Coil arcing w/o engine running
09-19-2008, 12:22 AM
#1
Coil arcing w/o engine running
Help needed.
Just put a BP4W head 1.8 into the 91. It had been running fine with the 1.6, MS PnP and no AFM.
Symptoms when I turn the key to the run position, not cranking:
Coils arc to the brake booster hard line
Can not connect to the MS
No engine check or brake light on dash, only charge light
What I changed:
1. Headlights & retractors removed. Harness still there bundled up.
2. Extended CAS harness, using matching gauge wire, heat shrink tube and soldered connections. Main power wire is one gauge bigger than the others.
3. Drilled/tapped hole in 1.6 engine hoist anchor for ECU ground. All WBo2 & MS grounds on separate lugs to the same ground location.
4. Two wire ground that is about 2" closer to ECU was grounded to manifold on stock engine so I did the same here. I used the aluminum EGR block off plate bolt.
5. FM 1.8 conversion kit minus their CAS extension wires.
6. No place on the BP4W head for the driver side ground strap so I attached it to the flex plate.
7. Rattle canned the coils a bit after wire brushing some surface rust off. Terminals are clean.
8. Tapped manifold for IAT. Made long extension and ran it along firewall. Cut relevant wires just before AFM connector and soldered in extension so those two prongs in AFM plug are now not connected to anything.
Didn't actually try to start it but obviously something is wrong with my wiring. I'm assuming I have a bad ground somewhere. The 94-97 seem to have another key ground between the TB and head but the 90-93 don't. Is there another ground I'm missing somewhere? If the flex plate isn't a good enough ground (probably not) I'll have to solder a bigger lug to fit over one of the header bolts.
Next step is to bust out the shop manual/madracki file and start checking all the grounds with a VOM unless someone sees something I missed.
Ideas?
Just put a BP4W head 1.8 into the 91. It had been running fine with the 1.6, MS PnP and no AFM.
Symptoms when I turn the key to the run position, not cranking:
Coils arc to the brake booster hard line
Can not connect to the MS
No engine check or brake light on dash, only charge light
What I changed:
1. Headlights & retractors removed. Harness still there bundled up.
2. Extended CAS harness, using matching gauge wire, heat shrink tube and soldered connections. Main power wire is one gauge bigger than the others.
3. Drilled/tapped hole in 1.6 engine hoist anchor for ECU ground. All WBo2 & MS grounds on separate lugs to the same ground location.
4. Two wire ground that is about 2" closer to ECU was grounded to manifold on stock engine so I did the same here. I used the aluminum EGR block off plate bolt.
5. FM 1.8 conversion kit minus their CAS extension wires.
6. No place on the BP4W head for the driver side ground strap so I attached it to the flex plate.
7. Rattle canned the coils a bit after wire brushing some surface rust off. Terminals are clean.
8. Tapped manifold for IAT. Made long extension and ran it along firewall. Cut relevant wires just before AFM connector and soldered in extension so those two prongs in AFM plug are now not connected to anything.
Didn't actually try to start it but obviously something is wrong with my wiring. I'm assuming I have a bad ground somewhere. The 94-97 seem to have another key ground between the TB and head but the 90-93 don't. Is there another ground I'm missing somewhere? If the flex plate isn't a good enough ground (probably not) I'll have to solder a bigger lug to fit over one of the header bolts.
Next step is to bust out the shop manual/madracki file and start checking all the grounds with a VOM unless someone sees something I missed.
Ideas?
__________________
Last edited by emilio700; 09-19-2008 at 12:51 AM.
Reply
-1
-1
09-19-2008, 01:04 AM
#2
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,026
Total Cats: 6,592
Boy, that's a dizzying perspective. 
In the 1.6 engine harness, there are two of the two-wire ring grounds in the same general vicinity. I believe that one bolts down to the head itself and the other to the back of the IM, or something like that. Are both of these twin grounds secured? I see only one in your picture- with the yellow arrow pointing at it.
edit: Is that it behind the new ring terminal on the hoist point?
edit 2: Is the rear ground strap (from the PPF to chassis) intact? Perhaps you could try supplementing the front ground strap by just clamping a jumper cable between the head and the batter (-) terminal or some convenient point on the chassis?
In the 1.6 engine harness, there are two of the two-wire ring grounds in the same general vicinity. I believe that one bolts down to the head itself and the other to the back of the IM, or something like that. Are both of these twin grounds secured? I see only one in your picture- with the yellow arrow pointing at it.
edit: Is that it behind the new ring terminal on the hoist point?
edit 2: Is the rear ground strap (from the PPF to chassis) intact? Perhaps you could try supplementing the front ground strap by just clamping a jumper cable between the head and the batter (-) terminal or some convenient point on the chassis?
Reply
0
0
09-19-2008, 02:30 AM
#3
Boy, that's a dizzying perspective.
In the 1.6 engine harness, there are two of the two-wire ring grounds in the same general vicinity. I believe that one bolts down to the head itself and the other to the back of the IM, or something like that. Are both of these twin grounds secured? I see only one in your picture- with the yellow arrow pointing at it.
edit: Is that it behind the new ring terminal on the hoist point?
