Building of my first tesla coil

[Skylex86]

Hello
 
I just started making my first DRSSTC
It is based on the onetesla CARD, with some improvement : - 1200µF capacitor
                                                                                       - rectifier weight of 50A in 1000v 
                                                                                       - 2*2 IGBTs in parallel (FGA60N65SMD)

I have already done the secondary winding, the torus, the power supply system with filter, circuit breaker and high power relay to disconnect the power supply
 
the thing is that I don't have much room for error because my budget is very tight.

and I would have liked to know if there were any improvements to be made to this control card ?

here are some pictures :






kind regards
alexis

[Mads Barnkob]

Hi Alexis and welcome to HVF

It looks like you have a plan, but please do tell more details about your resonant frequency, MMC size etc.

As this is your first Tesla coil, it is better to keep it simple and get it working before you think about improvements.

Your topload, secondary coil and primary coil looks fine, no critical errors but a few things to think about:
1: wooden primary coil supports can be prone to failure if the wood becomes conductive due to moist
2: the high voltage sticker on the secondary coil could also lead to flash overs simply from being another material, the adhesive changing the varnish underneath or just by supplying a edge along which the arc can track. High frequency high voltage works in mysterious way.

Looking forward to see your build progress and eventually some sparks

[Skylex86]

Thank you Mads Barnkob for your answer
it's true that I didn't go into too much detail

the primary coil measures 10µH
And the resonance frequency of the secondary is 130KHz
which must give me a 149nF MMC capacitor
what tolerance can be given to the capacitance of the MMC capacitor ?

my secondary winding makes about 2200 turns (cable 0.25 mm by 550 mm)
It is covered with 6 layers of polyester resin
the sticker is just a piece of paper, placed between the 5th and 6th layer of resin

how to know if wood is "conductive" ?
and what damage can it cause ?


kind regards
alexis

P.S : today I made the SD card reader

[Mads Barnkob]

Wood can become conductive when near high frequency high voltage because at high frequencies the insulating material properties changes from regular DC current behaviour. If it is moist, which most wood is to some level, it can conduct current. If there first is a single strike and a carbon track is formed in the wood, it will flash over the second after and start burning as long as you are pushing current into it. So avoid using wood in direct contact with high frequency high voltage.

The current through the primary coil is determined by many factors and no matter what design you have made, you can always limit it by lowering on-time to reduce number of cycles on which the voltage can build up or use a OCD setting that protects your IGBT within some limits you define after calculating the maximum current you will allow through it (example calculations here: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/igbts/ )

I have not yet finished my MMC guide, where I explain all the math regarding calculating the primary current determined on peak voltage from the on-time/cycle settings. Because you also need to worry about the peak voltage across your MMC and not just the peak current through your IGBT, its all interrelated. The below text is from that yet unpublished article:

To use the same example I will once again use the 0.45 uF MMC where I choose to use six parallel strings of two CDE 942C20P15K-F capacitors in series. Each capacitor has a DC voltage rating of 2000 V in a 4000 V series rating, but for protection I will derate that 20% so the calculations are done with a maximum allowed 3200 V across the MMC.

Peak voltage V_peak is given in Volt, Primary inductance L_primary is given in Henry and frequency F_resonant is given in Hertz.





The result shows with good precision why I ran my DRSSTC1 at 500 Ampere OCD settings, as this corresponds to the low voltage rating of the MMC, but not lower than it was a good match with the IGBTs used.

The maximum on-time to stay below 3200 V across the MMC, and thus also stay below 472 A peak primary current can be calculated, but it is dependent of the inverter type.

The maximum on-time for a full-bridge





5 half-cycles at 70 kHz on a full-bridge, according to table 2 is around 35 uS before we have either 3200 V across the MMC or 472 A flowing in the primary circuit.

