my version uses resonant primaries. Instead of a ferrite core I have 2 main 5 turns and 6 separate smaller windings 3 turns ea. Each of the 6 windings have a resonant cap of about 8000pf. the idea is to create a resonant trap or fly-weal affect. I chose a bunch of parallel resonant circuits built into the primary to increase the circulating current. I sett up the resonant traps frequency the same as the secondary. the point of this exercise is to improve the Q of the circuit and I got the voltage gain of about 80 with no double resonance on the 2 exciting primaries yet. The first picture is the tesla the second is a sweep measurement. The yellow is the toride output and the blue is the current pickup on the bottom under the secondary pict 3. the pickup will eventually be the feed back for the bridge. this coil is pearly experimental and I have had some successes and failures. It took 3 secondary's to get the secondary F res just right. In order to tune I just add a bigger toride to bring down F res. if I added some variable caps to the 8000pf I could make it even sharper. any questions or comments welcome

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #10 on: August 03, 2018, 05:48:07 AM »
I like the idea of your mods, never thought of going this route.
At least not with additional coils in the primary circuit.

I have to assume that you used secondaries with a much smaller diameter before.
With your nice stumpy design, how much of a difference do you think makes the bigger diameter compared to a longer and thinner secondary?
Asking as I am still playing with the idea of making a much shorter secondary than yours but struggle to make or obtain a suitable toroid.
Might have to make a pipe ring cage if all goes down south LOL

Offline Mads Barnkob

  • Administrator
  • High Voltage Senior
  • *****
  • Posts: 692
  • Karma: +4/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Primary coil designs and configurations
« Reply #11 on: August 03, 2018, 08:44:42 AM »
Steve Wards flickr account has some really good pictures of all his high quality projects: https://www.flickr.com/photos/kickermagnet/page3

Specifically you should check out the pictures of the "project titan" which ran with 4 parallel primary coils: https://www.flickr.com/photos/kickermagnet/5487414563/in/photostream/ (click back and forth from this picture to see the others)
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #12 on: August 03, 2018, 09:37:52 AM »
Thanks for all the info so far!
Seems I have to do a lot reading before I settle on where to start with my modifications.
Too many ideas, too much to try but never enough time :(

Offline Teravolt

  • High Voltage Enthusiast
  • *
  • Posts: 34
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #13 on: August 03, 2018, 06:23:51 PM »
If you wanted you could make the primary tanks a series circuit and feed each one with a bridge. this has all ready been done by Steve ward with his portable QCW back pack and large QCW on you youtube. this has the added benifit of destributing the currents in the primaries and making each bridge current equal if every thing is balenced propely

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #14 on: August 03, 2018, 06:45:21 PM »
Before I try complex configurations I really want to compare some basic design first.
For me it was always easier to understand these things by actually doing them.
You know, start with a square, round up some edges, make a hole in the center, round and smooth the outside a bit more and in the end I call it a wheel ;)

Offline Teravolt

  • High Voltage Enthusiast
  • *
  • Posts: 34
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #15 on: August 04, 2018, 05:45:49 AM »
so I found some pictures that shows what I like and am interested in doing. I don't know if I will get there right now I am just testing some ideas. testing ideas takes lots of effort and time. I have not bean able to find anything on sympathetically resonant primary for tesla coils on google so I had to try something. I am also working on a transformerless bridge drive that uses magnetic isolators and floating power supplies. the point of this is to use a mosfet driver directly to the gate to reduce rise and fall times of igbt's. so far I comfortably got to 500 Khz and that includes dead time.
 
