The code is small.....very small.
There are no security rules into the code without a *startup unload C* with grounding the scr.
If the magnet is into the neighborhood, the rule didn't work.

The hall need at least 25 degrees BTDC be placed.
The easystart into the untill 1800 rpm need this, the CDI start Point Of Pickup (hall) minus 25 degrees
Normaly a weeteather have a BTDC ignition of 36 till38 degrees.
2 stroke engines need more degrees, 38 + xxx, 4 stroke need less degrees, 38 - xxx

Use a degree wheel to set TDC and BTDC, dont use the flashlight in the first run.
The flashlight was made to controle the CDI if everything works and you will fine tune the engine.

PS, don't use the file I send to you.....there is a mismath in it, I hope I can fix it tomorrow.

That's ok..couldn't have used it anyway..
Assembly of the C-code to hex produced a few errors anyway.. working on rewriting a few lines myself, now that I sort of understand the timing sequence.

I put new files online to use only with the TCI (!!!) project.
http://www.modelbouw.gompy.net/tci/

New Exelsheet to use with the CDI (!!!) will be also (this week) today online.
http://www.modelbouw.gompy.net/newcdi/CDIcurve.zip

There is also a configuration setup howto use the compiler Micro-C like Donald Duck.....only pics !

Rob,

Nice! Thaks for sharing this.


Charlie Moore

Hi all,

There are new Exelsheets on the websites, did some bug-fix and manual inside.

Rob..
I now have 4 complete CDI boards with timer boards.. They all show the same problem. Battery voltage must be at least 5.3 volts and no higher that 5.6 or the units malfunction... Is this the voltage regulator chip (LP2950cz) causing this?
When I use my bench power supply set at 5.5 volts, everything works fine for hours. When I connect a charged ni-cad pack (4 cells) it will only work correctly for 3 minutes before voltage goes below minimum of 5.2 volts. Any ideas?

John

I think you have a battery problem or the small regulator can't handle the noice.
I use Lithium-Ion cels and don't have any problem with it.
Can you see how much current the CDI is using ?
Maybe you have to place some capicitors to eliminate electronic noice.
The capicitors must be placed between + input/ground and + output/ground

I just checked the datasheet for the LP2950cz. I agree with your assumption that it may be the problem. It is clearly low dropout, but it also shuts down at about 200 mA and has very little current capacity (100 mA) and heat dissipation in the TO-92 package.

I would try a standard LM7805 or similar VR, even a standard LDO and see what happens. With the 7805, you will need more head room voltage for the battery, but at least you can localize the problem. Another possibility is that your VR is oscillating. Do you have a scope to check it? Note, the pinout for the 7805 is different than the LP2950cz.

John

Or even try the LM340 or LM317T with the input voltage at least 1.5 higher than you need

LM 317 Calculator

http://www.electronics-lab.com/articles/LM317/

A high current drop from the timerboard isn't possible, the regulator isn't into the HV-circuit.
I think it must be ocsillating, sometimes one C is enough to eliminate the problem.
http://www.voelkner.de/products/8265...s_model=Q43848

As I looked at the voltage drop on my VOM, it showed me that my theory was correct. I jumpered out the regulator and all functions are now normal down to 3.8 volts, at which time the pic chip's internal brownout detection takes over and powers down the chip...
I am getting almost 3 hours continous running time now without a problem, on a 700 mah 4.8 volt nicad pack! Excellent performance as far as I am concerned..!
All I did was jumper the VR out of the circuit altogether, which eliminated the dropout problem.
T106d is a 400 volt-4 amp unit.
TIP42 is almost a direct substitute for the BD242 which I did for the fact that I can buy it at Radio Shack. I think the Hfe on the TIP42 was a little less but it functions well in it's environment.
Doing the 2 changes I made to the CDI board itself concerning components was to substitute a TIP42 for the BD242, and a T106d for the BT151...current consumption dropped from almost 900 mah to 300 mah with this change on one of the 4 boards. Maybe I should change all my boards? Voltage regulation doesn't seem to be needed.

