Sunday, August 2, 2020

On Subsidizing Millstone

July 2020 was the month that a lot of Connecticut residents noticed a significant increase in their electricity bills. The reason turns out to be that CT's Public Utilities Regulatory Authority (PURA) approved (and the governor signed) in September 2019 a requirement that Eversource and United Illuminating - CT's two major power providers - purchase electricity from the financially-torn Millstone Nuclear Power Plant.

Governor Lamont Applauds PURA Approval of Millstone Contract Between Dominion, Eversource, and United Illuminating
https://portal.ct.gov/Office-of-the-Governor/News/Press-Releases/2019/09-2019/Governor-Lamont-Applauds-PURA-Approval-of-Millstone-Contract

Of course, as soon as everyone saw their monthly bills increase - about 7c/kWh or almost 50% - the screaming started and "investigations" were called for and as of this post it is rumored to get rolled back, at least temporarily:

PURA to investigate Eversource Energy rate hikes
https://www.wtnh.com/news/connecticut/pura-to-investigate-eversource-energy-rate-hikes/

The idea of subsidizing Millstone seems to grate on a lot of people's nerves, especially given the significant diametric positions many have on nuclear energy.

But let's go back and read a quote from the opening paragraph in that first link:
“Had this contract not gone forward, the facility would be in danger of closing down which would have increased greenhouse gas emissions by 25 percent across the New England region,” Governor Lamont said. “This important step keeps Connecticut and all of New England from back sliding on addressing climate change. Now we can renew our focus on offshore wind and other renewable energy resources to fully transition to a clean energy grid by 2040.”
(My emphasis...wow, something I agree with Lamont on...is that bacon flying overhead...?)

The one thing we're forgetting about all this: the societal value of keeping Millstone Nuclear online.

How many of us have thought (or declared) that we're willing to pay extra for energy to support sustainable sources like wind and PV to replace fossil fuels? Probably a lot of us. Maybe most of us. And subsidizing energy from Millstone is a step in the direction of an environmentally-friendly future.

Wait, wut? Yep.

What's the biggest problem with PV and solar? It's inconsistent. You are highly unlikely to get power from it at night (zero chance with PV) so you either need energy storage for use at night or alternative sources of energy. The technology for macro-scale storage solutions are way, way off right now* so the only other alternative sources of energy are fossil fuel-based. If we want to transition to larger-scale PV and wind, then off-peak generation has to be by these fossil fuels: natural gas, oil, and coal. All exhaust carbon.

But wait...what about Millstone Nuclear? It's carbon-free, just sits there making power (and it's already there). With nuclear, we would throttle it down during the day when we're generating with PV and wind, and when the sun starts to go down we throttle it back up to provide nighttime energy. In the meantime, technology advances toward true macro-scale energy storage solutions. There truly is no other environmentally-friendly carbon-free alternative right now.

But due to regulations, nuclear energy is expensive (long gone are the days of "too cheap to meter"); Millstone is losing money, and is at risk of being shut down. So unless we drop our concerns with its existence (or allow further implementations using new technologies) nuclear is not getting any cheaper. Further, Millstone's life is finite; each plant has a regulated licensed life limit (I believe Millstone 2 is 2035, and Millstone 3 is 2045; Millstone 1 is already shut down).

If your end goal is carbon-free environmentally friendly production of energy, you'll never get there unless/until the technology is developed for reliable, affordable macro-storage -- or you embrace nuclear. Otherwise we're using more fossil fuel.

You've declared that you're willing to pay more to get to environmentally-friendly/sustainable energy. Subsidizing Millstone is one effective, immediate transitional step toward that environmentally-acceptable future...

- Greg

08/04/20 Edit: Another factor in Connecticut spiking energy costs is the state's decision to adhere to stringent Renewable Energy Portfolio Standards.

"Green energy" ain't yet cheap energy...

https://yankeeinstitute.org/2020/08/03/why-are-electricity-costs-spiking-in-connecticut/


*A nice "battery" solution is the old, but still functional, Rocky River Pumped Hydro Station along Route 7, below Candlewood Lake. The lake was created when they damned up the Rocky River to create the pumped hydro facility. "Pumped hydro" uses turbines powered by the cheaper overnight off-peak energy to pump water from the Housatonic River up into Candlewood Lake; during the day the plant releases water down from the lake to drive the same turbines to generate power and sell it at the higher day rate. Its efficiency is incredible, exceeding most other storage sources, and Rocky River is still functional nearly a century later.

To support PV and wind, that concept can be turned on its ear: we could use excess PV/wind-generated energy during the day to pump the water up into the lake, then at night drain the water to generate electricity when PV/wind cannot.

Alas, that particular station - the first ever pumped hydro in the USA, I believe - is not big enough for the electrical loads that we'll need at night, and there's just no way in Hades anyone would ever get approval today to build another one. So...off to further development of chemistry battery technology we go...

Thank you for your rapt attention on Yet Another Greg Rant. Time for beer.

Saturday, June 6, 2020

On "Microsquirting" the Porsche 914 - Part 11, Addendums and Updates

On "Microsquirting" the Porsche 914 - Part 11, Addendums and Updates

Back to Part 10

Lessons learned and subsequent changes made.

Figured I'd create a separate blog for "what I'm learning" and minor stuff. These are updated in reverse order, so if you come back on occasion the newest ones will be on top.

Last Update: 7/31/2020

Wideband Controller Failure

My original design used an Innovate LC-2 wideband controller with Bosch LSU 4.9 sensor. It wasn't too long into the game that I noticed the AFRs were not passing "the smell test" and were giving me onconsistent readings. In July the system quick working entirely.

Systems included erroneous readings, not powering on properly, and the dreaded "8 blinking LEDs" error, indicating that there was a problem with the sensor. However, the system would recalibrate properly each time, but upon engine startup would fail again. "The Internet" had a lot of comments on this, as it appears to be a common issue with this product. And, of course, the sensor is not covered under their warranty. As of this "press time" the company has not responded to my email inquiry.

The premise seems to be that the LC-2 is driving the heating part of the sensor too hard and killing it. One of Innovate's "solutions" is to purchase a $90 spacer that heat-insulates the sensor from the exhaust stream. For a unit that originally cost $145.

