Topic: CB Digital Ignition

[Flying J]

Wow! management is getting really sensitive. 

[pamcopete]

Wow! management is getting really sensitive. 

Careful.... 

[Bamboozler]

Haha, Calm down Pete.   I'll make sure this guy ends up to your liking.  One of the benefits of posting the projects is to bounce ideas and ask for suggestions to make it better and see if heading in the right direction, or favorable in peoples eyes.  This project is no where near done, if it takes a complete makeover to make people happy so be it.  I really appreciate your feed back and concerns, lets work to see if we can't make ya happy.  Feed back like you gave is exactly what I'm looking for.

To hit up some of your positive suggestions.

1. In it's current design it can optimize the dwell 3600 RPMs, but that is about it.  By nature it will be equivallent to stock, which was it's itention.  Two current controllers to be able to use the stock coils and a Dyna coil if you would like.

2. If it can send me a PM I'd like info about it and this guy will be dropping in price

3. Calm, I definately intent to test down with lower ohm coils and see what it allows.  I like Dougs idea and am going look into the sub 1 ohm coils.  Your not playing fair calling me out so soon.

4. Id love to offer this, but the cost of two automotive IGBTs, four individual coils (or two double output), and a vac sensor if controlled by your suggestion would blow this price tag skyward for the end user.  Not interested for a base.

5. No, I would like to offer a second unit that has more functionality and complexity which would cost more to make.

Not better than xyz.. you must be in a bad mood.  I think it's fairly competative to the Dyna2000 that's selling for ~$330, you don't think??


 

[pamcopete]

Well, I don't think that it is useful for you to tell me to calm down when I am already calm, and referring to me as "ya" doesn't move the argument further as well.

1. Right. As I said, your proposed product is no better than stock points above 3600 RPM.
2. Google it.
3. I think that you are really not familiar with coils, else you would have started with a lower primary resistance.
4. Your cost for the two additional IGBT transistors would be about $1.50. The end user buys the two additional coils, and if you had given thought to this idea of 4 coils, you would realize that you could offer the product with a selection of either two coils (wasted spark) or four and the end use could choose to buy the 4 coils later. If the pressure sensor is too expensive, then figure out some other way to determine the status of a piston. You're a smart guy.
5. Your product should offer the 4 coil option. Don't sell a product that is obsolete a year after purchase, you're not big enough to get away with that ploy.

As for the Dyna $300+ product, you are missing the point. You could be the Dyna killer if you solve the wasted spark inefficiencies, which they have not done.

I'm starting to get the feeling that you are a captive of some investors and you do not have the flexibility to change the design.

[Bamboozler]

Sorry ya feel this way Pete.  I think it's a worth while altertanive to a dyna  2000 at a much better price. 

[HondaMan]

Pete, whatcha talkin’ about, just crank up the timing advance dial and you’ll have a boat load more torque and HP, much more than any wiz bang spark can give ya… .  It’s obvious that you’ve been thinking about this before.  The very topic has been discussed when brain storming techniques to implement. 

Doug, that current does rocket up right quick.. One thing I wonder is the balance of the low resistance/low inductance coils vs the high res. higher L like the stock bikes use.  The low res./L charges faster but stores less energy resulting in a shorter duration spark (verses the high res/L coils) but higher voltage spark.  In a way is like debating the CDI vs Inductive ignition systems (short high voltage spark vs longer lower voltage spark). Interested in taking a look that those coils.  You are very much correct that they are cheap!!

I'm not quite sure I understand about Pete's comments, but could it be about adding more advance toward RPMs higher than the typical mechanical advance unit limits? When we were making the 14k and 16k RPM engines out of these (for midget racers), there was a bunch of people involved and tinkering. (A side note to viewers: before rocketing to those RPM, additional oil flow must be implemented, a different topic altogether...). The greatest difficulty we had in reaching those speeds was with the ignition advance and its controllability. With a standard advancer that could be altered to reach 50 degrees, the engines maxxed out around 12k because the waste spark causes burn back up into the intake tract a little bit (and fire out the pipes!). At least one company came out with a distributor (R/C, I think it was) to help solve this issue. But - the waste spark system also helps with several issues, including reducing bluing of the pipes on stock setups, fewer parts and current draw, and better MPG through longer spark discharge times. These engines greatly benefit from a longer duration spark because of the swirl-charge design they all have (a little less so in the post-1976 hemi head with domed pistons).

