Also guessing that little appendage down and to the right is a rotary selector for screen display?

I looked back through old posts, it would appear Julian has been holding out on us for a while, as this bit has been visible in a few pics.

Looks like TPMS readout, too!

I'm going to vote for Julian as the one who incorporates FSD into an aircooled car first.

been thinking about doing the same, was this straight forward switch wiring to swap? didn't want to upset the blinking/clicking function. would like to add a flashing LED indicator above as well.

Ok, so years ago I hacked together a remote version of TunerStudio to run on a little screen via a raspberry pi a slew of relays and general hackery. ... it worked but was glitchy. It was a good proof of concept and convinced me that having data on the dash would be a nice feature.

Fast forward and I've spent the past 2 years on and off working on this: it took about 9 months of starts and stops to make the display and the chip even talk to each other. I taught myself C++ Arduino and PCB design.

It's running approximately 3000 lines of custom C++ code that I wrote on custom hardware that I built and lives in a custom 3D printed housing.


*It receives a CAN BUS signal from the MS3ProMini (or any Megasquirt product) and can display any metric that the ECU has. So any base data point from the ECU or any data from additional sensors are available to display.

*There's a GPS module and antenna so that real-time speed, distance, MPG, trip data, altitude, heading, etc.. can be calculated and displayed.

*There's a built in TPMS system that uses off the shelf sensors that screw onto the tire nipples.

*There's a real time clock with a 10-year battery for accurate time display.

*It all lives in a custom 3D-printed housing with matching Porsche bezel. There's a hidden button that allows rotating of the screens. The bezel has a magnetic phone holder (or not, I printed both versions.)

*It has an anti-glare screen so that you can actually see it...

I've been playing around with what works and looks good and I've had versions with one screen, 5 screens 10 screens, etc...

Some of the screens I've made are:
*The above master display with 18 data points
*A VDO clock
*A trip screen with distance, mpg and the like
*A text screen with 20+ data points in text only
*A TPMS screen with a graphic of the car
*A shift light that goes form clear to yellow, orange and red based on the RPM
*A blank screen for night driving

I can make just about any screen with any graphic and any available data.

The gauge has active bluetooth for TPMS monitoring and can become a WiFi access point so that the user can interact with it and select from the available screens, set the time, set the injector rating, set the redline shift limit, and reset the current trip metrics.

Here's a video of it on the bench in a previous iteration where I had several 4-gauge silver dot displays that I've since consolidated into the above

<iframe width="560" height="315" src="https://www.youtube.com/embed/Z3VmrSoHy5U?si=s7357e8PniovJ3TH" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

Here's a video of the phone/computer interface.


<iframe width="560" height="315" src="https://www.youtube.com/embed/kAZEVGZ4hzY?si=zkL63iL0i-PNaPiZ" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

All that's needed is a CAN BUS H&L, 12v power and ground.

And here it is from a drive to the dyno. Cool to see real time MPG. The delta MPG for the current trip (since gauge power on) is incorrect as it's factoring in idle and non-moving time. I've corrected the code and just need to push it to the gauge.

http://forums.pelicanparts.com/uploa...1747673064.jpg

I plan on running it for a while to iron out any bugs and then making some more. I have the fixings for 10 additional gauges right now. Currently I'm only working with Megasquirt. While it's certainly possible to adapt the CAN BUS code to any ECU that outputs non-proprietary CAN BUS, I would need an ECU to test against and eventually a car running said ECU to finalize the code.

I've looked at some of the other ECU CAN BUS protocols and they're strange looking and not standard. That means it's a lot more work to adapt code. So for right now it's Megasquirt-only.

I anticipate starting to take "orders" in late summer/fall. Each gauge will be completely bespoke- every screen and feature, all the graphics, fonts and colors will be unique to each gauge I make. It's possible that I can make a very stripped down basic stock Megasquirt gauge but we'll see...

Amazing work of pure genius!

Would it be possible to display an iPod or phone screen for people using those for their music?

It's possible but not worth it. Same goes with maps and directions. We all already have supercomputers in our pockets so replicating those features isn't a priority.

I looked into turn by turn directions but you can't use apple maps or google maps- a third party app is needed and even then it's clunky and consumes a ton of memory on the chip so not really worth it.

Pretty impressive, well done!

Count me as a +1 for interest when you look to be closer to release!

One of the cool things that can be done with additional sensors that run through the ECU is things like oil pressure warning lights. Yes, the Porsche gauge has a very cute little light that maybe comes on and maybe catches your attention...

As this engine has (in addition to the stock Porsche oil temp and pressure sensor) a secondary oil temp and pressure sensor in the TOD, we can use that data to create warnings. Currently, the screen will flash bright red rapidly (5 times a second) and display a gigantic "OEL" just in case you weren't sure what the hubub was about.