I started at looking at ways to increase Q by adding capacitance directly to the secondary output using doorknobs with limited success and I fond that bigger torides give a sharper rise. Maybe because it helps create a fields around the coil. I got the idea of adding resonant primary from the ZCS circuit and tube tesla. I got the ideas of multiple primaries from Steve Wards designs. You wonder why I like squat tesla's? most all QCW have this shape because I assume to keep up the frequency up in the 350 to 600 KHz to create sward sparks. I was finally, I believe, able to make a secondary with the frequency that I want to ballpark target by treating a tesla 1/4 wavelength antenna and then derate the resonant frequency down about 15% because of the distributive capacitance. The distributive capacitance may be larger with tall coils. I am not shure how programs like java tesla deals with it, I am just trying to have ideas build and observe. I would always be accepting in input from others in my investigations and start my own thread if there is an intrest.

https://www.flickr.com/photos/kickermagnet/15705190011/in/photostream/lightbox/
https://www.flickr.com/photos/kickermagnet/5952449165/in/photostream/
https://www.flickr.com/photos/kickermagnet/9764514152/in/photostream/

hear are some of Steve's pics that shows where I want to go

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #16 on: August 04, 2018, 07:28:30 AM »
Nice designs and sparks indeed!
I see you are going in similar directions with your thoughts.
There must be a better way somewhere..... LOL
You already noticed that simply adding some caps or splitting coils not always results in all the desired effects.
Maybe something from my army times give you and idea where I am going here:
For our radio equippment we often used quite powerful transmitters.
And literally everything that was RF HOT was smooth and rounded with no sharp edges anywhere.
After a quick and dirty field repair the inside of our transmitter was glowing everyhwere.
Range was very limited, static in every transmission.
Only because we had to get some covers off and nothing to create smoth seals that are fully conductive when done.

Every straight piece of wire running at the resonant frequency will act as both antenna and receiver.
Means all connection should be rounded with no sharp bends or pointy bits anywhere.
Makes it hard to implement capacitors into the coil unless the coil itself is designed to have enough self capacitance to do without any additional caps in the resonant circuit.

A long time ago I experimented with magnet coils to see how different designs affect the magnetic field strenght.
Most of it was of course multilyered coils and quite often only for DC use.
And to judge anything I only had a compass or two and some fine iron dust in a clear plastic container.
The one thing I remember most is a coil from an old analog TV.
For some reason this old BW model had coils around the tube in seperate packs.
Without the curvature see them like a trapez in shape once flattened.
Like with conical primaries on a tesla coil the megnetic field could be elevated from the center of the coil by using different angles.
But if I used two of these coils with one fully vertical and the other angled, the megnetic field would be more like a funnel.
Bottom part basically like a tube and then expanding outwards on the top.
I think a similar effect coul be greated by using a seperated bifilar coil.
One winding cylindrical, the other half of the bifilar conical, so no flat coil anywhere this time folks....
I admit that I have no clue if the inductivites for both halfs should be matched.
Same for wire having the same wire lengths by the way.
But I do know that in general and with no regard of RF or coupling effects to each other that:
(only from my eperiements in the DC field!)
1. The magnetic field is very strong and powerful in the region from top to bottom of the vertical coil.
2. The field is quite uniform in strenght in this region too and in a circular configuration should be tunnel like.
3. Above the two coils the tunnel expands like a funnel, the angle of the conical coil affects the angle of that funnel.

If my thoughts are correct than, at least for a stumpy, it should be possible to have the magnetic field of the primary circuit overlap the electrostatic field of the topload.
It would be beneficial if the secondary is designed in length to match the wavelenght (at least IMHO).
We would have a magnetic field that full encloses the secondary before spreading out too much.
We would also have an electrostatic field, which I see as beneficial here, that overlaps with the magnetic field tight where coil and topload meet.
As everything in in resonance I assume the overlapping fields will extend acrs away from the topload.
So instead of allowing to go straight down with a risk of hitting the secondary or primary the arcs should act more like water coming out of a horizontal hose opening.
The more power the further they are driven out before going down to ground.
Well, at least in my theory of overlapping fields and harmonic relations ;)

