making HV bobbins for ferrite transformers

[Teravolt]

I am looking to create a bobbin for a ZVS and have chosen ABS and PLA. Any other suggestions are welcome. I thought that ABS would have the best holdoff but has a higher melting temperature. I am new to 3D printing but I am thinking that it may be the only way to make a high voltage bobbin. I have a shape picked out .

[davekni]

I'm interested in how your project turns out!  Every time I think through HV bobbin design I come to a different variation.  At the moment I'm thinking it might be best to have a very open structure, the minimum required to support windings.  That way there's ample paths for air to escape and potting material or oil to enter.  Of course, that would be different if you are trying to avoid potting and oil.  I haven't used potting (nor oil) yet, but plan to in the future based on what I've read on this forum.  So far the highest voltage transformer I've wound (not counting Tesla coils) is 8kV RMS.  It uses just tape and mylar sheets and spacers for insulation.  (6kW output into voltage-multiplier to feed my Marx generator.)

[johnf]

PLA is not a good choice as it absorbs water

ABS does not absorb water and is used in electronics for coil formers
PTFE is superior but does not print

bobbins can be made effectively by subtractive manufacture ie lathe or mill much faster and cheaper than 3D printing

[Teravolt]

my plan is to make a transformer that is has a grounded center tap so this transformer will have 2 secondary. I want to create a tapered bobbin wind it and fill it with epoxy in a mold and then put the ferrite through it. when these types of transformers are made arcing will occur to any low potential like the ferrite or primary making the need to put it in oil. it is two hard to find a ex-ray transformer so making one is my only option. the plan is to use Autodesk 123d to create the bobbins. the 3d printer will take a while to get to me while I familiarize my self with the cad software so maybe a service like Shapeways can print it.

[davekni]

Interesting that you are planning on both potting and oil.  I'd thought that most designs used one or the other.  Perhaps there's some advantage in the combination, in that good vacuum-potting could leave exposed leads separated enough that oil-filling could be done without vacuum.

[Teravolt]

I'm not shure if potting would be enough. I am planning to pot first oil second if necessary. I have a feeling oil may be necessary.  I also found out that places like shapeways will make objects from many different typse of materials like nylon. if I had a quick way to produce my ideas life would be good. I am self learning autodesk 123D witch has plenty of power to do what I want.

hear is the core

https://www.amazon.com/1pair-UF86A-Ultra-Transformer-ferrite/dp/B014A9GBRM/ref=sr_1_2?crid=377TIQZ7828AE&dchild=1&keywords=ferrite+core+transformer&qid=1597236824&sprefix=frrite+core%2Caps%2C209&sr=8-2

[SteveN87]

At the moment I'm thinking it might be best to have a very open structure, the minimum required to support windings.  That way there's ample paths for air to escape and potting material or oil to enter.

That's a great idea. Maybe something like this,

https://www.amazon.co.uk/Darice-Plastic-Canvas-White-13-Inch/dp/B0042LK6RA

if rigid enough, could be used to build up the layers.

[Teravolt]

hi Steve, creating a layered bobbin would work too. acrylic laser cut layers glued together might work as well but 3D is cleaner and easier, something like that was one of my first ideas. 

[davekni]

For an open bobbin, I was thinking more one-dimensional structure, such as adding thin glass or plastic rods perpendicular to the wire onto each layer of tape between winding layers.  Or some similar 3D-printed structure.  With the plastic-canvas, I'd be concerned about isolated holes w/o clear paths to the outside for air escape or potting/oil entry.

Concerning potting-then-oil, the potting needs to be good enough to prevent any internal breakdown by itself.  Defects (air bubbles) in the potting are isolated, so will never get filled with oil.  The only advantage oil can provide after potting is to prevent corona or breakdown around the outside of the potting.

[HighVoltageRulezz]

My approach:

I printed discs made of PETG and stacked them on a PVC tube- they will be held in a tank construction under pretty good vacuum (pumped down with a Leybold D4) and then filled with Shell Diala- I hope to get rid of most air pockets this way.
I planned for 5 chambers with around 70 Turns of 0.4mm wire each. Primary is 2.5mm solid copter tapped in 4 places.

When done I will let you know!

[Teravolt]

that looks good HighVoltageRulezz how much voltage do you expect, oil is the way to go and a round bobbin is easier to make. I would like to see how it ends up. I'm hopping a trapazoidal secondary will have better holdoff on the high side. what program did you use to create the shape?

[HighVoltageRulezz]

Thanks!

I do all my designs in Fusion 360 and use Simplify3D for the slicing.

Greets

[Teravolt]

hi all sorry for the delay but here is the bobbin shape that i am hopping to print. it may be a while before I get the printer so i will see if i cant get it printed online
if any body likes it wants the file ir print one let me know. I shaped it around the amazon cores I bought. any comment are welcome.

[davekni]

I wonder if it would be possible to make longer slanted slots between segments, and/or move the slots to corners.  The possible issue I'm seeing is how to get the wire from the top of one segment all the way to the bottom of the next segment within the slot.  Perhaps a dab of glue or tape could handle that issue during winding with your existing design.

If the wire doesn't start cleanly at the bottom of the next section, then that lead-in will be higher and therefore against later layers of the new section being wound, making a place for likely breakdown through just two layers of wire enamel.

[Teravolt]

so I cut a notch in the long side so the winding can be moved from one layer to another. hear is the vision

[davekni]

I'm not spotting the difference, beyond the helpful context for bobbin placement.

Context does raise a question:  The large end of each bobbin appears spaced farther from its inside core segment than it is from remaining parts of the core.  If the design if for optimum breakdown voltage, I'd expect roughly equal distance in all directions from the final HV segment to core and primary winding.

[johnf]

Nice render
but
the core is an aid not the b all and end all

I put my primary under the secondary split in two ie two primaries two secondaries primary under each half of secondary to limit leakage inductance --you are already on a hiding to nothing because of the turns ratio no need to make it worse by putting the primary on the the other core half away from the secondary.

[Teravolt]

Hi johnf. and all, primarily this is a thought experiment and I have not done any heavy calculations to create the bobins. I might add sepperator betwean the cores like kapton. I am hopping to make it a resonant system like a ZVS and drive it with a brick H-bridge setup. I hope to add some capacitors in parallel with the primary to increase the circulating current in the primary. My point
about making the secondaries angled is for holdoff because I would like to get to get about 25kv peak each out of phase whitch means they will probly have to go under oil. the end product is for a fullwave HV dublier. alot of thease jacobs ladder ZVS setups put out about 10-15Kv using a bobin like PVC tubing wich is awsum but I would like a little more voltage. the primary dose not have to go under the secondary so I have used my imagination to make it more resistant and with 3D printers this may be pausible. Any comments welcome