A lucky eBay find: NOS SWB Coil

[304065]

After years of scrounging for good electrical parts for my '66, I finally found an original '006 coil.

The primary resistance is 2.1 ohms. The secondary is 9.7K ohms.

I measured the turns ratio using 100Hz sine wave at 86:1. With a 200V primary voltage this coil is good for 18Kv. I have not measured the primary voltage, however. (That's very dangerous!)

Of course, on the primary side these were used with a ballast resistor for normal operation. Even though I measured 1.3 ohms in one test, the nominal value is 0.9 ohms I am pretty sure. So that gets you three ohms during normal operation. The coil charges through this resistor so it's an RL circuit, not simply possible to do a simple calc without knowing the coil inductance.

In the days before electronic ignition, more than about four or five amps would burn up the points in no time, even with the capacitor present. So the points were the rate-limiting factor on how low the coil resistance could be (and how fast you could charge the coil).

With modern solid-state ignition modules, lower-resistance coils could be used and much higher current used to charge them.

I always wondered about the evolution of the ignition system-- from Kettering to 3 pin CDI (points triggered) to 6 pin CDI (reluctor triggered) then back to a low resistance coil that was triggered directly by the DME. Then in the 964 and 993, external Bosch ignition modules were used. If you look at what Porsche/Bosch were doing with the water cooled cars of the period, that's also a useful guide (however, the 928 used two coils).

Anyway, Porsche didn't have an issue using a single coil all the way through redline once there were modern igniters to switch it. I think that's pretty interesting, that they discarded capacitive discharge ignition in favor of inductive.

Contrast this SWB coil resistance with the later CDI coil, which is basically a step-up transformer (Bosch Nr. 0 221 121 001). The CDI coil doesn't store ignition energy between pulses-- the ignition energy is stored in the capacitor, and that gets dumped into the coil when the CDI fires. The CDI coil primary resistance is between .4 and .6 and 650 to 790 ohms on the secondary side, according to Bosch. It's a smaller coil than the ones used before, or after, because it's being hit with a 460V pulse from the CDI, hence the turns ratio is probably lower than an inductive coil, with a lower secondary resistance as a result. This CDI coil was used unchanged from 1969 on the 911E through 1983 on the SC.

In 1984 with the advent of Motronic, the coil changed to part No. 944 602 115 00, Bosch No. 0 221 118 322. That coil had a primary resistance of between 0.4 and 0.7 ohms and a secondary resistance between 5.0K and 8.7K ohms. That same number was also used on the 924S, 944 (all versions including 968), the 964 and the 928 from 85 to 88. The inductance of this coil is claimed to be 3.6mH.

The 993 Coil (a dual pack that mounted on the engine mount crossbar) has a primary resistance of 0.58 ohms and a secondary resistance of 6.92 ohms (thanks Bazar1 from Leesburg, GA!)

See how the primary resistance dropped after the advent of electronic ignition?

Stay tuned. . .

[preS]

Nice find John!

Any chance you can reproduce the sticker for us?

Richard

[304065]

Richard, how can this be improved upon?

[preS]

Richard, how can this be improved upon?

The letter spacing is not correct.

For instance the 'n' should be more to the left. The 't' should have the little 'thingy' on the bottom, etc.

Richard

[304065]

I tested a few more coils for comparison purposes. Use these data at your own risk of death, property damage or personal injury. No kidding there Tesla, you can kill yourself or burn the house down if you don't treat high voltage with respect.

These are measured specifications with an LCR meter and an ohmmeter. Your results may vary.

Bosch "Black" coil 0 221 102 006 8.5mH, pri res 2.1 ohms. Original SWB coil as above.
Bosch "Red" coil 0 221 119 030 8.2mH, pri res 1.3 ohms. Requires external ballast resistor.
Bosch "Blue" coil 0 221 119 027 12.8mH, pri res 3.2 ohms. This has an internal resistor.
Bosch "CDI" coil 0 221 121 001 1.8mH, pri res 0.2 ohms. This is for Bosch CDI only.
Bosch "Motronic" coil 0 221 118 322, 4.5mH, pri res 0.4 ohms. Original on Motronic cars.
MSD "High Vibration Blaster" coil 8222, 8.0mH, pri res 0.7 ohms, sec res 4.5k ohms, turns 100:1

As you can see there is a huge difference in Coil specifications over the years.

A few observations.

Points had a max current limit of about four amps. With a +12V nominal voltage at the coil, that meant that three ohms of resistance was necessary in the ignition circuit to keep the points from burning up. Hence the SWB 006 coil was used with an external resistor for normal operation, which resistor was bypassed during cranking when the voltage was lower anyway.

Same thing applies to the Red coil. This must be used with a 1.8 ohm external ballast resistor for points operation. That gets you 3.1 ohms, see above. For electronic ignition it's a different story, see below.