Offline Teravolt

  • High Voltage Enthusiast
  • *
  • Posts: 34
  • Karma: +0/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #17 on: August 04, 2018, 04:55:04 PM »
Downunder35m in a lot of tesla's always the arc has a radiating pattern and I am not shure whether it is the electric field or magnetic field that directs it but from this picture   https://www.flickr.com/photos/kickermagnet/5952449165/in/photostream/   it seems to me that it might be the E field kind of like a girls hair on a van de graaff. like charges repel. In all cases in my observations tesla coils with a 300-600 Khz have straight sparks. I don't know why maybe interaction with the atmosphere. after all it is plasma. The magnetic field is responsible for induction of energy and resonance gives you amplification, I am just trying to find a way to compartmentalize with out having to do a tone of math right now. a good analogy of a tesla is a laser. the operate very similar  ways

Offline Downunder35m

  • High Voltage Enthusiast
  • *
  • Posts: 22
  • Karma: +1/-0
    • View Profile
Re: Primary coil designs and configurations
« Reply #18 on: August 04, 2018, 07:01:16 PM »
Might indeed by a problem of frequency and power levels too.
Maybe even how the coil is grounded.
Plasma is weird in this way...
But if we see it as a capacitive conductor it would make sense for the streamer to stay contained within the electrostatic field.
Within this field it would be easiest for the plasma to exist.
And the magnetic field could also affect the plasma to direct it away and up.
At least with a breakout point present.
Try a tiny sphere instead of a sharp breakout point to see if your arcs react differently.

I also think that your toroid is looking really good and certainly affects the field.
No corona effects, no uneven curves.
Mine is far from this quality but my small sparks also appear as straight lines.
Maybe worth investigating how to create a directional and focussed breakout point.

Offline Mads Barnkob

  • Administrator
  • High Voltage Senior
  • *****
  • Posts: 692
  • Karma: +4/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Primary coil designs and configurations
« Reply #19 on: August 06, 2018, 07:52:15 AM »
The fat coil and long straight sparks from Steve Wards QCW coils are a product of two things.
- Over 400 kHz resonant frequency
- Ramped DC bus voltage to grow the arc, this is how they get to incredible long to secondary height.

The parallel primary circuits, litz wire etc is all just to lessen the losses at the high frequency.

Directional breakout points can only be made with a pre-ionizing of the air in the direction you want the spark, with a laser or flame (but that is crude)
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

High Voltage Forum

Re: Primary coil designs and configurations
« Reply #19 on: August 06, 2018, 07:52:15 AM »

 


