EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Circlotron on August 28, 2014, 01:13:08 pm
-
On a General motors LSx V8 they use coil-near-plug ignition. Each coil has it's own inbuilt driver cct and is fed a 5 volt trigger pulse from the ECU. There are actually two wires that go back to the ECU from each coil, a trigger and a ground. This ground wire is separate and distinct from the high current ground wire that goes to the engine metal. This is because the engine ground and the ECU ground can't be guaranteed to be at the same zero voltage because of voltage drops in the wiring loom that would otherwise cause erratic triggering of the coils.
http://i227.photobucket.com/albums/dd86/phlouse/ls2_coil_schematic.gif (http://i227.photobucket.com/albums/dd86/phlouse/ls2_coil_schematic.gif)
The circuit they use presumably doesn't have an optocoupler because the ECU ony has the equivalent of a 2k7 @ 5V pullup at the output and so won't really drive an opto very well. Tried this already. Short of a differential amp, what else might they use?
-
Transformer with a DC blocking capacitor? (So you get a bipolar pulse on the secondary.)
-
I was able to get these coils to work on an sr20det. As with the ground i tied these together and had no issues with false triggering so it would be interesting to see what the voltage drop is currently.
-
The trigger line to that coil is somewhat high impedance doesn't take much current to enable the coil, take the trigger to ground and it should fire, the 2.7k pull up is ensure sharp square pulses from the ECU. The high current ground serves both as a return for the trigger pass element (probably a IGBT device) and the return for the high energy side of the spark coil. Those 'integrated' coils can be fun to play with if done with caution.
-
Hello,
usually a ASIC like CK200/CK110 or similar is used:
http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf (http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf)
Main features:
- Current source (pull up / pull down) for the IGBT which is within the coil.
- Diagnosis of misfires which are transferred over coil and IGBT backwards to the ASIC.
with best regards
Andreas
-
Hello,
usually a ASIC like CK200/CK110 or similar is used:
http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf (http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf)
Main features:
- Current source (pull up / pull down) for the IGBT which is within the coil.
- Diagnosis of misfires which are transferred over coil and IGBT backwards to the ASIC.
with best regards
Andreas
Yes, I did leave out the finer details..
-
Hello,
usually a ASIC like CK200/CK110 or similar is used:
http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf (http://www.bosch-semiconductors.de/media/pdf_1/einzeldownloads/engine_management/CK200_Product_Info_200705.pdf)
Main features:
- Current source (pull up / pull down) for the IGBT which is within the coil.
- Diagnosis of misfires which are transferred over coil and IGBT backwards to the ASIC.
with best regards
Andreas
No, sorry, that one is not suitable. It has six outputs so it would be put inside the ECU, not inside the individual coils.
Maybe some kind of differential input comparator with a great enough common-mode range. The comparator reference input DC connected to the engine block along with the high current ground but capacitively AC coupled to the ECU ground so that any momentary differences in earth potentials is cancelled by the differential input of the comparator.
Ideally a schmitt optocoupler would be perfect but a H11L1 for example needs at least 1.6mA but the ECU only has a 2K7 pullup so is not strong enough to drive one of these optos.
-
The high current ground serves both as a return for the trigger pass element (probably a IGBT device) and the return for the high energy side of the spark coil.
Sharing the return paths is just soooo asking for trouble. I have done a few SMPS pcb layouts, and one of the golden rules is to make the connection to the mosfet source lead as short and direct as possible to minimise this sharing of current paths. Otherwise, when the mosfet turns on the source pulls up away from earth and reduces the gate-source voltage. Worse still, when the mosfet turns off, any ringing caused by this so-so connection to the mosfet source will modulate the gate-source voltage and probably cause your SMPS to blow up. This ignition coil situation is no different in principle.
-
The high current ground serves both as a return for the trigger pass element (probably a IGBT device) and the return for the high energy side of the spark coil.
Sharing the return paths is just soooo asking for trouble. I have done a few SMPS pcb layouts, and one of the golden rules is to make the connection to the mosfet source lead as short and direct as possible to minimise this sharing of current paths. Otherwise, when the mosfet turns on the source pulls up away from earth and reduces the gate-source voltage. Worse still, when the mosfet turns off, any ringing caused by this so-so connection to the mosfet source will modulate the gate-source voltage and probably cause your SMPS to blow up. This ignition coil situation is no different in principle.
G'Day,
So what you're saying is Terminal A on the LS2 Coil does not go directly to the engine block, along with the IGBT or other device that is encapsulated in the Coil on top of the engine? (ie: sharing reference)
Maybe I wasn't clear in my statement. I would have thought you would want the 'call it what you will - pass element or switching device' source lead to closely follow the levels of the returns of the coil to keep the differentials in check. Now, I don't mean the signaling/trigger or translator grounds that are referenced in the ECU, just the high power stuff only. Now I couldn't agree any more with you regarding tying those signal conditioning grounds to the high current grounds.. creating a nice little loop, that would be trouble in river city! |O Sometimes it is difficult to actually translate a meaning into text on these boards.
-
No, sorry, that one is not suitable. It has six outputs so it would be put inside the ECU, not inside the individual coils.
Sorry that was not clear for me from your description. I thought you wanted to control the coil.
Within the coil another ASIC is used like BIP373 / BIP172.
Mainly a IGBT with some protection circuit.
With best regards
Andreas
-
These coils and their internal driver circuit I was working on have been all done and finished for quite some time now. https://www.facebook.com/ProRaceEng/videos/1363546977068400/ (https://www.facebook.com/ProRaceEng/videos/1363546977068400/)