PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2021-12-01, 04:30:14
News: Forum TIP:
The SHOUT BOX deletes messages after 3 hours. It is NOT meant to have lengthy conversations in. Use the Chat feature instead.

Pages: 1 ... 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 [66] 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 ... 103
Author Topic: Dally, Shark & Ruslan workbench  (Read 72400 times)
Newbie
*

Posts: 17
   

Group: Professor
Hero Member
*****

Posts: 2598
Looks to be tin plated copper tape with solid wires soldered along the length used for a ground strap throughout his circuit.
That ground strap has a lot of surface area.
I'd expect the proximity of the coils to such wide strap, to greatly influence their capacitance to ground, resonance frequency and standing wave geometry.
   
Sr. Member
****

Posts: 289
Hi Vasik,


yes, my feeling is that with 24V there should be much more voltage on the secondary top / antenna.

Presently i can pull a 0.2mm spark with a screwdriver which should be mm's i think, even with this low duty cycle the kacher is running on.

Itsu

Itsu,

Did you have more voltage on the secondary with the diode connected between the primary and MOSFET drain?  If so, perhaps consider using a higher voltage MOSFET.  The On Semi part #NVH4L160N120SC1 is a 1200 volt MOSFET with less capacitance and faster switching than your current part number.

If your 24V supply voltage is indeed remaining at 24V, about the only possible explanation for the "ratcheted up" voltage would be rectification of the primary oscillations by the primary-to-drain diode and storage of that rectified voltage via the MOSFET drain capacitance.

I see that other similar TC drivers use a diode in that location as well, and it is likely there to allow the TC primary to ring more freely during the MOSFET off time.  The "ratcheted up" voltage seen at the drain during MOSFET off times will bias and prevent the body diode and the primary-to-drain diode from being biased on during larger primary oscillation excursions. 

Clamping the drain as you did with the TVS was a good idea, and I would leave the primary-to-drain diode in the circuit with that "cure".

Another possible "cure" would be to add a diode between the MOSFET drain and the 24 volt supply.  Anything over 24 volts plus a diode drop will be bled off to the 24 volt supply.  The large filter caps on the 24 supply will "probably" not be charged up very much at the duty cycle and inductance involved.  However, no other TC designs related to this device use a diode so located and as stated above, many do use the primary-to-drain diode as per the schematic you followed.  And, if as above, that primary-to-drain diode is there to allow the primary to oscillate more freely during MOSFET off times, this additional clamping clamping diode would likely reduce the TC's output.

If the TC worked better with the primary-to-drain diode installed, and if the peak voltage observed at the drain was 1 kilovolt or less, consider using the higher voltage MOSFET mentioned above.  However, what the voltage seen at the drain will be with primary voltages above 24 volts is another question that needs to be considered.

As always, just food for thought...

PW

   

Group: Professor
Hero Member
*****

Posts: 2598
Or do you mean the primary windings spacing?
Yes
   

Group: Professor
Hero Member
*****

Posts: 2598
...and storage of that rectified voltage via the MOSFET drain capacitance.
Wouldn't that storage be discharged every time the MOSFET turns on ?
   
Sr. Member
****

Posts: 289
Wouldn't that storage be discharged every time the MOSFET turns on ?

Verpies,

The waveforms I have seen do appear to show the MOSFET discharging the drain voltage at every turn-on.

However, the TC secondary is  still ringing during the MOSFET off time and those oscillations, coupled to the primary, recharge the drain capacitance as soon as the MOSFET turns off. 

The TC oscillations increase in amplitude with each pulse sequence so the drain gets charged to a subsequently higher voltage at each MOSFET off time, up to some maximum, apparently.

My take on it...

PW 
   
Newbie
*

Posts: 17
Instead of 15kHz or 30kHz I set the freq to 1kHz (5 pulses) to see what things look like here.
Now there is no ringdown anymore from the Tesla as the next 5 pulses start.
Yellow trace is drain Mosfet
light blue trace is probe some cm from Antenna (x20 for actual Volts)
   

Group: Professor
Hero Member
*****

Posts: 2598
Now there is no ringdown anymore from the Tesla as the next 5 pulses start.