One clever individual made a complex mechanical advancer that could retard the spark a few degrees between about 14.5k and 15k RPM, and then advance back to its maximum: this let him stop burning off the intake tract charge as the inertia was building to push the charge in the valve and he was successful in reaching the 16k RPM smoothly and without the characteristic stutter his competitor's engines had up there (too bad he wasn't as good a driver, though...different story.). This let him run right up to the end of the intake capability, and by playing with it all we discovered that the waste spark was indeed the culprit (it also reduces emissions, if you care). But it also has no detrimental effect below 12k RPM, so it is a wise design for most bikes.

So, if your programmer has the control, maybe this little 'secret' could be built in to these boxes? It would give you a leg up in sales points to the very savvy, although I'm not sure how many might care on the SOHC4 bikes. I have no doubt that the DOHC bikes so prevalent today would still have this issue, and that it is not addressed anywhere else. Today's bikes typically have much narrower main bearings, though, and cannot take the hi RPM loads for long. Those that have roller bearings and high-flow oil systems might cope with it, though. I haven't kept up with the likes of the 16k Katana engines, can't say how they addressed it, but they do use short spark duration to avoid the back-burn.

[pamcopete]

HondaMan,

There are two disadvantages to a wasted spark system.

1. The coil fires every 360 degrees which shortens the available dwell time. A coil per cylinder would fire every 720 degrees, doubling the available dwell time. The coil will also run cooler with the same dwell angle because the duty cycle is doubled.
2. One of the plug wires in a wasted spark system is a positive voltage. Spark plugs work better with a negative voltage and the plug with the positive voltage requires 40% more energy than the negative plug, which is one of the reasons that dual output coils produce the high voltage that they do. It's not for performance, it's to provide the 40% more energy that the positive plug needs.
3. Exotic plugs, like a platinum or Iridium tip do not work as designed with a positive voltage. The ides for these plugs is to super heat the tip to promote the emission of electrons, but if the voltage is positive, the electrons are emitting from the grounded anode so the construction and additional cost of the precious metal tip is wasted. In fact, if you want to save some money, only install the precious metal tipped plug in the negative wire and install a standard plug in the positive wire.

Ford motor company equips their vehicles with a platinum tip on half the spark plugs and a platinum anode on the other half because of the positive / negative coil voltage of a dual output coil.

The latest coil on plug systems do not have this problem and that would be another option for an electronic ignition system that can drive 4 coils.

As for the comments that you quoted in your previous post, those were not my comments.

Any coil can be designed to produce a longer duration spark. That is not a characteristic that is unique to a dual output coil. Neither is the claim of reduced current draw.

There is no operational advantage to a dual output coil and a wasted spark system. It was introduced as a cost saving measure by the car manufacturers to eliminate the cost of the distributor, but now more and more cars are switching to a CIP system to eliminate the disadvantages of a wasted spark system, and bamboozle has a golden opportunity to introduce a coil per plug system to our world.

[HondaMan]

HondaMan,

There are two disadvantages to a wasted spark system.

1. The coil fires every 360 degrees which shortens the available dwell time. A coil per cylinder would fire every 720 degrees, doubling the available dwell time. The coil will also run cooler with the same dwell angle because the duty cycle is doubled.

While that's true mathematically, don't forget that the coil design itself, for any well-designed waste-spark system, is usually based on the available dwell. In the case of the stock SOHC4 coils, for example, dwell times available from the points to 11,000 RPM are sufficient to fully charge these coils: I have used the stock coils myself to 16K RPM with points and they worked fine (see the "Thoughts of Hondaman" post for some of the details). The waveforms I found during our testing for the Hondaman Ignition, and those Bamboozle shows above (and his research) shows that even the "slow" Dyna 5-ohm coils are being fully charged OK with these dwell angles. Now, if a slower coil were introduced, say a 7-ohm primary winding, the longer dwell available from some of the other offerings out there (like your Pamco units) can provide an advantage to bikes like the alternator-challenged 550, 350F/400F that wants to go all-electronic: of this, there can be no doubt.

2. One of the plug wires in a wasted spark system is a positive voltage. Spark plugs work better with a negative voltage and the plug with the positive voltage requires 40% more energy than the negative plug, which is one of the reasons that dual output coils produce the high voltage that they do. It's not for performance, it's to provide the 40% more energy that the positive plug needs.