I'll try and simulate that and take a video at some point.

We can add tire alarms for low pressure, temp alarms, manifold pressure alarms, speed warnings, RPM warnings, etc...

And if someone wanted a track-specific gauge with a shift light and an audible buzzer/alarm... that's possible as well.

Julian, you are f&#*ing nuts. Love it!

Slight departure form all the gauge talk...

Today was supposed to be a banner day.

This winter the car received quite a bit of work:

The Full Send Society showed up to assist in pulling the engine as I'd never done that before. The trans went home with Tony for a rebuild and a taller 5th.

The engine received M1 cams, again Tony swooped in from Wisconsin to assist.

The dizzy was replaced with wasted-spark COP. Maybe an odd choice but the goal was to control timing, not to go sequential spark. This necessitated a Clewett crank trigger, a hall sensor and a Clewett cam sensor.

The old MS2 was replaced with an MS3ProMini, all new electronics, motorsport-style harnesses and wiring and a general cleaning up of all the sensor wiring and whatnot. The entire engine electrical system is now separate from the original Porsche wiring (save for the oil temp, pressure, and idiot light) and all brand new.

The engine bay received a delete of everything unnecessary and a general clean up.

Small details like hard fuel lines on the engine, 3D printed harness holders, and the COP install were fun projects to figure out.

An IAT sensor and a combo oil pressure/temperature sensor was installed in the TOD to allow the ECU to get that data.

After getting the engine in, again with an assist from the boys, it took a few weeks to sort out some small issues- broken wires, crossed wires, etc... but eventually it fired up.

http://forums.pelicanparts.com/uploa...1747708802.jpg

http://forums.pelicanparts.com/uploa...1747708802.jpg

http://forums.pelicanparts.com/uploa...1747708820.jpg

http://forums.pelicanparts.com/uploa...1747708820.jpg

http://forums.pelicanparts.com/uploa...1747708820.jpg

http://forums.pelicanparts.com/uploa...1747708820.jpg

http://forums.pelicanparts.com/uploa...1747708820.jpg

Tuning was mostly done on the road with the left foot dyno and the results were pretty good but that last 10% was missing.

So, today was dyno day. Kind of.

After dropping the car off at the tuner's shop hopes were high. I was mostly hoping to receive bak a refined version of what I had created with a focus on the ignition table as I was running a stock 2.7 curve.

Upon returning the tuner shared the following:

*The car was running super lean 20+ AFR.
*The car was running so rich that the shop smelled of fuel.
*Adding and removing fuel from the VE table had no effect on AFR.
*The injectors were trash and unresponsive.
*The AFR sensor was garbage.
*The AFR controller was garbage.
*The car was leaking and burning oil so the AFR readings can't be used.
*The car had no 3rd gear; the trans was shot.
*Etc... etc... etc...

It became clear to me that despite professing knowledge of 911s, this tuner was in over his head and gaslighting me hard.

The tune that the tuner loaded was trash and promptly replaced with the tune I had created. The car ran smoother, AFRs were where they were supposed to be, the car ran smooth and we were back to square one.

Upon returning I went to address an exhaust leak on cyl 1. The underside of the car was covered in oil- heretofore after several hundred miles it's been clean and dry. 2 hours with a tuner and it's covered in oil... hmmm. Turning to the SSIs only to find that one of the studs was spinning in the head; the threads were completely stripped. Fark.


So, filthy engine, pulled exhaust stud, and no tune. What a day.

This was the ignition map that the tuner created and I'd love the Pelican brain trust's opinion on it. Any thoughts of suggestions would be greatly appreciated as timing is somethign I'm still learning about.

http://forums.pelicanparts.com/uploa...1747708390.png

Which is it? Good help is so hard to find.

That really sucks. Obviously he did more harm than good. I've heard this story before. Hopefully he didn't permanently damage anything.

The question is WTF did he do to cause it to leak oil?

You might check the oil. He might have, determined it wasn't full enough, and added a few quarts?

First things first. Every time I see a COP setup and a screen shot of a timing map, I want to know, first and foremost, how do we KNOW, for certain, that the theoretical XX.X of timing on the map is XX.X timing in reality?

Has this been tested? On all 6 cylinders? And how?

(As an aside, I don't see why you don't run sequential spark if you have a cam sync.. unless you only have 3 ignition outputs? Come to think of that's probably exactly why, never mind!)

To my eye that's a bit too much timing to be putting in as a baseline/ WTF with the timing going to from 34 to 38 degrees at full load at between 6k and 7K RPM?

