Breaking the silence again... Things keep happening but I keep forgetting to post.

On a flyer I tossed a set of B21FT pistons next to the Saab B202 pistons. Both pistons have 24mm pins but the Volvo pistons are 2mm bigger diameter and are quite a bit taller. The obviously huge dish in the Volvo piston would have to get cut down pretty drastically, which would remove a lot of the top ring land and would weaken that part of the piston somewhat, but it COULD be done if I wanted to undertake such a thing. Obviously the goal would be to increase volume and decrease compression ratio, and since Saab pistons are unavailable (unless I want to spend $800 USD on 8:1 forged racing pistons) one needs to think about these things...





While I think, I've bead blasted the Saab pistons after protecting the ring lands with a little E-tape. The skirts came up a couple tenths of a thousandth of an inch, and will retain a bit more oil this way. I've performed this method of reviving an old piston a few times with good results.





Bumped into this guy at the Rona. He's in the next neighborhood over. I forget the year of his car but it's quite similar to mine except that the original owner checked every option box he could - sway bars, center console, etc. I had my car in ACC and was listening to the stereo while I waited for him to come out, and while we chatted, and all the time I forgot that the stupid headlights stay on in this position. Battery was stone dead by the time I went to leave...



Not too much later I discovered that my dash repair wasn't foolproof. It started cracking again, right beside the bondo spots. There are days the crack is invisible and days that it's split wide open, so obviously the dash heaves and pitches as the temperature changes.

Time to fix up those Enkeis. I pulled the mismatched 205/50 tires off them and started looking at the wheels. Every single one of them was bent to some degree - two had pretty good flat spots and the other two were just warped a little. I screwed them down to my workbench and wailed away at them with hardwood and hammer until I was fairly happy. Roundness was verified on my lathe - there was only so much I could do...

I wasn't sure what I wanted to do about the coating - they'd all been curb rashed and they all had chips & gouges in them, and I knew it would bug me, but if they were powdercoated I knew it would be a PITA to deal with. I gave the backside of one of the spokes a quick scrape with some sandpaper and quickly got down to metal, so I decided to toss them in my blasting cabinet. That's when I discovered that while the backsides carried a light dusting of paint, the fronts were all powdercoated!!! I spent a good long time erasing my fingerprints while sanding these down. To topcoat I used a new-to-me product called Endura and I had a little trouble, so I ended up re-sanding and re-topcoating them again. The whole paint process ate up about 4 weeks and they're not perfect, but they're finished.





I tracked down a set of sway bars and paid dearly to have them shipped from Florida, but you'd be surprised to find how difficult it is to get someone to remove a set. I feared installation would require a lot of trouble, as the front one has to wind in and around the steering rack through the little holes in the frame, and the factory instructions I'd found included templates on where to drill for the mount bushing fasteners, but fortunately I'd removed all my power steering and by 1987 Saab was pre-drilling holes for sway bars, so the complete install took about 2 hours. No pictures.

With the bars in and the wheels on, other handling "traits" were more easily observed. I felt the next step was going to be stiffer control arm bushings, but poly bushings for these cars are quite dear and I'm not thrilled with poly in this location so I ordered some Delrin from McMaster-Carr and turned up some bushings on the lathe.





I also blasted, straightened, reinforced, primed, and painted the lower & upper control arms. The uppers can't go in until the engine comes out, so I didn't finish the bushings for this location either, but the lowers are done and installed. Steering response is much more predictable, the ride is slightly firmer and there's a little more NVH, but considering the bushings are solid plastic, it's pretty darn good.





(I did discover an interference between the pinch weld on the body of the car, directly below the driveshaft through-hole, that interfered with the top reinforcement plate of the control arm. I marked it and cut a slot for clearance, problem solved. Again, no pictures).

