RE: New CDI - opensource project JMJ and Bigboat
 

The old program is good. Bad control of the SCR!!!!!http://www.rcuniverse.com/forum/js/f...spinnyeyes.gif

RE: New CDI - opensource project JMJ and Bigboat
 

I think that a 18F pic would be a better choice. for one reason most if not all have 2 16 bit timers. which will allow for better resolution and a rpm measurement down to lower rpms. another reason is the architecture is better.

RE: New CDI - opensource project JMJ and Bigboat
 

<span class="short_text" id="result_box"><span title="&Eacute;n maradok r&eacute;gi projektn&eacute;l." style="background-color: rgb(255, 255, 255);">I'll stay an old project. </span><span title="Meg pr&oacute;b&aacute;lom megjav&iacute;tani r&eacute;gi program." style="background-color: rgb(255, 255, 255);">We try to repair the old program.</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

<span class="short_text" id="result_box"><span title="Mutatni." style="background-color: rgb(255, 255, 255);">Presented. </span><span title="Mit k&eacute;ne jav&iacute;tani." style="background-color: rgb(255, 255, 255);">What should be improved.</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

while(PORTB &amp; 0x02) // wait for low pulse on Hall switch- wait for high pulse on Hall switch= Good!!!!
<span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="Itt meg cser&eacute;lni kimeneti jelet.">Here is the input signal change.</span></span>(low&gt;&gt;high)

RE: New CDI - opensource project JMJ and Bigboat
 

PORTB &amp;= 0xee; // sends a ground to the SCR gate- take ground off the SCR= Good!!!
delay_us(56); // 56 us delay
PORTB |= 0x11; // take ground off the SCR- sends a ground to the SCR gate= Good!!!
<span class="short_text" id="result_box"><span title="Itt meg cser&eacute;lni kimeneti jelet." style="background-color: rgb(255, 255, 255);">Here is the output signal change.</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

@nyemi: would you care to explain what is wrong with using a bootloader and the EEPROM, so that people who want the use the CDI can load new software without a special programmer, and adjust the settings (spark advance, among others) without needing to recompile and reload the code?

RE: New CDI - opensource project JMJ and Bigboat
 

<span class="short_text" id="result_box"><span title="&Eacute;n nem vagyok az &uacute;j ellen." style="background-color: rgb(255, 255, 255);">I am not against the new. </span><span title="Jav&iacute;tom r&eacute;gi programot." style="background-color: rgb(255, 255, 255);">Corrects the old program. </span><span title="Mert van ig&eacute;ny." style="background-color: rgb(255, 255, 255);">Because there is a need. </span></span><span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="&Eacute;n is v&aacute;rom &uacute;j projekt.">I also look forward to new projects. Time is too long.http://www.rcuniverse.com/forum/js/f...eeth_smile.gif</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

I'm ready. Link:www.elektroworld.info/ftp/index.php
Pin 6=RB0= CDI SCR <span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="kimenet jel">Output signal</span></span>.
Pin 10=RB4=<span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="TCI vagy stroboszk&oacute;p (CDI) kimenő jel.">TCI, or strobe light (CDI), the output signal.</span></span>
Pin 7=RB1=Hall input signal (wait for high pulse on Hall switch).http://www.rcuniverse.com/forum/js/f...wink_smile.gif

RE: New CDI - opensource project JMJ and Bigboat
 

Is this software for the original timer board with the 16f628 pic controller?

John

RE: New CDI - opensource project JMJ and Bigboat
 

<span id="result_box" class="medium_text"><span style="background-color: rgb(255, 255, 255);" title="Igen.">Yes.http://www.rcuniverse.com/forum/js/f.../thumbs_up.gif </span><span style="background-color: rgb(255, 255, 255);" title="PIC 16f628 vagy16f628A.">PIC 16f628 or 16f628A. </span><span style="background-color: rgb(255, 255, 255);" title="Ez a r&eacute;gi program (C k&oacute;d) &aacute;talak&iacute;tottv&aacute;ltozata.">It' s the old program (C code) modified version. </span><span style="background-color: rgb(255, 255, 255);" title="Nincs tesztfunkci&oacute;.">There is no test for function.
</span></span><span class="short_text" id="result_box"><span title="Itt lehet v&aacute;ltoztatni fordulat tartom&aacute;nyt,hogy j&oacute; legyen Excel t&aacute;bl&aacute;hoz." style="background-color: rgb(255, 255, 255);">
</span></span><span class="short_text" id="result_box"><span title="Itt lehet v&aacute;ltoztatni fordulat tartom&aacute;nyt,hogy j&oacute; legyen Excel t&aacute;bl&aacute;hoz." style="background-color: rgb(255, 255, 255);">CMCON = 0x07; // delete comparator module
TRISA = 0x00; // delete PORTA input
PORTA = 0x00; // PORTA low signal
TRISB = 0x02; // set RB1 for input, all others for output
PORTB = 0x10;
OPTION_REG = 0x07; // </span></span><span class="short_text" id="result_box"><span title="Itt lehet v&aacute;ltoztatni fordulat tartom&aacute;nyt,hogy j&oacute; legyen Excel t&aacute;bl&aacute;hoz." style="background-color: rgb(255, 255, 255);">You can change the speed range to be good Excel table.</span></span> 0x07=917RPM&gt;468750RPM, 0x06=1830RPM&gt;937500RPM
<span class="short_text" id="result_box"><span title="Itt lehet v&aacute;ltoztatni fordulat tartom&aacute;nyt,hogy j&oacute; legyen Excel t&aacute;bl&aacute;hoz." style="background-color: rgb(255, 255, 255);">TMR2 = 0;
PR2 = 255;
T2CON = 0x07; // start TMR2 counter 1:16
INTCON = 0; // clear INTCON


But if you do not succeed. Send your ignition curve for me and I will help you.http://www.rcuniverse.com/forum/js/f...ular_smile.gif
</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

Thanks Nyemi !!!!:D
We have needed this for a long time.. you are right!

John

RE: New CDI - opensource project JMJ and Bigboat
 

John,

Did that hex file I sent you work?

RE: New CDI - opensource project JMJ and Bigboat
 

Modified my comments.Link:www.elektroworld.info/ftp/index.php

RE: New CDI - opensource project JMJ and Bigboat
 

Thanks nyemi,

what is the hight delay created from the table? And again thanks for sharing your code. I know this is a little off topic,but Did you use CCS C compiler? If so how do you like it?

RE: New CDI - opensource project JMJ and Bigboat
 

Offtopic
Please Iceland, send us cash - no ash !

RE: New CDI - opensource project JMJ and Bigboat
 

Hy all,

I'm still struggling on the HV generation with the "old" board.
No problem to generate 500 V DC, but the raise time after the ignition pulse is terrible high. I need about 25 ms for 200V. The raise from 0 to 100V is quite fast (7 ms) but after the curve becomes very flat. I tried different frequencies adjusted by the R2 (nominal 680R) but without effect to the raise behaviour. Power supply: 5V.

I used a ring core of Ferroxcube N30 with 400/8/16 windings. Core dimensions 20/10/7.6 mm.

Anybody an idea why I'm so far away? I would like to reach 200V in less than 5ms.


Best regards

Uwe

RE: New CDI - opensource project JMJ and Bigboat
 

Hi Charlie..
Maybe today I'll get time to program a chip and test it out. Working 3 jobs right now to get caught up from being laid off for 6 months.
Hopefully Nyemi's code will solve the wrong pulse from the hall effect, which caused problems with timing (static) adjustments to already installed mounts for the sensor. The old code sent a pulse as the magnet approached the hall.. the new code waits till the pulse turns off to send the trigger which is the way Synchro Spark works. This was why I went back to the Synchro Spark to test my cdi boards. If in fact the new code works correctly, ALL problems will be solved and everything will be perfect, and as a plus, we won't have to redesign and build new timer boards!
This is what we wanted all along and THANKS Nyemi for providing us with a solution to our problems and ending a very long wait for an end to the dilema, only requiring value changes for the advance curves. All my present cdi and timer boards work great, just requiring the program changes which Nyemi has made. Looking at his advance values in the code with his edited comments makes perfect sense to me..

Thanks again Nyemi ! And thanks Charlie for compiling it to hex for us.!!! I feel as if a huge weight has been lifted from my shoulders, and I can now concentrate on converting engines and possibly flying this year.