In the 1.6 engine harness, there are two of the two-wire ring grounds in the same general vicinity. I believe that one bolts down to the head itself and the other to the back of the IM, or something like that. Are both of these twin grounds secured? I see only one in your picture- with the yellow arrow pointing at it.
edit: Is that it behind the new ring terminal on the hoist point?
Yup. Did the steel wool/alcohol thing on the one near the diff too.
I'll try will soldering a larger diameter lug on the strap under the heater core pipes and running it to one of the header studs. The connection on the chassis side of that one, I didn't touch so maybe I'll clean that up while I'm at it.
So anyway, you think it's a ground issue or could the symptoms be something else? If it's just a matter of grounding the head better I'll just get medieval on it. Thanks for chiming in so quickly. In theory, I'm supposed to have this thing on a trailer going to an event tomorrow afternoon and I'll have to put fenders, hood, bumper on and build a base map..
__________________
Reply
-1
-1
09-19-2008, 02:41 AM
#4
Elite Member
Join Date: Mar 2008
Location: Enschede, NL
Posts: 2,053
Total Cats: 12
Arcing is of course not good, but you also say you cannot connect to the Megasquirt. Is it a known-good MS? Can you check in another car or do you have another MS around? Or a stim? Is is the PNP version?
Hope you make it to the event!
Hope you make it to the event!
Reply
0
0
09-19-2008, 11:39 AM
#6
Junior Member
Join Date: Sep 2007
Location: North of Atlanta
Posts: 78
Total Cats: 0
Looking at the wiring diagram, the 10a gauge fuse isn't tied to the MSPNP at all, so thats going to have to be a wiring problem somewhere.
It also looks like the coils ground through the bracket to the head, I saw you mention that you rattle canned the bracket, did you by chance paint over the mounting flanges? I'm not sure what it would do if it could ground through the head, but it would look for somewhere else to ground to.
It also looks like the coils ground through the bracket to the head, I saw you mention that you rattle canned the bracket, did you by chance paint over the mounting flanges? I'm not sure what it would do if it could ground through the head, but it would look for somewhere else to ground to.
Reply
0
0
09-19-2008, 12:37 PM
#8
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,026
Total Cats: 6,592
No, no, no. The coils do not ground through the bracket. They will work just fine while sitting on a Teflon plate.
If we're talking about the 1.6 coils: On the primary winding you have +12 going to one side of each coil, and the other side goes to the ignitor, which closes it to ground to charge the coil. On the secondary, both sides of the winding are connected to spark plugs. The secondary (HV) current path can be considered to be going out one coil, through a spark plug, across the head to the other spark plug, across that plug, and back into the secondary. This HV is not referenced to any other ground.
On the 1.8 coils (both NA and NB) you have a black ground wire that goes from each coil to ground at a common point (probably a ring terminal). This ground is there to make the igniter section of the coil work, and carries the primary current during dwell. The secondary path is the same as 1.6.
Which coils are you using?
This spark happens only once, at key on, right? Could you have one bad plug or wire on the engine? I'm thinking that if the secondary circuit is interrupted in one place, then the chassis would become a viable return path to complete the circuit.
(shifting gears)
The fact that you're also having other issues makes this complicated. Frankly, my approach would simiply be to attack the car with a multimeter, checking ground continuities and +12 supplies.
As an FYI, you mentioned the check engine light. Unless DIY has done some special workaround just to light it up for a second at key-on, the CEL should be dark. Megasquirt does not have any provision for lighting the CEL, be it an MSPnP or a regular MS.
Man, I wish I was still out in CA. I'd drive up and we'd have this bitch working in no time...
If we're talking about the 1.6 coils: On the primary winding you have +12 going to one side of each coil, and the other side goes to the ignitor, which closes it to ground to charge the coil. On the secondary, both sides of the winding are connected to spark plugs. The secondary (HV) current path can be considered to be going out one coil, through a spark plug, across the head to the other spark plug, across that plug, and back into the secondary. This HV is not referenced to any other ground.
On the 1.8 coils (both NA and NB) you have a black ground wire that goes from each coil to ground at a common point (probably a ring terminal). This ground is there to make the igniter section of the coil work, and carries the primary current during dwell. The secondary path is the same as 1.6.
Which coils are you using?
This spark happens only once, at key on, right? Could you have one bad plug or wire on the engine? I'm thinking that if the secondary circuit is interrupted in one place, then the chassis would become a viable return path to complete the circuit.
(shifting gears)
The fact that you're also having other issues makes this complicated. Frankly, my approach would simiply be to attack the car with a multimeter, checking ground continuities and +12 supplies.
As an FYI, you mentioned the check engine light. Unless DIY has done some special workaround just to light it up for a second at key-on, the CEL should be dark. Megasquirt does not have any provision for lighting the CEL, be it an MSPnP or a regular MS.
Man, I wish I was still out in CA. I'd drive up and we'd have this bitch working in no time...
