MTFCA Forum

I've been studying Heinze coil operation and performance using the ECCT and have come to a very surprising conclusion. The Heinze coil is actually a superior coil design compared to the K-W coil.

The Heinze coil is significantly easier to adjust and consistently out performs the K-W Coil. Coil firing consistency is typically better than 90% and is maintained to firing rates well above 3000 RPM. I believe the Heinze coil got it's notoriously bad reputation for being difficult to adjust for good engine performance due to the limitations of the coil test tools and method back in the day; not the Heinze coil or it's 2 element point design. The HCCT or other coil test tools that measure coil current as an indirect approximation of firing time are simply incapable of measuring the coil dwell time to fire the first spark. The displayed coil current is the average of ALL the sparks fired (including any double sparks) which makes average coil current a poor coil adjustment metric and typically results in poor engine performance.

Yes, the Heinze coil 2 element point design is vulnerable to double sparking but if the first spark occurs at the proper time, of the proper energy and does so consistently, does it really matter if there is a 2nd spark? NO! Ever hear of Multiple Spark Discharge (MSD)? I believe the Heinze coil will be vindicated as a superior coil design as more folks experience engine performance with their Heinze coils properly adjusted for equal and consistent firing Time rather than equal coil current.

Special Heinze Edition of the ECCT specifically designed to test and adjust Heinze coils:
Heinze coil firing time is very easy to adjust; simply turn the vibrator spring adjusting knob! Heinze coil firing consistency remains Excellent at firing rates well above 3000 RPM and often of up to 5000 RPM as shown below without having to fiddle with a Cushion spring tension; trying to find its sweet spot.

Mike,
I am not willing to go in discution with you.
You know much more about electronics as I will ever do but I know a lot of ignitions.

I can read the tension during the test is 12.435V, is this Dc or Ac ? At 12V about every coil work well.

The Heinze coils you show are starting to work at 1.2Vac.
This is needed when handcranking the car on the Magneto.
Did you try this? What are the results at this low tension?

Just a few questions I had while reading your post.
Andre
Belgium

My experience running heinze coils has been very good. I ran them for years in my 1911 torpedo and never opened the box. They simply ran well, started great and never needed attention. Perhaps the reputation largely stems from owners constant fiddling with coils during the active life of these cars. The old adage that 9 out of ten carburetor problems are electrical might be extended to 9 out of ten coil problems are in the timer! Note that in 1911 kingston was so desperate to keep car owners from tinkering with their coils that they put a metal cover over the points for awhile that had only a shunt button available to the owner to fiddle with. Ford added to the poor reputation by encouraging owners to turn in their non-kw coils when they standardized the design in favor of KW during 1913. The KW design is certainly sound and has been shown to be so through the excellent technical articles put forth by Regan, Patterson and others. However this does not mean that the other designs were not functionally satisfactory. I have found them to be very good in service during my years driving with them. The best coil you will ever own is the fifth one under your back seat because if you have it, you will likely never need to use it!

Sheri,
I agree with you. The last 6 years, I rebuild over 50 Heinze and Kingston coils. They are all still in the running without any problem.
Very often the high tension coil is bad, I use the Ford coil or the replacement some vendors offer. I always change the points.
The only thing I say to the owners of the coils I rebuild, is that the new contact point need a run in. They need to be reset after about 100 miles. After that no reseting is needed. They run as good as the others.
Don't forget that over 600,000 Model T's used them during the first years. If they were bad, they shouldn't stay that long.

Andre
Belgium

Great article! I am interested in getting one of these custom units!

Some questions: will the ECCT function okay with both coils that retain their original windings as well as ones that have been replaced with the regular windings the vendors currently sell or original windings out of regular style later coils? I remember RV Anderson saying the point current draw on originals and the replicas he built was 0.8 amps as opposed to 1.3 on the regular windings. I find I set my rebuilt coils at 1.0 amps for optimum sparks with my Strobospark. I realize your tester works different so I humbly admit my electrical knowledge is just enough to make me dangerous! Lol! Looking forward to trying out an ECCT for these in the future! Thanks!