The overiding response to many of these forum complaints was "get a Spartan Lamda Controller 3 from 14point7". $133 shipped, including a new 4.9 sensor. So I ordered one today and it should be installed next week. Watch this space...

https://www.14point7.com/

Ongoing Temperatures Battle

You've likely seen below where I've been battling under-trunk temperatures, especially with the oil cooling. I ended up buying a 4-probe BBQ temp sensor setup and doing some light evaluation. It's a lot hotter under there than I expected, and I learned that the duct I built made it worse. I think me and the car have come to "an agreement" and I've got satisfactory oil temps. Longer discussion of teh process and results on 914World:

http://www.914world.com/bbs2/index.php?showtopic=347282

Intake Air Temperature, Part Deaux


You'll note below that the IAT was not the problem with the cut outs...but it was still an issue. The engine compartment gets hot on this car. With the engine behind the vertical windshield there's a low-pressure area above the engine grill, right where you want high pressure, since air needs to go down into the engine compartment both for induction air and for engine cooling. A sub-optimal design. This car really should have been designed with a side air intake (or even both sides) like the first-gen Toyota MR-2 was...

I removed the plastic under-grill rain tray and that helped a bit, but not enough. And moving the sensor (see below) helped a little but showed me that the plenum and air fliter box are getting heat-soaked from the engine. They're bolted directly to the engine so I'm not clear how I'm going to resovle that...maybe some radient-reflective gold foil, yo...


I did effectively drop my IATs by 25-30F with this chintzy test setup. Haven't decided how I'm going to incorporate it permanently, as I need a way for water to collect and drain out before the airbox. Honda uses a plastic resonator in the fenderwell with a drain hole at its bottom, I may do something like that. I suppose I could drill a small drain hole at the lowest part of the metal airbox and that would do the same thing...after all, the air is pulling UP through the filter there so water ingestion is highly unlikely...and I think I'll modify one of the side grills to pull air from permanently.

Ignition Coil

After a period of driving, say a half hour or so, I started getting ignition stumbles. At some point it would get so bad that the car was not drivable. Replacing the coil resovled it for a short time but then it would come back. I'll save you the dreary descriptions of the troubleshooting and just say that the problem ended up being heat: the coil was physically overheating from the ambient temperatures. My location of mounting it directly to the block was a bad choice.

I tye-wrapped the coil directly to the fan housing and drove it for an hour without problems. I'll need to fabricate an appropriate bracket for it.

Oil Cooler Fan

You may recall that during the MS install proces I also installed an external oil cooler. Unlike some 914 oil coolers, which get installed in the nose for the ram/impact airflow, my cooler went under the rear trunk, and without a fan. I quickly found out, as ambient temps rose, that "no fan" wasn't gonna cut it; oil temperatures were running more than a tad hot. In fact, per my thermometer dipstick, I was seeing temps for 240+. Way too hot.

Installing a Derale 8" fan was easy enough, and it improved the issue. But how to control it? I could run it all the time with the key was on but that seemed inefficient; if the fan worked properly then the oil flow thermostat would effectively be controlling the temps (which is OK, I guess) but the fan is also running all the time when it doesn't need to.

Another option was to toss an inline thermostat switch on it, but the only ones I could find were 180, 190, or 200 degrees. Again, OK I guess but optimally I'd like more flexibility. The fan would probably be running almost all the time.

One other consideration is that I'm running a very small LiFePO4 battery, one with limited capacity (wanted to see how well it works; it does). If the alternator were to crap out then a fan running all the time may not give me the emergency reserve I need to get home. So maybe put the fan on a manual cut-off switch?

Then it hit me: use the Microsquirt! The MS allows extra sensor inputs and also offers output relay control. I can set the on and off temps to wherever I want them, and I can also make a condition that the battery voltage needs to be over a certain value.

My hangup is that I had initially removed those wires from the MS harness when I was a setting it up. That was short-sighted; I now needed the SPAREADC SPAREADC2 (spare inputs), and ALED and WLED wires (spare outputs).

But the one smart thing I did is make connectors everywhere so that the harness was easily removed.

So, on one of my COVID-caused "Furlough Fridays"  I removed the wiring harness and reinstalled those four wires to accept an oil temperature sensor input (SPAREADC) as well as use the ALED to control the fan relay (I also reinstalled SPAREDADC2 and WLED in the bundle for future use, maybe as a CHT input). Tested the ALED control with a "battery voltage>12.5" condition and it worked great. I installed the Race Technology oil temp sensor in an Improved Racing -10AN inline fitting and connected that into ADC6.

The MS cannot be programmed to recognize actual oil temps via a formula, but it does convert 0-5VDC to number 0-1203. I had to config the input for these "ADC" values and created a custom channel to convert to oil temps within TunerStudio (see .inc/cheater file here). I made the battery voltage a secondary "and" requirement (temperature over this "and" voltage over 12.5).

I can now control fan on-off based on temperature, and if my alternator craps out then the fan will stay off while I creep home on reduced RPMs.

I'm still fighting temps though. Measuring the cooler output temps I'm still seeing 215+ under load on the highway. I've now boxed in the cooler intake and heat-sheathed the oil lines; I'm not clear what else I can do.





Next up...CHT and alternate map?

I'm begin pretty aggressive on ignition advance and leaning out the engine. But I don't know what that means to the engine; is it too aggressive? It runs fine on the street but what will happens if I decide to do a track day? These air-cooled engines are basically limited by CHT and EGT and I'm not measuring either.

I have one more input available and I've got this idea of installing a CHT sensor. The Sensor Connection offers one with 0-5V output. My thought is that I could install that system, input it into the Megasquirt, and monitor temps. If I find that my aggressive tuning results in extreme CHT then I can tune a secondary map and have it switch based on CHTs. CHTs running fine? Run the aggressive map. CHTs getting too hot? Switch to the lesser-aggressive map.

We'll see how motivated I am over the next few months.

Intake Air Temperature

I've noticed something funky happening as the ambient air temps are gett hotter: a new either miss or some kind of load up at high temperatures and light throttle. It starts out as a stumble or miss and I glance over and see the AFR rolling back from extreme lean. If auto-tune is running then it starts to add a bunch of fuel in those areas. I only notice it at light cruise, 10% TPS or something like that, and if I go full throttle it goes away.

