Yes.
Ballast Resistors
- Thread starter 92b
- Start date
-
You are technically correct. However, my point is that DC meters are designed to measure constant voltages and currents. The ignition voltage and current is not constant, so the readings are not accurate in the way the meter was designed to be used. They are averages of the time varying current and voltage. The readings produced are based on the design of the meter for a digital meter, and can differ from meter to meter. That is why I used the word loosely.
Looking at ignition waveforms on the net, I wouldn’t characterize them as square waves, however.
AC meters are designed for 60 cycles, so that is not an option either.
The experiment, based on the average values seems, to be a good approximation as long as we understand that we dealing with average values. I am just drawing the distinction that the peak currents are likely in the ignition system is higher than the average current measured.
MDchanic
Connoisseur d'Junque
Not disagreeing, but I think the important point can be lost among the details.
In this case, we are absolutely not concerned about any instantaneous voltage at any time.
Partial duty cycles and variable wavefroms of varying shapes are all fine.
Why do we have a ballast resistor, and what are we trying to accomplish?
The ballast resistor is there to reduce current flow under conditions of high demand, to reduce the chance of the coil overheating, and also to act as part of the LRC circuit involving the ballast resistor, the coil, and the condenser, which provides its own frequency-based impedance to current flow.
The coil is drawing current a certain amount of the time, as described by the dwell angle, which is really just another way to describe its duty cycle.
For the six, the dwell angle is about 40°, out of 60° of distributor rotation per cylinder.
For the eight, the dwell angle is about 30°, out of 45° of distributor rotation per cylinder.
For both of these, the dwell angle (number of degrees the points are closed) is 2/3 of each timing interval, so it represents a 66.6% duty cycle.
Both coils are spec'd to draw 3A switched on constantly, 1.9A in use at idle.
Coincidentally, 2A (1.9A) is 2/3 of 3A.
Now, when you run a bunch of electric current through a whole lot of thin wires, they heat up.
There are formulas for exactly how much they heat up, depending on the diameter of the wire, its exact composition, and electromagnetic forces that may result from, say, being all wound up in a tight coil. All of this is for straight DC, completely disregarding frequency-related impedance effects.
So, if you run a whole bunch of current through a whole lot of wire (there's what? A hundred or more feet of wire in a coil, right), and that wire isn't laid out in the open, like strung along telephone poles, but instead is all wrapped up tightly in a can, the heat generated by that current has a hard time escaping, and the coil will heat up.
The longer you apply the current, the more the coil will heat up.
If you interrupt the current, the heat has a chance to dissipate a bit, but if you leave it on steadily, it will build up more.
So, at idle, with the coil "on" for longer stretches between sparks, the coil will heat up more, even though it does have longer periods to cool off in between. At higher RPMS, there is less time for heat to really saturate the coil, so it heats up less. Also at higher RPMs, you are likely to have greater impedance, so that will further restrict current flow and heating (I don't know the inductance of the coil, so I can't calculate just where the impedance is lower or higher in this circuit).
So, if the ballast resistor heats up more with increased current flow through the coil, and by heating up, its resistance increases, it restricts current flow through the coil at a time when the coil is only too happy to pass current and overheat.
If the ballast resistor cools down at higher RPMs, partly due to the increased impedance due to the increased frequency, it will allow more current to flow, helping to overcome the increased impedance and keep enough current flowing to run the engine.
What's my point? Actually, I forgot.
No I didn't. It is that what you want to do when measuring the resistance, current, and voltage here is not to measure the voltage at an instant in time during a long series of high-frequency oscillations (at 4,000 RPM in a \6, you are firing 200 times a second or 200Hz), but to capture the overall averaged values, which are what would be heating (or overheating) the coil, points, supply wires, etc. You can run a huge amount of current through a small wire if you do it briefly enough (there are formulas for this kind of thing). So long as you don't exceed the time required to soften the insulation or melt the wire, you're fine.
So in this case, where you're trying to prevent the coil from overheating, it's the big, mushy amount of current flowing that matters, not the little variations, whose effects will create little bits of heat which will be absorbed.
Analogue meters will serve this purpose. As Mike said, digital meters are running some sort of basic (not necessarily digital) programs that are likely reading the measured values at fairly high frequencies and then averaging them out according to certain rules, which may or may not provide a true dead average of the energy passing through, but an analogue instrument will do this automatically.
There will be a short quiz.
– Eric
Discussions like this usually seem to go down a lot of rabbit holes stray away from the original purpose (frequent habit of mine), but I find them very interesting and I usually learn something new every time.
I really like the ones when folks do experiments and get real data to talk about, like this one.
Great job by the OP!
I really like the ones when folks do experiments and get real data to talk about, like this one.
Great job by the OP!
MDchanic
Connoisseur d'Junque
That was VERY WELL explained.