I'm assuming if you let the car idle it bounces around in the cells on the upper left that are all 10.0s? That block of 12 cells and the block of 14.0s below are the only ones that look like they were generated by a human being, and not the smoothing function.

I hate to give advice too late but I don't care how good a "tuner" this guy, or any guy, is, I wouldn't leave my car unattended with ANYONE.

Well, bygones. Hopefully it turns out to be a not too expensive learning opportunity and move on.

Julian, you have proven yourself to be talented, hardworking, resourceful and detail oriented. I would be very surprised if you didn't already have it figured out better than almost anyone. You said you are going after the last 10%. There isn't 10% between your 2.7 stock curve and a perfectly optimized curve. It's more like 2 or 3%.

Hindsight 20/20...of course.

I didn't really want to leave the car but thus far it's been an incredibly difficult process of getting the car running; lots of problems (mostly of my own making) and I'm just tapped out mentally and emotionally (coupled with all the other stuff I have in my life...) The car is supposed to be fun and a distraction from all the other ills, and when it's adding to the stress, it feels toxic.

I figured it was going to be a minimal adjustment and dialing in. FWIW, the fuel map and the rest of the tune hardly changed at all which tells me that I did a pretty good job on my own and that this guy was fumbling around in the dark.

As for the oil... I haven't a clue and right now I don't really care.

Anyhow, spilled milk... on to fixing.

At this point, after talking with friends, I'm going to see if I can get the exhaust stud fixed before RR (or just run it leaky) and then try to build an ignition map that's safe and based on all the other 911 maps I've collected. Worst case I can default to a stock flat map until later. And like you said Jonny, the stock timing is fine- the car runs great with it now so I know that's an option.

To that last point... I'm in need of either a Stomski drilling joig to borrow or rent or the dimensions of the exhaust flange from a head so I can make my own jig. If someone can measure a head exhaust or SSI flange, or take a good photo with a ruler I'll design one and provide the 3D model for free to the community for printing.

I'd also really love an SSI wrench so that it doesn't take 5 hours and all the curse words getting them off and on...

Julian,
Is it possible to use the oil temperature and pressure sensors to drive the oil pressure gauge, oil temperature gauge, and idiot light through the ECU? Last week, I installed a Bosch temp/pressure gauge in my car, replacing the usual idiot light sensor. It's the same sensor I use for fuel pressure.
My plan is to activate the idiot light based on low oil pressure—say, anything below 10 PSI—using the ECU (Haltech R3). I'm hoping, though not entirely certain yet, that the DPO outputs can be used to control the existing gauges. I might even go a step further and use the ECU to manage the alternator.


Sorry about that not really a tuner guy. What a mess after all your hard work.

The stock Porsche gauges and sensors run 12v, the digital sensors run 5v but give signals usually 0-3-3.3v so no it's not a direct analog. Depending on your ECU, it might be able to receive the 12v signal and convert it for use digitally... then you can use the stock sensors with the ECU.

ADC (analog to digital converters) can be created and that would allow the stock sensors to be fed to the gauge or visa versa but it's a lot of work and it gets messy.

I don't know about the haltech but I' assuming that it has plenty of outputs that can trigger a relay to turn the light on.

I've found it's easiest to just divorce the stock Porsche wiring/electronics from teh modern digital stuff.... it's so much work to make them work together and there are tons of other issues like the fact that Porsche barely knew wahst grounds were and digital systems depend on the grounds tremendously so noise in the Porsche system really messes the ECUs up...

I've eyed up that Bosch 0261230340 sensor as a way to lighten and simplify the engine compartment - just as an FYI to Julian, they are known to be prone to leaking. I did find a recommendation to torque it to 24nm max or they can break internally and leak. I also found a Bosch document that specs 35nm which sounds high.

This is the sensor I'm using
https://www.wiringspecialties.com/bosch-145-psi-dual-fluid-pressure-plus-temperature-sensor-m10-x-1-0/

Interesting about the gauges. I've read about people having success using PWM to drive old gauges. I'll need to create a table which shouldn't be too difficult - assuming this method works. For the idiot light, I'm just using an unused output to switch to ground when the sensor reports low oil pressure. The R3 has power distribution built in so no relays needed.

For my combo oil temp/pressure sensor (that I use for the ECU) that threads into the idiot light sensor in the TOD I'm using one from Syltech:

https://www.syltechsensors.com/. (https://www.platinumracingproducts.com/en-us/products/mscs-10-bar-gauge-pressure-temperature-sensor)

They have a 10 bar: MSCS-4P-G10-N8

And a whole bunch of other cool sensors. Very nice piece of kit, highly recommend.