I think I understand your description of "positive" vs. "negative" voltage, and you're right concerning what Ford has done for their 100,000 mile sparkplugs (I actually was involved with a bit of the research for those plugs, building the testing machines for them). The first (dual-output) spark peak is opposite on the 2 plugs, but there are typically 7 alternating peaks per discharge spark while the plasma bridge is active (if the plug and cap are in good condition) at a very high frequency (relative to engine speeds), so I can't imagine that making any measurable difference?

Why do you think plugs work better with negative voltage applied? Is there a research about this that you know of?
Interesting concept.

[pamcopete]

1. If the coil is allowed to fire every720 degrees instead of 360 degrees, then the duty cycle is halved. In the case of the SOHC CB750 with a dwell of 190 degrees, the duty cycle is 190/360 = 53%, meaning that the coil is on more than half of the time, resulting in excessive heat buildup of the coil. The excessive heat buildup of the coil results in an increase of the primary resistance, which in turn results in a reduction of current through the primary windings and a reduction of the voltage output of the secondary. This hotter coil also means that it will take more time to “charge up” or reach saturation, so a hotter coil will stop producing it’s maximum possible voltage at a lower RPM.

By changing the wasted spark system to a coil per cylinder, the coil only fires every 720 degrees, so the duty cycle becomes
190/720 = 27%, meaning that the coil is only on for 27% of the time and spends most of its time cooling off rather than heating up, resulting in little or no increase in the primary resistance and no degradation of performance at higher RPM’s.

Whether or not dual output coils are more efficient could be argued for several days, but certainly any coil with a lower primary resistance will reach saturation sooner and not require as long a dwell angle. The dwell angle of the PAMCO ignition system is 120 degrees and was optimized at that to work with the “Ultimate” coil that has a 2.5 Ohm primary resistance. The result is a much cooler running coil that does not suffer an appreciable increase in primary resistance and will produce it’s full output up to red line.

The duty cycle for the PAMCO is:

120/360 = 33% which is a very good number, meaning that the coil is off longer than it is on, or, it has more time to cool off than heat up. This also means that the power consumption is less than a 190 degree dwell angle, which compensates for the lower primary resistance / higher current issue.

The general rule for coils is to have a duty cycle of less than 50%. Not only does a lower duty cycle allow the coil to perform better, but the lower temperature also means that the coil will last longer, an important consideration with today’s pricey coils.

If you are going to install a lower primary resisrtance coil, then it is essential to reduce the dwell angle or you will have a real firecracker due to the higher current and high duty cycle.


If you had a 720 degree firing time with a 120 degree dwell angle the duty cycle would be 120/720 = 17%.


2. You observation that the voltage on the spark plug is alternating is true, but the voltage is biased in one polarity and the excursions are predominantly either positive or negative.


Here is a sampling of some articles about the polarity of an ignition coil. If you are old enough to remember the cathode in the tubes of your TV, you will recall that the cathode was heated up and a negative voltage was applied to the cathode. The heat “encouraged” the electrons to move towards the plate, which had a positive charge on it. This is analogous to the hot tip of the spark plug and the relatively cold and grounded anode. The electrons, which are negative, will move more readily towards to the positive anode than the other way around.



Article 1:

Why then do we worry about coil polarity? Because the spark plugs do care which way the electrons are flowing in the high tension circuit. The spark plug has a thermally insulated center electrode (surrounded by ceramic). With engine running the center electrode runs substantially hotter than the exposed end electrode. Design of the ceramic insulator determines how hot the center electrode will run, leading to the designation of hotter or colder spark plugs. As electrons go, they love to jump away from a hot surface and fly toward a colder surface, so it is easier to drive them from hot to cold rather than from cold to hot. End result is a difference of 15 to 30 percent in voltage required to make spark  jump the gap on the plug depending on which way it is going.
 Article 2:

What difference does it make whether positive or negative voltage is supplied to the spark plug terminals?. It directly affects the amount of voltage required to fire the spark plugs. When polarity at the spark plug terminals is positive, it's harder for the voltage to jump across the air gap than when polarity at the plug terminal is negative. Just why this is so is related to a pair of electrical theories--the electron theory and the theory of thermionic emission. According to the electron theory, all current flows from negative to positive. The theory of thermionic emission states essentially it's easier for electrons to leave a hot surface than a cold surface. Combining the two theories, one finds that electrons will always leave a negative charged surface for a positive charged surface, and they will leave the negatively charged surface with more ease when the surface is heated. Spark plug design is such that the center electrode almost always operates at a higher temperature than the ground electrode. Since it's easier for electrons to leave a hot surface, it is preferred to have the electrons "jump" from the hotter center electrode to the cooler ground
electrode. When the center electrode is negatively charged (negative voltage at the spark plug terminals), this is what happens. Stated another way, putting the negative charge on the hotter center electrode causes the gap to be ionized at lower voltage. (Ionization is necessary to permit passage of the spark through the high resistance of the gases in the cylinder.) When positive voltage is supplied to the plug terminals the hotter center electrode becomes positive charged. Consequently, electrons must leave the negative charged ground electrode and move to the positive charged center electrode. But, since the ground electrode is cooler than the center electrode (and remember, it's easier for the electrons to leave a hotter surface), it takes more voltage to make the current jump the gap-in fact, up to 45 percent more.

Article3:

It takes 40 % or so more voltage to fire the spark plugs on a ignition system with positive polarity. On most bikes you would have to really work at it to do this. However, some four cylinder bikes like the early Honda fours are designed this way with double ended ignition coils. One lead is positive and one lead is negative. This means two of the plugs will require a lot more voltage to fire than the other two. Not much you can do about it, but if two plugs start fouling out on these bikes, this might be the reason.

[Bamboozler]

Pete, I think there is a way to satisfy both trains of though.  I agree with the technical benefits of a COP system.  No doubt that there is a benefit for the reasons you mentioned above; reduction in heat and improved spark quality.  However there is a substantial cost in buying a system built to operate four individual coils (current controllers, method for detecting cylinder firing order, and 4 coils that need to be purchased).

I really want to keep the cost of this guy at a reasonable point.  I’m thinking there would be split on those that would want to pay for the COP setup and those that would be fine using their stock coils to keep their investment minimal. 

What about the idea to create an add-on module to the main board that housed the two extra IGBT current controllers and associated electronics to run a COP?  I would think this setup would be a reasonable solution to this problem.  The main digital iggy would only need an extra two, maybe three wires, a switch to tell the micro to run either a wasted spark code base for stock equivalent coils or a COP code base for 4 individual coils, and maybe an EEPROM to house the two flavors of code. This would keep the cost of the main digital ignition relatively the same and allow those that wanted the COP option the chance to acquire it.

I will talk to my firmware friend and see what he thinks regarding the dual boot feature and look to see if there is enough I/Os with the current micro.

[pamcopete]

bamboozler,

Well, you could just include the two extra IGBT coil drivers and some programing to make it a user selectable operation to either have the traditional wasted spark / stock coils setup to start with and then he could purchase the four coils later and remove a jumper on the board to bring on the 4 coils program. I don't see where you would need an extra EProm for this feature and the IGBT transistors themselves are about $0.75 each in 1,000 lot, which is the way that I buy them.
The additional time available with a 720 degree 4 coil  system could be used in one of several ways:

1. Leave the dwell angle as is and benefit from a cooler running coil with longer life without degradation of the primary resistance, plus eliminate the positive voltage penalty.
2. Increase the dwell and still have a sub 50% duty cycle and no positive voltage penalty.
3. Increase the dwell based on RPM's without having to worry about coil heat and primary resistance increase and eliminate the positive voltage penalty.

If you elected to use a vacuum sensor to determine the piston position, then that too could be an add on item that would not be part of your basic system.

So, you could achieve your objective of introducing a basic system for relatively low dollars that has future expansion capabilities. We all love products like that.

[MasterChief750]

just a thought for a vacuum sensor, where are you going to hook it up? if you use the vac ports then how are you going to sync the carbs and have the bike running at the same time. your other option would to be drill and tap somewhere and how many people are going to be willing to do this? just an example the ecotec i-4 platform gm has the 2.2l originally had a wasted spark system. once it was switched to a cop system a gain of only 2 hp was reached. more drastic hp improvements came with the addition of vvt. bamboozler i like your current setup ideas. things to ask yourself would be what does the market want. and how cost prohibitive would a four coil system be. adding four coils would add price and what about mounting are you going to supply this or leave the end user to fab something up? a system that is expandable would be an exelent product.