[Skylex86]

hello

what should be the minimum and maximum value of dV/dt?

because I would like to use these capacitors
https://www.digikey.fr/product-detail/fr/R474N315050A1K/399-12724-ND/5731215/?itemSeq=274045822

I will connect them six in parallel and six in series
which would make me 0.15µF for 6000V

according to this formula :



I find 550 A peak

What do you think?




in another subject I read that you should not use simple DRSSTC circuits
but these use expensive fiber optic respirator emitters that are expensive

except this one (and others) :
http://kaizerpowerelectronics.dk/wp-content/gallery/2011_08_14_-_kaizer_drsstc_ii/IMG_5158.jpg

but every time there's no schematics.
so there are two possibilities for me :

or I find the schematic of one of these control charts   

or I adapt an OCD system on the onetesla card
as here : https://www.google.fr/search?q=ocd+onetesla&safe=active&client=opera&source=lnms&tbm=isch&sa=X&ved=0ahUKEwiss_2K8oveAhUFxxoKHWOhAwEQ_AUIECgD&biw=1280&bih=883

what do you think is best ?

kind regards
alexis

[Mads Barnkob]

hello

what should be the minimum and maximum value of dV/dt?

because I would like to use these capacitors
https://www.digikey.fr/product-detail/fr/R474N315050A1K/399-12724-ND/5731215/?itemSeq=274045822

I will connect them six in parallel and six in series
which would make me 0.15µF for 6000V

according to this formula :



I find 550 A peak

What do you think?

As long as you use MKP/PP type capacitors, dV/dt will never be an issue, since these capacitors ratings for pulse currents is already in the range of what we need.

You found a X1 safety capacitor used for EMI input filters, while this is a regular MKP capacitor, their datasheet does not specify its current capabilities and their dissipation factor also does reveal that it might not be useful in pulse application, 0.1% at 1 kHz is already very high and now at 100 kHz that will get even worse, my best guess is you would set there capacitors on fire if you run that coil for too long. Find a capacitor that has a datasheet with RMS and peak current specifications and how to derate that to frequency etc.

550Apeak is properly right if you put in 6000Vpeak, so this would be your maximum OCD setting in order to protect your MMC from overvoltage, but you still need to find out if your IGBTs can handle 550A, which I do not doubt, but you can always do the math to be sure

in another subject I read that you should not use simple DRSSTC circuits
but these use expensive fiber optic respirator emitters that are expensive

except this one (and others) :
http://kaizerpowerelectronics.dk/wp-content/gallery/2011_08_14_-_kaizer_drsstc_ii/IMG_5158.jpg

but every time there's no schematics.

so there are two possibilities for me :

or I find the schematic of one of these control charts   

or I adapt an OCD system on the onetesla card
as here : https://www.google.fr/search?q=ocd+onetesla&safe=active&client=opera&source=lnms&tbm=isch&sa=X&ved=0ahUKEwiss_2K8oveAhUFxxoKHWOhAwEQ_AUIECgD&biw=1280&bih=883

what do you think is best ?

kind regards
alexis

Fiber optics is not cheap, not even the cheapest, but it is the best choice for human safety and operational stability.

All my coils on kaizerpowerelectronics.dk uses Steve Ward's drivers like the UD1.3, which is the one you linked a picutre to, https://www.stevehv.4hv.org/new_driver.html or the newer driver with phase compensation circuit, https://www.stevehv.4hv.org/leadcomp/

[Skylex86]

I agree that fibre optics is the best solution in terms of security
but on the picture above the receiver used is comparable to the one used   
https://www.digikey.fr/product-detail/fr/IF-D95T/FB123-ND/243780/?itemSeq=269844015
and it costs less than OPF 2412 (and other)

so if I replace an OPF 2412 with an IF-D95 will it work ?

Alexis

[Mads Barnkob]

They are both give CMOS/TTL compatible outputs, but the OPF 2412 has a NOT gate on the output, so polarity of the two receivers is opposite, so you need to add in or remove a hex-schmitt trigger gate on the driver board, there should be one available when its the regular Ward UD1.3/2.1.

[Skylex86]

Mads Barnkob 
thank you for your answer

would you still have at your disposal the diagrams and files used to make your DRSSTC 1 OR 2 ?
it would save me the trouble of having to do them again.

another question, I've been running your youtube channel for a while now and I saw that in one of the video vaux you were "dismantling" a server inverter
where did you find it ?   

thank you in advance
alexis

[Skylex86]

hi

today I'm almost done making my control box 





it includes the entire power management system of the DRSSTC (a large relay controlled by several switches, the 230v filter and an emergency stop) 
it also includes the interrupter (which is not yet finished for him)

[Skylex86]

hello
I have just finished modifying the schema to use a TTL fiber optic transceiver
He's on my drive


https://drive.google.com/open?id=1IeWnqPCSSu1kxaGyyZuOufIun5g1LoJA

Could someone please check that I didn't make a mistake ?
(except MOSFETs and resistor and condenser values)

thank you in advance

[Mads Barnkob]

I never shared the board files as I never got them up to a public state in being error free and its not really a priority for me to get that work done. You can find eagle boards files on Wards home page for his drivers.