In my brain I understood every word but no way in hell I could have explained it that well!
There was another reason for a ballast resister.
Say you leave the ignition switch in run and you have a point ignition and the engine stopped with the points closed.
Now you have a heater just burning energy and converting it to heat. The ballast will get very hot and it's resistance will change more and help keep the coil from over heating.
See I bet you can explain that much better!
MDchanic
Connoisseur d'Junque
Thank you! I'm glad it was helpful. I'm not an expert, so am always grateful for any corrections others might have.
Well, that's kind of like the slow-RPM case. Lots of On-time without a chance for the heat to dissipate. And I would bet that the ballast resistor does get a bit warmer and resist a bit more, but, as we all know (but some know better than others), even with the ballast resistor, if you leave the key on for too long with the points closed, you'll overheat your coil or burn up your points. I've been lucky enough not to know this from personal experience.
If we're messing around with ballast resistors, it might be interesting to extend the resistance / temperature test to a constant-On condition.
According to the manual, the coil draws 3A, so, at 12.5V, that would make it about 4.2Ω, so one would need to set up a static 4Ω resistance that could withstand 3A, or use a known-crappy coil that was already going in the trash, then leave it on for about 15 minutes and see how hot the resistor got and what its resistance was.
But I'm not doing it. Got a whole lot of other things to do...
– Eric
Anyone have an explanation of why the average current goes down at higher RPM? Seems counterintuitive. You would think that with the coil firing more times the average power consumption would go up.
Two things I can think of:
Is it a measurement issue where the meters are not catching all the current?
Is it real and the on time of the coil prior to flyback shorter at high RPM?
Maybe something else?
Two things I can think of:
Is it a measurement issue where the meters are not catching all the current?
Is it real and the on time of the coil prior to flyback shorter at high RPM?
Maybe something else?
Just a reminder. This is the ballast resistor that decreased in resistance when tested with an ohm meter and a heat gun. I still want to "live" test the ballast that under heat gun /ohm meter conditions increased in resistance as it was heated. Just to see if there is a difference. I'm also trying to round up an analog meter to compare to the digital meters I am using.
MDchanic
Connoisseur d'Junque
The behavior of coils (inductive elements) at high frequencies is kind of its own science, with lots of fun calculus (which I don't understand, Sorry, Mrs. Abramson), but as frequency increases, reactance (resistance to flow) increases.
Figuring the coil's inductance at about 5mH (I can't find a specific spec for the specific coil in question), it probably comes to 4-8Ω at 277Hz, the firing rate for a V8 at 4,000 RPM.
Also, the skin effect means the center of the winding wires conducts less and the surface more as frequency increases, which makes the wire more resistive, as though there was only the outside of the wire and not the center At 277Hz, it equals 124µM, which is to say 0.124mm or 0.005", which, if doubled, means that most, if not all, of the diameter of the roughly 0.010" wire is the "skin," so it may not really figure here.
Also, the proximity effect describes the impedance created when you have wire wrapped in layers. The changing magnetic fields of each layer affects the others, making the coil more resistant at higher frequencies the more windings it has. From what I understand, it can easily multiply the regular DC resistance by 10.
So, the regular DC resistance of the coil of about 4Ω, plus another 4-8Ω from inductance, plus an unknown amount for skin effect, plus multiplying it from 5 to 10 to maybe even more times for proximity effect, you get maybe 1,000Ω at 4,000 RPM, which would substantially reduce current flow (and my common sense says the increase in resistance is less than that, but you get the idea – it's something).
I'm sure a member who is an electrical engineer, and actually understands it, could explain it better.
– Eric
I think the frequency is way too low for skin effect and other RF effects. Generally skin effect and other RF phenomena become significant in the megahertz range. However, the change in frequency could certainly affect current. But a factor of 4 isn’t a lot of change in frequency.
Many years ago I could do the math, but it has been a very long time….
The nominal inductance of an old style ignition coil is approximately 100 mH.
Many years ago I could do the math, but it has been a very long time….
The nominal inductance of an old style ignition coil is approximately 100 mH.
MDchanic
Connoisseur d'Junque
If they're exclusively RF effects, then, yes.
I tried to get some numbers at the frequencies we're talking about, but those numbers may be wrong.
And if the skin effect at those frequencies is 0.005", and the wire is 0.010" (just a guess there), then the wire is all skin, and there's no effect.
But there's definitely an inductive effect.
It looked like it was around 5 when I tried to find it, but, as I said, I could find nothing specific for the coil types we use. It'd be great to find an accurate spec. (or someone with an inductance meter).
– Eric
I think I misspoke on the inductance. I don’t remember it. The stored energy in the coil is 100 milli joules. Getting my Henry’s and joules mixed up, been a long time…..
MDchanic
Connoisseur d'Junque
This is getting interesting. Wish I was not jammed up for Carlisle prep!
-
