The Stovebolt Site Forums The Shop Area The Electrical Bay Ignition condensers...a suitable alternative...

In another post mentioned I am in the process of contacting manufacturers of auto condensers (capacitors to be honest here) to discuss where they're made and specifically if they're made in China. Comments here and on other forums have left me thinking there could be an out-of-box failure rate with the Chinese made condensers as high as 50%. This seems unbelievably high even for China. Knowing how manufacturing of small electronic parts not enjoying high consumer demand is conducted in China, I am 99% positive these condensers were produced by one company in one large run (possibly years ago) and have been and are being distributed by one or two Chinese jobber-type companies from existing warehoused supply. I further suspect these are failing for 2 reasons: poor manufacturing and heat...maybe in equal parts. So far I've learned Standard brand of condensers are not made in China but made in Mexico and the man I'm talking to there says he knows of no failure trend in them and has asked for part numbers if I can produce them. He also added "keep those Chevy and GMC sixes running."

If you have ever torn one of these condensers apart you will see a design which is simple and subject to a few different types of failures. You'll see a cylinder which is made of a strip of aluminum foil with either gel or some type of film on one side which has been tightly rolled. It sits inside the canister but only touches that canister at one end (to keep it from shorting out along the side). The size, the length of foil, the nature of the coating...these are the things that determine the capacitance rating, voltage limit and temperature tolerance of this sort of device. Age and heat will eventually cause failure. The case provides one contact point---ground, and there's a wire soldered to a copper or brass disk which is insulated from the case by a rubber grommet contacting the other end of the internal cylinder. Dead simple. What can go wrong? I won't waste the electrons necessary to discuss everything but the main 2 things of rapid failure would be: contact is lost either where the case contacts the little aluminum foil/film cylinder or where the wire end connects it...creating an open capacitor or an intermittent one and/or heat and current causing the gel or film to degrade, collapse or shrink to the point the foil layers touch each other creating a short circuit. And heat combined with current could cause an instant failure in one made improperly or it could take a little while for this to occur. Heat for any capacitor is detrimental and in fact students of electronics are taught to begin their circuit troubleshooting by spraying a chiller on any capacitor they see...particularly electrolytic capacitors. Many times the reduced temp will make the device start working again. Let's go back to the lost contact issue. Let's suppose just for grins the little foil cylinder inside was made just a wee bit too short for the canister it fits inside of or let's say the canister wasn't crimped to the right point to press the cylinder into contact with both ends...let's also say there is little to no testing or quality control in China for these. Easy to see how intermittent or no contact (or even a short circuit) could happen, yes?

Bottom line is a better, longer lasting and much cheaper alternative exists...an alternative very common today but which would not have been feasible for the consumer market 60+ years ago. A common film capacitor of the proper uF and voltage rating which is high temperature rated. Most condensers (those used for inline 6 cylinder engines, anyway) seem to work fine at .22uF and a voltage rating of 500v would be plenty. A quick look at Digi-Key brought up this one:
https://www.digikey.com/product-detail/en/kemet/R75QN32204000J/399-18311-ND/9449366 There are others, but this one costs about $1.50 and fits most of the needed values perfectly. I'll keep looking for a more high temp capacitor, but I'll tell you up front those are more $$$...and some are larger

Note: it won't look like a typical auto condenser and there is the chance it might have to be mounted outside the distributor (to reduce heat), but I'm certain it would work. And it probably would never need to be replaced.

Here is my question: This capacitor is fine up to 220 degrees F. Your distributor could get hotter than that inside (hopefully not, but it could if you're running your engine on the hot side of things). If it needed to be mounted elsewhere (say over by the voltage regulator or in some other cooler spot) with one wire connected to the points side of your coil and another wire to ground, would this be ok with you? In this application you would have no condenser inside your distributor.

Mount it on a heat sink near the fan- - -even on a hot-running engine there should be a little safety margin at 180+ degrees, and if a stovebolt engine ever gets that hot the average forum member with lace on his drawers will call 911 because his blood pressure is spiking- - - -the engine won't care!