Darren, they draw .8A because that is what they are set to draw. There are differences between original and new windings, but those differences are not fundamental to the .8A setting. This setting is consistent with the original spec'd setting of these or these earlier type of coils (no cushion spring). Very low amperage draw leads to easier successful starts when cranking by hand. The higher amperage draw setting, the fiercer the pull must be on the hand crank to achieve rotational speed to create a spark when starting on magneto.

As an aside:
It is the cushion spring tension, travel, etc., which combines with the rest of "modern style" points to often sound remarkably different to each other even when adjusted for the same amperage. Earlier coils were far simpler in point design, with fewer influences into how they sound and is the basis of "tuning by ear" that so many people attempt to do on 1914 and later coils (with very mixed results). All of the early references in Dykes and others to "tune by ear" to an equal pitch were written at the time and in the context of Heinze and similar coils (which very frequently sounded similar when correctly adjusted). Those same admonishments from Dykes, et al, do not carry over into later editions of the manual, when speaking to more modern coils and once those coils (with points cushion) were in production, all such suggestions and advice to "tune by ear" vanished from texts. It is unfortunate that some few folks fail to recognize the difference and cling to outdated technical advice for a coil for whom the advice was never envisioned, much less intended.

Well said Scott, Top of the class, One jelly bean for you....
I would love to see the Scope trace of the HEINZE coil compered with a post 1914 style coil.....n

Good to see this post inspiring thought and dialog. Here is some more to keep it going.

12.435V, is this Dc or Ac ? At 12V about every coil work well.

12.435V is the DC voltage powering the tester. The stimulus applied to the coil under test is derived from this DC power but is NOT a steady, fixed voltage DC. The test stimulus is a very fast pulse of electricity which emulates the magneto pulse at very high engine speed where ignition timing is most critical.

A common fallacy is: The Model T magneto puts out an AC voltage that operates the coils, therefore, coils should be adjusted using an AC voltage like that produced by a HCCT." Here are a few facts why this reasoning is flawed:

1. The magneto pulse that fires a Model T coil does NOT alternate polarity during firing. It will be either a positive pulse or a negative pulse depending upon spark lever setting. Accordingly, the coil current does not alternate between positive and negative when firing a particular spark. Hence, alternating current does not occur during coil firing.
2. The HCCT contains no timer function to permit the coil to rest in between firings and return to steady state. The coil is subjected to a continuous succession of positive and negative voltage pulses from the magneto during coil adjustment but will never occur during normal operation.
3. The HCCT is hand cranked at abnormally slow engine speeds (60-120 RPM) in which the car will never operate. The resulting magneto voltage rises abnormally slowly to an abnormally low in voltage value is used to adjust coils but the coils will never operate during normal use.

Very low amperage draw leads to easier successful starts when cranking by hand. The higher amperage draw setting, the fiercer the pull must be on the hand crank to achieve rotational speed to create a spark when starting on magneto.

This makes perfect sense but, is it true? Does anyone have hard data on this? Its something I have often pondered but have yet to actually test. I suspect this common reasoning is also a fallacy and here is why with the help of some math rather than just conjecture:
1. Spark energy depends upon the amount of coil current flowing in its primary winding at the moment the points open to generate spark (note this is NOT the average coil current read on the HCCT meter; typical coil firing current is ~6A). The spark energy can be approximated by the following equation:
Where E=Spark Energy, L=coil primary inductance and I = coil current flowing at the time the points open.

The lower the current flowing in the coil primary at the time spark fires, the lower the spark energy. Note that the current term, I, is squared (I^2) which means spark energy decreases significantly as the points are adjusted to open at lower firing current. The lower the spark energy, the more difficult it may be to start the engine due to insufficient spark energy to reliably ignite the air/fuel mixture (ie more vulnerable to partial or incomplete combustion; misfire).

