Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Death of a Condenser (with Autopsy)

  1. Jul 6, 2016 #1
    So, a few days ago I was beetling down the road, and out of nowhere, the engine started all kinds of unacceptable tomfoolery: spluttering, loss of power, misfiring, hesitation, refusal to accelerate. I pulled over and fiddled with various things for an hour or so, and ended up with it running fine again.

    Couple days later, same thing again. I checked vacuum lines and the needle valve especially. After a while, I started it up, and it was fine. Later, though, on the return trip, all the problems returned: sort of a mixture of what it's like when you're running out of gas and only running on 3 cylinders. Luckily, it was down hill all the way home, and I got it back home despite the ailment. But a new symptom cropped up: on top of the other stuff, it backfired a couple times.

    I got out my manuals and fired up the Samba archives, and the internet in general, and read for a few hours. Basically, everything said this could either be a fuel system or electrical problem. My points, condenser, wires, and plugs are all less than two months old, so I attacked the carb: pulled it off and sprayed out every orifice and jet 20 times. Then I put it back together, but when I tried to start the car, all I got was a big backfire.

    More reading. I ran into reports that brand new, but poor quality, condensers sometimes fail after a very short time. On the remote off chance I might have bought the worst brand new condenser ever, I took it out and put the old one back in.

    She started right up, and 30 miles later with no return of the problem, I think that was probably it: the new condenser.

    I performed an autopsy on the brand new Autozone Duralast condenser that seems to have failed and found some damning badness inside. Here is some photographic documentation of the procedure:

    I cut the rolled over "lip" off the can off with a hacksaw. That's the ring around the green wire. Once that's off, I pulled out the rubber plug which insulates one end of the condenser. That's the black thing the green wire goes into.
    conAut1_zpsjyv4hzmt.jpg

    Pull the rubber plug all the way out, and out drops a little silver "soup can" looking type thing. That is the condenser plates, all rolled up. Notice they are not soldered to anything, or even mechanically connected in any way.
    ConAut%202_zps6yx9jve6.jpg

    Here is a shot of the inside of the can and one end of the condenser plates. They are clean and shiny: no problems here.
    ConAut%203_zpsotlyilvi.jpg

    Here is the other end of the plates and the steel contact that is what is on the inside end of the green wire. These don't look so good.
    ConAut%204_zps2hk6mz67.jpg

    Close up of above elements. You can see this is where the condenser fried. There has been a lot of high voltage sparking between these two things.
    ConAut%205_zps9erpk7bt.jpg

    This is how the steel contact plate is attached to the green wire.
    ConAut6_zpssjflfyyx.jpg

    Starting to unroll the condenser plates. The plates are, as far as I can tell, common aluminum foil. I miked one and it is about .0003 thick. The clear plastic insulation between the foil is about .0005 thick. There are two layers of foil and two layers of plastic. The two layers of foil are offset from each other: one sticks out about 1mm on one side, the other about 1mm on the other side. Once they are rolled up, the excess that is offset is just crushed to the bottom on the can on one end of the roll, and to the steel contact on the other end of the roll.
    ConAut7_zps8fqctfjl.jpg

    Here's all the intestines unrolled. I didn't find any evidence of sparking or burning inside; no insulation breakdown between the plates.
    ConAut%208_zpsmynj5dhf.jpg

    The aluminum foil (one plate or layer) is about .670 wide and about 137 inches long. That's an area of 91.79 square inches. It would be a square measuring only about 9.5 inches on a side (though it seems like it'd be larger than that from the pile you see.)

    So, what this autopsy reveals is that this is very old, uncomplicated technology. The basic design of these condensers prolly hasn't changed since the 1930's when the car was first designed. Now they use plastic to separate the plates. Back then they used wax paper. It ain't no kind a fancy.

    Why did it fail? I don't know. Maybe it was just manufacturing procedure: the rubber insulator wasn't pressed in far enough, leaving a poor contact between the steel and aluminum, inviting a spark to jump and burn the metal. That's plausible, but really just speculation.

    Anyway, this is what is going on inside an Autozone Duralast condenser. On Beetle forums I found reports of other makes of condenser failing prematurely, even Bosch, which is alleged to be the highest quality. On top of that, I happened upon an HVAC tech forum thread where a tech was complaining that more and more of the service calls he makes end up being caused by a failed capacitor. Another tech chimed in in agreement and said he was no longer confident about the caps that were traditionally known to be high quality, and didn't know what to tell his customers who were wondering why the caps kept failing.

    Is it your experience capacitors are being more and more poorly made?
     
  2. jcsd
  3. Jul 6, 2016 #2

    phinds

    User Avatar
    Gold Member
    2016 Award

    Outstanding autopsy. Wish I could be some help but my only speculation, first thing I thought of, is what you already guessed might be the case which is loose fit or one that worked a bit loose over time, thus not giving a good contact and allowing the sparking that caused the obvious darkening where the cap contacts the upper plate.
     
  4. Jul 6, 2016 #3
    All I could do is to speculate and I assume you are more or less on track with your ideas of why it failed.

    While it is true that this design is old technology it has been in use and working for a long time with "normal" replacement times. So...just because the design is old does not in and of itself create the failure.I really think that many companies in years past tried harder to make good products than they do today.

    I find caps for electronics made by different companies today range widely in failure rates. For the most part you get what you pay for. It also depends on the style of cap. F&T a German cap MFG makes good quality axial electrolytic but their "can" type caps containing two caps in one housing are reported to be less than good. I also assume these conditions change from time to time as companies make improvements or go the other way and make less good products.