Thus far, after about 300 miles, no leaks.

tools
 

"To that last point... I'm in need of either a Stomski drilling joig to borrow or rent or the dimensions of the exhaust flange from a head so I can make my own jig. If someone can measure a head exhaust or SSI flange, or take a good photo with a ruler I'll design one and provide the 3D model for free to the community for printing.

I'd also really love an SSI wrench so that it doesn't take 5 hours and all the curse words getting them off and on..."


I have the SSI wrench and Stomski exhaust drilling jig that you can borrow.

regards,
al

I wish I'd known about that product. It's less expensive and uses the Deutsch connectors. I've become a big fan of the those connectors.

I just did a quick test of using a DPO outputs to drive the oil temperature and pressure gauges. It worked!!!! The idiot light is also working.

Does your ECU have DPO outputs?

When Well Enough Can't Be Left Alone; Backdate, EFI/ITB, AC, and more!
 

That’s pretty cool.

You’re just a few steps away from divorcing everything engine related from the original wiring which would be very cool given the state of most 911 wiring.

All you need is a PDM and ignition system and the original wiring will be for lights and whatnot.

My ECU has PWM outputs which can create an “analog” signal, or one that looks like an analog signal to an analog receiver like a gauge, and a few other low voltage outputs for driving relays, etc… I have one reserved for triggering the oil cooler fan or something in the future that I haven’t thought of yet.

Mine is a stripped down version of the MS3 pro evo which has more features than I needed or wanted (along with more than double the cost).

For this car it’s more than sufficient. For a more high strung engine or twin plug it wouldn’t cut it.


Sent from my iPhone using Tapatalk

I'm fairly far into getting rid of the old wiring. I'm running Denso COP's in sequential mode. Not much left now, it's basically just lights, oil level sender, and the alternator field wire at this point.

Hopefully, you'll be able to drive the gauges with the PWM outputs. Here's a shot of how I setup the temperature gauge. The oil pressure gauge still needs some dialing in. The values for the oil pressure gauge are inverted. http://forums.pelicanparts.com/uploa...1747775715.png

I can't recall exactly as it was a while ago that it was converted, but from my loose memory and looking at the flow chart I recall it wasn't too complicated.

I had to use a more robust toggle switch with 8 poles and some of the poles housed two wires... Again, not terribly complicated but reference your flow chart and diagram it out...

Here's the back of the switch (the wires are mine and don't necessarily reference the Porsche wires.)

http://forums.pelicanparts.com/uploa...1747840812.jpg

Count me in for +1. Currently running a round AFR screen but this is just awesome!

PART 1

It's 4am in Chicago right now and I should be en route to Indianapolis to rendezvous with Chris, Nate and a few others and pointed to the mountains of West Virginia for Ruchlos 5.

Instead, I'm on my couch because Tuesday, this happened:

http://forums.pelicanparts.com/uploa...1749115858.jpg

-----------------------------------------------------------------------

A long long time ago in a galaxy... Wait, wrong story...

This is a story all about how my life got twisted, turned upside down... No, that's not right

Here's the story, of a lovely lady... yup, that's not it...

Sit right back and I'll tell you a tale, a tale of a fateful trip... Sure, close enough.

-----------------------------------------------------------------------

This is a story of grit, tenacity and ultimately, failure.

Let's back up for a minute. After the dyno fiasco I spent a few days re-tuning the car and found that in fact there was noting wrong with it. I tuned it on the road to a very respectable place and spent many days ripping the one nice curvy road we have here in Chicago. I have found that the M1 cam is really quite mild below 4000 RPM but once you hit that, it's heaps more power and that's just lovely.

Last weekend in anticipation of Ruchlos, I cleared my calendar and devoted two full days to driving. Refining the highway fueling to give me nearly 30mpg, dialing in a wnadering idle and just shaking the car out. IT was wonderful- me and the girlfriend took senic drives, I made her giggle with a 40-120mph rip and life was feelign good.

I decided to commute to work this week becasue why not- it's a 4 mile trip through a forest preserve and spring has sprung so it's quite lovely. Monday was flawless. Tuesday was flawless... until it wasn't.

Midway through the drive the car died while going 20mpg. I got it restarted instantly but then it promptly shut down again. No noises, no smoke, no nothing. I pushed on to a side street and began diagnosing it.

I found a blown fuse on spark bank 2 (456) and changed that. Success! THe car fired up and I was on my way... or so I thought. It was idling rough and the AFRs were spiking to 20 but I could smell the fuel... Fals lean... Ok. In gear and under acceleration, it stumbled and nearly shut down. More RPM and it seemed fine... Until again, it shut down.

Tow called. Morning shot.