Back to the expert. Peter Florance quickly diagnosed that it's possibly due to either an overrun condition but more likely that my intake air temperature is showing really high and leaning out the engine. I had been noticing the high IAT; I figure it is because I have the sensor mounted on an aluminum plate directly to the intake manifold. I figured it was heat-soaking, showing high temps. Per Peter's motivation, I'm moving the IAT sensor to the bottom side of the air cleaner, right where the snorkel is pulling in air. It's a good place and close enough to the current location (in the port for the D-Jet's cold start valve) that I won't even have to make any wiring changes. The old plate will be a 3/8" NPT plug for now (maybe a new block-off plate later).



Peter also sent an adjusted tune for it. We'll see how it goes!

Tuesday, May 26, 2020

How To Shut Down Your Local Society Without Being a Governor

Contact Tracing and Exposure Notifications on your phone. Hoo-boy. I'm far less concerned about the privacy aspects as I am about the potential abuse.

For example, what stops someone from intentionally answering the app's questions with COVID-positive information and pretending to have the disease? Then they just saunter among large populations, now broadcasting those symptoms via Bluetooth? Say they go to Home Depot, or Walmart, or Shop-Rite, or walk into your local restaurant, or wherever. And now everybody's else's phone will get notitications that they've been exposed to someone who is positive for the disease and notified they need to stay home for 14 days. And then everybody that was exposed to them would get the same notificiation.

The result would be a logarythmic increase in the number of people being notified to stay home for 14 days, along with employees of all these facitilities, and (assuming people followed the instructions) society would again be shut down  All over a prank.

I can easily see privacy groups doing the above intentionally to make a point.

OK so, will there be legal force behind the answers you give to the app (e.g., threats of perjury)? If so, will police be able to legally review and enforce your answers? Your data (incl health symptoms and locations) certainly can't be anonymous for that to happen. What are the implications of that?

Will there be legal force for you to stay home if you receive a 14-day quarantine directed by the app (even if it was done as a prank)? Again, then your health and location data certainly can't be anonymous.

If not, how long before such laws are made and the data released to enforcement agencies?

And if there's no legal enforcement behind your answers, and no legal enforcement behind the app telling you to quarantine...then why have it? After a few of these pranks get going then users' faith in the information would be quickly diminished and society will ignore the apps and/or stop using them. They'll quickly become pointless.

Right now the apps' API are "opt in". Great! But (and here comes the tinfoil hat) as with everything Silicon Valley (looking at you, Apple and Facebook), participation will assuredly, in some future version, quietly revert to "opt out" instead of "opt in" and Silicon Valley (and thus governments) will have databases of our location and travel information b/c most users will neglect to (or not be aware to) turn the "opt out" off. And data is easily sold...

There's a ton that could go wrong with this. And not much good to come of it.




5/27 Update: How to Disable COVID-19 Exposure Notifications in iOS 13.5

https://www.macrumors.com/how-to/opt-out-covid-19-exposure-notifications/

Sunday, April 19, 2020

On "Microsquirting" the Porsche 914 - Part 10, The Design Details

Part 10: The Design Details

Back to Part 9

So your first question will be, "can I do this?" Yes, you can. I did it and I'm not a professional mechanic, but I do have an engineering background and fairly solid mechanical and electrical skills.

Your second question is, "how hard is it?" Well, that depends on your skill level.

The majority of the components are bolt-on, such as all the sensors - I tried as hard as I could to make it so - but you'll need three basic abilities:
  • Ability to remove and install the engine.
The biggest issue here is access, with the hardest being getting to the back side of the fan housing to install the crank position sensor. I guess theoretically you could remove the front air cooling sheet metal to get to the front of the engine and remove it all while working from the underside. But having the engine sitting on your toolbox or on an engine stand makes the processes soooo much easier.

Removing and installing the 914 engine is really not that hateful. Basically speaking, you remove the shifter rod, remove the eight bolts holding the CV axles to the transaxle, a couple wiring harness, clutch and throttle cables, four bolts holding the drivetrain to the car and it's out. I have the advantage of a lift at home so I raised the car and lowered it down on my big 44" Harbor Freight toolbox, unbolted the four bolts, and lifted the car out of the way. If MS was I was going to do then I'd just work on it while on the toolbox then use the lift to put it all back in. A buddy and I removed the drivetrain in a couple hours, including beer.

Look for Series 9 Episode 5 of Wheeler Dealers where Edd China did just that to a 914.
  • Ability to do some minor fabrication skills or access to someone that can
You'll need to fabricate three plates: an aluminum plate to adapt the Bosch TPS to the D-Jet throttle body, another aluminum plate to adapt the IAT to the intake plenum, and one steel sheet metal plate to mount the coil. The designs for those are listed below. These can be done with a series of hand tools you may already have, or contact Chris Foley at Tangerine Racing; he's the one that made mine and can make them for you.

You'll also need to weld in the O2 bung into the exhaust.
  • Ability to read electrical charts and fabricate the wiring harness
This one's the biggie. Fabbing the wiring harness will require the ability to read and understand (a generally-offered) electric design, specialty tools for crimping/terminating electrical terminals, and specialty tools for installing and removing electrical terminals. Not everyone has this capability, but I'll discuss it more in detail below.

I tried to find a business that would have interest in providing all this as a kit, but no one seems to be interested. I'm guessing the volume would be too low to make it worthwhile. So I'll offer to you all of my design details and you can have at it.

I've been keeping track of everything in a spreadsheet: costs, part numbers, prices (but not time; per Wheeler Dealer rules that's "FREE"!!!) Blogger's vertical format doesn't lend itself to showing spreadsheets so I posted it on Google Docs. Grab that and follow along...

https://drive.google.com/open?id=1jVNLFQIDbmaDh81jwmIC3dq7t5Yz6RkA

But I bet you're looking for the bottom line on cost. I'll spoil it: I missed my target design cost of $1500, primarily for three reasons:
  • I did not plan for the costs of some out-sourced work;
  • I did not plan for the costs of required tools, mostly for electrical work;
  • I did not plan for the costs for all the electrical harness build stuff.
Total out-of-pocket costs were $1,719.79, including the costs of the electrical components. It does not include the costs of the tools, some electrical supplies, and the out-sourced work. Also, none of these prices include tax and shipping.

Outsourced Work

As decribed above, the out-sourced work, done by Chris Foley at Tangerine racing, was welding the O2 sensor bung into the exhaust and fabricating the mounting brackets for the IAT and TPS. How much of this you can do yourself will decide additional costs.

