COP Ignition Development

-
nuttyprof said:
Kit, I think you contacted me about some heat sinks a while back, are you still interested?
Click to expand...

Not at this time, I am finding not much heat generated. I have more testing to do, but need to set up in drill press to get higher speeds.
 
The weather turned nice to do some needed tree work. I put the ignition testing on hold to cut down some problem trees at my place, for what seemed like weeks. I do not burn wood, so I took it to a neighbor. I am muscle sore, and ready to get back to electronics work.
View attachment 20160323_132218.jpg

Yesterday was the first day to do some testing. I verified the plug firing order is correct at coils using a timing light. I put 8 marks at the base of distributor, and used a timing light on each plug. I also made a crude timing tape that shows timing in 10 crank degrees per increment. A small pop rivet pushed into the roll pin serves as a pointer. It is simple, use drill to spin shaft, timing light to view timing, RPM is read via user interface.
View attachment 20160324_163934.jpg

I also set up an external single current shunt using a 0.1 Ohm resistor, to monitor all coil currents with scope. The purpose is to determine dwell time is sufficient, to reach desired peak current under various operating conditions.
 
Just wanted to chime in to say I've been following this with a ton of interest. Would make an excellent upgrade for any engine with performance and drivability in mind.
 
I put the unit on today. I found that my test distributor shaft was too long. A SB is 3 7/8", but the shaft measured sightly more than 4", so it was from a RB. I changed the shaft out, and found my other NOS distributor shaft is 3 1/2" for a BB. The good is I have 3 different lengths for reference. I found info here: http://www.hotrodreference.com/1310/mopar-distributor-identification-by-shaft-length/
View attachment 20160405_172519.jpg

Installation was simple with the pre-wired harness and coil assemblies. Drop in distributor, static set base timing, connect ground, hook power to bypassed ballast ignition source. Connect coils to plugs.
View attachment 20160405_155645.jpg
View attachment 20160405_160015.jpg

I got the car started, some trouble with that, but once running, it ran great! I took it for a short drive, it was flawless. The settings replicate stock tune for a 66 273 2B.

I need to focus on the starting problem, I think it might have to do with starter noise or power supply noise immunity issue. I tried to use the regulator that is on the Nano circuit. Typically I use an automotive rated part, with additional parts for transient protection. I have parts and circuit space, so can add a better regulator. I will do some probing with a scope, and see what I find.
 
I bet you are jacked. Always satisfying when a bunch of work starts to come together!!
 
I was happy it runs well, but not jacked enough to continue working to solve the starting issue. I noticed that the starting RPM jumps around and seems low. I am going to to try a simple fix, moving ground connection from engine to battery (-), and ign relay supply from starter relay stud to battery (+).

I have a bunch of stuff going on, so it might be a day or two, to find time again.
 
Sweet! I noticed there was some latency in the rpm numbers on the computer screen, is that normal or could the baud rate be changed?
 
That is not related to baud, it is the update setting. Set slow for now. With a PC application it can go as fast as 10ms, at that rate numbers are a blur, so bar gauges are used. I am using a dumb terminal emulator at this time, to save development time. Later I will write new PC graphical interface. I am moving from VB6 because it requires runtime apps on PC, my new one will be JAVA based and will run on many platforms, windows, Mac, Linux. The application drives the update rate, it will be variable like on my EMS2.
 
Any idea when or if you might have beta versions for sale or testing?
 
KitCarlson said:
That is not related to baud, it is the update setting. Set slow for now. With a PC application it can go as fast as 10ms, at that rate numbers are a blur, so bar gauges are used. I am using a dumb terminal emulator at this time, to save development time. Later I will write new PC graphical interface. I am moving from VB6 because it requires runtime apps on PC, my new one will be JAVA based and will run on many platforms, windows, Mac, Linux. The application drives the update rate, it will be variable like on my EMS2.
Click to expand...

Cool
 
mguner said:
Any idea when or if you might have beta versions for sale or testing?
Click to expand...

I have to get past the hard to start issue first. I am working on means of trouble shooting to find the cause. I have not experienced this issue on similar 4 cylinder systems, but the V8 crank current is likely double.

At best, from a great working prototype that is ready, there is a month or two of work for, PCB artwork for production. And at least that time for application software. If there are FABO members, with PCB layout, software dev, and machinist skills for distributor mods, time would be reduced. I do not work full time on this, but typically do work when the weather is too bad to work outside. If anyone desires to help, please PM me.
 