Back at my work where the car lives, I pulled out my laptop and started triaging the situation. I used TunerStudio to confirm the injectors were working and then moved to spark. I was able to get cyl 1 to test positively but none others. Moving back to cyl 1 and it too was now not responding. I confirmed all the wiring and continuity just in case something had become loose or melted and then called AMP/EFI the makers of the ECU to troubleshoot. The tech was great and we worked through a bunch of tests. Having no success with spark, he suggested sending the ECU in for evaluation just in case as he started to worry about a transistor or something…

So, I pulled and packed the ECU and relayed the news to the boys who had been texting all day trying to brainstorm: no Ruchlos for me.

I cleaned up my work, threw away a bunch of crap and went home gutted.

Sigh.

Wednesday morning comes and I’m pissed. I had been anticipating Ruchlos for a full year, the car had new cams, a new ECU, a brand new trans with a taller 5th and LSD. I was ready to rip! I started thinking about the system and what could have gone wrong. Gabe suggested that I go buy another coil and test it with the ECU just in case. Well it turns out that I had three VW/Audi coils… or I thought I did… where were they… IN THE TRASH and I COULD HEAR THE TRUCK COMING!

Like a raccoon in the early hours I was in my dumpster digging around for a bag of coils whine the truck waited. In what would be the first of many triumphs I shot up and exclaimed “not today motherfu…” with coils in hand.

ECU unboxed, and installed, coil rigged up as the connectors on my Denso coils don’t fit the Audi coils (dear reader, note this moment of foreshadowing) and into TunerStudio test mode and WE HAVE SPARK!!!

Ok, it’s 9:00 am and I have a plan: order more coils and connectors, build new harnesses and reap the rewards! Unfortunately, none of those parts are in stock locally at any stores. But NAPA will have then delivered for same-day pickup. Order placed. Now to wait.

And wait.

And wait.

I suppose the wait was good as in the ensuing 6 hours I completed the rest of the week’s work, cleaned my studio, and caught up on emails.

2:30pm rolls around and I receive a text that my order is ready. I hightail it to the NAPA store, giving my A6 Allroad a workout that undoubtedly left the turbo glowing red. 20 mis later I was back at work with PARTS IN HAND!!!

Success!!

PART 2

Except not.

The connectors did not fit the coils. NAPA website confirmed the fit. NAPA customer support confirmed the fit. It ain’t fitting. Back online and scouring for new connectors or coils; I was ready to throw money at the problem, but alas, nothing would arrive in time.

Game over. Back to gutted.

I picked up my kids and we went out to dinner. I had a big beer and settled into the reality that once again, I wasn’t going to Ruchlos. In explaining to my girlfriend what the problem was she asked me if I could just make a connector; after all I’m pretty handy and I do have a studio full of tools and machines. I tried to explain the subtle intricoes of automotive connectors to her when these exact words came out of my mouth:

“I just wish I had a way to make a plastic housing…like a mold or a… 3D printer”

Holy cow, batman I do have a 3D printer, a brand new Bambu Labs H2D, probably one of the nicest ones on the market. Holy cow, there’s a chance...

After dispensing with the kids I raced back to work, again putting the Allroad through its paces and whipped up a 3D model of a connector that I whiped would fit. 20 mins later and I had a PLA test piece and….

IT FIT!

But PLA isn’t suitable for an engine bay. Luckily for me I have a roll of PA6-GF: glass reinforced nylon; the stuff that half of my Audi is made of! YES!

Print time for 3 connectors: 1.5 hours: just enough time to cut the harness open and begin wiring. And wire I did. As the printer finished up so too did I and sitting in front of me was a new harness awaiting connectors which were cooling in the printer.

I peeled off the supports, cleaned them up and went to put the pins. But they didn’t click into place. I tried again to no avail. It turns out that in an attempt to speed up the print time from 3 hours to 1.5 I had changed some of the settings and while that made the print faster, it resulted in less fidelity. Detail and resolution that I desperately needed to ensure the pins would snap into place.

Ok, no problem. Except time was running out. It was now near midnight and I had to he on the road in 4, maybe 5 hours.

Screw sleep; if I got tired, Nate would trailer my car and I could sleep in his Cayenne.

So, I fired up the printer again and set it to fine as hell, we’re talking Jennifer Lawrence, Amber Heard levels of tight. I figured that I’d sleep on my couch for a few hours- the harnesses were ready and all they needed were plastic housings. It would take 20 mins max and then I’d be on the road, literally.

I settled into the couch ready for some rest as I had been burning through adrenaline, cortisol, serotonin and whatever else pulses though your brain when you hit the FULL SEND button when I heard a beep beep beep from the printer.

This printer is sensational and I highly recommend Bambu Labs. I’ve thrown everything at it and its never blinked.