What about the flywheel time on subsequent charge/discharge cycles? Condenser operation is not a one-time thing- - - -it repeatedly charges and discharges as the induced voltage in the primary winding from the coil firing must be dissipated by cycling from the insulated side of the condenser to ground and back while the points are opening and closing. Can the capacitor you're recommending be charged from either side of the dielectric?
Jerry

Jon, Thank you for the great research. Here is what i have used in the past for fuel lines. It may work on distributors. A heat sink found in electrical items like on circuit boards (many sizes) may work on a distributor. Ed

Like you, I wonder about the quality of parts that come from China. Just for discussion purposes.

Inside engine distributors can be heat, freezing cold, hopefully dry air but also high voltage (typically 12 to 25KV). I suspect the outer metal cover of a automotive condenser/capacitor provides some shielding from EMF (the arc across the gap between the rotor button and the wire terminals comes to mind)? For high voltage military power supplies, I've seen removable metal covers over the capacitors.

Almost any .22uF 600V capacitor would work in a clean environment. An external capacitor to the distributor could be exposed to a splash of water from a mud hole, oil leaking off the valve cover, a coolant leak etc. all of which would alter it's capacitance value so it would need to be completely insulated or in a protected box.

If you move the capacitor behind the firewall, to hide it, you need to determine if your length of shielded copper-stranded wire adds/subtracts any capacitance to the original .22uF value.

My experiences in the late 60s and early 70s : the american made 216/235/261 condensers had a relatively high failure rate (maybe it was just my bad luck).

I have always carried a spare condenser (and, a spare "configured" complete distributor and set of ignition wires).

I once had a problem with the turn signal flasher when I was at 6 volts. The new 6-volt flashers that I bought flashed way too fast. I went to my local junkyard and found one which happened to be in an Army AD. No charge and it worked fine until I went to 12-volts.

Hi Jerry and thanks for the comments. As for the flywheel time, we'd have to know the resistance value, yes? At least I'll say the formula I have requires it. However, simple charge/discharge time lacking resistance for this type of capacitor (with a capacitance of .22uF and assuming voltage of 1000) is as I recall a fraction of a microsecond. I can't imagine the primitive foil/gel or foil film condenser would be any faster or more dependable than this newer type of capacitor.

This morning I found another one I like better. It is roughly 1 inch square by 1/2 inch tall (a specialty thin dielectric type) with 25mm long leads. Also, it is rated up to 260 degrees F. That is about the top range for standard grade capacitors. And it is one of the type designed so that it will "self-heal" shorts caused by transient over-voltage spikes. But I don't think this will ever be a factor here...it is rated up to 1000 volts dc and 700 AC. The dielectric will charge both ways (which as you pointed out must happen).

For the people looking for a traditional condenser not made in China, Borg-Warner advises me they make these (BERU brand---Germany) but not specifically for Stovebolt engines. Possibly another condenser could be adapted to work but the case, mounting clip, connection lead, size would be different and possibly a condenser for a later year Chevy engine might adapt. The man in Germany has given me the name of a person to contact in the USA. I also spoke with a friend (retired from NAPA) who said all these being sold in the USA have been made for 25 years or more in Mexico or China with the possible exception of a couple of special ones. He said the ones from Mexico might be less prone to failure (and he thought some were once made in Brazil) but he also added they've been hit or miss from the standpoint of accuracy and quality for a long time.

Jon, you're probably one of less than a dozen people on the forum who actually understands the function of a condenser in an ignition circuit. Since the primary ignition circuit also includes the ignition switch, the battery, and every hot and ground wire in between the condenser's components, figuring the circuit resistance would be quite an exercise in higher mathematics. How about including the inductive reactance of the coil windings in the mix on the second or third oscillation of the primary waveform?? Few, if any people realize that the condenser discharges itself by flowing current backwards through the battery, several times a second, to charge from the ground side. How many of us know that the little sine wave at the "points open" part of an oscilloscope pattern is the voltage slowly dissipating as the primary voltage spike from magnetic induction bleeds away? Now, let's do that something like 60 times a second at cruise RPM!

"Voltage transient spikes"- - - - - -like maybe a loose plug wire that dangles out there in midair and causes 20KVDC to look for somewhere to go- - - -and fires down the coil high tension snout to the primary terminals, for instance? I've only seen that happen a few thousand times!
Jerry

Yes, at least a few thousand times. But you can always grab it and give that 20KVDC a temporary home.

