Original ignitor (TCI)

I made XJ's original TCI named TESLA. It can change advance angle on beat with controller, so you can setting advance angle as you like from 2000rpm to 11000rpm per 500rpm.

If you want to get pictures and more information, please check "aobadenso.blog" .

I'll upload here micro computer (PIC) program files, you can get codes if you want to make this.


tesla_std_3.asm　: main program
tesla_std_timer.asm : timer program
tesla_std_sw.asm : swich input program
tesla_std_lcd.asm : controller's display program
tesla_std_intr.asm : interrapt program
tesla_std_eewr.asm : EEPROM write/read program
tesla_std_dividor.asm : division program
tesla_std_calc.asm : calculate advance angle program
tesla_std_calactor.asm : caractor data for diplay
tesla_std_hed.asm : config hedder
macrop16f1938k.asm : macro program

And I'm trying to make injection drive project.
If complete that, I'll release program and wiring diagram here.

 

Interesting information. Maybe someone can layout a basic PCB which can be DIY build, then load Tanaka's code..

 

This is circuit diagram.
P11, P12, P14, P15 is connected to the controller's push switch. If switch is pushed, P11,12,14,15 is connected to GND.




Controller display sample



Controller



PCB

 

Wow! I can't quite see the schematics (too small) but I'm guessing teh 14-way connector is for the controller/programmer. Are you looking to sell these or make it 'open-source'.

 

Wow this is very nice.. are you selling any? i would like to try this.

 

a100man, I uploaded circuit diagram as attached files, so you can get original size there.

I distribute the TESLA program as open source, so anyone can be used freely for free .I'm glad if someone resurrected broken XJ by the TESLA.

 

Thanks - OK my electronics is weak but I see from left to right the two input channels form the pick-ups via AC coupling and filtering then some opto isolators
into the micro. The programming inputs are from P11, p12, P14 and P15 and I'm guessing the (to coil) outputs ar P13 an P10 via FETs. The power suppy is bottom left so what are P16-P21 for, the display ?

Cheers

 

Hi, a100man, I'm glad to your challenge for making TESLA.

1. Your question's answer
P9 and P16-P21 are for LCD display, it should connects to a display by wires. I'm using a LCD display called SC1602. Pin names of P9 and P16-P21 (R/W, E, RS...) are input name of SC1602.
And P5 and P7 are power supply for display. Controller can be driven with 5V power, but max output power of U1 (78L05) is not enough to drive LCD display.
So supply the 14V from the main circuit, controller have another 5V regulator to drive LCD.

2. Unnecessary parts
P6 is vacuum sensor input for revolution-vacuum advance angle map. And D5, R5, R8, R9 that connected AN3 of micro computer are reference voltage supply for A/D converter.
But I've not completed revolution-vacuum advance angle program yet, so these parts are unnecessary now.

L1,L2,L3,L4, C1,C2,C3, D2,D3,D4 are unnecessary if you make it simple. Because these parts are for noise reduction to stable drive, it can operate has been confirmed without these components.
L5 , L6 is also one which is installed for the same purpose, however you should install these parts because very effective for stable drive.

3. About parts
IC1,2 (PC817) are parts called photocoupler. AC input is translated to rectangular pulse output by this parts. If you cannot get same model(PC817) of parts, it can be replaced with specifications of component similar.
Q1,2 (2SC1815) are transistor to amplify the signal. It's OK in a typical small-signal transistor.
P8 is pin header to connect pickit (program writer for PIC).
Q3,4 (FGP3440G2_F085) are igniter dedicated IGBT. This is difficult to get, it may be replaced by another IGBT or FET. If you want to use the FET instead of IGBT here, the Gate of FET should be connected to GND by 10k ohm resistor.
The FET that I have to check the operation in this circuit is 2SK447.

Please contact me if you have any questions!

 

Thank you for the explanation of your circuit. I hope you receive further interest in your project. There are many motorcycles from 1980s that use this ignition system so hopefully more people can see this alternative solution. Good luck!

 

Neat project. Thanks for sharing, and thanks for offering up the details as well.

I have a couple quick questions/comments:

I can see that D3 and D4 would be unnecessary since they are on the regulated side of the power supply, but I'm not so sure I would skip D2. It's on the uncontrolled side of the regulator and who knows what's happening over there. Same with L3.

Also, I know the adjustable shunt regulator D5 is planned to be used when you incorporate the future vacuum sensor, but I'm not sure I understand the resistor value choices. It looks like you're trying to create a 5V reference for your A/D (2x Vref) from a 5V power supply and I'm not sure that works. What am I missing?

Did you actually put this thing on a bike and ride it? That would be awesome!

 

Thank you for your comment. I'm not electrical professional, so I think there is a part not optimal in this circuit.

D3, D4 are Transient Voltage Suppressor (TVS).
Certainly D3 and D4 are not required theoretically, but spark noise came from Vss or 5VD actually.
So I installed D3, D4 for protect a computer.
But now I think these parts are not needed because I've been driving more than 10,000km without D3 D4.

I'm sorry I couldn't understood your question about D2 and L3 because my English skill is poor.
L3 is common mode choke coil, and D2 is TVS for spark noise reduction.

I made mistake about shunt regulator D5. Cathode of D5 should be connected to 14V (Battery) through a 1.2k resistor.
I want to make correct 5V reference voltage by using shunt regulator, because the output of 5V regulator (U1) is not so correct using for A/D reference.

Engine start movie. My XJ has traveled more than 10,000 kilometers in a non trouble with a TESLA.
Wiring is dirty because I'm trying EFI project now...

 

My only comment about D2 and L3 was that I think they would be more important than D3 and D4. If you're getting spark noise on Vss and 5VD, it might have more to do with the PC board layout, or it might be coming back through Q3 and Q4. High frequency and high current is a difficult combination to deal with. I see that you've got the S of the FETs connected to a different ground than the Vss. Are they tied together at one point inside the Tesla, or are they tied together in the wiring harness somewhere?

That's excellent that you've traveled more than 10K kilometers on one of these ignitiors. And I saw a picture of a fuel injector sticking down into the carb body where the slide normally goes. Looks like a neat project there as well!

I'd love to talk about some of the electromecanical aspects of the timing. Did you test an original ignitor module first to see how it performed?

And don't worry about your English. Your English is 10000 times better than my Japanese. You're doing great and I appreciate you're taking the time to discuss in my language.

 

I could understand. Vss and AGND(S of the FET) tied together at one point inside the PCB. AGND and Vss are completely separated until it's connected to the reference GND. Based on my experience, removing spark noise by some parts(TVS, choke coil...etc) is too difficult. Then spark noise should be reduced at noise source, I think the most effective parts for reducing spark noise is resistor spark plug & cable & cap.

Thank you for watching my web sight. You can check here my XJ's EFI project (Sorry, Japanese only)
http://xj750e.web.fc2.com/

The original ignitor's timing is written in service manual. Spark timing at idle is BTDC 7 deg. Advance is start at 1300 rpm, and it increase linearly until 3300 rpm. The timing at over 3300 rpm is 37 deg.

XJ is sold very well at 80's, however it's not popular in Japan nowadays compare to SUZUKI GSX, KAWASAKI GPZ, HONDA CB-F etc. So I cannot talk about XJ's in Japan very often, I'm glad to talk with XJ lovers like you.

 

I'm not sure how closely the ignition actually matched the diagram in the manual.

I did test the original TCI on the bench and looked at the advance over the range, but I'm not sure how good my notes were. I'll see if I can dig out my notes the next time I'm in the shop.

 

Having run a few manual spark-advance engines (twist a grip, or move a lever to advance the spark; big fun in traffic I imagine), I can say that the advance curve dosen't have to be exactly the same as the ideal that is specified; it just needs to be ballpark (though the closer to ideal the better)

 

I made little bit mistake. Original spark timing is here based on XJ750E(5G9)'s manual.

It's interesting data. Where did you get this data? Is this a manual of US model? I've checked that original ignition timing is roughly the same as manual when I carried out the test of timing by using timing light.

I'm also the same idea as you. But I think ignition timing is important for enjoy riding. Original ignition timing is not optimized because computer technology was immature in order to realize the ideal ignition timing.

 

I don't have anything constructive to add to this conversation as this is all way over my head, but I just wanted to say I think it is really cool that you are doing this.

 

The ignition timing seems to be different as emissions standards change for the different markets, though it may be that the manuals have printing errors.

 

what compiler do you use?
MPLAB? Arduino?

 

I use MPLAB.