The famous "Blue" coil teaches us a lot. It has both an internal ballast resistor and a very high inductance. This will produce more ignition energy, but this coil's output will fall off dramatically as RPM increases. This coil would be fine for four-cylinder, low RPM motors such as the VW beetle, in which community the Blue coil is frequently found.

The CDI coil has very little inductance because in the Bosch CDI system, the coil does not store ignition energy, it's stored in a capacitor, which discharges through the coil. So the coil's purpose is to serve as a transformer only to step up the voltage from the 460V output of the CDI.

The Motronic coil is an interesting one. Looking at this, we see the tradeoffs made by Bosch and the Factory to get the right balance between ignition energy and high-rpm performance. Fairly low inductance and very, very low resistance-- this is due to the fact that the coil is switched by a transistor inside the Motronic unit, which can probably switch 10 amps or so. Having an onboard coil driver probably isn't great for the longevity of the control unit, however, which is probably why the Factory and Bosch started using external coil drivers with the 964 and 993.

[304065]

OK so the above data aren't published anywhere as far as I can tell. That's what makes Early S different!

Where it really gets interesting is when you begin to compare the performance of various coils.

You may have read that the big limitation of using a single coil on a high-performance engine is high-RPM performance. This is due to the fact that with a 38 degree dwell period, or a 63% duty cycle during which current is flowing in the coil primary, there simply isn't enough time at high RPMs for the coil to charge, field to break down, spark to jump and then repeat itself-- above a certain RPM, the coil doesn't charge all the way, which dramatically reduces ignition energy.

The way we calculate this (Thanks LorenFB for the guidance) is to assume that the coil eventually charges to 100% of the max amperage after five time constants or so, but that for purposes of dwell vs. ignition energy, we're really concerned with the amperage at the first time constant. (1-(1/e))

The amperage at the first time constant (63% of the maximum approximately) is T = L x I / V where L is inductance, I is amperage, V is voltage. But R= V / I by Ohm's law, so T = L / R.

So let's take the SWB coil example, L = 8.5mH, R= 3 ohms (including the ballast resistor). So T = 8.5/3 or 2.83 ms to charge to 63% of the maximum amperage of 4 amps (12V / 3 ohms), or 2.52 amps.

So if there's more than 2.83 ms in between the end of the closing of the points and the opening of the points, then the coil can charge beyond the first time constant and get closer to the maximum amperage. Remember, however that as the engine speeds up, the time available gets shorter and shorter, and remember further that only 63% of each ignition event is available for coil charging due to the Dwell. From a practical standpoint this means that the ignition energy begins to fall off as RPM begins to rise. This is depicted graphically in the chart below.

In terms of assumptions on the chart, it made sense to use real-world equipment-- so the SWB coil and the Bosch Blue Coil have a maximum of 4A in the ignition circuit. In practicality, only the SWB coil gets to 4A assuming it's being supplied with +12V, as the Blue Coil's internal resistor limits the current that can flow.

For the TCI Coil and the Bosch Red Coil, the assumption is electronic ignition with a solid-state coil driver-- either the Motronic Control Unit or an external driver like a BIM 124 or BIM 137. I assume that the coil driver can supply 10 amps to the coil. Again, given the cutback of the time due to dwell and the time constant, we never quite get to 10 amps.

Does anyone (a Motronic hacker perhaps) have an amperage value for the coil drivers on the Motronic unit? I think they are on the order of 7-10 amps. I know for sure that the BIM 137 module will do 8-10 amps as long as you use a heat sink, and the BIM 137 actually does active dwell control i.e. it will start limiting the current at low RPM or if you leave the ignition on with the engine not running, so you don't burn up the coil.

Turning back to the comparison chart, you can see that Motronic does a pretty good job of maintaining in excess of 30mJ of ignition energy up to approximately the 6600 rpm redline of a 3,2. You would get pretty good results from the Red Coil without a resistor, so long as you had an electronic coil driver that could give it the current, and you would want both coil driver and coil mounted on a nice heat sink.

Perhaps this is the reason the factory moved the coil over to the fan shroud in 1969? To put it in the cooling airflow? That certainly makes sense to me. They moved it back with the Carrera and had it stand vertically against the electrical console/heat sink.

I guess it goes without saying that it was impossible for the Factory to achieve modern performance using points. With the advent of hall-effect triggers driven off the crank, with timing set by the Motronic control unit and with the coils switched by a solid-state driver, at last enough amperage could be put through the coil to get it to charge to a level that was acceptable. These days you would use a hall-effect trigger like a hot-spark or pertronix to trigger the coil driver. I am by no means certain that a pertronix can switch seven amps on its own, but I'm willing to try it.

The enormous limitation of CDI is not its complexity or its cost, although those are both factors: it's the very short spark duration. With inductive ignition you can get a spark that is as long as a microsecond, providing plenty of crank angle for the mixture to ignite. It is the recognition of this limitation of CDI that is behind MSD's "Multiple Sparks" at low RPM.

Any electrical engineers out there have a comment on the above? Thanks in advance for any ideas.