I find all the equipment in my videos on privately owned metal scrap dealers grounds.

Comparing your schematic to Wards UD1.3b ( http://www.stevehv.4hv.org/newdriver08/DRSSTC_pndriver1_3b.pdf ) you seem to have left out C3 and you did not connect pin 2 and 4 on your IC5A to +5V.

[Skylex86]

Thank you Mads Barnkob
indeed it was missing C3 and the +5v on IC5A
I based myself on your UD1.3b
I just added a hex-schmitt trigger gate at the output of the fiber optic receiver 

another modification I will make is to separate the different parts of the circuit on several PCBs 
it will be easier to do (of course all PCBs will be in the same metal box) 

alexis

[Skylex86]

Hello

I finished making 2 PCBs ;





I still have one more to do.
celui avec le LM311, le détecteur d'intensité et celui pour désaccordé avec la fréquence de résonance.

Alexis

[Mads Barnkob]

Nice boards, I assume they are all homemade.

How did you do the silkscreen? And a little funny with a mirrored silkscreen sitting on the opposite side of a surface mount component board

[Hydron]

Nice boards, I assume they are all homemade.

How did you do the silkscreen? And a little funny with a mirrored silkscreen sitting on the opposite side of a surface mount component board

I'm guessing that silkscreen is toner transfer? If so, nice trick, never tried it when doing my own PCBs.

[Skylex86]

Nice boards, I assume they are all homemade.

How did you do the silkscreen? And a little funny with a mirrored silkscreen sitting on the opposite side of a surface mount component board

I'm guessing that silkscreen is toner transfer? If so, nice trick, never tried it when doing my own PCBs.

That's exactly what it is.
I thought to myself if it works on copper it must work on bakelite
so I tried it and here's the result 

[Skylex86]

Hi
After a short pose in the realization of my drsstc,  I restart

I first redid the primary coil support, then I added a second torus to drop the resonance frequency and to finish, I finished the security part of my control desk.



I still have to order the components to make the electronic cards
but first I have a few questions :

          - for ferrite rings are there any characteristics respected ? 

          - how do I know if my IGBTs will withstand 550 A peak ?
because on this DRSSTC, I peak max is 500 A, but these IGBTs only resist 300A max !

          - what do you think of these capacitors (https://www.mouser.fr/ProductDetail/KEMET/R474N31005001K?qs=sGAEpiMZZMu3dWSqd4Tl0Fd7dHVopjbiki9cBgNAjKv9dySeLvLI4A%3d%3d) ?
because Mouser no longer has a Kemet R474N247000A1K in stock


thank you in advance
Alexis

[Mads Barnkob]

Here is a good guide on finding the right material and making your GDT: http://thedatastream.4hv.org/gdt_index.htm

You can either calculate what kind of load your IGBTs can withstand or you can base it on others experience with pushing that part to a certain peak current: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/igbts/

I would not use those X2 Kemet capacitors for a MMC, they are not made for the job. They have a low dV/dt rating and its hard to get any other specifications on it. You can properly compare it to the Kemet capacitor I added in this list: http://kaizerpowerelectronics.dk/theory/good-mmc-capacitors/ and you can try to give it a design go at: http://kaizerpowerelectronics.dk/calculators/mmc-calculator/  but be aware that there is currently a error in the MMC calculator regarding temperature rise, so it might be 10-20 times too small, I am looking into that issue.

[Skylex86]

Hi

Since I'm not the budget for big IGBTs, I have chosen to use FGA60N65SMD.
According to Mads Barnkob's painting, those if can hold 300 A Peak (http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/igbts/).
So I'm going to put three in parallel which should make me 900 A peak
Here is the circuit :



What do you think?

Sincerely
Alexis

Merry Christmas and a Happy New Year !