Today, my friends, at nearly midnight, it blinked. A beep, a flashing red light and an error. The print head something or another. A quick search on Bambu’s wiki told a story I didn’t want to hear: the print head was either clogged or had failed and would need to be disassembled.

I’ve been here before and the process to remove and disassemble the print head, while not terribly difficult is time consuming. Last time it took over an hour. An hour I just didn’t have to spare; it was 1am, I was wrecked exhausted and even if by some miracle the prints worked and I was able to get the car running, I’d still have to cannonball to Indy solo with no backup and no room for error, then drive another 8 hours to West Virginia and recover to be sharp enough to rip super technical roads with the homies.

And though my girlfriend who was with me at the studio, cheering me on, feeding me water and encouragement pushed me to keep going; clean the printer, use the other print head (the H2D has 2 print heads) hardwire the injectors, use hot glue to hold the pins, do something DAMNIT, you’re on the precipice of the greatest comeback in automotive history (she doesn’t know much about cars or well, comebacks) I knew this was the end of the road.

-----------------------------------------------------------------------

The drive home at 1:30 was serene and calming. The cold air in the windows felt good on my sweat stained and greasy face. The lack of traffic in the forest preserve felt a lot like the twisties in West Virginia and for a moment I was there, awaiting the next rest break where I’d trade stories with the boys.

-----------------------------------------------------------------------

This car has put me through the wringer this winter- I didn’t post about it for a lot of reasons but at many point I was dejected and thought about tarping it. I felt/feel that way now. But I’m also angry as hell. I had the thing that I was looking forward to snatched away from me. And it wasn’t just about fun driving with friends. These past few years have been brutal for me and this last one in particular has tested me to the core of who I believe myself to be. Everywhere I turn are fires, arrows, and firey arrows.

The one refuge that has allowed me to escape myself, to be free of all that ails me and to be comprehensively present was ripping the car around, smashing the accelerator and engulfing myself with the deafening roar of the engine.

Ruchlos was a heavy, potent dose of mainline self care that I desperately needed. Still do.

Lst night I decided to trailer the car to storage and leave it for a week, month, year.. who knows. I just couldn’t take that pain anymore. I was up at 3am this morning because see above. I decided that this story wasn’t over and that if I’m to be beat, it’ll take more than a failed coil.

Parts ordered, weekend cleared, and as Tony often says when it’s go time. LFG

Holy crap Julian, I had to read every word of the last couple posts. A riveting story with a (currently) unhappy ending. We've all been there in one way or another, and when you stick your neck out sometimes it doesn't go your way. But these things DO get solved. You're tenacious and talented, you'll get it sorted. Sending good vibes your way SmileWavy

Me too! What an amazing piece of kit!

Unusual to have multiple coils fail. Are these dumb coils that perhaps your hitting with too long a dwell period?

Denso smart coils.

Bench tested a bunch of them for resistance and they were shot. I didn’t test all of them because at that point who cares.

Dwell was 3ms, cranking dwell 6ms.

My guess is that they were either crap coils or the heat caused them to fail. Dunno. Theories welcome


Sent from my iPhone using Tapatalk

What a story. Old cars giveth and old cars taketh away.

Hope you get yours back on the road soon and are able to enjoy the smells, vibrations, and sweet sounds these cars make.

What was the "shot" resistance? Typically with a later COP system (5-12V input signals), you can't really measure the primary coil resistance.
So your coils are most likely conventional coils requiring an ECU output coil driver capable of 5-8 amps for the 3ms or longer dwell time.
If ECU provides a too short of a dwell time, an inadequate spark will result. Hopefully the coil supplier provided the proper coil for your ECU setup.

But when you note, "Denso Smart coils", this implies the ECU is providing a low level switching voltage (5-12V).

So what type of COP system do you have? For proper troubleshooting, you need to use a scope to check the coils' primary signals.

Post 545: "Well it turns out that I had three VW/Audi coils"

So with just three coils, it's a wasted-spark system, conventional coils (primary ~ 2-3 ohms)?

Dave,

These are Denso smart coils requiring 5v logic signal from the ECU to fire them. No output coil driver necessary as with "dumb" coils.

You most certainly can measure resistance on a smart coil and I followed multiple procedures for these coils and of the several I tested, they were out of spec.

I had initially purchased the Audi pencil coils for the car but went with Denso as they were smaller and fit the engine better and have been used by many folks on 911s, including TurboKraft which sells them as part of a kit. As a Toyota product it also seemed logical to think that they were reliable, ubiquitous and economical. I couldn't return three of the the Audi coils (I did return the other three) so I tossed them into a box.

The dwell time I had for these coils was 3.5ms, but I saw that the cranking dwell was 6- I checked my notes and it should have been 4 and I can't recall why it was set to 6, so that may have contributed to fatigue on the coils...