I found a small amount of time to measure battery cable voltage drops while cranking. The service manual suggests the drops should be no more than a few tenths of a volt. I found the drops were excessive, with about 1 volt on the positive lead and about 0.5V on the negative. I replaced the cables about 35 years ago, with inexpensive universal fit ones of the correct length. It appears most of the drop is in the cast lead terminal ends. I plan to replace the cables, I am not sure if I will build my own with modern stamped and plated copper terminals, and quality bulk cable. I might used stud terminals, with lugs crimped on cable. It will be easy to tap on start switch and other supply taps, or change out terminals.

I also measured the starter current to range from a high of 250 A, and a low of 138 A. The current increases as cylinders come up on compression, the meter is digital, so results are somewhat uncertain. Digital meters capture over a period of time, so when the current varies, results vary. I wish for a needle gauge.

I have heard good things about min-starters. I found a place in TN, called DB electrical. They sells a universal fit unit for $69 new. I plan to install that, at the same time as the cables.

I also measured the distributor drive backlash to be just under 5 degrees. I used a dial indicator to measure the tangential travel of 0.055". The wheel diameter is 2.6", with a circumference of 3.1416 x 2.6" or 8.168". The angle is estimated as 720 x 0.055/8.168, or 4.8 degrees. I have purchased a new mopar performance intermediate shaft and new drive bushing. I am also making a tool to pull the bushing. I will install the new parts and check the backlash. There is some wear in the original shaft, I am a bit surprised by that.

I also captured the distributor signal and decoder output while cranking using my logic analyzer. I expected to find faults there, but did not. I will report back in a week or two. Most of my time now is spent with spring farm chores. I have more trees to cut, and garden to plant.
 
I made a tool to pull the distributor drive bushing. The tool was made from 10 mm threaded rod, nuts, a couple washers and a small piece of steel. The rod and a VW cylinder head nut, we're chucked in a 4-jaw, so the nut could be turned down as an eccentric. The diameter was sized to just fit the bushing bore. I then made a part that served to push the rod to the side, so the eccentric locked in the bushing. It installs with the tab in line with eccentric bump, then the rod is rotated 1/2 turn, a nut was tightened to hold in place.. A large washer, socket, and nut on top, turned to pull bushing. It worked great.
View attachment 20160417_110016.jpg
View attachment 20160417_105706.jpg
View attachment 20160417_105810.jpg

The drive was replaced, the measured slop was about 2 degrees, the rest is due to the distributor drive tang. I have an idea how to reduce that. I will post if that works.

View attachment 20160417_141537.jpg
 
Find any more time to play with this? Would be interested in hearing the outcomes.
 
You may have mentioned this, but it helps during cranking to delay the spark timing to TDC or even later. That way the cylinders don't fight the starter. I recall that Holley's Commander 950 does this. If the sensed crank speed is <500 rpm, the controller assumes the engine is cranking and delays the spark. Also, to get a stable idle speed, the spark table usually needs a "saddle" around idle. Otherwise, you can get surging where as rpm increases, it gives more advance which makes the engine run faster, ...

You probably already know that you can get a cheap mini-starter by using one from a V-8 Magnum engine (~1990 - 2000 trucks & Jeep GC, perhaps even V-6). I put one in my 1965 273, but also fits slants and BB/RB engines.
 
I have not found much free time. I have had time to think about it, while working on other things. I am doing a DIY A/C system for my kitcar. That should keep me busy for a few weeks.

I converted the Barracuda back to the Mallory UniLite, and discovered the float level in the Rochester 2GV carb went high. I replaced the foam float with a new brass one, and wonder if a pop-back when cranking bent the float, or the brass was too soft from initial anneal for fabrication process. Brass work hardens, so it might hold after re-adjustment. I want to drive more as time permits, to evaluate float before testing the ignition again. I also want to change the starter to the mini and report on the crank current and RPM. The battery cable work has resulted in excellent starting of the Mallory, it springs to life on first cylinder it hits. The electronic will take a minimum of 2 engine revolutions. With second sensor I can get that to one revolution, still thinking on a scheme to reduce sync detection interval. My prior work has been ignition with EFI, it is easy to avoid over fueling when cranking, with cranking fuel settings. With a carb and choke, extended cranking (even the 2 revolutions) without fire seems to result in flooding. I am thinking about setting the choke open some, and relying on accelerator pump squirt for initial enrichment as needed. Or just pressing the pedal some for unloading on initial start.

The pop-back is ignition event happened at key-off when cranking. I have not had the problem with my other ignitions, so I did not implemented a soft shutdown. In doing research, soft ignition shutdown is a feature of modern integrated ignition drivers. I have done some design work, for ways to implement a cost effective, soft shutdown without the expense of integrated ignition drivers. I need to order some chips, write software, and make some hardware changes. My best hope is mid summer when it is too hot here for much outdoor work.
 