[TwoTired]

Much has been said about coil heating.  Nothing has been said about coil heat dissipation, which varies by design.  It doesn't matter to the coil how much power you put into a coil if the heat generated is removed in an efficient manner, so as not to raise internal temperatures.  If coil heating is a real issue, I might seek to improve the poor heat transfer characteristics of the coil, particularly if the coil characteristics change it's own operational characteristics, (beyond the environmental application space, of course).

I will point out that the stock coil's core is thermally connected to both the coil windings and the frame, which give it a large heat sink to transfer heat into.  The frame also gave a very large area with which to transfer the heat to the atmosphere.

A coil on plug has no such aid and relies on the thermal characteristics of its construction, as well as the temperature of the air flow over it.  The alternative is to restrict the power input the coil.  You have to choose which trade offs better fit the application space in any design.

Just some thoughts,

[pamcopete]

HondaMan,

Just some corrections that I feel are appropriate for your comments in your last post:

"While that's true mathematically, don't forget that the coil design itself, for any well-designed waste-spark system, is usually based on the available dwell. In the case of the stock SOHC4 coils, for example, dwell times available from the points to 11,000 RPM are sufficient to fully charge these coils: I have used the stock coils myself to 16K RPM with points and they worked fine (see the "Thoughts of Hondaman" post for some of the details). The waveforms I found during our testing for the Hondaman Ignition, and those Bamboozle shows above (and his research) shows that even the "slow" Dyna 5-ohm coils are being fully charged OK with these dwell angles. Now, if a slower coil were introduced, say a 7-ohm primary winding, the longer dwell available from some of the other offerings out there (like your Pamco units) can provide an advantage to bikes like the alternator-challenged 550, 350F/400F that wants to go all-electronic: of this, there can be no doubt."




1. It's not just true mathematically, it true in the mechanical design of the points cam. In other words, it real, not just theoretical.

2. It's obvious that the Honda engineers could not find a suitable dual output coil for the CB750 when it came out in 1969. They violated the golden rule of not having a duty cycle greater than 50% and they did this to accommodate a high reving engine with available coil technology. In fact, I could not believe it when I first learned that the dwell angle was 190 degrees. My disbelief was so intense that I actually argued that it could not be 190!! The choice of 120 degrees for the PAMCO was mainly based on our success with a 60 degree dwell on the cam mounted XS650 system.
3. PAMCO does not offer a longer dwell angle, just the opposite. The PAMCO dwell angle is 120 degrees instead of 190 degrees and is optimized for coils with a lower primary resistance. The shorter dwell angle causes the PAMCO and a lower resistance coil to run cooler and draw no more average current than a stock set of points.
4. Having a set of points driving the stock coils at 16,000 RPM is not the same thing as having the ignition system work better at 16,000 RPM. When you say it worked at 16,000 RPM, I think that you mean that it worked at 16,000 RPM, not that it worked well.

[pamcopete]

TwoTired,

Some comments on your last post:

"A coil on plug has no such aid and relies on the thermal characteristics of its construction, as well as the temperature of the air flow over it"

All coils on plug are either sub 1 Ohm or they are a Piezo device, neither of which develops much heat at all, and they all work in a 720 degree environment with very sophisticated current and duty cycle management. So, I don't think that heat is an issue with COP designs.

[pamcopete]

masterchief750,

If the ignition system does not sense vacuum because you have removed the vacuum sensor to do a carb sync, then it will assume that the vacuum sensor is defective and automatically revert to the wasted spark system driving both coils at once. The system would have to do this in any event for a "limp home" mode in the event that the vacuum sensor fails.

[Bamboozler]

just a thought for a vacuum sensor, where are you going to hook it up? if you use the vac ports then how are you going to sync the carbs and have the bike running at the same time. your other option would to be drill and tap somewhere and how many people are going to be willing to do this? just an example the ecotec i-4 platform gm has the 2.2l originally had a wasted spark system. once it was switched to a cop system a gain of only 2 hp was reached. more drastic hp improvements came with the addition of vvt. bamboozler i like your current setup ideas. things to ask yourself would be what does the market want. and how cost prohibitive would a four coil system be. adding four coils would add price and what about mounting are you going to supply this or leave the end user to fab something up? a system that is expandable would be an exelent product.

