Tell me more of this site!
Tell me more of this site!
In reply to Pete. (l33t FS) :
It's a drag strip on the western side of Cincinnati called Edgewater. https://www.motorsportreg.com/events/may-4th-be-with-you-rallycross-edgewater-sports-park-scca-western-ohio-977967
That's what I get for not being on Facebook, I guess
They changed the 1st date to May the Fourth, when Star Wars cosplay people are stupid busy. I'll be at the next one in June. Yesterday's test was supposed to be a few weeks ago, but Edgewater was UNDERwater.
They're trying to run an non-SCCA autocross the same dates now, and we're curious how that's going to go. I'll be looking for your feedback!
I brought the car into the shop this morning to take a look and make sure everything was ready for the event on Sunday. I wanted to figure out exactly what relay I had in there so that I could replace it, so I ordered up a box of 5 or 6 of them from Amazon. I'm not really sure that I need a 60A relay, but that's what I went with, just to be sure.
I put the car up in the air and gave it a once-over. One of the things I wanted to look at was the fuel filler and the fuel level sender. The fuel filler has been banging around loose for a while, so I pulled it through the quarter panel and got it secured. I don't think it's going to be a long-term fix, though, because it seems like the plastic has deformed enough that it wants to slip back through. We'll see how it does - I may need to either secure a new one or find some way to keep it in place. I also would like to know how much fuel I have on board at any given time, so I took a quick look at the connections to the pump and sender. I'm not 100% sure what's going on back there, but it looks like there's a blade connector that doesn't have anything attached to it, so I'm wondering if I'm missing a ground or something. I'll have to look at some diagrams and see what's up. The sender will just randomly give me either full or empty, but I don't know that it's ever actually correct. I got distracted from that job, though, because in the front, I saw this:
I don't know if that was always there from the original crash damage, but I sort of doubt it. I feel like even I would have noticed that at some point in the last 20 years. I had some 14ga steel blanking panels from an Hitachi rack that I salvaged from the data center garbage about a year ago and used it to make this:
Then with some hammering and whatnot, I started welding that in:
It's not quite finished yet, but I should be able to wrap that up tomorrow and I'm hoping that will hold things together.
I also replaced all my vacuum caps, although I understand that I can get some silicone ones that will be less prone to failure so I'll look into that later.
And as you now know, you have extra time... (I might actually be able to make June 1.)
In reply to Jerry :
Yeah, I guess we're supposed to get an inch of rain on Saturday into Sunday so it's not too surprising. Hopefully you'll be able to make the new June 1st date!
Even though tomorrow's event got rescheduled for next month, I figured I would go ahead and get back at it since I had the time set aside anyway. Very unlike me, I know. I used a torch to heat up the patch panel and then pounded it with a mallet to get it to conform to the shape of the strut tower so that I could finish welding it in.
Yeah, the welds are ugly. But with a tire in there, you'll never really be able to see it. I probably could have done a much better job if I would have pulled the strut out, but I thought I was in a bit of a hurry to get the job done and getting these particular struts out is kind of a pain in the rear. I think it's because I have them extended a bit, so to get the control arm to drop out of the way in order to pull the strut out without destroying the CV boot, it takes a serious amount of prybar work.
Anyway, I hit it with a wire brush and then put a little paint on it on each side to keep it from getting rusted up. I might get really motivated at some point and pull the engine back out so I can open up the transmission and fix the VSS drive gear and if I do, I'll probably put a corresponding patch on the inside of the strut tower, but there's not enough room to work there with the engine in place.
Once that was done, I wanted to try to attack the janky shifts. The 1-2 shift is almost impossible when you're under power, so I've been launching in 2nd. The 2-3 shift is also not always so great and on the street I find myself working pretty hard to find 3rd and occassionally 5th. It could be a problem with the transmission itself, but before I worry about installing my spare, I figured I'd try to make sure the Maddog short shifter was properly adjusted.