So now the voltage level of the steady states increases every 4-5 bursts.
Phase offset rollover ?
« Last Edit: 2021-07-25, 10:04:32 by verpies »
   

Group: Professor
Hero Member
*****

Posts: 2598
However, the TC secondary is  still ringing during the MOSFET off time and those oscillations, coupled to the primary, recharge the drain capacitance as soon as the MOSFET turns off. 
I wonder how it would look with the drain diode in place but the TC/Kacher secondary removed.  (secondary slid out of the primary completely).
« Last Edit: 2021-07-25, 10:01:44 by verpies »
   

Hero Member
*****

Posts: 747
Hallo Itsu  I see now where we have different outcomes.
I follow a bit Alexeew`s setup ..  or at least what i think is like his set up of coils.
I have a 19m Tesla that runs at a higher freq of its own .. but with shnyaga and antenna connected it resonates at about 1,76MHz.
I can put a long ferrite rod (1cm x 15 cm) in de Tesla and than go down to 900kHz res freq (Tes, Shny, Ant).
If i want max output at antenna  than i also have to adjust the puls freq. down to 900kHz.

I noticed that when doing so i got less pulses  because 4 pulses  at 1,7MHz need a less amount of on time than 4 pulses at 900kHz. differently stated: at 900kc 4 pulses take more time than 4 p. at 1,7MHz ( so i had to readjust to get exactly 4 pulses again )
Thats also probably the reason why you use more amps at 25 volts (lower freq bigger on time) than i use (higher freq smaller on time) . ( and also because my former measurements were with 4 pulses and you used 5 i think.

I also noticed that my drain signal is getting les clean with the ferrite rod in the Tesla  .. I think than i also heve to adjust the duty cycle from the pulses a bit for optimum performance.

Furthermore i did some measurements with 4 pulses compaired to 5 pulses. at 1,7MHz Mosfet frequency ( even though res freq of Tes. Shny. Ant. was 1,76MHz as I noticed only later. )

Volt      mA   Vtop drain  Antenna   pulses

25V      60     316V        1760Vpp      4
40        91     480          2990           4
70       150    592          5720           4
100     245    784          8kVpp         4
119     300    864          9040           4
 
now with 5 pulses notice the drain voltage is not bigger.

25V     69mA  312V       2120           5
40       125     480         3680           5    0,5mm streamer, spark
70       214     588         6440           5    4   mm  ..  but you have to start by touching and then move away to get 4mm
100     314     768         8800           5
119     370     848         9840           5

Utopia Now,

Interesting data, thank you for sharing.
Have you noticed at what voltage sparks become white ? (did they actually ?)

Thanks,
Vasik



---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Hero Member
*****

Posts: 747
We don't have much information, but it seems that this pattern present in both
Stalker's and Alexeev's devices

- ground wire length = Tesla coil length
- gradient coil length = 2x Tesla coil length

Itsu and me perhaps need unwind half of Tesla coil.

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Group: Professor
Hero Member
*****

Posts: 2598
- ground wire length = Tesla coil length
- gradient coil length = 2x Tesla coil length
The speed of propagation along a coiled wire is not the same as along a straight wire.
In other words: electrical length is not the same as physical length.

Besides the wire's geometry, the properties of its insulation affect the propagation speed, too.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3270

Utopia,

thanks for the data, so it seems there is some pivot point between 40 and 70V input voltage to boost the sparks / streamers from 0.5 (mine presently) to 4mm.

Perhaps i just need some more input voltage for this pulsed Kacher to deliver.





PW,

Quote
Did you have more voltage on the secondary with the diode connected between the primary and MOSFET drain?


Yes, there was, like about 750V, but that was before i installed the 440V TVS, so not sure how it goes with TVS AND drain diode installed.
I do have some 900V MOSFETs ( FQA11N90C like suggested by Utopia), and will look into yours as it looks a good one too.


Thanks for the "ratcheted up" voltage explanation, seems plausible to me.

I see in Utopia his table that the input voltage does not linearly increase the drain voltage, so it might stay within the 900 or 1200V limit of the MOSFETs.





verpies,

i am experimenting with tighter (both diameter wise as spaced wise) primaries, and a 3 turn very tight / spaced one does seem to increase the secondary output (1mm (white) sparks @ 24V).