I too think a the idea of a 4 coil driver an excellent idea to make as an expandable option.  While the merits of a four driver system is noted I don't see it as a 'gota have' item by most.  Maybe more data or opinions would convince me otherwise, I could be wrong here?  Worth taking the time and effort to make upgradeable/expandable?... sure thing, it makes this iggy a better product, and if it satisfies a portion of people while keeping the goal of keeping the price down I'm all for it.  If this is a feature that is able to be accommodated, the coils and mounting methods will have to be looked into.  I do believe we have a member here sporting a COP system on his EFI bike, maybe he would be a source of insight?  If there is fab required it is something that could be offered to complete the offering.

Pete, you may be right regarding not needing an eeprom, I'm not sure how the firmware for two split programs would be handled, need to talk to my firmware friend.  But thinking about it a bit more, it may still be one hex file flash to the micro and not two separate config files that the micro pulls from a eeprom .  Keeping the code one piece would allow for a redundancy scheme as you noted without a restart to the micro.

Regarding adding the extra two coil drivers on board, still not sold on the idea.  If someone wanted the iggy for reasons other than four drivers with no intention of ever using the four drivers, they're paying for something they don't need.  If it could be used to aid the stock wasted spark system I could get more excited about it.  I'd figure it best as a four driver upgrade and leave it as such. 

Above all, before getting too excited it has to be shown to work reliably first. 

[Bamboozler]

To hit up a bit more regarding this ignition here are a few pics of the rotor. 

This is the first 7075 aluminum prototype, it fit reasonably well for a first turn part.  For the most part it's a dupe of the existing stock advancer, you can see the similarities.  The stock washer and fastening nut are able to be retained to rotate the motor.  Slight adjustments where made with the index pin and magnet location based off the prototype. 

The bottom of a few numbers didn't come through due to a incompatibility with the conversion of the 3d file and the CNC computer program.  Instead of fixing the drawing it was OKed to proceed anyways.  As noted prior, the small order sample will be anodized black and laser engraved with machine engraving as an alternate if laser doesn't come out as desired.

In the first small order there will also be a second magnet added 180 degrees from the existing hole to eliminate setup offset.  The current code based uses two halls and one magnet.  My firmware friend noticed that there was a slight offset based on where you positioned the magnetic hall sensor board that surrounds the rotor (think stock points plate in size/shape).  If the hall sensors sits at 9:00 and 3:00 in relation to the rotor, and the hall sensor board is positioned slightly north or south (not perfectly centered around the rotor), the time it takes the magnet to travel from sensor A to B and again from B to A can be slightly off (i.e. 182 degrees/178 degrees).  This slight error can lead to a spark triggering/timing error.  To fix this, one bipolar hall sensor will be used with two magnets with alternating fields (one north, one south).  With this setup, even if the hall sensor board is set to TDC with a slight offset the two magnets will be picked up evenly spaced.

When the next rev of firmware that tackles accommodating acceleration is finished there should be a specification for the degree of accuracy that can be expected with the ignition.

[MasterChief750]

looking good cant wait to read mroe, if the price point is at or below 200 ill deff consider picking one of these up as i want a digital tach

[Bamboozler]

Hey MasterChief, I hope I can help you out and give you something you’re looking for.  Really would like to come in at that price point too, we’ll see how it shapes up as time goes on (hopefully sooner than later).

I'm updating this thread with a bit of test results and feature tweaking heads up.  Nothing too much new, other things have consumed my last few weekends and my week days have been rather busy.  My firmware friend has been too busy with his business and family to get a crack at the RC2 release, but said he’s looking to free up shortly, which is understandable, it’s a side project.

I did manage to get 2 low ohm coils to run a current profile on and one higher ohm Dyna (2.2 ohm).  One of the low sub 1 ohm coils is common GM C849 by Wells Manufacturing Corp. that Doug suggested.  The other sub 1 ohm coil was a Ford Motorcraft DG508 coil on plug; the same COP that 23tbucket tried on his EFI bike and seems to fit real well with the available room on the CB. Both of these coils are readily available, inexpensive, and should provide a decent look at what charge time (dwell) will satisfy them.

I used a different Oscope and the actual ignition instead of a test circuit with the same IGBT, so I ran all the coils over again.  The coils are being fired with the ignition loaded with the stock 190/170 charge/discharge firmware, this is to allow the higher ohm coils enough time to fully saturate.  I loaded the RC1 8.2mS dwell firmware for the quick charging sub 1 ohm coil so as to not heat the electronic up too much, though testing at a higher RPM would have work similar I suppose.

Below are the results, very much similar to the screen captures posted earlier.  The equivalent current per division is around 1.33 amps/division based on the gain of the current measuring circuit.