While I was thinking about control arms and bushings and such, I was thinking about springs. With a little research I found a JY spring that, when cut to height, should very closely match the SPG specifications. The front springs are from the rear of a '92-96 Olds 98 and the rear springs are from the rear of an '02-07 Buick Rendezvous AWD. The rears ended up a bit stiffer than stock SPG and the fronts ended up a bit softer than SPG so the ride balance is a bit truckish, but the handling is great! I cut them to published free lengths of 11.7 front and 11.5 rear, and the car sits with perfectly even fender gaps front and rear. If the stiffnesses were TOO far off, the fender gaps would be whacko and they're pretty close so I guess we're in the ballpark. I may build new upper & lower mounts for the front springs that incorporate coilover springs and adjustable perches, and will probably try some 450 in-lb 10" long coils, but they're pretty excellent for now. ZOMG!!! I cut my springs! F**k off, get over it.

Before cutting:





Before and after, forgive the different perspective but you get the idea.

I guess one more picture-free post: We've decided to enter this car in the 2018 GBC event. Things are going to get a little silly. More on that later.

Last weekend I did some basic maintenance to the car - new plugs, new wires, new injector seals, new belts, and a new hood release cable. While I was in there, and while it's summertime, I pulled the heat stove assembly and cobbled together a cold air intake from junk I had lying around: a 2.5" 90 degree silicone elbow and a chunk of Volvo intercooler pipe I'd cut a piece off and tossed into my stash. The elbow fits perfectly and requires no more than light clamping to the K-jet airbox, and the tubing runs down to the bumper grille area like it was designed that way! It was immediately apparent that when hot, the car had more snap. I also discovered (shame on me for not discovering this earlier) that the vacuum advance in my distributor has a torn/rotted diaphragm and was causing a huge vacuum leak at part throttle. I have no spare distributor so I capped the TB and my light cruise stumble is all but gone - go figure!

ZOMG!!! I cut my springs! F**k off, get over it.

Lol!!

Nice work as always Matt! You and Craig amaze me with all the repurposing/fabricating ideas you come up with. Retail!! We don't need no stinking retail

Thanks, Jim. I'm not sure all of my (or Craig's) ideas are actually good ones, but they're usually good exercises. It sure made the car perform better at Streetwheelers than it did at one of the LSCC autocrosses last summer, and on my trip from Lethbridge to Writing-On-Stone to Waterton to Calgary it got better than 30 USMPG at Vanagon speeds, no sore back, no problems.

Once I got home I figured out my Megasquirt issues, so it's now running with MS for ignition and idle control. The vacuum canister on the distributor has been dead for quite a while, so all that cruise advance has been lost to me - now that I've got MS simulating that, I may do even better for mileage! But the important thing is now I have the ability to retard timing under boost, which will come in handy for... oops, I shouldn't say anything.

Megasquirt notes to self:

Dizzy has a hall sensor that requires a pullup resistor - I used 12v from the S12 source. 5v worked sporadically but may have been caused by other issues (more later). Maybe I should try a 5v pullup to see if that's reliable... I'm driving the coil directly - 12v from a relay, switching ground to MS. My board is a V3.0 with a MS2 daughtercard, so I'm using JS10 and the high current circuit I build it when I populated the board in '06 or so. JS10 is jumpered to OptoIn and OptoOut is jumpered to IGN, the board has a VB921 on it (may switch to the upgrade transistor mentioned in the assembly manual). S12C is jumpered to JS9 but without using the stepper motor circuit this is probably not required.

I machined an adapter to mate a Hyundai TPS that I had on hand to the stock throttle body. I'm using a Bosch CT sensor with the larger thread (M12?) in an unused coolant hole. TunerStudio allows for a quick thermistor value change. I'm using a caged GM IAT sensor zip-tied to the bumper braces to gather ambient air temperature. I'm using a VW/Audi 2 wire idle air valve but may switch to something smaller and/or more airtight (it bleeds air through the valve when fully closed).