I used to let my high school auto mechanics class get into a big circle and hold hands with me while I grabbed a plug wire from a HEI distributor running on a test bench. Those big, bad football players and ghetto gangbangers weren't nearly as tough ad they thought they were! Then I'd stand there holding a spark plug in my hand and let them watch the spark jump the gap while I told them what a bunch of pansies they all were! "You're crazy, man!" "Yep, and you'd better hope you never get to see how crazy a Southeast Asia veteran can be when he really wants to!"

LOL!
Jerry

Dirty shame schools don't allow teachers to "teach" students real life lessons any more.
We had a shop teacher who taught us how to weld. Someone got real good at it. He welded the jaws shut on the shop vice.
The next day, the shop teacher had us all go over to bench to see the vandalized vice.
He simply asked, "Who did this?" No one fessed up. He said, "Too bad whoever did this didn't come clean because he would have received an A in this class. That is some of the best welds I have ever seen. Look at the perfect bead from one end to the other, and the good penetration."
He never said another word about it. The next day, that vice was returned to its fully functioning state.
High level teaching method right there.

Tesla coils work too.

A few other replies to my inquiries have been received. Interestingly some who appear in ThomasNet and other industry lists as makers of condensers tell me that information isn't correct. Also interesting those who I am pretty certain are simply distributing Chinese produced products haven't replied. So currently for the typical condenser product (non-Chinese), we have only Standard Motor Products (made in Mexico and with a helpful and sympathetic fellow there) and BERU (Germany) who may possibly have an adaptable condenser somewhere in their product list but not one made for these engines. In fact I'm putting BERU on the doubtful list as it seems nearly everything they make is for European vehicles. I'm still awaiting replies from others and will let you know as they arrive.

Just to throw a bit more confusion into the mix (When have I ever done that?)- - - - -Using one of the "no ballast resistor" ignition coils changes the entire overall resistance of the ignition circuit, along with the inductive reactance I mentioned before, which will change the capacitance required to assure complete dissipation of the primary voltage spike before the points close again, and also prevent material transfer from one contact point to the other.
Jerry

Have you read this article? http://www.nonlintec.com/sprite/cap_failure/

He has a couple of suggestions for a replacement capacitor. The Porsche one might work.

Be careful to wear gloves and try not to breath-in fumes if you want to open up one of these capacitors for inspection.

Excellent Thread !!

Jon,

Can't thank you enough for all of the leg work that you are doing on this!! It will go a long way toward helping us keep these old systems running. And, even more importantly, providing insight into what's going on when they don't.

For the last 20 years or so, I have typically used the Echlin line of ignition parts from NAPA on any breaker point ignition vehicles. I have not personally experienced any condenser failures - although I will be the first to admit that my sample size is very small, and do not put nearly the number of miles on these vehicles as many others on this site.

Regarding the quality of available condensers today, I appreciate your statement about the potential issues related to the overseas "manufacturing of small electronic parts not enjoying high consumer demand". Very well stated. Unlike a smart phone or TV, which is made in some huge, modern, automated factory, it would not surprise me if these are made in some guy's garage, with minimal understanding of the fundamental design concepts, using 50 year old tooling and technology.

I think you are right on regarding your search for capacitors that use newer capacitor technology. No secret that significant improvements have been made over the last 20 - 30 years.

I also think that the main challenge will be packaging. It's hard to beat the convenience of having that nice compact metal encased single lead capacitor sitting there on the breaker plate.

I'm not quite as concerned about other factors such as temperature. Many of these capacitors are used in motor controls or other circuits that may experience elevated temperatures. How hot does a distributor really get? If you throw water on it does it turn to steam? Probably not, but it would be interesting to see some actual temperature data. Same with voltage rating. Lots of offerings out there in the appropriate range.

It would be interesting to see the original requirements from the likes of Delco Remy, or other ignition manufactures for the capacitors used in their distributors. I would be willing to bet that most are way below what is available today, or the offerings that you are investigating.

It would be great if some entrepreneur somewhere would take this on. I would easily pay several times the current cost of a condenser if I new that it offered better reliability.