Dyna 5 ohm


Stock Coil ~4.5 ohm


Dyna 3 ohm


Dyna 2.2 ohm


Ford DG508


GM C849


GM C849 at 250uS/div, this shows the charge ramp much better.


Both the GM and the Ford coils fully charge by around 1.4-1.5mS and draw around 6 amps.  I was kinda surprised by the sharp knee where the sub 1 ohms saturate; much sharper than the lazy rounded bend on the higher ohm coils.  My hardware friend wanted to see the circuit to verify that the IGBTs weren't saturating, the ramp does look very much similar to the other coil ramp screen captures on the web; thinking there is nothing misleading.

What I would like to do is get a modified version of firmware with a dwell that is fixed at 1.5mS instead of the 8.2mS and do some average current measurements and map it like other 190/170 and 8.2mS current vs. RPM charts.  This should give an idea of what they would be consuming real world through out the RPM range.

The Dyna 2.2, these guys started to raise the temp on board quite a bit; I shot 202F at it’s normalized temp in still air with a infrared thermometer with around 78F ambient.  Granted it was with the stock 190/170 at 1k RPM (worst case), but a looking into the heat sinking maybe need (i.e. AL lid coupled to the current drivers).  This is something that will be looked into when hitting up the heat chamber.

How I can see the iggy shaping up is to add a 4 position DIP switch on the board, there should be enough room on the micro to accommodate it (I/O wise), though the switches will need to move a bit.  With this the following functionality could be tentatively added (barring no issues that come up with the firmware side):

o   Dip#1 – Vacuum or boost switch installed / No VOES/Boost switch installed.  This just moves the external wire that needs to be grounded to enable the timing retard/advance option onto the DIP switch instead of grounding out a wire (convenience). 

o   Dip#2 – 2 coils / 4 coils.  Lets the micro know there are 2 or 4 coils installed and that an add on board is installed that provides 2 extra coil drivers for a COP or individual plug setup. 

o   Dip#3 – High ohm coils / low ohm coils.  Choose weather you would like a 1.45mS dwell or a 8.2mS dwell depending on what type of coils you are using (i.e. stock or high ohm Dyna vs a sub 1 ohm automotive coil).  Note, those dwell times are not yet set in stone but are close and cover a reasonably large range of coils.

o   Dip#4 – Third rotary switch function select: adjustable RPM window switch (shift light ect..) / Adjustable dwell control.  If a dwell control different than the two default choices of 1.45ms or 8.2mS are desired you can choose from 16 different preset coil charge times that should cover a wide array of coils, but you would have to sacrifice the RPM window switch (and vice versa).  Choosing dwell control would inactivate DIP#3.


I think these choices would add a decent amount of flexibility and choice while keeping the cost relatively the same.  I talked with my firmware friend and he noted that there wasn’t anything he could think of preventing those features to be added.  We’ll have to wait and see it implemented to see for sure, but its looking doable.

(sorry for the wall of text)

[Pinhead]

Any updates on this ignition?

[23tbucket]

Bamboozler; I'm the fellow that "had" 4= 1.5 ohm coil-on-plugs on my efi 750 with the power/arc optical ignition. This was still in a wasted spark format. They looked real sweet.....got about 10 hrs and one side of the power/arc module had a melt-down. Power/arc replaced the module under warranty. I removed the COP's and I now have the coils that came with the kit installed.
 I still would like to have coil-on-plugs. I'm not as familiar with this stuff as TT , Pamco, Hondaman, others, or yourself. Most of this stuff is waaay over my head! For sure, to get a true sequential ignition you have to know where each piston is on compression. The vacuum sensing from a cylinder may work...better would be to pick up from the camshaft somehow? I thought about making a 1/2 crankshaft speed toothed drive (hall effect or VR), using a small timing belt and a couple of toothed pulleys from the existing points side crankshaft. It would be fairly easy to do and wouldn't take up much room. The bigger of the two pulleys could have a magnet embedded or machine off one tooth. The microsquirt ecm can control ignition for two drivers. I haven't done much research, but I think you can add some electronics to make this work with four separate drivers?

Take care; Clifford

[Bamboozler]

Sorry for the lack of updates on the ignition.  I’ve been rather busy the last weeks with a new job opportunity that needs the better part of my day as well as other summer activities (been out of country for 10 days, family events, local trips/events ect). 