For now I'm leaving the mechanical advance active and letting the dizzy do what dizzys do. I just twisted it 15 degrees advanced so that I had a little overhead for "vacuum advance". For now the trigger return is set to an arbitrary number of 40, and the timing map is built up as a delta advance table. A value of 25 on the table is Delta 0 (40-25=15 degrees of electronic retard to negate the 15 degrees of physical advance I gave it). I've added the full 15 degrees of "vacuum advance" at engine speeds above idle and at max vacuum, tapering down to no advance by 70 KPA or so. I've included 10 degrees of retard for boost pressures up to 140 kpa, tapering down to no retard by 100 KPA. I can add or subtract timing wherever I need to, but the dizzy curve is probably fine for now. I've pulled 5 degrees at all the idle cells to compensate for the leaky idle valve. Cranking timing is set to match these cells at 20 (5 degrees of retard). Too much cranking timing and/or too much cranking dwell causes a no-hot-start condition. At the moment I'm using 5ms cranking dwell, 2.5ms running dwell, and 0.1ms spark duration.

I'm using closed loop idle control, though I'm having difficulty tuning it due to a bad USB/DB9 converter creating lockup of the MS unit. So far the closed loop idle control works okay, though if I drag it down too far for too long the idle hangs with the valve open, and doesn't seem to slow back down again until a few seconds after I blip the throttle. More tuning required when I get a better USB/DB9 converter. This converter may have been the cause of my unpredictable 5v Hall sensor pullup resistor troubles... or not.

I've got an output to drive a single fuel injector - I'll probably add another cold start injector in the intercooler pipe for a little extra fuel when required. I thought MS would deliver a fixed amount of fuel on cranking so I was going to hook this to the original CSI and use the same injector for both uses, but MS2 seems to use a proportion of one of the fuel tables to calculate required fuel for cold starts. Since I'm not using the CSI for fuel and the fuel map has values of 0 pretty much everywhere, I can't use a proportional cranking value. More research needed.

O2 sensor input wire is ready to go for when I can start datalogging.

Maybe I will include an input for a launch control switch, because the transaxle needs more abuse. More research needed on how to do this.

Great Beater Challenge 2018 was a near total success. Craig & I were edged out in a tight battle with one of the people of the tundra who got major bonus points for dragging their sorry selves down to Lethbridge... in an old mail truck. They won by 24 points out of 3300, the next nearest team was at 2875 or so. Craig & I misinterpreted a couple of sneaky clues and we didn't do any of the optional stages, which we learned were worth BIG points (plus unlocked some other points that we were killing ourselves trying to find on the "normal" route). We had two failures on the drive - the ground strap on the alternator broke about 200 km into the trip, and then about 100 km from the end of the trip the big power wire to the alternator sheared off from the lug. I had brought all my electrical fittings and gear so we were back up and running in a total of 20 minutes for the two failures, and we were lucky enough that another competitor was just behind us both times.

https://youtu.be/7NJYCNdBbIE

I'll let the video and pictures tell the rest of the story, ask any questions you like!









And here's an example of thinking "outside the box" (NARF!).



The guys in the 142 ... oh man. How bad did that car sound at the START of the event??? It had a serious rod knock - they started the car at Silver Automotive before we all got underway and everyone stopped what they were doing to watch and listen. It sounded like a bag of hammers had been dropped into a rock crusher. But it only got worse as time went on - we caught up with them late Sunday at our fuel stop in Bragg Creek and not only was the rod knock worse, but the timing gear had also separated. Craig had to yell this to me over the noise the car was making... it was epic.

Awesome job guys, that is hilarious!! I love in the video when the kid in the blue Carhartt hoodie comes over, the look on his face says he has no idea what he’s looking at.

A typical well engineered/built effort from you guys. The Pacer +T was gobbed together, albeit hilarious in its own right.

Thanks Jim.

Hey I’m a dope, never even asked how it performed? Guessing lots of lag with all that piping, what did you run for boost? Any plans for a more “conventional” +T?.

How did it perform? About as you'd expect.

I had seen fairly rich air/fuel ratios under power with it naturally aspirated and with the big VW airflow plate I hoped that it would continue to deliver until at least 2psi, which would match the fueling requirements at sea level. I did add a bung for a second cold start injector and I was planning on running it from Megasquirt as required. The unknown here was how "loose" was the control pressure regulator and how high was it letting the airflow plate lift? I could assume that since the VW was only rated for 90 horsepower it would take advantage of as much stroke as it could for as little restriction as possible, and with the Saab being rated for 110 horsepower (maybe?) I wasn't sure how close I was to the limit already.