2. What coil current can we generate via brisk hand crank? The magneto voltage output is a triangular wave (not sine wave as is another common misconception). The resulting coil current rises exponentially with the linearly rising triangular magneto voltage as the crank is pulled swiftly. The coil current can be calculated by the formula: Where L is the coil primary inductance, Vcoil is the stimulus voltage applied to the coil and dt is the change in time.

So to calculate the value of coil current, we need to know have fast the applied coil voltage rises with time (VL dt) term. Lets assume we can pull the crank a 1/4 turn in ~30ms (That's equivalent to completing 1 full rotation in 1 second or 60 in 1 minute; 60 RPM. Thus, 8 magneto voltage cycles produced per in 1 second revolution or 1/8 = 0.125 seconds to complete 1 complete magneto voltage cycle that of two voltage pulses (one positive and one negative) and lets say the voltage pulse is a relatively weak 2Volts cranked that slow. Here is what that would look like in time
so the time it takes to produce a single magneto voltage pulse to rise 0 to 2V takes 1/4 of that time (1/4 crank pull) or 0.125/4 = 0.03125 seconds for the magneto to produce a weak 2V voltage pulse (see dashed red box) . So then the magneto voltage (VL) rises from 0V to 2V in 0.03125s so the linear slope (m) of that line is 2/0.03125 = 64. The voltage applied to the coil can be written as: Where m is the slope (rate of rise) and t is time.

Plugging that back into our equation for coil current equation and solving: 9.47A is more than enough coil current to pull open the coil points to fire spark. A properly adjusted Model T coil firing current is about 6A on the ECCT so should have no problem starting an engine via hand crank.

Not sure what the correlation is between ease in engine hand crank starting and what the average HCCT coil current reading (0.8A vs 1A vs 1.3A vs 1.5A) measured while hundreds of sparks are firing continuously. There are so many variables involved with the average coil current reading on an HCCT, I don't know that you can correlate a solid relationship between average HCCT coil current reading and ease in engine starting. You would have to know the correlation between average coil current measured and the coil current flowing in the coil primary at the moment the coil points open to produce spark. That is the coil current that really matters.

Thanks Mike for the very informative info! For those of us who are a little on the dumb side like me, will your tester work reliably with the Heinze coils with original and or newly installed windings? Or so you have to pre set parameters in the tester to accommodate the windings? Thanks for lowering your expectations with rubes like me!

Darren, the Heinze version does have different Pass/Fail criteria due to the Heinze coil properties being different from the KW coil parameters. I just started to accumulate feed back from a few folks who I have been working with. So far, the feedback has been excellent providing motivation to proceed with making a separate tester specifically designed for Heinze coils. Time will tell if my suspensions about the Heinze coil are accurate as more folks experience engine performance with their coils adjusted for equal and consistent firing time as opposed to average coil current.

RV would have been the one to add his comments to this post. As I recall, he was the living expert on Heinze coils. Until he starting making points, you couldn't get them anywhere.

I was in contact with RV just before Christmas to share my findings with him. He was looking for a few good Heinze coils to put together for me to buy and test. RV used an HCCT to adjust coils via equal average current and was very interested in my preliminary findings. I had hoped to collaborate with RV further, he is sorely missed. Hopefully, someone else will take on the challenge of making Heinze coils and replacement points.

Got them on my '13. I'm not "up" on the coil stuff, but I do know I like 'em on my car. Run great, and quiet also.

Mike,
Your post is quite interesting. Some pre-T’s used both Splitdorf as well as Heinz coils (Model K in particular). Any idea how the two compared? Both ran on 6 volt batteries (K had independent magneto and battery/coil dual ignition systems). Other pre-T models used both type of coils, with 6 volt batteries.
Thank you for your post and research,
Rob

Rob, thanks for information the inquiry. Unfortunately, I don't have any information or data on the Splitdorf coils so can't provide any performance data comparison.