TPS mount was per EricP. Details available on his Microsquirt build thread on 914World. Chris Foley of Tangerine Racing has these designs, and I'm sure if you call him up he can fab something up for you. Or you can do it yourself like Eric did!

IAT mount is not drawn up, but it's a small plate with three holes, one big and one small. Chris has that, too.

Can you weld? If so, you've got that whole O2 bung thing figured out. If not, find someone who can. Chris installed mine in the rear muffler, after the two header tubes converge but before the muffler.

The steel flat-plate coil mount bracket is easy. Here's a link to a photo of my coil mount bracket. This one I did myself. It's a simply piece of sheet metal from Lowe's fabbed into a bracket with a hacksaw, drill, vice, and hammer. Easy to make.

One other part you're going to need for the electrical harness (presuming you use my design) is one of the factory T4 pin housings, also described as the "D-Jet interface plug". See description and photo in Part 6. These are not available for sale however there should be some floating around out there, as all 4-cyl 914s were shipped with them. You can also de-pin yours (mark the wire locations in case you want to go back) and re-use that; replacement electrical terminals are available from NAPA (see the parts list).

Optional Parts

Here's some optional, but recommended, parts for this project:
  • Throttle body gasket, Pelican, $6
  • Intake manifold gaskets (runners to head), 2 needed, Pelican, $18
  • Intake runner tubes (the block hoses on the runners) 914Rubber, $17
  • New 5/16" rubber fuel line, various sources
  • Stainless 5/16" fuel injector hose clamps, Amazon B073X2W399, $14
Recommended Electrical Supplies

...not included in price sheet. I consider them tools and supplies (and I have a lot left).
  • Cloth wiring tape, Amazon B07Y9VTPTF, $9
  • Scotch 33+ vinyl tape, Amazon B00004WCCL. $4
  • PET Expandable Sleeving 1/4", Amazon B07NZLR8HS, $8
  • PET Expandable Sleeving 1/2", Amazon B07P1QRLJ4, $9
I also used a Deutsch 3-pin, connector, DT06-3S-E004-Assy and DT04-3P-E004-Assy, to have a single-plug-connection for the wires that go through the rear bulkhead: the two reverse light wires and the starter solenoid wire. This gives me a one-plug disconnect for removing/reinstalling the engine.

But here's a top tip. Did you notice that I spec'd out each individual Deutsch connector in that spreadsheet? They add up to a strong number. You can reduce your costs on a lot of these Deutsch connectors by purchasing a couple of the kits I linked to in Part Six; that's what I did and reduced my connector cost by about $125. But there are a couple of caveats on doing so:
  • You will have to file down the pin housing "keys" in the Deutsch 8 connectors that go into the two bussbars; they are not keyed correctly for those;
  • You will not have unique pin housings when you use two of the same kind of connectors, such as the Deutsch-8. What that means is you need to ensure that you mark each so there's no confusion as to which plug goes where; plugging in the 12V Deutsch 8 into the other Deutsch-8 might blow out a bunch of sensors or maybe even the Microsquirt itself...so be careful.
Maybe you use the "A" and "B" keys for the Deutsch -8s, and those lesser-expensive kits for everything else? Your call.

Tools

I purchased the following tools, also not included in the price sheet:
  • Distributor drive puller, Amazon B07622JJYN, $21.27
  • Sargent Open Barrel Contact Crimp Tool, Amazon B00CIRFUMQ, $34.95
  • Deutsch terminal release tool, Amazon B00A6AS4G6, $26.07
  • Bosch terminal release tool, eBay 292357522562, $6.49
  • Deutsch Pick Removal Tool, Amazon B00O7S43AE, $12.65
  • USB-Serial Adapter, USB-2920, $22.49 at DIYAutotune
  • RS-232 extension cable, TuneCable6, $5.50 at DIYAutotune
  • TunerStudio MS Ultra, Tunerstudio.com, $99
I also bought the Eastwood Weatherpack crimper kit, PLA00010, for $86 but that's when I thought I was going Weatherpack on everything. The tool still works OK on Deutsch stamped connectors but I suggest it's redundant to the Sargent tool above. I'd suggest finding a good ratchet-action dual crimper designed for the Deutsch pins (or, even better, go with Deutsch barrel-type connectors).

I think that were I to do this all over again, I'd just do it with Deutsch barrel-type connectors. Love those things.

Wiring Harness Build

By far, other than working through the general design, the most difficult and time-consuming part of this project was fabricating the wiring harness. I've got half the battle fought for you: I've already done the design. Now you just need to fabricate it.

Wiring harness pics

Harness pics can be found in my Google Drive folder for this project.

Harness lengths chart

There is a worksheet in the Excel Workbook linked above, labeled "Wire Lengths." That chart shows distances between connectors, referencing the connector numbers as labeled in the "Wiring Harness Design Board" photo.

Connector Pinouts

There is a worksheet in the Excel Workbook linked above, labeled "Connector Pinouts." That indicates the wiring locations within each pin, referencing the connector numbers as labeled in the design board photo.

The Build

My suggestions.

Print out my wiring harness pic as large as you can, and tape it to the wall where you can easily glance at it.

Print out each and every one of Mario's wiring diagrams for the Microsquirt and tape them to the wall. If you don't have a color printer then bring them to FedEx Office or something and do them in color:

http://thedubshop.com/microsquirt/

Get a big piece of plywood, at least 4x4 if not bigger, and prop it up somewhere vertical where you can stand/sit/work comfortably. Use that design board photo above and my Excel spreadsheet to draw on that plywood, in 100% scale, exactly where these connectors will go. Write in the connector numbers and what they're for and drive a nail into each location where you can wrap wires ends there. Draw it out so you can run the wires to that location on the plywood. Resist the temptation to add too much length to each (though an inch or two won't hurt).

Add more nails where wires make turns so you can bend the wire there.

Next, decide where you're going to mount your Microsquirt. If you're going to the same place as I did, then no additional length will need to be added to the drawing. However, if MS will be somewhere else then you  will need to add that much length to the mounting point of the Microsquirt ECU. BTW, my wire harness was about 3-4 inches too long but still manageable, and it gave me plenty of slack to accomodate engine movement. I'd not change it if you're going to the same place.

