A New Idea For Ferrite HVHF Transformer Construction / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

A New Idea For Ferrite HVHF Transformer Construction

1 2 3 4 next

Move Thread

LAN_403

jpsmith123

Sat Jul 02 2011, 10:47PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I wonder if the following idea is feasible?

The idea is to take a cheap off-the-shelf Hammond #1591 series plastic (ABS) enclosure and modify it by boring two holes through it, spaced 3.0 inches apart on center (in this case to accomodate a ferroxcube U100/57/25 double "U" core).

Then you would turn two cylinders (coil formers) to fit snugly into the holes.

Then wind your HV secondary (or secondaries) on the coil formers. Now if you had some 26 gauge teflon wire, you could wind, say, one 300 turn coil on each leg, for a total of 600 turns. Or you could wind one 600 turn coil on one leg. (The total build would be about 1.0 inch so you could have 2 * 0.5" or 1 * 1.0" to get your 600 turns). If the coil winding width was kept at 1.0 inch (or less), you wouldn't need any layer insulation, except maybe for some teflon thread-sealant tape to help hold it together.

Then you would epoxy the bottom part of the cylinders (with the coil(s) wound on them) into the enclosure, and put the lid on (to hold everything in place while the epoxy cures).

Then after the epoxy is cured, you'd take the lid off and pot the assembly with a low-viscosity RTV or wax or something. You wouldn't have to fill the whole thing up; rather, you'd just cover the coils to a depth of maybe 1/8" to 1/4" inch; this way you would more than adequate insulation and also room for thermal expansion of the RTV or whatever.

In this case, you wouldn't even need to pull a vacuum on the encapsulant, since you're using it basically just to protect the assembly from the environment and to make it electrically and mechanically rugged and reliable.

Here's a picture of what I envision it looking like:

As I see it, the benefits would be:

(1) No problematic layer insulation (except for teflon tape which should be very easy to work with in this case).
(2) High Frequency capability (Should have low capacitance and be able to be driven at 30 to 40 kHz at least).
(3) Doesn't need to be vacuum impregnated (with the coil width limited to 1 inch, the field stress along the edge of the coil should be well below 30 kv/cm.
(4) High power throughput and low loss (The 26 gauge wire should be able to handle 0.5A rms with no problem, the teflon insulation has a very low dissipation factor, and the low field stress and high frequency capability should easily allow 5 kw or more.
(4) With the RTV potting, it should be rugged and reliable and relatively safe.

One question I have is, can you bore a hole through ABS plastic with a hole saw (the cheapo kind you'd find at Harbor Freight)? I've never worked with ABS plastic so I'm not sure what's the best way to put the holes in it. Does anyone have any idea about that?

jpsmith123

Thu Jul 28 2011, 07:36PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

After more thought about this kind of construction, I think that unless the goal is "extreme" high voltage right off the transformer, where both legs of the core would have a HV secondary winding, it would be better (as in more flexible) to leave one leg of the core outside of the box, so as to have an accessible primary winding.

This would also allow the rest of the box to be used to house a voltage multiplier if necessary.

Being that I have enough 26 gauge teflon wire to make two 600 turn coils, I'd like to make one as a transformer, and one as a transformer/multiplier unit.

Patrick

Thu Jul 28 2011, 08:13PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

yes you could pot everything in the case with epoxy or silicone. the diodes the caps the coil everything would be in one compact structure.

Proud Mary

Thu Jul 28 2011, 09:02PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

It's not usually a good idea to generate (non-industrial) high voltage in a single step, which streses both components and insulation. Much better to have a number of small steps to minimise the stress between each stage.

Ash Small

Thu Jul 28 2011, 09:56PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I'm sure you won't consider this post 'off topic', but I assume the core you mention is square section. Is a round section coil former ok, or are better results achieved with a square section former? (I'm currently working on square section clear polystyrene bobbins)

Patrick

Fri Jul 29 2011, 12:19AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

well Ash, the square cross section is greater in SA than a circle of the same diameter as square of the same side length due to A = PI x r^2, or A = ~ D^2 x 0.8, but generally there is more field intensity at square coners , so thats the reason i prefer round instead of square, but there no huge difference between the two. I prefer the round ones whenever possible.

jpsmith123

Fri Jul 29 2011, 01:46AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Ash Small wrote ...

I'm sure you won't consider this post 'off topic', but I assume the core you mention is square section. Is a round section coil former ok, or are better results achieved with a square section former? (I'm currently working on square section clear polystyrene bobbins)

When you say "square section" do you mean that both the inner and outer surfaces are square? I've tried putting windings on square bobbins (square inside and outside) and I don't like it. One problem is that the winding tension varies as it turns, and that irritates me. Another is that it takes more wire for a given number of turns. It may also cause some field enhancement. All things considered I don't recommend it.

Ash Small

Fri Jul 29 2011, 09:26AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Patrick wrote ...

well Ash, the square cross section is greater in SA than a circle of the same diameter as square of the same side length due to A = PI x r^2, or A = ~ D^2 x 0.8, but generally there is more field intensity at square coners , so thats the reason i prefer round instead of square, but there no huge difference between the two. I prefer the round ones whenever possible.

jpsmith123 wrote ...

When you say "square section" do you mean that both the inner and outer surfaces are square? I've tried putting windings on square bobbins (square inside and outside) and I don't like it. One problem is that the winding tension varies as it turns, and that irritates me. Another is that it takes more wire for a given number of turns. It may also cause some field enhancement. All things considered I don't recommend it.

According to my calculations a round former uses 11% more wire than a square one.

4 x sqrt ((r x r) + (r x r)) = 5.66r

2 x Pi x r = 6.284r

All the commercial bobbins I've seen for square section cores are square section. You can also get more turns on a square section bobbin for a given window size.

It seems you both agree that a square section is more efficient.

I agree that round 'might' be easier to wind, but that appears to be the only advantage.

Please correct me if I'm mistaken about any of the above.

Patrick

Fri Jul 29 2011, 09:41AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Ash Small wrote ...

According to my calculations a round former uses 11% more wire than a square one.

sqrt ((r x r) + (r x r)) X 4 = 5.66r

2 x Pi x r = 6.284r

All the commercial bobbins I've seen for square section cores are square section. You can also get more turns on a square section bobbin for a given window size.

It seems you both agree that a square section is more efficient.

I agree that round 'might' be easier to wind, but that appears to be the only advantage.

I was meaning surface area of the cross section circumscribed about the circle, your showing it circumscribed within.

Ash Small wrote ...

It seems you both agree that a square section is more efficient.

I dispute this claim. Ill use either round or square (what I can find), but prefer round (no corners).

Ash Small

Fri Jul 29 2011, 10:12AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Patrick wrote ...

.
I was meaning surface area of the cross section circumscribed about the circle, your showing it circumscribed within.

We are discussing using a square section core though, aren't we?

Patrick wrote ...

Ash Small wrote ...

It seems you both agree that a square section is more efficient.

I dispute this claim. Ill use either round or square (what I can find), but prefer round (no corners).

Do I understand correctly? You are of the opinion that a round bobbin on a square section core is more efficient, electro-magnetically speaking, that a square section bobbin on the same core?

1 2 3 4 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob