7MGTE Ignitor Operation & Alternative Ignition Coils

Nick M

Black Rifles Matter
Sep 9, 2005
8,894
38
48
U.S.
www.ebay.com
jetjock;1982758 said:
^ Huh. Everything I've seen from Toyota and aftermakert docs say dwell is set by the igniter.

I miss read what he wrote. Shame on me. The igniter sets it according to the 850 book, and I thought he was saying that.
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
Added charge times and current curves to the second post. Will add more observations when I have time. Maybe a few of you guys with more knowledge can take a look at the burn time/line and oscillations.
 

supraguy@aol

Well-Known Member
Dec 30, 2005
4,235
37
48
Atlanta
Forgot to msg you, Grim- got the coils back, thanks.

PS- Is it just me??? I can't open a single picture from post two.
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
Small success in getting main portion of the code working. About 25% there for a programmable ignitor for the stock TCCS. The biggest hurtle is coming up with a coil ON signal. Nothing out of the stock tccs can be used for this so it has to be calculated based on the previous ignition event.
The previous IGT period is measured via a timer, call this timeOld. Desired dwell is subtracted from timeOld which = the time to wait to turn ON the next coil. SO the coil turn ON signal is now controllable! Lots of testing to do through out the 'rpm' band but with this under control it should be possible!

Also have a few ideas on how to convert to a true sequential setup but not sure there is a benefit to that at this point.



View attachment NewFile1.bmp
Yellow is simulated IGT signal from TCCS/ECU, ignore exceptionally long duty cycle!
Blue is coil output signal or DWELL/coil charge time

This is with 2ms of dwell programmed at a simulated idle rpm.






View attachment NewFile2.bmp
Yellow = Interrupt triggering showing timing between coil start charge and then ignition event
Blue = simulated IGT signal from TCCS/ECU

interrupts triggered based on previous IGT signal (simulated) period. First one turns the coil on, second turns coil/ fires ignition
4ms dwell programmed

View attachment NewFile3.bmp
Yellow= Interrupt triggering showing timing between coil start charge and then ignition event
Blue= simulated IGT signal from TCCS/ECU

Interrupts with 2ms dwell.
 

supraguy@aol

Well-Known Member
Dec 30, 2005
4,235
37
48
Atlanta
Grim-

I was wondering - are you still planning to test the DH61 igniter, to see if it can be repinned to a 7mgte for use with the stock ecu?
Or do you think that it might be unnecessary, since you're making progress with a custom igniter?
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
The stock tccs wont work with the DH61 by itself. or any other ignitior that I know of. In order to us the DH61, we need some variation of the code I have been working on. My goal is to convert the tccs signals to a form that allows any 'dumb' ignitor to be used with the stock tccs. Then we can mix and match coils and ignitors as desired since dwell will be controlled too. This also allows us to use 'smart' coils as well such as the 1zz, LS2, Audi etc...

For 'dumb' coils the DH61 is a good candidate for the high current or power stage with my board feeding it the necessary signals. Tach and IGF are already there and I can tweak any of the other signals to fit our needs. Would really like to find an affordable 6 channel ignitor like the 2jz variant. The stock coils can still be used as well and a bit more output may be seen. Assuming the stock coils have it to give.

Take for example the ford COP, with a 6 channel ignitor + this project, we can do away with the series wiring. Coil dwell is precisely controlled allowing complete charge from idle to redline. Those ford coils should fully charge in less than 2 ms when properly driven with 1 coil per driver.

Thanks Rollus!
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
More success. One last timer to add in the code and then on to board layout! So far everything has bench tested perfectly. Hope to have a proto board setup for testing in parallel with the stock ignitor sometime next week.
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
Thanks guys, the idea is to produce these in small numbers once any future kinks are worked out. No eta, but sooner rather than later! Goal is to keep this project affordable.

Demo board testing is under way. Good results so far. Response to changes in timing is working as expected. Below are a few more scope images with it wired in parallel to the ignitor and engine running.

View attachment NewFile0.bmp
Output of my board in BLUE,
Stock coil output in YELLOW

Here you can see the programmed 7ms of dwell is dwarfing the stock ignitor to coil output. Probably not the best setting for most coils! Forogt I had that much programmed in. Was previously testing what happened at high rpm when the programmed dwell took up over 99% of the available time. Entire code takes about 15us to run, will take longer once battery offset is added in.

View attachment NewFile1.bmp
My board in BLUE
IGT shared with the ignitor in YELLOW

This is the board mirroring IGT at the time the TCCS calls for the coil to fire. The blue line is an internal signal I tied to a pin to make sure the interrupt was catching on every IGT negative slope. What ever coil was selected off of IGA/IGB would be getting pulled to ground to trigger the ignition event.

Trying out a few voltage regulators and decoupling setups on the MCU pins next. Then getting my hands on a DH61/62 and seeing how things really function!
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
not 100% sure. A lot of the costs will be in how the end product is mounted. An internally mounted version keeps cost down by using the ecu peripherals and wiring. An external version requires more PCB space and additional circuitry + some form of case and connector. I can build both styles but havent priced out each side by side. After I finish the board layout I'll have a better idea.

Are you guys wanting something that requires soldering/desoldering and is cheaper? or a bit higher cost that can be wired in by repining the stock connectors?
 

grimreaper

New Member
Jul 2, 2008
2,180
0
0
Dallas
Waiting on terminals to build a jumper harness. Picked up a 6 channel 'dumb' ignitor from a 1994-1995 camry. Tach and IGF are built in. Tested regulators and so far everything I have on paper or the bread board is working as it should. Adding all of this to my car and ecu to test with stock coils. There will be some more development on the code but as it stands, it should be good enough to run with out issues. Just fine tuning. limited on time with school and family lately.
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Dec 30, 2005
1,177
69
48
39
MKE, WI
A V6 ignitor? Like this?
http://www.ebay.com/itm/1994-96-Toy...Parts_Accessories&hash=item461412dfec&vxp=mtr

Are you intercepting the IGF, IGT, IGA & IGB from the ECU -> Feeding it into your custom PCB -> Modifying the signal to feed into a new ignitor.
I am happy to give you any help with this project if needed.
I've made a few one off custom boards/microcontrillers for different vehicles in the past.
Would be happy to see this project materialize into something very useful.
Doing some great work there Grim.

Curious what the BOM is for this board you are using :D