After everything on that plywood is marked and nailed, review all the wires in the MS harness and split them between what you'll use, and what you'll not use. Take some painter's tape or Velcro strap to keep them separated. Check, recheck. Once you're satisfied that you've got it right then de-pin unneeded wires from the Amphenol connector and put them away in a safe place (they're easy to add back in later if you make a mistake, or change your mind, or want to add a feature.)

All set? OK. Drive a nail in the board where the Amphenol connector belongs and use a tye-wrap or Velcro strap or tape or something like that to secure the connector to that nail. That's "home base".

Now it's time to start running wires. Starting from sensors the far end and working inward, take each wire from the MS and route it to the appropriate components connector nail, one by one, bending around turns to that connectors and then. Wind a few turns around the nail then trim off the excess. Leave yourself some good slack as you'll tidy 'em up later but don't feel the need to make the wires either too slack or too tight. Make 'em just right.

Trust me, that first cut will take some courage, but it'll get better. But remember that at this point all you're really only risking is the $79.95 MS long harness. Any terrible errors at this point can either be repaired, or the harness replaced.

After all the sensor/MS wires are done, then do the same for all the 5V power and grounds for the sensors, coming into and out of connector C4. Use the excess trimmed to run it to the other sensors from there.

A note on the sensor 5V ground wires. The MS uses the black/white for sensor grounds. However was not enough excess wire for me to use that color for all sensors grounds. So, you will see a note in my chart that I used some black same size wire that I had for the remainder. There's nothing wrong with that, as long as you notate it and ensure you keep track of it. And remember that MS also uses black for chassis grounds for things like the O2 controller, coil, and the MS itself. That is a distinctly different ground situation than the sensor grounds so don't get them confused. I suspect if I could have bought some white/black wire to finish that but hey, "he was on a roll."

Next. run the wires for all the 12V-powered components: injectors, coil, wideband. Those all home back to C3.

Once all wires are run, and you've double- and tripled-checked everything, now you can start working on trimming, stripping, and terminating wires. Don't worry about wrapping the harnesses in the sleeving yet, because you've bought the correct tools to de-ping the terminals which allows you to put the protectives sleeves on later...which you'll do when you're done terminating.

Trimming, terminating, sleeving done, grab some cloth wiring tape and wrap up the edges; try to get it as weather-resistant as possible (water loves to capillary to odd places).

All done? Grab a beer and enjoy your work. You deserve it.

Chassis Wiring

You'll note there's some wiring that needs to be done on the car for the MS to interface, on the car side of connector C2. Your C2 connector will have wires that needs to go to the T4 relay plate plug, tach to the T12 plug, and (optionally) the T12 plug for starter, oil pressure switch, and reverse lights. You'll also need to install a relay to take power from the battery and drive the aux relay. All that is in the chart.

You'll also need to install a Tach-A-Dapt behind the factory tach in order to get it to work, and a diode behind the gauges from the GEN light to the engine bay so that the engine turns off with the key.

I told ya: the hardest part is the wiring. But you're almost there.

Installation

OK, job well done. Install all the major components and sensors, ensure proper wiring protection from crimping, bending, chafing. Install the wiring harness and secure it properly. Look over your work and get ready to re-install in the drivetrain.

Note, you'll probably need to leave the injector rails off  for drivetrain installation, as the hoses stick out to far and will interfere with the body of the car. I'd suggest buying a set of the factory angled fuel injector hoses to assist with that.

Installed? Ok, turn the key on (but don't start it). Your first clue that things are likely good is that you'll hear the fuel pump run for a couple seconds then turn off. That indicates that the MS has power and that it is controlling the fuel pump. If that didn't happen then recheck everything.

Jump the fuel pump relay and check for fuel leaks. Crank it over to see if it cranks.

Tuning

I'm guessing you're aware that your Microsquirt likely does not have a configuration, yes? Without that, it won't start. I'm hoping you have a basic understanding of EFI and Microsquirt and can config the box to at least get things rolling. Here's some basic resources you'll need.

TunerStudio software. The tuning application. I purchased the TunerStudioUltra MS version to have full access to all features.

USB Adapter. The MS comes with the RS-232 tuning cable, but not too many laptops have RS-232 ports any more. Use this adapter to interface with your USB port.

Alternatively, use EFI's Bluetooth adapter. I bought this, too, and ran the MS cable through a firewall hole I drilled near the engine lid release cable. I then use Bluetooth on my laptop to connect to it.

Megalog Viewer. Reviewing all the logged data afterward to work on your tuning. I paid for the HD version.

For basic startup, go to Mario's TheDubShop pages and review the specs for each sensor. He lists in each how the products should be set up. I chose to pay Mario to give to me a basic startup file, but I've tweaked it quite significantly since then. I've also contracted PF Tuning to help me improve the file significantly.

Or, you can take it to someone that has access to all these tools, can configure the unit for startup and proper operation for you.

What Now?

As I write this, April 2020, we're in the midst of the COVD-19 lockdown. I've gotten out a few times to do some tuning and I really do like the way the car drives. Starts nice, drives smooth, shifts nice, snappy throttle, and if the VE chart is a good indication then the torque has been improved and moved upward around 750RPM higher (and I can feel it). That means better driveability and more horsepower. Once this breaks I'll be looking to take it to a dyno and get some hard numbers and finer tuning. I'll update info here as I have it.

I know this is but a shell of what you're looking for, but I hope it's a good start for you. I'll be re-reading asnd editing these posts for errors and clarifications, and I'll watch the comments section for questions.

If you're interested in trying it yourself, don't hesitate to give it a shot!

Greg

FAQs

Would you be interested in building a harness for someone?

I'm hesitant to do that for a few reasons: one, I'm not a professional though I believe I do professional-quality  work; as such it could be a big liability if something goes wrong (what if someone sets their car on fire?)

Two, all I can offer is the exact design I have with no customization, so you'd have to want to do exactly what I did, in the way I did it.

Third, and most importantly, it wouldn't be cheap. I'm guessing I have at least 10 hours in this harness design/build, and I'm not going to do another one for minimum wage. So you would have to expect the cost would exceed what you'd want to pay for it (and certainly more than what you can do it for yourself!). With time and materials the cost could approach $750-1000; you can do it yourself for a lot less. But I'd consider it. Or maybe we can find a company that is interested in the work?