BillGrissom said:
You may have mentioned this, but it helps during cranking to delay the spark timing to TDC or even later. That way the cylinders don't fight the starter. I recall that Holley's Commander 950 does this. If the sensed crank speed is <500 rpm, the controller assumes the engine is cranking and delays the spark. Also, to get a stable idle speed, the spark table usually needs a "saddle" around idle. Otherwise, you can get surging where as rpm increases, it gives more advance which makes the engine run faster, ...
Click to expand...

Bill I captured the sensor data and ignition events during cranking and confirmed the ignition timing is as desired by the spark tables. My cranking issues related to loosing strands on starter cables, and flooded engine.

The 950 Commander is likely a manifold ported MAP sensing. This system as tested is for 2D RPM and ported vacuum, cuts the vacuum advance component of timing because the ported vacuum is near zero, hence 0 contribution for advance. Your are correct in 3D manifold MAP systems the tables need to have less advance in the quadrant of low RPM, and low MAP, and in the axis of low MAP for all RPMs.

It is possible to do idle control with ignition timing, but it is helpful to have throttle closed sensing, and even speed sensing or brake on sensing. Sensing is necessary to know when to engage closed loop idle control. I use a combination of idle air control and temperature based timing control for idle speed control in my EMS2 system. I plan to add idle speed increase when A/C compressor is on, on my kitcar.
 
It got miserably hot here. I found time to work on the ignition system user interface software, and also found a way to simplify the construction of the distributor timing sensor.

I found a reasonably priced MOPAR CAM sensor distributor on ebay. It was only $21.15 shipped. I tested it and it works fine. The plan is to double stack the cam and 8-tab timing sensor. This type of sensing means was in my original plan in the ECU capability. With the CAM sensing, the 8-tab shutter does not need modification, and it can be replaced with a 4-tab crank sensor for even more accurate timing control.

While first tests of the COP electronic advance revealed excellent performance when running, engine starting needs improvement. The CAM sensor provides information to sync the cylinder index once per revolution, and in a simpler way than decoding the windowed tab reference that occurs at the interval of two engine revolutions. It is a win-win, simpler to make sensor mechanism, simpler software, quicker starting capability.

The user interface under development is going to be graphical. The actual timing curves for RPM and Vacuum are visible, and the means to adjust is simple by selection plot values and bumping them up or down a fraction of a degree at a time. The changes are visible, and can be done in real time with no ill effects. The interface will also show RPM, vacuum, real-time ignition timing, engine status and other controls such as fuel pump status.

The application software is being developed with B4J by Any Where Software. It will run on Windows, Linux, and MAC systems with a USB port for connection to ECU. I have some of the individual parts working, but more work is required to combine parts an improve the visual presentation.

I still have by hands full with farm chores, and the A/C project, so development might be slow.
 
I forgot to show the url for distributor/cam part. Cardone 30-3899 Distributor fits Dodge/Jeep B150,250,350/B1500-3500 1992-1997 It is a Cardone remanufactured part, but there is not much to it. A solid shaft, semicircle shutter, and Hall Sensor.

Here are a couple draft software items.
AdvPlots.PNG
changetable.PNG
 
Yes it is, I checked measurements of drive tang width, length .... It is more compact than starting with mechanical advance distributor. It would be nice to have a custom nipple less cap. If there is interest in the COP system, might be worth the risk of distributor purchase.

I need to take about 0.005" off the upper shaft diameter, and make an aluminum flange adapter, for the shutter wheel attachment. I will mount an aluminum plate over the OEM Hall Sensor to mount the optical pickup for 8-tab wheel. I plan to cut the odd 3-pin connector off, and replace with a MX150 4-pin sealed connector for use with both sensors. They will share +5V and ground. The 3-pin connector pigtail seems unobtainable.
 
Last edited:
Wow.. Excellent work!! sure ya didnt get hit with a beam of light, like from the flick, Phenomenon?

definitely interested in this. wish I could help ya out!
 
mopardude318 said:
Wow.. Excellent work!! sure ya didnt get hit with a beam of light, like from the flick, Phenomenon?
Click to expand...

No beam. When I was 7 years old some neighbors gave me old radio's. I fixed some, others I took the parts and made transmitters and other things. I learned electronics myself by doing projects. Later I went to college and became a EE. My work experience over the years and companies, in product development and R&D, prepared me for my automotive electronics hobby, that started after retirement.
 
You must log in or register to reply here.
-

Back
Top