epoxy vs epoxy failure analysis / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

epoxy vs epoxy failure analysis

Move Thread

LAN_403

Goodchild

Mon Oct 29 2012, 05:20AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Hello good folks of 4hv, it's been a while sense I posted about work I have been doing mainly due to school. But today I have a good old failure analysis for you all regarding epoxy and how they are not all created equal.

The other night we are running the two large DRSSTC, playing MIDI all is good and coils are pushing about 10 feet each. About half way through the song we smell burning, but can't seem to find the source. After a couple minuets of this it was finally spotted at the bottom of my secondary, we had a fire!! Power was cut fire was put out, but the damage was done.

8133868815 Df65832555

Under closer observation it turns out that the wire from the winding had snapped from mechanical fatigue from years of moving the setup and running of the coil. As a result of this arcing between the broken wires it had set the surrounding PVC and epoxy on fire.

An observant eye however will notice that the epoxy used to coat the main winding and the smaller loose wire are of different types. The main winding is covered with counter top type two part epoxy while the loose GND wire is covered in 60min two part epoxy.

After sanding and cleaning it became obvious that only the 60min epoxy burst into flames while the main winding stayed completely intact. To put this into perspective this fire was so hot that is discolored the primary (2 inches way) and chard the PVC underneath. The the main winding epoxy was untouched along with the wire.

After fixing the secondary it is better than when it was originally constructed.

8133868027 C27814c58d

After the fix the coil in now happily pushing out 14 feet without breaking a sweat. If you look close the 14 footer is near horizontal in this photo.

8133894706 8e9aa56b09

The moral of my story is watch out for the epoxy you use with HV they are not all the same. Also metal fatigue is a killer for secondary wire. But for a coil that is nearly 2 years old now and has close to 100 hours of run time and traveled hundreds of miles through a rain storm in the back of an open pickup track, it's doing rather well.

Hope you can learn something from my failure!

Also high res images can be found here:

nixie

Mon Oct 29 2012, 05:27PM

Registered Member #3908 Joined: Tue May 24 2011, 09:40PM
Location: Gilbert, Arizona USA
Posts: 68

Nice detail and images!

I'll never forgot that smell...of pvc

We'll know if there's ever a next time.

Jeff

Ben Solon

Wed Oct 31 2012, 01:10AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

When I get to winding the coil of my new drsstc in the next week or two, i'll test some of the adhesives I'll be using(depending on the results). Maybe we could use this post as a list of results for flammability tests? I think that to mimic the type of arc a tc would produce we could use some sort of hv pulsed current such as an ignition coil? it would better imitate the capacitive bursts of energy seen better than a hot arc like a flyback would produce.

Herr Zapp

Wed Oct 31 2012, 05:31PM

Registered Member #480 Joined: Thu Jul 06 2006, 07:08PM
Location: North America
Posts: 644

Goodchild -

You were lucky that you were able to salvage the secondary after that much local damage ..... did you have to "unwind" several turns to perform the repair?

I would like to offer some alternate failure analysis scenarios .... would it be possible for you to provide some additional background info?

In the first photo, the bottom-most turn of the secondary wire comes off the close-wound area, passes "loosely" through the first epoxy blob (the wire was obviously not under tension when epoxy was applied), and enters the burned area. Then, it looks like the wire possibly exits from the LH side of the burned area, passes through another blob of epoxy and continues on to disappear under the head of a plastic screw. Is this just an illusion, and in reality the wire just makes a 90 deg turn under the burned area and proceeds across the end cap to the bolt? Or was there some sort of junction in the wire here, directly under the burned area?

There is no sign of any discoloration of the epoxy surrounding the wire in the "blobs", so there is no evidence that resistive heating of the wire was a factor in the failure.

Was the segment of wire that was routed across the end-cap to the central bolt also encapsulated (and strain-relieved) in epoxy?

You conclude that "the wire had snapped from mechanical fatigue". Given that the wire appears to have been fully encapsulated in epoxy and could not have been exposed to "bending" stresses, and was not under tension when the epoxy cured, and that there is no way that the secondary coilform could have expanded (thermally) enough "stretch" the wire, I think mechanical "fatigue" of the wire was probably not a factor. Depending on how the end of the wire was connected to the central bolt, it is possible that the wire got stretched and failed in tension if the bolt got rotated during mounting of the secondary, and "wound up" the wire until it broke? Do you install (and firmly tighten) another nut on the bolt to secure the secondary to the coil's base structure?

Do you know what the average ground current is, when the coil is operating at full power?

What gage wire was the secondary wound with?

As for the suspected difference in flammability between the epoxy you used for coating the secondary windings vs the 5-minute epoxy you used for securing the ground wire, I think there may be a simple explanation.