Will see what happens with "the drain diode in place but the TC/Kacher secondary removed".

 

Itsu
   
Sr. Member
****

Posts: 289

PW,


Yes, there was, like about 750V, but that was before i installed the 440V TVS, so not sure how it goes with TVS AND drain diode installed.
I do have some 900V MOSFETs ( FQA11N90C like suggested by Utopia), and will look into yours as it looks a good one too.


Thanks for the "ratcheted up" voltage explanation, seems plausible to me.

I see in Utopia his table that the input voltage does not linearly increase the drain voltage, so it might stay within the 900 or 1200V limit of the MOSFETs.

Itsu,

Consider testing your older "wound on plastic" secondary with the your new TC driver...

PW
   

Hero Member
*****

Posts: 747
Vasik,

What wire size and number of turns did did you use for the TC secondary in your replication?

Is your TC similar to Itsu's as well?

Are you still using the self resonating driver (as on your PC layout) and what kind of secondary voltage (spark length...) are you seeing?

PW

original post here: https://www.overunityresearch.com/index.php?topic=4154.msg93856#msg93856

PW,

Due to various reasons my replication moving forward slowly.
I am trying to make "home work" first, get better understanding how things are.

I have wound TC about 38 meters, similar to Itsu's but on plastic tube.
But it seems that I need unwind about half of it.

BTW I think that cardboard could be better than plastic.

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   
Sr. Member
****

Posts: 289
original post here: https://www.overunityresearch.com/index.php?topic=4154.msg93856#msg93856

PW,

Due to various reasons my replication moving forward slowly.
I am trying to make "home work" first, get better understanding how things are.

I have wound TC about 38 meters, similar to Itsu's but on plastic tube.
But it seems that I need unwind about half of it.

BTW I think that cardboard could be better than plastic.

Vasik

Vasik,

What size wire did you use for your TC secondary?

Why do you think cardboard would be better than plastic?

PW
   

Hero Member
*****

Posts: 747
Vasik,

What size wire did you use for your TC secondary?

Why do you think cardboard would be better than plastic?

PW

PW,

I use 50mm plastic tube.
Cardboard could be better because we don't know what in the plastic.
There could be some resistive additives etc

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Group: Professor
Hero Member
*****

Posts: 2598
Cardboard could be better because we don't know what in the plastic.
There could be some resistive additives etc
Cardboard is hygroscopic. Damp cardboard is slightly conductive.
The dielectric constants of dry cardboard and plastic are different.
   
Sr. Member
****

Posts: 289
PW,

I use 50mm plastic tube.
Cardboard could be better because we don't know what in the plastic.
There could be some resistive additives etc

Vasik


Vasik,

What was the diameter of the wire used in your TC secondary?


Regarding cardboard versus plastic.  A simple test.  Using a 50mm cardboard tube with 4mm wall thickness, three turns of bare (uninsulated) wire were wound tightly around one end to make contact with the tube.  A second similar coil/contact is placed 10" from the first coil/contact to make contact at that position.

165VDC is applied to one coil/contact and an electrometer is connected to the other coil/contact: 

Measured leakage current through cardboard tube was 5.5 microamps.

A similar test was performed using 2" schedule 40 PVC pipe:

Measured leakage through PVC tube was 3.5 picoamps.

This test was only performed at 165VDC.  Although this says little with regard to the dielectric constant, etc., of the two materials regarding high frequency AC characteristics, I would suspect that at higher voltages and with high frequency AC, leakage would likely be even worse for cardboard than as measured above.

And, as mentioned in a previous post, untreated cardboard is hygroscopic, so leakage, and very likely other AC parameters as well, will vary a great deal with humidity.

PW   
   

Sr. Member
****

Posts: 343
   PW:
   Ruslan had shown on his videos that he used sink drain pipe for the grenade/Kacher former tube. That is what I used, also. But, as the grenade and inductor coils can get quite hot, it's best to support the middle as well as the ends of the grenade, so it won't sag. It might melt, but it won't sag...
   So, the drain pipe was used and tested by Ruslan, and it worked for him. It is thinner than the regular pvc pipe.
   The humidity here right now, is 85%. And it's not even raining.  It's not just cardboard that gets damp. Everything does here, and HV runs on the outside of the coils. Mine all full of humidity and salt.
   Most self runner are shown using PVC pipe, although Ruslan used the sink drain pipe.
   If the Kacher is not on the same tube as the grenade, things might work differently. And, if they are too close to each other, that can also affect things.
   There may be a good reason to place both grenade and kacher on the same tube.  But, who knows.

   In the video below, you can see that Stalker also used the sink drain pipe, as well as the cardboard Kacher former. At least that's how it looks to me.
Here he is showing how to tune his Kacher circuit to the grenade.
   https://youtu.be/v6FrGTF731o

   NickZ
« Last Edit: 2021-07-25, 17:52:10 by NickZ »
   

Hero Member
*****

Posts: 747
What was the diameter of the wire used in your TC secondary?

I can't find 0.75mm wire, for now I am using 0.5mm

Quote
Regarding cardboard versus plastic.  A simple test.  Using a 50mm cardboard tube with 4mm wall thickness, three turns of bare (uninsulated) wire were wound tightly around one end to make contact with the tube.  A second similar coil/contact is placed 10" from the first coil/contact to make contact at that position.
165VDC is applied to one coil/contact and an electrometer is connected to the other coil/contact: 
Measured leakage current through cardboard tube was 5.5 microamps.
A similar test was performed using 2" schedule 40 PVC pipe:
Measured leakage through PVC tube was 3.5 picoamps.
This test was only performed at 165VDC.  Although this says little with regard to the dielectric constant, etc., of the two materials regarding high frequency AC characteristics, I would suspect that at higher voltages and with high frequency AC, leakage would likely be even worse for cardboard than as measured above.
And, as mentioned in a previous post, untreated cardboard is hygroscopic, so leakage, and very likely other AC parameters as well, will vary a great deal with humidity.

It is true in a way.
However, usually, if you want make a good Tesla coil, you take good dry cardboard and put a layer of acrylic paint on it.
Then you wound your coil and put few more layers of acrylic paint.
It will prevent humidity issues and significantly improve isolation strength.
Plastic seems to be good solution when you need make it fast, works ok if you have good quality tube.

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3270
I wonder how it would look with the drain diode in place but the TC/Kacher secondary removed.  (secondary slid out of the primary completely).

Drain diode in place but the TC/Kacher secondary removed.  (secondary slid out of the primary completely)

Screenshot 1 shows first few steady states, screenshot 2 a more expanded view and screenshot 3 a zoomed in view.


Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3270

Not sure this is relevant, but i just want to mention it.

While experimenting with the kacher secondary (old PVC secondary installed now) and primary (tightly wound / spaced), i noticed some oscillations on the secondary / antenna with the yellow probe close by the secondary / antenna but WITHOUT 24V input voltage on the kacher drain, see screenshot 1 where red is the kacher input pulses (4), and in yellow the probe laying on top of the secondary top.
Screenshot 2 is a zoomed in view of those oscillations.

So i was wondering where these oscillations come from as they only occur when the secondary is attached to the ferrite / antenna.
When disconnecting the secondary from the ferrite / antenna, these oscillations are gone.

I have the drain diode inserted again for this experiment.

These oscillations have a frequency of 1250Khz which is the same as the Grenade resonance frequency found earlier so i suspect they come from the grenade.


Video here:  https://youtu.be/f-y9H6dNO9A     

Itsu
 
   

Hero Member
*****

Posts: 747
Itsu,

I think this is HF oscillations inside grenade captured by "shniaga" coils.
If you place one of MOSFET drain signals on same screen, you can confirm it - oscillations will be in the area of BEMF.

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Hero Member
*****

Posts: 747
One thing which was discussed already, just want to remind about it.
Tesla coil pulses need to be unipolar, antenna coil should be positively charged.
(Supposingly) easy way to achieve this is to add diode.
This is a key to working device, so it is very rarely mentioned.
Attaching two variants which seems reasonable for me, but experiment will show...

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   
Pages: 1 ... 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 [66] 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 ... 103
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2021-12-01, 04:30:14