The little 8mm bolt that's sticking out down there is the crossover adjuster which is the standard Neon thing. I popped that loose and then put a 1/4" drill bit into the adjuster arm on the transmission to center that end. I cleaned up the mechanism on the the shifter as best I could and shoved some grease in there, then loosed the bolts that hold the shifter to the bracket and that lock in the front rod brake. Once both rod brakes were loose, I cinched up the rear and set the front brake, then tightened the front, loosed the rear, and set the rear brake. Then I got the shifter tightened down, centered it, and set the crossover.
When I get back, I'll put everything back together and see if that takes care of the shifting issues. If that doesn't fix it, I may grab a stock shifter assembly from one of my bins o' Neon parts and see if that works okay, just to rule out shifter adjustment as a problem.
Well, that escalated quickly....
I popped over to the shop this morning to go ahead and replace that relay that's acting up. But before I could do that, I thought that maybe it would be nice to make a little bracket to mount that relay on. And if I were to make a bracket for that relay, maybe I should mount the fan relay on a bracket as well. And if I were going to re-mount the fan relay, why not also replace it with one of these new ones? But this is the spot that I wanted to place that bracket:
I've been meaning to clean that mess up for a while. The closer you look, the worse it gets.
I started by de-pinning the ECU connector and removing it from the car
Clearly I was drinking the day I put this butt connector on. I have no idea what the hell I was thinking...
I also opened up the power distribution center and removed some of the wires from it for systems that are no longer in use. After unwrapping the harness as best as I could, I pulled a number of wires back through the firewall. Once that area was cleaned out, I decided to bend up a bracket that I wound up tossing back into the metal scraps bin. Instead of that, I decided to make a bracket that would stabilize the PDC. It was originally part of the battery box, but since the battery is in the trunk now, the plastic mounting bit isn't very solid. I used a bit of steel to make a bracket that would give the PDC a firm connection, create a channel for wiring, and allow me to mount the relays.
There's still a significant amount of wiring to clean up. I'm not entirely sure of the best way to go about it but I'm trying to pull out anything that isn't used and route what's left a little more cleanly.
The previous wiring was pretty ugly, but it had one very large benefit: It worked.
In the process of trying to clean up and remove some of the unused wires and re-route and tidy up the ones that are still in use, I have created a situation where the fuel pump relay will remain on when the ignition is off. I'm in the process of diagramming out exactly what I have so I can understand it better, but I strongly suspect that in the process of removing wires, I inadvertently clipped a ground somewhere. To replicate the problem, I take these steps:
ASD = Neonspeak for "Auto Shutdown Relay". Basically, it is the +12V feed for the ignition coil, injectors, alternator (maybe in my case), and some things that have been missing from the car for a long, long time.
I believe similar behavior occurs when the ground source for the relay trigger is disconnected as well. The difference being that the fuel pump doesn't run on initial key-on, but despite the coil trigger only having a +12 connection and nothing connected to the GND side, it will click on and run the fuel pump. I need to verify that, though.
If I unpin the ASD output circuit from the relay and power it directly from any +12 source (with an inline fuse, of course), the function of the fuel pump is normal. That is with the ASD relay removed.
Tomorrow, I'm going to go back over and work on diagramming the circuit and I have a couple more tests to do:
I'm also sort of wondering why I had my separate relay in place to begin with and I'm considering removing that and just using the original relays for the ASD and fuel pump in the PDC. One of the weird quirks of how I'm set up currently is that I'm relying on Megasquirt to trigger my relay which then triggers both the ASD and fuel pump relay. But the MS trigger I'm using is the fuel pump control. That means that the ASD system (again, injectors and coil basically) comes on for a few seconds at key-on and then turns back off until MS sees about 300 RPM. Which means a slightly weak battery makes it very hard to start the car because there's no fuel or spark except at 300+ RPM. So I might pull my relay out of the system, use MS to control the fuel pump relay directly, and then wire the ASD relay to be powered when in ST-RUN, like it was originally designed. I really don't know WTF I was thinking when I wired this particular relay into the system.
But I'm not touching that at all until I figure out what I broke.
In reply to wae :
Ah, I remembered why that relay exists. The ASD relay is triggered by being grounded by the PCM whenever the PCM is powered up. With the PCM gone, I didn't have anything that could ground the ASD, so I used the fuel pump trigger from MS.