Would you be willing to consult/assist someone with theirs during their build?
Sure. Leave comments here and I'll respond when I can. If you want one-on-one consulting services we can discuss it.

Can I have your msq file?

As with building a harness, there's liability involved. I'm concerned that someone may just dump the file into their own car and not use the tools to ensure it's applicable; a bad tune can wreck an engine. But as with one-on-one consulting, it's something we can discuss.

Do you know if anyone can offer this as a kit?

I hope to find someone someday, but I have not as yet. If you know anyone that's interested, I'll be glad to give them all my design details and history not included here, as well as offering consulting services to ensure they're successful with this design.

Go to Part 11, Addendums and Lessons Learned

Friday, April 3, 2020

This situation will end in one of three ways...

This situation will end in one of three ways:
- Polical easing of the restrictions;
- Civil disobediance;
- Civil unrest.
We are already starting to see the second one happening, currently in limited ways. However, its increase, absent the first option, will result in increasing authoritarian response. Which, of course, will result in the third option.
We only have a very limited window of opportunity to emerge out of this gracefully; that window is quickly closing.

5/29/20 update: Just as governors were finally relenting and beginning to "allow" society to emerge, the death of George Floyd happened. By then, the societal rage was too pent-up. The third wave has begun and the civil discourse and further societal splitting has begun.

Friday, March 20, 2020

On "Microsquirting" the Porsche 914 - Part 9, Tuning

Tuning

Back to Part 8 

Wow. You've made it this far...? Good for you, you're no "TL;DR"! Or did you cheat and skip ahead...?

Tuning. Another "oof, I have no idea what I'm doing." Gotta figure it out though.

I began with a baseline tune from Mario at The Dub Shop (if you're gonna do this, buy from Mario.) That's what I used to fire up the car. But Mario has not done a setup like mine; Mario usually does his individual throttle body setup for Type 4 engines. That's overkill for a dead stock engine like mine (and outside of my target budget). So I needed to improvise. But, he knows these engines so I requested a base tune.

I adjusted Mario's tune for a few things like injectors and fuel pressure. The stock FPR would only do 36psi; the injectors I used were rated at 3 bar (43.5psi). So I used the TunerStudio program and an online calculator to adjust the injector size and edited that into the tune. I reviewed all the other settings and for the most part they looked OK (not that I really knew what I was looking at...)

You saw the start in the last part. I was quite surprised and happy. But after start, the car kinda/sorta idled but I could tell the AFR was all over the place. I could drive it up and down the driveway but driveability was very poor. I guess that should be expected since we really didn't know where this combo needed to be in the fuel table. I tried to take it around the block but it was horrible: running like crap at partial throttle, bucking, stumbling, really undriveable. AFRs bouncing all over the map. I nursed it home and took the whole map down a few notches and that seemed to improve things but it was still not good. Going further down didn't improve it. I let it sit in my driveway and warm up and I played around with the cells where it was trying to idle, messing with timing and fuel to get it to idle nicely (it helps that this engine starts to RPM surge a bit when it gets too lean; just fatten it a tad and it's happy; 13.8 AFR or fatter cold is about right). It was starting and idling OK. 

But when I took it out on the road again, it still drove like crap during throttle transition and mid-range. I flipped on auto-tune to get a rough idea where it needed to be, and while I could get it to run at idle and at full throttle, anything in between was terrible. The brother-in-law and I tried a few radical changes in timing map, AFR targets, nothing worked; bro almost lost his lunch from the bucking!

The Dub Shop Fuel FPR
I started reviewing the logs (excellent bedtime reading) and noticed that the TPS map was inconsistent. Fue enrichment was scattered. I'd also noticed that I had to set and reset the TPS range a few times as it didn't seem to come back to exactly the same ones. I then logged just the TPS trace without the car running and even at the stops of idle and full throttle it was wavering a few percent. I began thinking that maybe I had a bad TPS which was causing accel enrighment to kick in even with steady throttle. So, on a hunch, I changed my accel enrichment from 100% TPS-based to 100% MAP-based and took the car around the block...and the car smoothed out and ran like a champ! AFR targets started coming in and the fuel map started to get smooth and consistent.

So was the problem was bum TPS/accel enrichment? I began questioning the value of having a TPS at all! Or, I'm wondering if the cheapie eBay $15 TPS was at fault.

I had already planned to buy a new Bosch 3-bar FPR from Mario, so I decided to go ahead and order a new non-generic Bosch TPS from him ($54, ack). Parts came in soon enough (thanks again, Mario!) and were a quick install. I doubled-checked I had a consistent 3-Bar fuel pressure (I did) and I reset the TPS range in TunerStudio MS Ultra (check) and did some neighborhood auto-tuning runs.

It ran about the same as it did with the old TPS. Initial auto-tuning showed big changes in the fuel map; not surprising, given we had a baseline map that could not have been expected to be accurate. plus, the poor driveability was gone so I could see that things were progressing. Having smoothed out the map, both automatically and manually, and proven that the driveability issues were gone, I adjusted my targets for AFR and ignition timing to be a bit more aggressive. Car was running better. Things were looking up!

Oil, oil, everywhere...
But there was serious unrelated problem: I had one HELL of an oil leak. I mean, enough oil that the exhaust on the left side was getting covered, and burning oil. Right side was bone dry. Sounds like a valve cover, right? I replaced the VC gasket and the problem was exactly the same. I had enough of an oil leak to where I'd stop, look under the car, and could see the oil drip-drip-dripping down. I was unamused.

I commiserated with Chris Foley via text (over a beer; we like beer) trying to brainstorm lwhat this could be. I put the car on the lift and could see oil everywhere on the left side, seemingly more toward the front. Right side dry. But I had replaced all the seals up front, the VC gasket looked good, the pushrod tubes seals were new and looked good, and what could there be over there causing that oil? Where else can oil come from?

It wasn't doing this before I pulled the engine so I started thinking about changes I had made. New crankshaft seals, I'm sure they're good; new Tangerine stainless pushrod tubes and seals, those look fine. I replaced the oil cooler with a Tangerine remote oil cooler block but that was over on the right side, so that's not causing a leak on the left side. And since I was doing the remote oil cooler I removed the factory flapper and thermostat bellows system so...hey, wait a sec...
"The Hole"

Greg: "Hey Chris, is the hole where the thermostat bellows mounts a blind hole...?"
Chris: "No. The bellows mounts to two studs but the pulley screw hole goes through."
S**t.