[Bullethead]

Agreed with previous suggestion/critiques, and curious if you're still tinkering with this project... if so, 2 cents from another graphic artist/designer: carefully remove the decal and scan it. Anyone with a little computer skill can clean up the image and the result will be exactly as Bosch produced it.

FWIW, these were likely printed on Mergenthaler presses, so ink, decal material/color and cutting die are your next hurdles for perfection.

That's bunch to go through, so good luck!

[gled49]

I wanted to move this thread to the top. This I know about electricity, +,-, and 500ma is 1/2 amp I think. Great info, trying to absorb it. Between Kettering and CDI, Porsche used a Transistor ignition in limited production cars like the 911R. A system M.B. used as well with a different box but with the same Bosch part number. Loren called it an Inductive Discharge Ignition. Any info/help in understanding how this fit in or how it worked. Gordon

[304065]

Gordon,

Yes, this is the elusive finned black box that some people have discussed. It's basically a transistorized electronic ignition that switches the coil.

Recall from the other thread that the max current limit for points is about four amps-- more than that and they burn up for a handful of reasons. Bosch, and others, figured out that if you use a transistor to switch the current you can switch more amperage, without the issues associated with points "bouncing" (or multiple open-close cycles).

If you use more amperage to charge the coil you can charge it faster. Without going into all the math, the more amperage and the LOWER inductance, the more energy is built up in the coil at the first time constant. You can sort of infer that from the difference between, for example, a Bosch Blue coil and a Motronic coil-- the Blue has high resistance and lots of inductance-- would be great for a slow-revving 4-cylinder VW motor, for example, where there is plenty of dwell time to charge it.

On the other hand, a six-cylinder motor like ours doesn't have that much time to charge the coil, particularly at high RPM. By the time a big coil charges and discharges, the next cylinder has already come around. The practical effect is that the coil energy just doesn't get that high, resulting in weak sparks, right when you need them most: High RPM.

How did the Factory cope? Porsche took the conservative route: in 1969 they retained the points, but only as a trigger for the CDI, and with only 420mA running through the points, they would last a long time. The sacrifice with CDI was the extremely short spark duration-- even though the CDI charges the capacitor and dumps it into the little CDI 001 coil with plenty of time to spare, the sparks are so short that you're at risk of misfires depending on load, mixture, etc. But it generally worked well, and in 1978 the Factory went to a magnetic reluctor triggered six-pin CDI to eliminate the points entirely. This is, by the way, the reason that the CDI coils have such small primary resistance and are so physically small-- the coil doesn't store energy, it's just a transformer to step up the 350V voltage output of the CDI up to ignition level. You actually do not WANT a bunch of resistance, I think it would be BAD if you used a Bosch Blue coil with a CDI, might burn things up.

Coming full circle-- remember that I said if you use more amps you can charge faster. Well, with the advent of transistorized electronic ignition, Porsche gave up on CDI in 1984 when it was integrated into the control unit for Motronic. These can switch on the order of 10 amps (don't quote me on the source). This allowed Bosch to charge up the coil faster and still generate enough ignition energy to reliably light the mixture with fewer emissions-causing misfires.

So that's that's all the box is-- a big transistor with an even bigger heatsink attached. Probably done in the days before CDI as an attempt to switch the coils faster to avoid high-RPM low ignition power.

The M-B folks show it like this:



So that's what it is and how it worked-- as far as its usage on race cars, it's just a footnote-- I've only seen a couple photos of those boxes over the last 10 years or so-- I keep wondering whether they will turn up in some other Bosch application, like a Volvo or something, but they are quite rare.

Hope this helps.

epilogue:

With the 964, Bosch moved the coil switches off-board, to a separate "igniter" like a BIM 124, looks like this below. They are even smaller and better and more reliable today-- these are found in junkyards all over the world. THIS is what Bosch was after with that old box in the 911R-- but it took twenty years of semiconductor development to get there.

I have a few of these and the matching coils and a plan in my head. . . but it will have to wait until LONG after Parade 2013. . .

double epilogue:

Today we have inductive pencil coils, which fit on top of the plugs-- with six individual coils, there is plenty of time to charge even a small coil to an energy level high enough to reliably ignite the mixture AND do it in a way that doesn't misfire and push unburned hydrocarbons out the pipe. For REALLY high RPM applications (think sportbikes) there are Pencil Coils that are CDI-- they are actually driven with 90V input from a CDI ignition. When we start hacking coil-on-plug for 911 motors, this is the limitation-- easy enough to find coils that fit down on the spark plugs, hard to find a box to charge them. It can be done, just time. . . and lots of money.

[718RSK]

Checking in on this thread. Curious to learn if the decal has been reproduced yet?

On another note the 006 coil appears to be the the new Bosch part number equivalent of the TK12A10 coil. This coil was used on all 12 Volt 356 era cars including the 904 and C2's. I am not sure when Bosch phased in the new part number but am sure that most/all 901's and a bunch of the early 911's likely had the TK12A10 coil and not the 006.