So I'll go ahead and change things up so that the MS fuel pump control actually controls just the fuel pump. Then I can either use what I've dubbed my "main relay" to connect the ASD trigger to ground when the ST-RUN circuit is hot. Or, I can pull the ASD output wire (dark green/orange) out of the PCM, and connect it to pin 87 on my "main relay", connect the red/white wire from the PDC - which is the ignition/fuel injection circuit on fuse 21 - to pin 30, then ground pin 85, and connect 86 to something that is powered in ST-RUN. Basically, just replace the stock ASD with my own.
Could be a Megasquirt thing? All things that the Megasquirt controls (fuel pump relay, injectors, solenoids) must be cut from power when the ignition is off.
That's usually where a bunch of added relays comes in.
wae said:
- Does re-installing the ASD relay in step 9 above cause the fuel pump to come back on?
- With the ASD output circuit powered directly from the horn circuit, what happens if I leave the ASD relay in place?
- Verify that I get fuel pump and/or ASD relay function when the relay coil circuit is not completed
Yes, re-installing the ASD relay, even with the ASD output wire disconnected, does cause the fuel pump relay to activate.
With the fuel pump relay removed, the ASD relay remains off until I turn the ignition to run. When the fuel pump relay is installed, the ASD relay clicks on with the ignition in off.
When the ASD relay is installed, the coil pins on the fuel pump relay socket draw .14 amps. With the ASD relay removed, it drops to zero.
I think that this means that whatever is causing the issue is not on the load side of either relay but on the control side. And that means it's something on the +12 side, since I've removed all three relays, jumped pins 85 and 86 on the fuel pump relay, and then put a test light across pins 85 and 86 on the ASD relay socket. That means that the load side of the relays are completely disconnected from anything and the only thing to which pin 85 on each socket is connected is pin 85 on the other socket.
That's telling me something really important, but I can't quite get my brain to focus in on that. Time for some thinking juice.
Pete. (l33t FS) said:Could be a Megasquirt thing? All things that the Megasquirt controls (fuel pump relay, injectors, solenoids) must be cut from power when the ignition is off.
That's usually where a bunch of added relays comes in.
Theoretically, the Neon PCM is set up the same way and I can just use the ASD relay instead of installing relays for megasquirt to turn power on to the injectors, coil, and O2 sensors, etc. I still don't know why I put a relay in place to connect the fuel pump and ASD relay triggers to ground when megasquirt grounded the fuel pump relay output pin on the DB37, though. Instead of connecting the fuel pump and ASD relay triggers to each other and then to a tertiary relay, I could have just connected them directly to the fuel pump ground pinout...
That, however, was working before I cracked into it, so it might have been redundant -- and the cause of an occasional no-start condition -- but it's something that I can embetter once I've figured out what I broke.
If, with the ignition off, you have 12v on the coil side of any relay controlled by Megasquirt, that can keep the Megasquirt awake.
All inputs and outputs should be electrically dead when you shut off the ignition.
In reply to Pete. (l33t FS) :
Maybe that's why I put that relay in place, because the ASD circuit is always hot whereas my "main relay" is only hot in ST-RUN. The fuel pump relay, however, is only hot in ST-RUN, so I could connect MS to that directly and then leave out the Mopar ASD relay and just use my own.
In reply to wae :
Ah-HA! Okay, I have restored normal function to the fuel pump and ASD triggers, so I know where to start looking. This is the way the PDC is wired, with those dark lines representing what I'm referring to as bus bars that are inside the PDC itself.
The fuel pump relay uses connectors 52, 53, 55, and 56 while the ASD relay uses 57, 58, 60, and 61. Connector 71 is the DG/WT wire that comes from the passenger compartment fuse box. That's the positive feed for the fuel pump relay coil, is fused, and only has power in ST-RUN. If I disconnect that DG/WT wire from position 71 and use the +12 feed for the horn relay coil - connecting position 62 to position 33, which is part of the same bus bar on which position 56 resides - the fuel pump and ASD relays all work like they should.
That tells me that what I have screwed up is something that lives on the fuse #10 circuit. Time to go trace that one back and see what I've done there.