With regards to whether or not the capacitor would be fast enough, assuming no change in the coil or other primary ignition system components, the value of the capacitor is what really determines that, right? It's the "C" in RC time constant - which is indicative of how fast the capacitor in an RC circuit can charge / discharge. So, I'm thinking that as long as you keep close to the original value, you will be OK.

Jerry's point regarding the function of the capacitor is well taken. Back in the Dark Ages, when we first learned about breaker point ignition systems, the purpose of the condenser was usually just explained as: "It keeps the points from burning up." In reality, it is more complicated then just that.

This link is from a college EE lab that actually deals with the physics of ignition systems. It is pretty interesting.

Physics of Automotive Ignition Systems

The paper states that the sizing of the capacitor is important, in order to match the resonant frequency of the coil. The idea is to create an RLC circuit which increases the emf in the primary, which in turn increases the available voltage in the coil secondary. It would be interesting to scrounge up some of the test equipment shown in the lab, and verify this assumption.

I have BTW, seen this in other papers on ignition systems. However one thought always comes to mind: What about electronically controlled ignition systems such as HEI? I don't ever remember seeing a condenser attached to them - just the control module. Based on the size (in uF) of the condenser, I doubt that it could have been packaged inside the module. Maybe they found another way, or were just able to get around the problem entirely by increasing the coil primary current.


Best Regards….

Thanks Bob...and also thanks to Gord and Bob for the links to those articles. It was especially interesting to see that fellow had a way to hook up a distributor and an oscilloscope, wasn't it? Yes, the C in the RC time constant is the capacitance and as Jerry pointed out we might have a bit of work measuring the R value.

As for the HEI, since that's a point-free (notice I didn't say pointless) distributor, a condenser wouldn't be needed for the same purpose. However there was a condenser in the early HEI distributors and it looked like the typical one in the points distributor. I want to say it was used to help control radio interference.

My thought is that there is room to mount the capacitor in the stock 235 distributor and doing so would not be difficult. One lead would attach to the mounting tab and become connected to the breaker plate (ground) and the other would be connected to an insulated wire which would attach exactly as the original condenser mounted. I think it should help points last longer and do a more accurate job. The next time I have to order anything from Mouser, I'll include a few of these in the order and modify them as needed. Maybe then someone here can give it a try. Personally I'm using one of Tom Langdon's mini-HEI distributors and it is working great. Now that we're back, I'm going to see if I still have an old points distributor here so I can see how it would fit.

I will tell you this: if you ever have to replace an ignition module (especially in a Ford product), use acetone to clean the mounting surface really well and get some extra thermal grease to mount it. A couple of months ago I replaced one in a friend's Ford over in NM. He mentioned they were going bad about every 10 days to a couple of weeks and he couldn't figure out why. Turned out somebody had left a horizontal ridge of dried thermal grease about 1mm thick when the first module failed and the new modules (apparently a bit wider than the original one and made in...you guessed it...China) sat on top of that ridge...meaning they were not making full contact with the heat sink and would burn out pretty predictably. We were there last week and I noticed his Ford is still running fine.

The surge current through an HEI coil primary as the module begins to conduct (before the inductive reactance in the primary windings builds up) is about 15 amps- - - -approximately 3 times what a set of contact points can handle. That's also why a heavy wire from the ignition switch to the coil is needed. The transistor amplifier can also handle the surge voltage as the magnetic field collapses better that points can, so there's no need for a capacitor to shunt the surge away from a mechanical switch. The "condenser" in a HEI distributor is just for radio frequency damping- - - -similar to the condenser that's found on some DC generator armature terminals.
Jerry

Hi Jerry,

Yeah.... I knew that primary current on HEI systems was considerably more than breaker point systems. My poor tired brain did not remember it being that high, but it makes perfect sense - considering how much larger the distributor feed wire is on HEI.

This can have ramifications, however. Back in 86 or 87, my 85 Corvette was recalled to replace the HEI coil and module. My understanding was that under certain conditions, the coil in that integrated distributor design was getting too hot, so they had to throttle back a little on the primary current. I'm sure that at the time, the Corvette was probably worst case, since the reduced air flow through the engine compartment probably led to some fairly high under hood temperatures. Combine this with a distributor tucked in close to the firewall, and a customer that lives in Arizona, and the result was probably a design that was pushed to the limit.

