1988 Turbo 'Vert Project

[toplessFC3Sman]

I've been very hesitant to borrow tools or anything since I really have no idea when I'll be finished - taking care of my son Paul, work on my thesis, the house, and all sorts of things keep getting in the way. Anyhow, I've been slowly chipping away at things like getting the radiator and oil cooler cleaned and pressure-tested, more disassembly and cleaning of individual parts etc. I haven't started cleaning the engine bay yet, but individual parts are coming out ok.

Basically, a giant tub of simple green and water that I let stuff soak in for a few days, then scrub with a wire brush for a little bit, and re-soak if necessary. The two smaller cans are white vinegar, diluted to about 25% to eat away rust on bolts, the front pulleys etc. This has been working well, especially combined with a bit of wire brushing too. Then, all the clean aluminum manifolds, brackets etc in the top tub and box.


Parts storage, after cleaning. Some of the bits have been painted after de-rustifying them, like the water pump, block-off plates and other steel/iron pieces.


Definitely found the source of the coolant leak - these coolant seals were really toast! Both were already broken when I lifted the end iron off the stack, and the inner one had mostly separated into three individual strands. At least this rotor and housing was very clean - yay water injection?






I've been trying to keep everything well-organized through liberal use of trays, boxes, zip-lock baggies etc, to make sure that I know where everything goes back to when it finally comes time for reassembly. So far, all the internals look like they're in good shape, with only a little bit of marking on the end plates where I imagine the rotor sat for a while when the car was parked. No grooves (especially not big enough to catch a fingernail on), and nothing else that looks worrying on the housings or bearing surfaces.



And thats where I had to stop yesterday; about half done. I've had to do a bit of drilling and tapping on various bolts that have broken off in the block, like oil pan studs, one in the front iron for the water pump, etc. I also want to plug the EGR ports and coolant ports into the LIM more securely. No plans for porting - I'm happy with the power potential of the mostly stock porting, and don't want to trade off engine life, idle, or low-load drive-ability for more power at this point. The guy who rebuilt it before said that he did a very mild bit of porting on the exhausts - I don't really see any signs of it, but then again this is the first rotary I've torn down, so I don't have a basis for comparison.

[murz]

Awesome! Are you buying new hard seals? How do the bearings look?

If you do buy new hard seals, I'd recommend FD corner seals and springs. Should take some pictures!

[toplessFC3Sman]

I haven't seen anything that would suggest that I need new hard seals, everything looks like its in good shape. I think that the engine already had the FD side seals and springs when it was rebuilt the first time, but I'll take some pics of all the seals just to make sure I'm not missing something. Bearings look great too, but I should get some pics for documentation. I still haven't gotten into the front rotor yet - maybe this weekend.

[toplessFC3Sman]

Well, I think I spoke a bit too soon in the previous post. Not only did I find a chipped apex seal in the front rotor (fortunately the chip was held in the rotor tip groove)...


... but the bearing in the front rotor looks pretty bad.


Lots of copper showing through, with much of it looking like normal wear, but some sections (especially in that second rotor pic) that look like the babbitt material stuck to the e-shaft of something and got torn off. The E-shaft looks and feels fine, but I'm really not liking that rotor bearing.


Have any of you guys replaced a rotor bearing before? It looks like its just a matter of pressing out the old and in the new, but I know how sensitive bearing tolerances are. Speaking of tolerances, I haven't measured them for the front rotor or any of the other bearings, but I really should.

Also, any preferences for apex seals? I would lean towards OEM since I don't plan on pushing this engine too hard, and would prefer a longer life between rebuilds, but I don't really know what else is out there.

[toplessFC3Sman]

Just fixed the pictures... google must have changed some of the details of how they host and share photos. Anyhow...

The rear rotor looks fantastic, with no spots where the babbitt has worn through, although there was one line where it was slightly rougher than elsewhere. This may have been from when the car was sitting for about a year between draining all the fluids and getting around to pulling the engine?


There was less than 0.06 mm clearance between the e-shaft and rear rotor bearing (that was the smallest feeler gauge I had), and there was about 0.09 mm of clearance for the front rotor. Too much? The max spec seems to be 0.1 mm, but that's pretty close to the max for an engine that I'd like to get a lot more mileage out of.

All the main bearing surfaces look ok, with some patches of rub-through but nothing that looked too concerning.

[toplessFC3Sman]

I ended up getting the front rotor bearing pressed out and replaced with a new one since it was looking so beat up. The stationary gear bearings were a little worn but I wasn't as concerned with them. I got new apex seals, corner inserts, and all the o-rings, crush washers etc required to rebuild the engine a while back, so after a long day the weekend after the 4th of July, the engine was reassembled. I didn't take many pictures of this process since I pretty much had my hands full trying to get it done in a day. Spinning the e-shaft by hand, it moved smoothly and gave distinctive compression pulses on all 3 faces of each rotor. The manifold, OMP & lines etc got assembled over the next few weeks, with one stripped hole in the OMP needing to be drilled out and get a threaded insert installed. I found some extra-wide crush washers that would bridge over the threaded insert, and so far it doesn't appear to be leaking.