The system is wasted-COP. I considered going full sequential with spark but the cost for an ECU that was capable (and had the other features I wanted) was more than double the cost of the ECU I went with and for a single spark low compression engine I felt the benefits to sequential spark weren't necessarily worth the cost.

Now, perhaps with COP sequential really is necessary as it reduces the frequency of firing and load on the coils. and if I find that to be the case (all research and documentation conducted thus far runs contrary to that) then I'll pull the eCU and wiring harnesses and switch to one that is capable of sequential spark...

New connectors are arriving today and I've already remade the harnesses- I have this weekend cleared to work on the car and hopefully solve this supremely inconvenient problem.

Julian,
I'm using a copy of the Denso coils.
https://www.amazon.com/dp/B091YJ9B3T?ref_=ppx_hzod_title_dt_b_fed_asin_title _2_0

For cranking and dwell, I'm using these values. The coolant temperature is actually cylinder head temperature. I'm using a cheap circuit board to convert a spark plug thermocouple to an output that the Haltech ECU can interpret.

http://forums.pelicanparts.com/uploa...1749242653.png
http://forums.pelicanparts.com/uploa...1749242653.png

I'm surprised you didn't just stick the pins on the coil without the connector - it's just a holder after all. It would be fine temporarily.

Trust me, I consider doing that. I also considered soldering the pins to the connectors, using plasticine clay, my girlfriend even suggested filling the void with hot glue. And while all of those seemed like perfectly reasonable solutions in a pinch, starting off a 2000 mile road trip with that as the basis seemed a bit reckless.

It turns out not going was for the best. Even after going home and getting four or so hours of sleep on Thursday, I was a wreck. I ended up having to take three naps at work and going to bed at 8 PM; certainly not the state one desires to be in while piloting a 50 year-old car in the mountains and inclement weather.


Sent from my iPhone using Tapatalk

Yes, but not actually measuring the primary of the coil itself, you're measuring the input circuitry for the 3-5 amp power device that drives the coil primary (<5 ohms).

Keeping the dwell time too long causes the voltage of the primary winding of the coil to go to zero volts, over-stressing the internal power transistor resulting in a failure. Assume a coil primary resistance of 2 ohms @12V equals 72 watts @ COP transistor switch once the primary of the coil saturates.

Your existing wasted-spark system should be more than adequate, as long as you don't over-drive the ECU output pulse width to the COPs. You should use a scope to verify the actual signal pulse width the ECU supplies to each COP,
and determine what the minimum is to avoid a misfire. Most importantly, the COP datasheet should reviewed! Each COP only sparks once per engine revolution, so @ 6K RPM that's every 10ms.

Well, the new connectors arrived and fit…phew. So first thing tomorrow I’ll be wrenching on the car to get this hopefully sorted. I have an exhaust leak to chase and a stripped head that needs a timesert so tomorrow will be a project for sure.

It’s amazing how quickly things can go from good to bad to good to bad to good, etc…

This morning, after sleeping for nearly 10 hours on Friday night I woke at 5am thanks to the newly hatched birds in the nest outside my bedroom window. The miracle of life is something to behold… anytime other than 5am. It was almost the miracle of death but I refrained.

Instead, I took the opportunity to hightail it to my work studio to begin either wresting the car into working order or planning its demise.

The tasks at hand were not terribly complicated, yet we all know hoe that goes:

Finish the wiring harnesses for the VW/Audi 1.8 coils,

Get the coils and harnesses installed and button up the engine bay

Remove the SSIs

Drill, tap, timesert and install a new exhaust stud on cylinder 1.

Start the car.

The harnesses came together quickly and I’m satisfied with them. I made them removable from the main harness with Deutsch connectors (something I didn’t previously do… d’oh) There’s a real zen in making harnesses and doing a nice job- it’s one of those detail that I think communicated the care that went into the build. Sometimes it’s the little things that speak volumes.

http://forums.pelicanparts.com/uploa...1749335504.jpg

In any event, with the harnesses built and the coils installed the work inside the engine bay was complete and it was on to the real labor of the day: removing the SSIs.

These were installed by the shop I first took the car to and among a litany of things they did wrong, sloppy, or just didn’t do even though they charged me, was ordering thin flange SSIs. This car is a 1977 and the exhaust studs are long… like 58mm long so ordering the thin flange will of course work but it leaves a LOT of exposed stud. This means that it’s damn near impossible to get a wrench on the stud due to the limited space between the end of the stud and the SSIs.