I also received a bit of a setback this last Saturday.  As I was riding into work with the CB I happened to lose my chain (master link failure?, recovered chain has none, the clip was install in the correct orientation..?).  Upon exiting the primary sprocket, the chain decided to punch a hole into the motor case directly in front of the primary sprocket.. :/  It’s truly disappointing, as the weather has been exceptional as of late.  Before pulling the motor to fix the hole I will most likely rebuild the whole shooting match with intention to withstand boost (forged pistons, rods, trans tweeks ect., head goodies all ready purchased).  We’ll see when the spare funds become available to complete the wish list. 

However the ignition project isn’t dead in the water.  I’ve been running the ignition on the bike for the last month, maybe 750 or so miles logged.  It has been running without any issue and I'm please with its performance thus far.  I was going to be dedicating this last weekend after returning from the 10 day trip to push on final functionality testing, but the motor received damage (that very day).  I may be asking local friends and board members for help in testing and logging run time, but haven’t gotten any plan worked out as of yet. 

The rotors where pushed back due to a delay in processing the PO and a upswing in orders for the machining company that will be creating the CNC rotors.  I was promised late July after the order was officially placed.  They emailed me last week asking for confirmation of the final drawing and noted material was being ordered, so I’m assuming they are on schedule. I will be trying to complete layout and order a run of rotor PCBs sooner than later, time permitting. Then it will be the main boards that will be left for final tweaking and ordering, if I’m able to devote time this could be within this month, but I’m not committing on a date seeing how free time has been scarce as of late.

In regards to the technical details that have been hashed out as of late.  One thing is for sure and that is the COP and or sub one ohm coils will not work on the current board design.  The amount of heat that is generated by the IGBTs from these hot coils heats up the board much too great (IGBTs: current switching devices, all the current flowing through the coils packs flow through these two devices, each IGBT handles one coil).  I’ve even made the 3”x2” lid out of aluminum and heat sunk the copper pads of the IGBTs to the lid and it still was hitting near 300F with these hot coils within a few minutes of run time (IR gun shot on the IGBTs side of the PCB); much too warm, inadequate thermal management.  However I ran a pair of Dyna 2.2 ohm coils with positive results never seeing over 170F after 40 minutes run time in a potted unit at worst case RPM (4K).  I understand that the COP would stretch the cycle time from 360 to 720 degrees, giving more time to dissipate heat; which might be something to still look into and look for improved results.  The testing with the COP coil and chevy sub 1 ohm coil was preformed with a 360 cycle with 1.45mS dwell.

I discussed this with the firmware friend and it was determined to push the sub 1 ohm coils to a second version as I noted prior and build the case out of aluminum (extrude or CNC maybe) to assist with shedding heat and helping the existing version to be the most cost effective as possible.  With that being the case it would be PC configurable and have a pressure/vacuum sensor on board for boost retard COP if that method proves worthwhile.  I have ran a few tests on pressure sensing the vacuum and it looks like it may be worth the time to code it and try.  See the screen capture below, you can clearly see the dropping voltage when the valve opens.  Triggering on a falling edge should be possible, it’s just a matter of coding to accommodate any false triggers or odd vacuum signals that are possible by implementing a coding plan to work around or see through these not so perfect vac signals that could arise.

vacuum sensor hooked up to carburetor #4's synch port


Clifford, thanks for taking the time to reply with your suggestions to get the COP workable.  Interesting suggestion regarding the half speed pulley, I like the idea!  If you set up both the crank and the half speed gears to use hall sensors you would should very well get your orientation that you are looking for, or every other crank rotation would be a positive trigger for the half speed gears..  Great suggestion, I like that it is mechanical as well!

[Flying J]

Not the best idea,You would lose your tac, but the tac ratio is 4:1. so you could use 2 sensors on that to give you the 2 triggers per revolution.  Kinda like what acewell uses for their speedo. Not exactly sure how that works. Bad idea?

[23tbucket]

ffjmoore: Perfect find. That is almost TOO simple that it may just work? If they are able to get a "dump" or "trigger" a signal once every 4 crankshaft turns......should be just as easy to enable 2 signals..... Electronic gurus wanted!!!!

There may be a bit of gear backlash and slop on the slotted end to overcome...but certainly do-able.

As for the tachometer....any electronic one should be easily adapted...just need a signal from one coil.

Good stuff my friend.........thanks....

Take care; Clifford