I personally always liked that integrated design. I thought it was clean, self-contained, and did not have a coil wire. However, I always got the feeling that the ignition system guys preferred the eventual evolution that utilized the separate coil, which provided much better cooling, and was probably easier to package.

One of the interesting take-aways from that paper is on page 4: "To answer this question, you have to appreciate that the expression for the induced emf tells us it is not possible to switch off the current instantaneously: as Δt → 0, emf → ∞! What happens in practice is that, when the voltage gets high enough, the air breaks down (ionizes) in the strong electric field near the point where the current is interrupted, and a spark is produced "

Although they do not explicitly include it, they are obviously referring to V = L (di/dt). So, as the author states, as the value of t goes down, the value of v goes up - to the point where it sufficient to ionize the point gap, and act just like a spark plug.

As you mentioned, the switching circuitry in the HEI module must be designed to handle the "V" caused by the coil being switched off. My only question still, is that according to the author of the paper, that capacitor is needed to provide a resonant RLC circuit in order to boost the coil output. I'm wondering if this is perhaps not as important of a factor as the author maintains. (You know those academics…. ).

I too, have seen those RF suppression capacitors on vehicles as well. I used to see them sometimes on alternators, or the old electro-mechanical voltage regulators. Some vehicles even used them on the brake light switch. The inherent inductance in the brake light bulbs (coiled filaments) would cause enough surge when the pedal was released to cause an audible "switch pop" in the radio speakers.

My truck, which is a former military vehicle, is loaded with suppression capacitors. They are everywhere, along with star washers on the front sheet metal bolts, and extra ground straps to help ensure ground integrity. In fact, if you look at the registration number on WWII vehicles (the number typically seen on the hood) the "-S" suffix stands for "suppressed". The Army was obviously very concerned about radiated emissions.

Lastly, if you are trying to determine total primary circuit resistance (wires, ignition switch, connectors, etc.) this measurement is not difficult to make. Just detach the ignition wire at the coil, and insert an amp meter and power resistor in series to ground (you could even use the coil, I guess). Activate the circuit and measure the current, along with the voltage drop across the entire circuit (or individual parts, if you are looking to isolate certain sections). Using ohm's law, you can then calculate the circuit resistance using the measured voltage and current. We used to do this all the time for running cranking system control circuit resistance tests (aka the starter solenoid feed circuit).

Best Regards....

Unfortunately, an ignition system is not a static environment. Total resistance in the circuit changes from moment to moment, depending on the state of development (or collapse) of the magnetic field in the coil, the inductive reactance of the coil windings, and the state of charge or discharge of the condenser. While the ignition system might be technically called a DC circuit, there's a huge AC component involved which varies continually and affects the overall current flow. My brother gets paid a thousand dollars a day to run computer modeling simulations of proposed new electrical parts for automotive applications, and the manufacturers who purchase his consulting services are glad to pay his fees. He's got a master's degree in electrical engineering, and a doctorate in plasma physics. He was one of about half a dozen engineers worldwide who were working on computer control programming for the fusion reactor program a few years ago. That gives him the math background to figure out all those variables with some degree of authority for his findings. BTW, he's also an ASE certified Master Automotive Technician!
Jerry

Correct. The total circuit impedance is determined by the total resistance, the inductive reactance of the coil, and capacitive reactance of the capacitor.

For any non-electrical types, impedance is essentially a fancy term for the opposition to current flow in an AC circuit, where the effects of capacitance and inductance (capacitive reactance & inductive reactance) are present. This is why test equipment , like a multimeter or oscilloscope, might call out a spec for "input impedance" as opposed to "input resistance", since they usually provide AC measurements as well as DC ones.

Inductive reactance is represented by the equation XL = 2*pi*f*L where XL = the inductive reactance (in ohms), f = frequency, and L = inductance. Capacitive reactance is represented by the equation XC = 1/(2*pi*f*C), where XC is the capacitive reactance (in ohms), f = frequency and C = capacitance.