The turbo came from an '01-02 Cummins 24v, and that wastegate can started cracking the flapper at 25 psi or so. Fortunately it was a slam dunk to install a Garrett 5 psi canister: I just had to slot one hole in the bracket cast into the compressor housing, and carve up the compressor housing a little for clearance. The unknowns here were that the turbo was pretty small for the original application and the turbine housing generates lots of EGT when towing, and that is on an engine 3x the volume turning at 1/3 the RPM, but at 5x the boost pressure. Low turbine inlet pressure (this thing's way bigger than a Garrett T3) requires a huge wastegate hole to control boost, and this one's flapper is pretty tiny. I had no idea what would happen.

What happened was this: It was safe until about 4000 RPM where it it started to go lean, but it still pulled pretty strongly up to 5800 RPM where it started to lean misfire. I connected Megasquirt to one cold start injector and got it to run safely up to about 5300 RPM and it would pull to redline but by 5400 the injectors were at 100%, so I added a jumper to the injector wiring and cobbled up a fuel line for both the original and my auxillary cold start injector, and ran them both. It didn't seem to make a difference - I'm not sure why, but by the end of the trip neither injector was working and I was back to going lean by 4k. Maybe I was running out of fuel flow or fuel pressure, maybe the injectors were failing (is known to happen, though I'd bench tested them before using them), or maybe the sinful way I had them wired was a problem (they were both high current injectors so I wired them in series so I didn't zap MS - bad practice because if one fails, the circuit stays open).

Boost was really weird too. In 2nd gear at full throttle it would begin to build boost by 2500 RPM. In 5th at 3000 cruise (110 kph) it made 2-3 psi in the intercooler pipes. But the wastegate always seemed slightly open and never seemed to fully open - my datalogs confirmed 2500 boost threshold and by 6k RPM it was making 10 psi, but at no point did the boost stabilize as though the wastegate were actually doing something. At 3000 RPM cruise one would expect it to generate pretty good boost when you went WOT, since it was able to make boost at 2500 in 2nd gear at WOT, but that 2-3 psi that was in the intercooler pipes under cruise went away and took forever to come back.

I also found out that I'd done something else rather sinful and wrong: I had the wideband installed in the exhaust manifold. I've since learned that it will not operate properly if it sees significant backpressure, and it needs to be downstream of the turbo. There's an air pump cell in the wideband that brings fresh air in to compare to the stuff in the exhaust stream, but it can't overcome the turbine inlet pressure. Some of my boost air/fuel readings were surely garbage as a result.

I was thinking that if I had a massive boost leak, as many have suggested would be the case for the erratic boost, I would have seen really rich air/fuel ratios from all that "metered" air being dumped overboard. But two things counter my logic now that I think about it: the wideband reading all goofy, and the stock fuel system being maxed out by 4k RPM. All the fuel Megasquirt was supplying is NOT being metered, therefore if there's a boost leak the wideband wouldn't show it and who knows what the wideband was showing after all.

So yes, it ran fine and it even made more power than stock at a lower RPM, but it ran out of fuel before it could make a whole lot more power than my car makes naturally aspirated. On the highway it was a total dog, a lot of the performance drain coming from the massive dump in aerodynamics the car took. For reference, bucking a headwind on the way home I used 33 litres for the 206 km drive (14.6 USMPG) and for the whole trip we averaged just over 20 USMPG. I generally average 23 USMPG in my daily commute, driving the balls off it, and I got 33 USMPG cruising on the highway this summer.

Plans for not such a stupid turbo? Yeah, probably. More on that later.

I'm going to just leave this here!

Only 27 people watch that show, I figured nobody would have seen this.

“Dodgy done well”

They didn't even see the true detail and they're commenting on it. Buddy, this is fantastic! So happy that you sent in this video.