Last edited by Joe Perez; 09-19-2008 at 01:24 PM.
Reply
0
0
09-19-2008, 01:17 PM
#9
1.6 coils. still using the ignitor (waiting on Savingtons COPS). Same set that was in the car. I have piles of coils around but wanted to reduce variables. The arcing is continuous, random at 1~2hz. No wires, arcing from coil upwards to brake booster hard line.
I couldn't remember if the CEL came on with the 1.6 or not so I sorta ignored it. Just including anything I can think of to help diagnose it.
Just got in to the shop, I'll redo the ground strap on the hotside and see if that has any effect.
If no joy then the multi meter comes out. That's bums me out because I don't have the knowledge to intuively work my way through the system quickly. It's poring over the schematics and FSM deciphering what to me is, a different language. What takes you 15 minutes will take me all day :(
If you were in Cali I'd be hiring you to work on a few different projects I have on deck.
I couldn't remember if the CEL came on with the 1.6 or not so I sorta ignored it. Just including anything I can think of to help diagnose it.
Just got in to the shop, I'll redo the ground strap on the hotside and see if that has any effect.
If no joy then the multi meter comes out. That's bums me out because I don't have the knowledge to intuively work my way through the system quickly. It's poring over the schematics and FSM deciphering what to me is, a different language. What takes you 15 minutes will take me all day :(
If you were in Cali I'd be hiring you to work on a few different projects I have on deck.
__________________
Reply
-1
-1
09-19-2008, 01:23 PM
#10
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,026
Total Cats: 6,592
Get some plugs and wires hooked up. Seriously.
The design of the driver circuit on the MSPnP causes the ignition coils to discharge once immediately at powerup. It's not really a "feature" per se, more like a hardware bug. Don't feel bad, most of us use the same design on regular DIY MegaSquirts as well.
Anyway, here's my ultra-far-out theory:
1- You power up, with no plugs or wires.
2- The MS boots up, and causes an ignition event.
3- Since there are no plugs or wires, this event causes an arc that discharges through chassis and causes a huge spike in potential throughout the harness.
4- This causes a processor reset.
5- Goto 2.
What happens if you power up with the ignitor and coils completely disconnected? Can you talk to the MS then?
The design of the driver circuit on the MSPnP causes the ignition coils to discharge once immediately at powerup. It's not really a "feature" per se, more like a hardware bug. Don't feel bad, most of us use the same design on regular DIY MegaSquirts as well.
Anyway, here's my ultra-far-out theory:
1- You power up, with no plugs or wires.
2- The MS boots up, and causes an ignition event.
3- Since there are no plugs or wires, this event causes an arc that discharges through chassis and causes a huge spike in potential throughout the harness.
4- This causes a processor reset.
5- Goto 2.
What happens if you power up with the ignitor and coils completely disconnected? Can you talk to the MS then?
Last edited by Joe Perez; 09-19-2008 at 02:44 PM.
Reply
0
0
09-19-2008, 03:16 PM
#12
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,026
Total Cats: 6,592
Ok, just as an FYI concerning the other problem you mentioned- the 10A METER fuse goes to a lot of different places besides the instrument panel:
1- It directly powers the backup lights.
2- It powers the turn signal / hazard warning control unit.
3- It powers the "Sound Warning Unit" (the buzzer that annoys you when you have the key in the ignition and the door open or your seatbelt off).
4- It powers the main cruise control switch.
5- It powers the whole A/T control system and part of the shift interlock system (which hopefully your car never had).
6- It powers the control section of the power antenna relay.
1- It directly powers the backup lights.
2- It powers the turn signal / hazard warning control unit.
3- It powers the "Sound Warning Unit" (the buzzer that annoys you when you have the key in the ignition and the door open or your seatbelt off).
4- It powers the main cruise control switch.
5- It powers the whole A/T control system and part of the shift interlock system (which hopefully your car never had).
6- It powers the control section of the power antenna relay.
Reply
0
0
09-19-2008, 07:40 PM
#13
Changed a few things.
Apparently I removed the hotside ground strap when I touched up the engine bay paint so there was paint underneath. Fixed that and attached it to a manifold stud.
bat (-) terminal was not torqued. Friend who helped pull old motor must have done that.
Turns out the PnP won't communicate if the "fuel inj" relay isn't in. Doh!
Started after cranking for about 2 sec.
Thanks!
Apparently I removed the hotside ground strap when I touched up the engine bay paint so there was paint underneath. Fixed that and attached it to a manifold stud.
bat (-) terminal was not torqued. Friend who helped pull old motor must have done that.
Turns out the PnP won't communicate if the "fuel inj" relay isn't in. Doh!
Started after cranking for about 2 sec.
Thanks!
__________________
Reply
-1
-1
Thread
Thread Starter
Forum
Replies
Last Post
eddy1128
Miata parts for sale/trade
8
12-10-2015 05:34 PM
Motorsport-Electronics
ECUs and Tuning
0
09-05-2015 08:02 AM