See that little threaded hole in that photo, way up there? That's the mounting stud hole for the pulley for the small cable that ran to the air-control flappers. And Yours Truly had removed that pulley, along with its mounting bolt and the thermostat bellows, not realizing that that little hole opened all the way into the oily parts of the engine innards (Why? I dunno). I'd been dumping a lot of oil out of that hole, and it made a freakin' mess. Leak fixed with a bolt and a copper sealing washer. Amazing how much oil comes out of that little hole...

At this point I had a lot more confidence in the car and took several opportunities to setup and use the auto-tune to build a solid fuel map. I even got brave and took it to work a couple times! The setup is still at 100% MAP-based accel enrichment* and I'm pleased with how it's driving so I'm not clear how much TPS could improve it. However, I'm working with Peter Forance of PF Tuning and he's going to take my current tune and work some of his magic with it. From there I'll work on warm-up enrichment (the engine really likes to be fat when cold, but tolerates leaning out when hot) and then play around with AFR targets and decide how aggressive I want to get on ignition timing. Maybe I'll even build a separate map for 87 octane versus 93 octane (though, as little as I drive it through the year, I won't save much $$ with 87...but hey, tuning is free!) That's probably something I want to work on via some dyno tuning; John Malapetsis of Performance & Styling has a DynaPack, my preferred dyno. I've worked with him in the past on race car tuning. I probably should also get a CHT gauge...

One other design note: the factory tach does not work with the Microsquirt. There's an easy enough fix: Pelican Parts sells Ashlock Tech's "Tach-Adapt" tachometer adapter. It took 20 minutes to install with basic wiring hand tools and I mounted it behind the factory tach. If you want to get fancy, there are forum threads out there on modifying the gizzards of your Porsche tachometer to install modern electronics while keeping it looking stock, but for $70 the Tach-Adapt was my easy button. Yeah, the stock tach is still lazy (it's just sweeping past 5600 when I hit the 6000RPM redline) but mine's a street car, it'll work great for me.

Or, if you'd like a modern tach that will pretty much slide in, SpeedHut makes a 4-1/2" tachometer that looks really close to factory and can be easily installed. I bought one of those for the race car; I'll post some photos when I install it later this year.

So, that's pretty much where I am at the moment. It's running (yay!) and dependable (yay!). I'll also try some tuning changes to see what happens and to learn more. I suspect my hardest part will be getting into the 914 and not feeling compelled to connect the laptop to do that last nth tuning...

Someone online asked me how it drove. I replied,
Hard to say since I've not fully tuned it, but it drives a lot like...a solid perfectly-working D-Jet system. It shouldn't be surprising, given that the D-Jet system was pretty good, especialy in context. Despite being from 50 years ago.
Startup is smoother and easier, no throttle needed. I removed the aux air regulator and did not install an idle air control valve, so I have to tickle the throttle for about 30 seconds before it will idle, buit I think I can tune this with ignition timing and warm-up enrichment. No external idle control needed.
Tip-in with the engine cold is very good, no stumbling at all like I used to get with a cold D-Jet engine. Throttle response, even dead cold out of my driveway, is improved.
The one thing I really noticed is improved driveability, both in neighborhood streets and on the highway. The car seems to be running much, much smoother.
My confidence level with the engine management is significantly improved; no more occasional stumbles, wondering if it'll catch (the D-Jet left me on the side of the road one time last year, wire problem).
This modern system has some advantages:
  • Easily-obtainable, lower-cost replacement parts. For the vast majority of the design, I can replace anything from a FLAPS;
  • Electronic ignition. No more disty and its associated mechanicals (vacuum and centrifugal);
  • Related to the above: built-in rev-limiter;
  • Tunable. If I make any changes to the engine (cams, compression, displaement, etc) I can whup out the laptop and change things to accommodate;
  • Economy/power compromises. I can move it toward economy at the low load/throttle settings, while moving it toward performance on the higher-end. Or create a different setting for 87 or 93 octane.
So, is installing Microsquirt "better" than the D-Jet for a dead-stock engine? Sure. The vast majority of the D-Jet system is available and the 914 community still supports it (and, frankly, prefers it to carbs for a stock engine). But all it takes is one D-Jet MPS rebuild ($450), one 1-2-3 distributor replacement ($500), and one new wiring harness ($500) later and you're at the cost of a complete Microsquirt system. But for a modified engine? It's absolutely better than D-Jet. And for a highly-modified engine, Mario's ITB Microsquirt setup is the way to go.

Would I do it again? Yep. I'm very pleased with it, I enjoy the tinkering, and I learned a lot. And I trust my car a lot more.

I'll post more tuning details here as I learn.

On to Part 10: The Design Details (coming soon)

* 3/22/20 Update. Peter Florance sent me an updated tune that adjusted quite a few things, including the TPS/MAP enrichment. Car is improved, still no more bucking/stumbling and a bit snappier on the throttle tip-in.

Warm-up enrichment (WUE) is pretty good. Our ambient temps are now 30s to 40s at night so I can't tune warm-up for below that, but honeslty I rarely if ever drive the car when the ambient are much below 40F. I'm still trying to find a good balance for WUE that allows the car to idle on its own withing 30 seconds - remember, no idle control valve - and idle nice when it's hot. We're picking at the edge now.

I also added a Qstarz BT-Q818XT GPS to my laptop to use while tuning, so now I have data such as speed, acceleration, and even estimated fuel economy (low- to mid-30s on the highway right now). I may look into some kind of bluetooth capability for the Microsquirt; first, I need to get rid of the cable that I have coming through the passenger window; second, I'd like to use the iPhone as a head-sup data source as it's hard to watch the laptop while driving.


TDC - Top Dead Center, the firing position for #1 cylinder
CPS - Crank Position Sensor, also known as CAS, Crank Angle Sensor
TPS - Throttle Position Sensor
MPS - Manifold Pressure Sensor
IAT - Intake air Temperature sensor
CLT - Coolant temp sensor
AFR - Air-Fuel Ratio
FLAPS - Friendly Local Auto Parts Store
FPR - Fuel Pressure Regulator

Thursday, March 5, 2020

On "Microsquirting" the Porsche 914 - Part 8, So, Will It Run?