Rob, it is a battery supply problem, I agree..
I might try 2 cell Lipo pack with an external regulator (LM317T) as suggested by Kermit, to keep the supply steady at 5.5 volts. I love to experiment, but the cost of parts is killin' me right now..$50 this week and since I am currently unemployed, that hurts a bit. Still have to change the initial curve in the PIC16f628 I am using..my 20cc motor with the 15" prop can take it but the 2.4 cu. inch chainsaw motor I converted doesn't like the steps in timing..

In a twostroke race enige, they use a timing of - low rpm / 36 degrees BTDC and high rpm 15 degrees BTDC.
Normaly into a Chainsaw engine the distance isn't so big, I think - low rpm 35 degrees BTDC and high rpm 28 degrees

If you don't use other batteries high then 4,8 Volt, a regulaor isn't needed.
I put this regulaor into the schematic for those who use 6 Volt batteries.
If the Voltage is to high, 5,5 Volt, you will blowup the pic.
It's also possible to use a 5Volt diode (.....?) and a resistant to reduce the Voltage for the pic.

Bigboat, the advance is low at low RPM and advance is higher at high RPM. You got it the other way around in your post.

Advance is wat you make into your timetable.
There is only one thing we don't fixit, belowe 1800 rpm the CDI makes his own time of 15 degrees.
The CDI start allways at 15 degrees BTDC, abouve 1800 degrees you can make your own timetable.

Rob, I jumpered out the regulator in the timer boards.. All working perfect now! Since I only will use 4.8 volt battery packs, it won't be needed.. thanks !
Question for you.. what does the 2 pins marked BREAK on the component picture do? can't get anything out of these two pins.. other than my initial problems with battery voltage, everything is working great now..except for timing light feature, but I don't use that anyway.
Making more boards for the rest of my planes that had fixed timing. What a joy it is to have the engines easily start now.!

John

Break is a push button to short the CDI, the engine will be stopped.......we hope
The flashlight is a white LED with a resistor, nothing more.
If you mark the cranck, you can see the marks with this flashlight.

The BREAK pins are working.. Thanks Rob ! This makes another useable feature on this great Project.. a KILL button !

I've still a problem with the C-code.......it doesn't generate a table like I want.
Not a big problem, the CDI will work also with this *bug*.
From 1800 till ~8000rpm the table is OK, but over the 8000 till 16000 there is a mismath of a few degrees (~5).
Most weeteathers don't have a problem with it, but fast running engines can have a problem with it.

Hi Rob,

I don't think there many engines that will reach 16000 for very long.

Hi Charlie,

Zenoah 22ccm and 26ccm making ~15,000rpm and some can let them run 23,000 rpm !!!

Charlie.. have you tried to interface the new timer board with another brand of ignition yet?
I found that there is something loading the optocoupler chip and it's not sending a full signal to triger the SCR on my other ignitions I make here.. maybe the coupling capacitor and resistor.. not sure.. it might just be a resistance value too high on my boards..
I used a 180 ohm on mine and the new timer board /CDI value is 150 ohms.. I can't trigger the new CDI either by just plugging in a hall effect to the proper parts of the circuit..I'll have to change a few values to try it I guess..

My design board schematic I made uses a 680 ohm before c3 and a 200 ohm after it for the coupling of the hall effect to the SCR gate..maybe I have to reverse those values since the optocoupler is using a phototransistor output..

Any ideas?

Hi gr8flyer55,

I have tried the auto advance with my design and it works great. On my setup I don't use a opto. So I can't say what you need to do.

So you are taking the output from pin 6 on the PIC chip and directly tying it to your trigger circuit? Hmmm , might try that to see what happens...more chips coming in the mail if it fails later.[]

If you put the two schematics together (TCI and CDI) you canb see the differend output of the opto.
The opto was placed to protect the pic for backslash from the scr.
You can try to remove the opto, the signal from the pic will be the same of the signal out of the opto.
A better idea, take signal from the flashlight for test.
A resistor of 150 Ohm between and the flash-out, from the pin flash-out you have now the same signal.