The 5-minute epoxy at the failure initiation site (presumably where the wire was broken) was surrounded by air on one side, and by PVC pipe on the other side. Neither of these have very good thermal conductivity, and it would have been fairly easy to raise the temperature of the epoxy to its ignition point at a small, localized site, especially if there was a continuous arc present to ignite the flammable vapor from the epoxy as it thermally decomposed. On the other hand, the thin layer of epoxy covering the windings was backed by a nearly solid layer of copper, which would provided a much more efficient "heat sink" than PVC. The reason that the epoxy over the windings never "caught fire" may have simply been that (because of the copper heat sink underneath) it never got hot enough to reach its ignition point.

It is very possible that the two different epoxies you used do have different thermal decomposition temperatures, and would start to liberate ignitable vapors at different temperatures, but the heat-sink provided by the windings may have been a larger factor.

Home improvisation of standard "flash point" testing would help determine which epoxy started to thermally decompose first, which one had the lowest ignition temperature, and if there was a significant difference between the two.

Herr Zapp

Dr Hankenstein

Fri Nov 02 2012, 05:46AM

Registered Member #1642 Joined: Sun Aug 17 2008, 11:36PM
Location: Black Canyon City
Posts: 96

Yes! I think Herr Zapp was there when we fried the secondary on Zotzilla. We just broke out the cement grinder and ground off the scorched insulation, chiseled off some of the bad turns, cleaned the affected area with some lacquer thinner to remove the carbon, reconnected the wire with a cutting torch, re-insulated the secondary with some thick electrical tape; and to this day, Zotzilla performs better than ever...just looks sort of crappy where the damage occurred, but I'm afraid to re-wire because it runs so dog gone good...so much better now...go figure! There was an entire army of folks there helping to fix the old girl; and we won!

Eric. don't listen to all that long winded, unnecessary ridiculous analysis BS about how the fumes are going to kill ya and bla, bla, bla,....just chisel off the bad area out and fix it, brotha! It worked for us...it can work for you.

BTW Scott; where's the coil, man! We're still waiting. Arm chairing it; eh?

Hank

Kizmo

Fri Nov 02 2012, 06:01AM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

Hehee secondary repairs :)

Few bad turns is not worth of rewinding but slightly larger accidents may require more advanced surgery

Even though some time ago i found out how little secondary coating really matters when everything else is done right:

Dr. Drone

Fri Nov 02 2012, 03:42PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

Goodchild

Fri Nov 02 2012, 04:42PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Thankfully we didn't have to remove any windings, not even a single turn. Only the yellow colored epoxy burnt. The clear stuff that covers the main windings protected the wire like a champ! We simply just removed black sludge with acetone and sanded off the burnt epoxy and it looked good as new.

I always did wonder about the black tape on Zotzilla, I'm going to need to get one of those "Shit Happens" stickers now.

Herr Zapp,

We know for a fact that it was the bottom wire breaking, it was going right over a 90 degree corner with very little epoxy and to no surprise it snapped right at that corner. It was a silly mistake made 2 years ago. Even though it had some epoxy on it, it was still not enough to prevent the problem.

As for heating of the wire I find that to be very unlikely, although the ground current may be 12 amps or more in most cases, the current in the secondary is out of phase with the voltage by almost 90* making the reactive power high but the "real" power rather low. This is because when the current is at 12 amps or so the voltage is almost 0 and vise versa. Making the heating of the wire not an issue at this frequency.

For the heat sinking of the wire theory I don't think I agree with that ether, because even the areas of good epoxy that had no wire under it didn't burn. It's clear from my photo that the yellow epoxy cooked, it is pitted and discolored. The other epoxy is just black but smooth. This is just from fumes deposited from the burning of the epoxy directly below it. To further my point to this the primary 2 inches away changed color from the heat of this flame and melted the lexan cover. The primary has way more thermal mass that the 24AWG dose; making it unlikely that the primary could heat up that much without the secondary doing the same.

Steve Ward

Thu Nov 15 2012, 09:39PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

As for heating of the wire I find that to be very unlikely, although the ground current may be 12 amps or more in most cases, the current in the secondary is out of phase with the voltage by almost 90* making the reactive power high but the "real" power rather low. This is because when the current is at 12 amps or so the voltage is almost 0 and vise versa. Making the heating of the wire not an issue at this frequency.

You are confusing yourself here. Current is what causes heating, we dont care about the coils voltage or its phase. The RMS current is likely 1-2A i would estimate (based on your 12Apk number). Obviously there was some defect that caused 24awg wire to fail at this current.

Ive seen ground connections fail on secondary coils similarly to this failure. The failure i observed was on a poor solder connection to magnet wire that was not properly stripped. The connection was fine for years and eventually began arcing and burned up the area around the connection. This is a notorious problem in pulsed power where seemingly "good" connections can fatigue over years of use and eventually go open circuit from arcing and pitting of the surfaces.

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