Damnit. It escalated again.
If I'm going to be tracing and cleaning up wires... It's easier with the dash out of the way. And if the dash is out, I can figure out why the blower stopped working. And maybe look at why the tachometer needle came off. And maybe put the glovebox back in. And maybe get the gauge backlighting to work right. And maybe properly mount the fan switch. And maybe connect the fan relay to megasquirt so it doesn't have to be manual. And maybe clean up all the wires going into megasquirt. And maybe connect the horn button.
It'll be a miracle if I have this back together for June 1st.
As long as you don't start 3D printing circuit boards for Arduino controlled gauges, you'll be fine.
There were a bunch of wires that I managed to run around the under-dash, so in order to take it out, I had to make a few cuts.
Even the knock sensor module had to come out.
All of that and I was finally able to wiggle out the rest of the dash.
I'm starting with this:
It doesn't look like it, but this is actually an improvement after a couple hours' of work:
When I originally installed Megasquirt, I made my own harness to keep costs down for the Challenge. Later on, though, I bought a pre-made harness that had more and better wire colors. That original harness was still in place behind the dash, so the lowest bit of fruit to pick was to pull those wires out. There were a few other wires that I de-pinned and removed for systems that I don't have on the car anymore like the fog lights. The gauge wiring was all in a giant knot, so I started to pick that apart. The wiring for Megasquirt, the wideband controller, and the ignition module is all straightened out and rerouted to go over the dash and down instead of being a little tangled in the brake pedal. It's almost to the point where I can go ahead and start bundling the wires and reconnecting what I snipped.
As part of the process, though, I'm considering my options to mount Megasquirt And Friends. The plan I devised this afternoon was to cut a panel of steel to cover the hole in the dash where the glove box would normally be. Then, with some rivnuts and maybe some rubber isolators, I can mount everything against that panel in order to keep it looking tidy and making it easier to service in the future. Once that's complete, I'd also like to make a panel to fit in the radio hole on which to mount the fan switch, the lightbar switch, and the knock sensor LED.
Oh. And I still haven't figured out what component is backfeeding voltage....
Welp... Harness is completely out...
The good news is that if I can find a grommet that will cover a 3" hole, I don't think I need to depin anything to run it back through.
Time to start putting things back together. I started the the starter wiring, then the horn and the fan, and am basically working my way forward from the front left corner of the car. With the PDC and relay bracket mocked up, I can now get the rest of the relay wiring organized up front before continuing towards the firewall.
Before I could proceed making progress on the wiring harness, I absolutely had to run out to one of the local junkyards. They recently got a pair of 1gn 2drs on the lot and surely I can't proceed with putting my car back together until I've acquired some spare interior parts, right? I believe the kids would call this a "side quest". I picked up another spare dash top section and a spare dash center section since those are somewhat prone to breakage. I also grabbed a spare gauge cluster since the tach needle popped off of mine. Maybe it can be fixed, but spares are getting harder to come by so at least I've got an extra now. When I was removing mine from the car, I took a little chunk out of one of the HVAC control cable holders and I'm not sure if it's still going to hold the cable and/or be repairable so I grabbed one of those as well. Interestingly, it appears that in 1998 they added some vacuum components to the system instead of it being all cable. The one from the 1995 was identical to my 1997 one, but the '98 was pretty different. Both cars had broken window guides, of course, so no joy on those, but I should be able to 3D print what I need for that.
Back to the problem at hand, though, I've definitely begun making progress at putting things back together:
I'm working from the engine back into the car and patching wires back together as I go. I had another little side quest to remove the electrical tape wrapping for the speed sensor, IAC, IAT, and TPS so that I could use the braided loom instead and I like the look of that. Eventually, I'll continue across the right side of the car and make it all look uniform, but for now I'm getting my focus back to the problem at hand.
There are a few wires that are left hanging on purpose which I will bundle up and secure. I've got a single wire that is in place just so that I have a wire already passed through the firewall for some future need and there are a couple wires in the megasquirt DB37 harness that are unused right now. I also have the wires coming from the IAC on the throttle body. For some reason, I tried turning on IAC valve control in MS a while back and it didn't work, so I snipped the wires and just let it go. I may try to get back to that at some point, but for now I've taped them off, labeled them, and positioned them outside of the loom so I can find them later.