Even with the SSI wrench that Al loaned me (lifesaver) it was painful and arduous. I cut up a few other wrenches and made some custom ones; what works on cylinder 1 doesn’t work on cylinder 3…

Eventually, after much cursing the SSIs were off. Clearly there were some exhaust leaks. I knew about them- I could hear them after all but this was absurd. Brand new gaskets were used and the car has maybe 500 miles on it since the motor went back in…

http://forums.pelicanparts.com/uploa...1749335504.jpg

Al also sent me the Stomski tool which is designed for drilling out broken studs. For that, it does an admirable job but for what I needed it to do, some modifications were to be made.

I first milled a drilling busing out of some spare aluminum I had lying around. This allows the 10.2mm drill for the Timesert to be held perfectly perpendicular to the surface of the exhaust flange.

http://forums.pelicanparts.com/uploa...1749335504.jpg

With the bushing in the Stomski tool, drilling could commence. I measured the depth of the hole and made sure to not extend past it with the bit. That’s what the very scientific piece of tape on the bit was for.


With the hole complete it was then time to tap it for the timesert. For this I 3D printed a PAHT-CF sleeve that was the same diameter as the tap, ensuring that it too would be perfectly perpendicular to the exhaust flange.

http://forums.pelicanparts.com/uploa...1749335504.jpg

Tapping the magnesium wasn’t hard at all save for the limited space in which I had to work. Keep in mind I was doing all this on my back under the car; I figured it would be faster than pulling the motor.

http://forums.pelicanparts.com/uploa...1749335504.jpg

http://forums.pelicanparts.com/uploa...1749335504.jpg

Some high-temp red locktite (272 IIRC) and the timesert was in. Easier than I thought it would be but of course having the right tools makes hard work deceptively easy. You can see the delta in the length of the studs. I ordered 12 studs and was very, VERY tempted to try and extract the others and replace them with shorter studs until I hear in the distance the words of John Walker: “Don’t” and that was enough, so I left them…

http://forums.pelicanparts.com/uploa...1749335504.jpg

http://forums.pelicanparts.com/uploa...1749335504.jpg

It was then time to put the SSIs back on and I will say this; if they ever have to come off again, they’re staying off for good. Yes, I like heat and want to be able to drive up until the first salt, but these SSIs are a nightmare. Well, now they’re a bit less of a nightmare thanks to one shorter stud that went on rather easily. I was going to mill some spacers to slide onto the studs but I was out of steel and I didn’t want to use aluminum, nor did I want to wait to buy or receive steel so the studs will be a problem for future Julian, F that guy.

I decided to sand the surfaces of all the flanges- on the heads, on the SSIs and on the muffler, just to ensure good contact with the new gaskets.

This is a photo of the AchtungKraft flange that mates up with the SSIs. Notice how buttery smooth and free of imperfections it is… this yields a phenomenal seal to the gasket.

http://forums.pelicanparts.com/uploa...1749335504.jpg

This is a photo of the same mating surface of the SSIs. Notice all the deep vertical gouges. Well, that's what's left of the deep, deep gouges in the surface, like .25-.5mm deep- deep enough that I had to spend a good 40 mins sanding them down. And while an exhaust leak right before the muffler isn’t the end of the world, these gorges were on all of the flanges that mated to the heads. No wonder there were leaks. In addition, the MFI bung weld was open and not complete around the nipple and worst of all, the holes that allow you to feed a socket through the body of the SSIs to tighten the nuts don’t line up with the nuts, making it impossible to tighten them without some very custom offset wrenches. And even then it takes a helluva lot of twisting, bending a contortion the SSIs so that they line up and fit.

http://forums.pelicanparts.com/uploa...1749335504.jpg

Perhaps this is just the norm with SSIs but for the price, these are trash. I can’t wait to get rid of them and go with headers. I’ll run an electric heater or make my own head exchangers.



Anyhow. That work took from 7-about 1pm. Way longer than it should have but also infinitely shorter than it could have.



And then I sat in the seat, heart pounding, hand trembling and turned the key.



And she fired right up.



I took the car out for a 40 min drive to log some data and my initial impressions are that the car is a totally different animal. It’s way smoother, the acceleration under load is much more aggressive and it just feels way tighter and more powerful. The exhaust leaks are gone (oh the joy!) and everything was doing what it should.

I’m convinced that some if not all of the child I bought were duds or fakes. That, combined with a too-high cranking dwell just put too much stress on them and they failed. These VW/Audi could were 2x the price and came from NAPA with a warranty and if first impressions are any indication, they were worth every penny.

It’s too bad this couldn’t have happened 1 day sooner as I would have made RR5 and well, who knows. But, I’ll take the win today and live to fight another…

Man I’m just glad you got to drive the car after all that! My car has been similarly testing and I folded and handed it over to a professional. I commend your persistence on this. Hopefully you get to enjoy a bunch of easy miles on it now