* Recent Topics and Posts

post Re: Slow rise & fall time
[Solid state Tesla coils]
Laci
August 13, 2018, 10:20:43 PM
post Re: EMI & placement of driver, interrupter and MIDI controller
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
August 13, 2018, 11:29:30 AM
post Do you really know what 3, 6 and 9 stand for?
[General chatting]
Downunder35m
August 13, 2018, 05:19:03 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
the_anomaly
August 13, 2018, 02:18:42 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
T3sl4co1l
August 12, 2018, 11:35:33 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
the_anomaly
August 12, 2018, 10:25:17 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
T3sl4co1l
August 11, 2018, 10:19:18 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Downunder35m
August 11, 2018, 04:31:42 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
the_anomaly
August 10, 2018, 10:36:15 PM
post EMI & placement of driver, interrupter and MIDI controller
[Dual Resonant Solid State Tesla coils]
DaiKen
August 10, 2018, 09:34:15 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Downunder35m
August 10, 2018, 05:18:22 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
TDAF
August 10, 2018, 02:30:23 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Mads Barnkob
August 10, 2018, 12:00:38 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Laci
August 09, 2018, 10:07:01 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
T3sl4co1l
August 09, 2018, 09:53:57 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Downunder35m
August 09, 2018, 07:11:40 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Hydron
August 09, 2018, 06:21:27 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Mads Barnkob
August 09, 2018, 03:07:50 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Laci
August 09, 2018, 02:52:15 PM
post Teardown of a Eaton PowerWare 60kVA UPS system
[Electronic circuits]
Mads Barnkob
August 09, 2018, 12:25:32 PM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Mads Barnkob
August 09, 2018, 09:10:09 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
oneKone
August 09, 2018, 04:23:26 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
TDAF
August 09, 2018, 02:30:24 AM
post Re: Slow rise & fall time
[Solid state Tesla coils]
Downunder35m
August 08, 2018, 11:34:19 PM
post Slow rise & fall time
[Solid state Tesla coils]
Laci
August 08, 2018, 10:08:48 PM
post Re: Drsstc 3
[Dual Resonant Solid State Tesla coils]
oneKone
August 08, 2018, 06:54:28 AM
post Trying to understand the inner workings of the resonance between coils
[General chatting]
Downunder35m
August 07, 2018, 11:52:20 PM
post Re: Primary coil designs and configurations
[Solid state Tesla coils]
Downunder35m
August 07, 2018, 04:27:26 AM
post Re: Is it possible...
[Light, lasers and optics]
profdc9
August 07, 2018, 03:21:31 AM
post Re: Drsstc 3
[Dual Resonant Solid State Tesla coils]
oneKone
August 07, 2018, 12:58:50 AM
post Re: Is it possible...
[Light, lasers and optics]
Hydron
August 06, 2018, 11:08:25 PM
post Re: Is it possible...
[Light, lasers and optics]
Mads Barnkob
August 06, 2018, 08:37:39 AM
post Re: Primary coil designs and configurations
[Solid state Tesla coils]
Mads Barnkob
August 06, 2018, 07:52:15 AM
post Re: Is it possible...
[Light, lasers and optics]
profdc9
August 06, 2018, 03:20:35 AM
post Re: A little help with first Tesla Coil
[Solid state Tesla coils]
Teravolt
August 06, 2018, 02:44:53 AM
post Re: Is it possible...
[Light, lasers and optics]
FilipŠebík
August 05, 2018, 08:34:12 PM
post Is it possible...
[Light, lasers and optics]
Longranger
August 05, 2018, 07:40:01 PM
post Re: How to machine end-caps for secondary coil
[General chatting]
Downunder35m
August 05, 2018, 07:37:22 PM
post Re: How to machine end-caps for secondary coil
[General chatting]
TDAF
August 05, 2018, 06:13:11 PM
post Re: How to machine end-caps for secondary coil
[General chatting]
Downunder35m
August 04, 2018, 07:25:14 PM
post Re: Isn't there a better way to make a custom toroid?
[General chatting]
Nijin
August 04, 2018, 07:22:38 PM
post Re: Isn't there a better way to make a custom toroid?
[General chatting]
Downunder35m
August 04, 2018, 07:16:57 PM
post Re: Primary coil designs and configurations
[Solid state Tesla coils]
Downunder35m
August 04, 2018, 07:01:16 PM
post Re: How to machine end-caps for secondary coil
[General chatting]
Hydron
August 04, 2018, 05:35:03 PM
post Re: Primary coil designs and configurations
[Solid state Tesla coils]
Teravolt
August 04, 2018, 04:55:04 PM
post Re: Choosing Primary impedance and Fres
[Dual Resonant Solid State Tesla coils]
Nijin
August 04, 2018, 04:48:00 PM
post Re: Copenhagen Suborbitals: Launch of Nexø II rocket (new date! 4th August)
[Science, research and news in other fields than electronics]
Mads Barnkob
August 04, 2018, 04:32:49 PM
post Re: Choosing Primary impedance and Fres
[Dual Resonant Solid State Tesla coils]
Hydron
August 04, 2018, 12:47:06 PM
post Re: Choosing Primary impedance and Fres
[Dual Resonant Solid State Tesla coils]
Nijin
August 04, 2018, 12:15:39 PM
post Re: Choosing Primary impedance and Fres
[Dual Resonant Solid State Tesla coils]
Uspring
August 04, 2018, 11:44:50 AM

Sitemap 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30