John

RE: New CDI - opensource project JMJ and Bigboat
 

Hello Aerodactyl..
Looking at your ferrite core size you are using , I think your problems come from the physical size of the core.. Using that size and keeping the same number of windings will make for a slow recovery time that you are experiencing. It is like simply not having enough windings on the core to produce the amperage required for a quick recovery after the spark. If you can obtain a smaller core and keep the windings at the number you are now using, I think your problem will be eliminated. The size I use is about 3/4 the size of yours... .5 inch OD - .25 ID inch ID and .1875 inch thick.

John

RE: New CDI - opensource project JMJ and Bigboat
 

Ihaven't chosen my core yet - still haven't gotten that far. I'm the fellow who posted the 4-channel circuit on the previous page, and I'll be feeding 4 capacitors from the transformer, with a 12v feed. At 18500rpm Iwill basically have 37000 sparks per minute, or one every 1.6msec (6..5 msec per capacitor).

Am I liable to have any issues?

RE: New CDI - opensource project JMJ and Bigboat
 

John,

Yes this could be the reason why. Unfortunately I missed all the magnatic issues in school 30 years ago :eek:;)
I have a smaller toroid available with about 3/4 of the actual size. I will try this one again [:@] Damned, another 2 hours for the bloody transformer. It makes me mad [&o][&o]

RE: New CDI - opensource project JMJ and Bigboat
 

Hi nyemi,

I don't get any spark when using your code. I think the trigger logic is backwards from what I need. I need a low going pulse. I will play with the code and see if I can get it to work.

John,

I think you will see the same problem.If you get yours to work with nyemi's code let me know.



Charlie

RE: New CDI - opensource project JMJ and Bigboat
 

Charlie, my cdi boards can use either by switching the resistors per the schematic Rob did from it. It seems to work as good in static testing as a Synchro Spark so far.. Haven't put on the engine yet to run.


John

RE: New CDI - opensource project JMJ and Bigboat
 

<span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="K&ouml;sz&ouml;n&ouml;m a dics&eacute;retet.">Thank you for your compliment.http://www.rcuniverse.com/forum/js/f...ssed_smile.gif
</span></span><span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="K&eacute;rd&eacute;sek:">Questions:</span></span>
"what is the hight delay created from the table?" TMR2 8bit <span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="sz&aacute;ml&aacute;l&oacute;">Counter. </span></span><span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="T&aacute;bl&aacute;zatban a maxim&aacute;lis &eacute;rt&eacute;k.">Table, the maximum value.</span></span><span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="sz&aacute;ml&aacute;l&oacute;">=255 x16us=4,08mS~ 917RPM= 22,6&deg; maximum delay.
"</span></span>Did you use CCS C compiler?" <span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="Nem.">No. </span><span style="background-color: rgb(255, 255, 255);" title="Osho PIC SimulatorIDE.">Osho PIC Simulator IDE. </span><span style="background-color: rgb(255, 255, 255);" title="MPLAB v8.10.">MPLAB v8.10. </span><span style="background-color: rgb(255, 255, 255);" title="&Eacute;n ezekethaszn&aacute;lom.">I use them.
"</span></span>I don't get any spark when using your code." <span id="result_box" class="short_text"><span style="background-color: rgb(255, 255, 255);" title="Akarom l&aacute;tni &aacute;ramk&ouml;r&ouml;d.">I want to see circuits. Then I can help.</span></span>

RE: New CDI - opensource project JMJ and Bigboat
 

Hi Charlie..
Nyemi's code works perfectly with my cdi boards and the timer boards! I am so glad it works.. Thanks again Nyemi !!!:D:D

Just ran the 25cc with it and also my Cox 1.4...works great..timing is right on. No more jumping around with the timing light! Throttle transistions great and I actually picked up about 300 rpm in the deal. Idle is now able to go lower than before and runs steadily without a single misfire. I am getting 7300 top end and 1450 idle on the Cox 1.4...with a 15-8 Master Airscrew wood prop.

You must use the timer board in it's unaltered state hooked direct to the CDI board and it works. Switch the resistor as per the diagram of the CDI with Hall effect. I also tried Rob's design board and same results.. just about perfect for my engines. The pulse from the hall effect is now correctly applied..as the south pole of the magnet is leaving the face of the hall effect I get the spark. Now it works correctly!

Just for a test, I plugged in my Synchro Spark to run the engine....no noticable difference in timing with the light and just as smooth as Nyemi's code. There is no test function in the code, but if you make one of my 3600 rpm simulator boards with the 555 chip and plug it in, it also works fine.

John