    Billy
     
  5. Jul 7, 2016 #4
    Thanks!
    After I posted this same thread on the Beetle forum, someone linked to a similar autopsy on two failed condensers from a different make/model of old car. That author came to the same conclusion about why they failed.
    I didn't mean to imply the technology was old, therefore crude. What I was aiming to convey is that this technology is simple enough to have been mastered a long time ago, that there's no excuse for a modern manufacturer to not be able to produce an extremely reliable version of this simple component.
    This. There is either an unacceptable sloppiness at work here or a deliberate effort at planned obsolescence.

    As I mentioned, I solved the problem by returning the old condenser to the car. I had only replaced it for "routine maintenance" purposes, along with the plugs, wires and points. I do not have the original distributor, but the one I have is probably at least 20 years old, and this condenser (the one that works) might well be the original condenser that came with the distributor. (I've only owned this Beetle for a year and a half and can only guess at the history.)

    A guy on the Beetle forum claims he runs a condenser that was original to the car: over 40 years old. Other people claim that, when you can get hold of condensers that were manufactured in the 60's and 70's, they last years without failing. (These unused, original things come up on ebay now and then; cases and boxes of them discovered in the dusty back shelves of old dealerships and repair shops.) Apparently people threw away really superior condensers by the millions over the years on the principle they should be changed out after so many miles of use for "routine maintenance".
     
  6. Jul 7, 2016 #5
    Even though you found the problem I can reassure you with my own experience.
    If i'm correct your driving a WV Beetle?
    Anyways the same style round capacitor comes also in other older typically mechanical distributor ignitions , my older retro car has one too , once i was driving and even though the engine is fresh and all is tidy after all I done the whole engine job myself , the car started to jerk in acceleration and the engine was literally not running whenever it was taken out of gear.
    I noticed that this happens more when there is vibration and less when i drive at lower speed.
    The capacitor has a screw which attached it to the common chassis ground of the engine block , that screw simply had come loose , as soon as I made it tight again and made sure there is a good electrical ground connection all problems faded.

    But at first I was rather terrified by this problem , I thought it must be something big and I can't find the source , it's common that such small things that are hard to see and find cause big problems.


    BTW , that capacitor you have taken photos of seems almost like home made stuff , which is a bit weird.
     
  7. Jul 7, 2016 #6

    Averagesupernova

    User Avatar
    Gold Member

    I have swapped out numerous condensers that were new a short time before. There truly is something to the theory that condensers made many years ago will outlast what is built today. A failed condenser can be hard to diagnose. Breaker points can just about always be 'brought back' until there is little left of them. An ignition coil is often thermal when failing. Knowing these 2 things often leaves only the condenser left as the suspected fail. I can't say I have ever replaced a distributor cap or rotor and solved a problem. I have dried a handful out with a hair dryer due to condensation and replaced them as part of a regular tune up but never one day had an engine suddenly run like crap and had the rotor or cap fix the problem. I always thought that a breaker points ignition system would not leave you on the road as often as an electronic ignition module but with the quality of condensers we seem to be getting I am not so sure anymore. You can clean up a set of points good enough to run with a pocket knife but there is little to be done with a failed condenser while stranded along side the road.
     
  8. Jul 7, 2016 #7
    well even with bad capacitors the old mechanical contact ignition can't leave you on the road unless you are really out of touch with mechanics.
    I simply have the rotor contact , a few fuses and an extra capacitor with my wherever I go in a small box so even if they would go bad , unlikely , I would have a replacement , a small capacitor is no problem to take with you in a car.
     
  9. Jul 7, 2016 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    What do you get for capacitance by εε0 Area/distance ?

    I'm of the longstanding belief that around 0,22uf is typical for condensers

    Looks like typical ε is between 2 and 3 ?
    http://users.tm.net/lapointe/Plastics.htm

    I would have expected them to spotweld the endcaps to the foil by passing a high current somewhere in the manufacturing process .Industrial caps i've unrolled use spotwelds at connection point. Could be a glitch on the assembly line .

    No chemicals in that thing ? If not,
    I don't see a time related age failure mechanism. Keep your old cap.

    Nice Post ! Look what we all learned ....Thanks -

    old jim
     
    Last edited: Jul 7, 2016
  10. Jul 8, 2016 #9
    Yes, it's a 1972 VW Beetle.

    Your story was interesting. I can see it's quite possible to have all the same symptoms from a mere loose screw. And the symptoms are very alarming. You start to imagine you need a whole new engine.
    After this, I am going to be sure to always carry a spare.

    Here's a complaint about condensers I found today that echos my suspicions:
    http://www.distributordoctor.com/distributor_condensers.htm
    This guy either knows or supposes the dielectric material is mylar:
    http://www.nonlintec.com/sprite/cap_failure/
    The 3 capacitors he tested were:
    This next guy says:
    http://mgaguru.com/mgtech/ignition/ig129.htm

    The foil in this one I took apart is probably too delicate for any kind of connection other than the one they used. I would bet it's actually an OK design provided they stick to tolerances: the "soup can" should be a certain length, and the rubber plug should be held in the can a certain distance such that it puts a certain pressure on the plates. The rubber is softish and would constitute enough of a spring to be able to handle a bit of excess pressure. No reason I can see for them to be erring on the loose side rather than the tight side.

    Only hint of a chemical is that they might have used some silicon on the rubber to help push it past the retaining lip. It has to be squeezed to get past that, then it would expand back out and be held in place by the rolled over metal. Otherwise nothing at all liquid inside.

    Glad you liked it!
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Death of a Condenser (with Autopsy)
Loading...