As I looked at the PDC diagram, I decided that both my main relay and the ASD relay were a bit unnecessary. As the car was originally built, there was a single feed from the inside fuse box (fuse 10) to the PDC which provided +12V for the ASD relay coil (86) as well as the power source (30). If I want the coil and injectors to be powered up in ST-RUN, then there's no need for any sort of relay, I can just directly connect the DG/WT wire -- which comes from fuse 10 to the PDC -- to the DG/OR wire which powers the coil and injectors.
For now, there's no pressing need to worry about getting the lightbar installed, but when I do, I'll use the relay that I had originally mounted as my main relay for the light bar. In retrospect, I could have also just rewired the PDC slightly to allow me to use the stock fan relay, but that relay is already wired up and installed so it stays for now.
I added a 3-pin connector to the LC-1 wideband controller so that it can be easily disconnected from the harness so that in the event that I ever need to pull the harness back out, the wideband controller doesn't also have to be disconnected. I'm also going to add a connector on the top of the dash that will be all the wires needed for the gauges up there. I'll use a single connector -- 6-pin, if my math is right -- for all the gauges since they each already are set up that the wiring can be removed from the gauges themselves. That way I can disconnect one connector that is easier to reach so that the dash can be lifted up. I also have ordered a pair of switches and a voltmeter to mount in the space where the radio would normally be mounted. Those, too, will have a single connector for that panel to facilitate removal. The voltmeter is actually somewhat important for me since the PCM controlled the charging system failure light on the instrument cluster.
Also on the way is some larger wiring loom -- the combined Dodge, switchgear, and megasquirt harness is just way too big for the 3/4" loom that I had. In addition, I've got some of the little harness clips coming so I can secure the wiring and prevent it from rubbing around. I'm actually starting to get pretty excited about how nice it's going to look.
The larger loom arrived so I went ahead and started putting things together today. Before I went too far, I needed to solve the grommet problem. I got a 12x12 sheet of 1/6" thick rubber and cut a circle out of it to match the inside of the OEM grommet.
After cutting it in half and putting some slits in it, I used rubber cement to attach it to the back of the grommet.
It isn't perfect, but it's really really good. I wrapped everything in the new loom, ran all the connectors and loose wires through the grommet, and pushed it through the firewall.
Once I verify that everything is working, I'm going to close up the gaps there and tape things up with the cloth harness tape. I also got the wire harness clips today, so I'll be able to secure things.
With everything in the car, I went about getting things plugged back in so I could start testing to make sure that I didn't remove any wires that I actually need. I don't want to put the dash in until I know everything is working, so everything is just sort of hanging around for now, but that's okay. I hooked up the battery, turned the key, and....
nothing.
Totally dead. I had 12.6 volts at the battery and at the PDC, but turning the key did absolutely nothing. It was when I went to verify that I had power at the ignition switch connector that I realized I hadn't connected it to the ignition switch proper, so of course nothing was happening! After plugging that in, megasquirt powered up, the wideband came to life, and I got the priming pulse from the fuel pump. Turning the key to start, however, accomplished nothing. So I went hunting again and realized that my neutral safety ground wire was hanging loose. It isn't quite long enough to reach the grounding point that I want to use and I forgot that I need to extend it. For testing purposes, though, I went ahead and stretched it across to bolt it to the fender. With fingers crossed, I turned the key and the beast roared to life.
So far, the headlights and taillights work, reverse lights operate, brights work, there's power at the cigar lighter, and the gauge cluster appears to function properly.
Things that I know aren't working:
Things that I haven't been able to test yet:
Next steps:
High beams issue sounds like the common RX-7 (and other Mazda) malady with dirty/corroded contacts in the multifunction switch. That you have high beams one way but not the other clinches it - somehow, dust or dirt may have gotten in there.
Cleaning the contacts is an every five or six year thing on the various RX-7s I've had. Mazda was kind enough to use screws to hold the switch together for ease of disassembly.
You'll need to log in to post.