Before everyone reading this starts to nod off (unless they already have), the main point here is the location of the "f". Note that for inductive reactance, as the frequency goes up, so does the impedance. In contrast, for capacitive reactance, as the frequency goes up, the impedance goes down.

Electrical components contain elements of each to varying degrees. For example coils have often have a fair amount of resistance in addition to inductive reactance. Other times, one element is so dominant, that the others are considered insignificant. For example a switch or wire has capacitance, but the value is so small that it's effects are probably negligible for applications such as this, leaving the resistive element as the primary factor in determining circuit impedance. There was a question in an earlier post concerning the calculation of primary circuit resistance (wires, switches, grounds, etc), so this is what my suggestion was trying to address.

When the points close, you really only have to deal with the coil and related wiring (RL circuit) - since the capacitor is essentially shorted out. The voltage transition as the points close looks like a relatively high frequency input to the coil. Since the coil is an inductor, if the f is high, so is it's inductive reactance, so it is essentially trying to resist a change in current. This is why if you ever see a plot of current vs. time for a coil charging, it starts out low, and gradually builds up until finally leveling off. The rate of the build up is a function of the inductive reactance, while the steady state after that is due to the resistance in the circuit.

Granted, once you throw both the coil and capacitor (aka condenser) into the mix, things get a little more complicated. Once the points open, you now have an RLC circuit. The capacitor sees the rapid voltage transition caused by the field in the coil collapsing as a high frequency input. The "f" in the capacitive reactance formula is on the bottom, so high frequency = low impedance. The capacitor temporarily acts almost like a short circuit, keeping the voltage potential across the points low enough to suppress any arcing.

At the risk of beating a horse that is not only dead, but hauled away and sold to the glue factory, the question I still have concerns this business about the capacitor forming a tuned resonant circuit, which supposedly increases the primary voltage thus increasing the coil output. I see it in some documents, while it is totally ignored by others. I am just curious as to what real effects it has.

Jerry, your brother sounds like a really sharp guy!! (Must run in the family) Maybe he could create a Saber, or Matlab model of a breaker point ignition system and let us know what the real answer is.

Lastly, in case anyone is interested, I happened to stumble across this site which has a high level schematic of an HEI ignition module. It appears that they use a Zener diode to suppress the peak coil primary voltage.

HEI Info


Best Regards...

Hi Bob,
Interesting that article was written in 2003. Much has changed since then even though 2003 doesn't seem that long ago. Yes, a zener diode was used to suppress peak voltage in the primary it seems. Could have been a couple of reasons I can think of for that. The author indicated he would follow up with more info but evidently didn't.

As for resonance, don't you think they'd be talking about the resonance of the RLC circuit? When the points are closed, the primary is part of an RL circuit, yes? Then when the points open, an RLC circuit is created. From there I think it has to do with running the secondary at the proper resonance to make the most of the coil's energy...and the only see-saw comparison I can think of quickly here is tuning the idle ratio of fuel/air in a carburetor. Because the inductance is fixed (the air intake in the carburetor) and the capacitance can be changed (the idle richness screw), the capacitor becomes the regulator of that resonance. Maybe the use of the word "tuned" means nothing more than choosing the right value to get the most "oomph" out of the coil.

Coils can require a lot of thought no matter where they are. You will recall I'm sure the trouble I had with the intermittent windshield wiper circuit (relay bounce) and the capacitor I ended up using to solve that and finally make it work correctly.

Those of us who grew up looking at oscilloscope patterns can recognize an original equipment GM HEI ignition module just from looking at the primary waveform on the scope. When the coil reaches full magnetic saturation, the module kicks in a "current limiter" that creates a step in the horizontal scope sweep partway to the equivalent of the "points open" part of the sweep where the coil is supposed to fire. Since that limiter incorporates some fancy (translation: E$$$PENSIVE) electronics, most aftermarket module manufacturers don't bother to include it in their products. Original equipment Delco-Remy parts are good. I don't like the recently-made stuff that's labeled "AC-Delco"- - - -guess where it's made?
Jerry

I know that the condenser keep the points from burning and through some recent troubleshooting I learned that my international cub will not start without out it. I was surprised by that so I removed the condenser from the 250 in my 66 pick up, same thing will not start. So what all does the condenser do?