So, Will It Run?

Back to Part 7

Spoiler: ...it runs! But first...

The drivetrain install went fairly smoothly. If you've removed/installed one of these, you realize it's really not so bad of a job. I hate, hate, hate the concept of removing a drivetrain to repair something (Lotus manual for changing the oil: Step One, Remove the Engine) but the 914 is really not that hateful. A couple wires, a couple cables, the eight (yes only 8) drive axle bolts, and the whole thing is held in by four more bolts. I used to laugh at Edd China casually removing a drivetrain on a car to do some work on it, but after watching him remove the 914 drivetrain in an episode (Season 9, Episode 5, worth the watch), I'm kinda there (and I use his "drop on my toolbox and raise the car off with the lift" trick). Yeah, it's a pain in the ass to a certain degree, but the amount of time you save having things right there in front of you and accessible makes it so much morez betterz....anyway...the bro-in-law and I had the drivetrain installed in an hour or two, including beer (exact time lapse unknown, mostly due to "including beer"). He went away and I spent the next couple of evenings after work tidying up stuff like wiring, hoses, cables, axles...and I spent some effort installing a Tangerine Racing rear-mounted oil cooler (took a little engineering but job done).

Firing up a car for the first time can be stressful. There could be fuel failures (leaks, failed components) electrical failures (sparks from bare or improperly-terminated wires) -- or worse, both (top tip: fuel leaks and sparks in combination are bad, m'kay?) So I kinda did it in stages.

First, I pulled the fuel pump relay and briefly connected the battery ground wire then quickly removed it, listening for sparks...nothing. Then did it again, slightly longer...nothing. So then I screwed it together and tehn looked all around for heat, sparks, hot wires...nothing. Good so far.

Then I went in the car and turned the key on and off quickly...nothing. Turned it on for longer...nothing. No sparks, no noise, no hot wires...so far, so good.

So next I turned off the key and inserted the fuel pump relay and turned the key on. The fuel pump ran for two seconds which is exactly what was supposed to happen. That told me that the Microsquirt ECU was getting power and properly controlling the fuel pump. So what could possibly go wrong...?

I wanted to next test for fuel leaks so I cut the top off of an older relay I had to it as a jumper switch. Once the top is revealed you can push down on the solenoid and it'll engage the relay. I installed it in the FPR slot and used a piece of paper to keep it on so I could measure fuel pressure and check for leaks. I was surprised to find that the stock adjustable fuel pressure regulator (FPR) would only screw up to 36psi...I was hoping it would go higher. The injectors I chose were spec'd flow at 3 bar/43.5psi so I really wanted it to be there, but I figured 36 wasn't that bad so I went with it.


Everything was set and checked. I got the bro-in-law to bring over some tasty local Connecticut beers for the reveal. We got my two fire extinguishers in key locations (by the two garage exit doors) and we jumpered the fuel pressure relay and let it build up and checked for fuel leaks. None. All set.

So, armed and ready, we cranked over the car and let's see what'll happen...

I'll be damned if that thing didn't actually fire right up! Well, "right up" is a bit of an exaggeration, but not much: the first time I cranked it we got a "pop!" from it but no start...but the second time it fired up with some throttle and I could keep it running.

The damned thing actually started. It started! With oil pressure! I was so ready to have to troubleshoot wiring, pump, oil, ignition, injector, whatever problems. But nope, all the homework paid off. No one was more surprised than me.

Yes, by bro-in-law took a video (he's a Honda guy, into that "engine start up video thing") but I hate the way I look in it so you're just gonna have to YouTube search it for yourself. But I did have another beer.

(Edit: eh, screw it: https://www.youtube.com/watch?v=tc_WD-302Oc)

It actually started...damn. With no problems...
.
.
.
.
.
.
.
.
.
Well...except for one little minor miniscule silly nothing not-a-big deal problem: the car, once idling, wouldn't shut off with the key. It would start, it would run, but when I turned the key off...nothing changed. Regular noises continued. Somewhere in my electrical design there was a flaw. Either the alternator was still powering the system or I failed to accomodate an unintentionally-latched relay of some kind and the car was running on the alternator. I pored over my "CAD" wiring diagram ("Cardboard Aided Design"), doing some FEA ("Ffffffriendly Everything Analysis") to figure out my "fale".

The T4-I relay panel pin that I'm powering the Microsquirt with is the same pin/circuit that powered the D-Jet system and it was used for both the D-Jet ECU as well as the coil and injectors. It was done the same as the D-Jet. I was kinda stumped.

After some basic diagnostics, I found that this circuit was still hot when the key was turned off...but it shouldn't be, it's fed by the ignition switch. And that circuit is dead when the car isn't running...normally. It gets powered by the original "power relay" which supplies power for everything in the engine compartment. I pulled the power relay and the car died. I pulled the fuel pump relay and the car died. I pulled the plug at T4-I and the car died. I pulled the alternator plug at the relay panel and the car died. So the alternator was keeping those circuits powered up. But how...?

I'm not gonna bore you with the troubleshooting details, but I took a WAG that the alternator/GEN light on the dash was involved. Somehow. So I removed the GEN bulb from the back of the gauge, started the car, turned the key off...and the car died. So, electricity was back-feeding through the warning lamp and keeping the electrical system energized!

But that left me with two questions: why did it not do this with the D-Jet system, and how am I going to fix it? On the first question, the only thing I can figure is that between the old D-Jet ECU, coil, and injectors, that maybe they pulled enough amps collectively such that there wasn't enough flow through the dash lamp to adequately power it all so it died. Because otherwise, it's the same exact wiring design that was used for the D-Jet system.

And as for how I'm going to fix it...the answer was "a diode". Specifically, a solar panel diode I had laying around in a tool box. I added it to the D+ wire from the alternator and all is well. I just won't have overvoltage notification (a very rare failure).

But the damned thing runs. How about that?

So on to Part 9: "What About Tuning??" 


TDC - Top Dead Center, the firing position for #1 cylinder
CPS - Crank Position Sensor, also known as CAS, Crank Angle Sensor
TPS - Throttle Position Sensor
MPS - Manifold Pressure Sensor
IAT - Intake air Temperature sensor
CLT - Coolant temp sensor
AFR - Air-Fuel Ratio
FLAPS - Friendly Local Auto Parts Store
FPR - Fuel Pressure Regulator