Possibly the points were arcing as they opened, causing the current thru the coil to slowly drop instead of going to zero nearly immediately. In order for the coil to generate the high voltage to fire the plugs, the current and magnetic field needs to collapse quickly. Most likely there wasn't enough secondary voltage to generate a hot spark for starting.

I believe the idea about the points arcing without the condenser in the circuit has some merit. Just as the points break open from the action of the distributor cam, the collapsing magnetic field in the coil induces a voltage spike of 100 volts or so in the primary (low voltage) winding of the coil where either 6 or 12 volts was present a fraction of a second before. As the condenser charges with that voltage spike, the coil winding sees that as a continuation of current flow and the points are protected from flashover as they're opening up and increasing their resistance. Since electric current flows at approximately the speed of light (186,000 miles per second) the condenser gets its maximum charge in a few milliseconds. Then the high voltage goes racing backwards through the primary winding, effectively slamming the door on any further current flow. That causes an instantaneous collapse of the rest of the magnetism and a spike of several thousand volts is induced into the secondary winding to fire the spark plug. Without the condenser opposing the primary current flow, the magnetic field doesn't collapse quickly enough to develop that really big voltage pulse.
Jerry

Ditto that the condenser protects the points. I hope Jerry doesn't mind me modifying what he posted with my own twist.

The points close. The condenser charges up to near battery voltage and at the same time current flows (ramps up) through the coil primary winding and creates a field. Next the points open but the condenser (temporary battery) discharges and continues the primary field current flow for a extremely short time. This allows the points to safely open (minimal arching). The condenser has now discharged. The current flow through the primary winding stops. The primary field flips polarity (the magnetic flux stops but opposing EMF remains). This causes an "inductive kick" to the coil secondary winding which spikes to high voltage on the coil output to fire the spark plugs. The coil primary current now flows in the opposite direction through the capacitor to ground. The field in the primary winding has now completely collapsed and the cycle is ready to be repeated.

If the condenser is missing or open, there is no current flow path that allows the primary field to quickly collapse. That's why the engine doesn't start/run.

Interesting idea- - - -except that the condenser charges and discharges repeatedly- - - -20 or more times- - - -while the points are in the process of opening and closing. Each cycle has a slightly lower voltage peak due to circuit resistance, and by the time the points close again, there is no high voltage left in the primary winding to cause a "points close" arc which would be just as destructive as the one that can happen on "points open" without the condenser in place.
Jerry

There is no current flow through a capacitor.

While a capacitor (condenser) is charging, it appears to flow current. As it approaches full charge the "current" flow tapers off, leaving the condenser charged with much higher voltage than the battery. At this point, the condenser discharges back through the battery and ground circuit, and charges itself from the ground side, losing a little voltage along the way. Full charge- - -discharge- - - -charge from the insulated side, etc. until the voltage dissipates entirely. Ideally, this happens before the points can close again to prevent an arc discharge as the points make contact. With the points closed, the condenser is shorted out, with both sides at the same potential- - - -ground. It's a simple circuit that becomes complex when people try to describe it in terms of a straight resistive situation without induction and capacitance involved.
Jerry

I need pictures. I've never looked at primary voltage and current with an oscilloscope. I copied the attachment off the Internet and added arrows pointing to discussion points A, B, C, D and E. First, will someone please verify that this is suitable waveform for a points distributor and not the HEI or Hi-Tech stuff of today's modern engine?

Questions:

A = points closing?
B = current ramp?
C = points open?
D = inductive voltage spike?
E = negative current (coil primary field reversed polarity)?

Are condensers polarity sensitive? A friend has an old 1951 MGTD that is a 12 volt, positive ground ignition system. Would one of your condensers work in his system?

No, since the polarity changes several times a second as the condenser charges/discharges repeatedly while the points are opening and closing. The diminishing-amplitude sine wave that's visible on a scope pattern represents the charge and discharge cycle of the condenser that protects the points from burning. Jon's improved capacitor should work well on either coil polarity.
Jerry

Thank you for the information. I will tell him as he has been trying to get some information on the condenser for quite awhile. I guess you just have to find the right forum and it is not always the one about your particular make or model. Thank you.