Everything got a nice cleaning while it was disassembled too... I hope the cleanliness lasts a little while at least. I filled the ports in the intake manifold that used to be used to pass EGR to the intake and coolant to the turbo as well; I've eliminated the EGR system a while ago, and re-routed the coolant to take it out of the manifold so I wouldn't be unnecessarily heating the intake air. This was all filled with a high-temperature putty; i think it was sold for patching exhaust components, but seemed like itd work well for the manifold too. It was like a giant clay stick, but needed to be kneeded to really stick well.


Two weekends ago, I got a day to actually install the engine and hook up the oil & cooling systems.



One of the oil cooler brackets needed a bit of... reinforcement.


Everything seemed to go very smoothly, but the cooling system leaked from a number of places. A few years ago, I had tried to go through and replace a lot of the smaller hoses with bent steel tubing to clean things up and remove failure points, but at the turbo I couldn't find any 10mm compression fittings so I drilled out a few 3/8" tube fittings to fit. They held for a couple years, but didn't survive disassembly well.


This time I ended up buying some 10mm compression fittings and pieces to adapt them to the 3/8" tubing I had from mcmaster.com, and bent/assembled most of the smaller cooling system parts again this past weekend.





Basically, the coolant to the turbo comes out of the nipple on the top of the rear iron pointing upwards, travels under the manifold and out near the firewall, then turns to go through the pipe bolted to the BAC, then back down to the turbo on the passenger's side, and out of the turbo between the turbo and the engine down to the small nipple on the pump inlet.

[murz]

That's nicely set-up, I ended up bypassing the BAC completely, since it will never really be driven in the winter. That rotor chip could have been really bad, glad it stayed put!

[fidelity101]

i love mcmastercarr for things like this - excellent solutions on the fly! lol

[toplessFC3Sman]

Ok, I've been pretty negligent in updating this recently - lots of other things have had my attention. Anyhow, the engine has been in the car since early Sept, and running since mid Sept. She started up with the first key-twist like nothing had happened, which was very reassuring. At that point I made the choice not to finish things up and get her out on the road so that I could tackle a few other projects that have been nagging at me.

Basically, I haven't been happy with the air-to-air FMIC for a while now, due to the huge amount of piping and intake volume it adds (which is only going to create lag when pressurizing all that volume), the fact that I had to remove the front bumper bar and supports, and that it meant I had to use a smaller battery and a pretty flimsy battery tray that came with the Corksport FMIC kit (unless I wanted to do a lot of re-wiring, which I may still do... Project creep...). Add to that the belt slippage on the coolant pump pulley (which isn't helped by the mechanical radiator fan), and the desire for a functional cold-air intake (not just a pipe and filter sitting where the original airbox had been (also evicted to make room for the FMIC piping), and I decided it was time to tackle the whole cooling system & intercooler air flow problem at once.

First, the radiator e-fan & switching. I started out roughly following some of the You are not allowed to view links. Register or Login You are not allowed to view links. Register or Login on RX7club.com, and settled on the 2-speed Taurus e-fan since the shroud seemed to fit well, the fan area was huge (so no flaps would be necessary), and it was 2-speed to more gradually adjust cooling to the needs. I've actually had one of these kicking around in my pile of parts for a couple years now, so time to use it!

The only real tight spots were where the rad inlet & outlet are, and the stock fan shroud has big dimples to get around the hoses.


At first, I tried to heat up a bit of pipe and use that to just deform the plastic around the inlet and outlet. That didn't work so well, so instead I cut out the two corners in a pattern so that I could flip them over and use them on the opposite corner to make little cut-outs.


After that, it was just a matter of making up some brackets from angle-iron & stand-offs to allow the whole thing to be bolted securely to the radiator mounting brackets, and cutting out insulation foam to stick between the shroud and the fan to prevent them from rubbing and to seal up the fan area.

To control it, I found some 2-stage BMW thermo-switches that would trigger at 90 & 98 C, with a similar switch that triggers at 80 & 88 C in You are not allowed to view links. Register or Login. Since the threading was very similar to the single switch hole in the S4 thermostat neck, the connectors and pins were available new, and the ranges sounded about right, I got a few of them to play with. The threading was the same pitch as the hole in the thermostat neck, but 2mm narrower in diameter. There weren't any threaded inserts that would bridge this gap, but since the pitch was the same, I could make one as long as I got the threads cut on the outside and inside started at the correct point.


This took a few tries with hand-tools on the bench, but eventually I got something that worked.


More to come on the intercooler-side of the project soon...

[toplessFC3Sman]

Now for the intercooler... I wanted to go with something that would allow a much more compact intake system, allow me to get rid of the big air tubes running up to the front of the car, and reinstall the front bumper support. This basically ruled out most air-to-air intercoolers besides a compact V-mount setup, but that would require another fan or two & ducting the hood and a different radiator (or at least different radiator outlet plumbing and angles), both which I also didn't want to do. So, water-to-air intercooler it was.

You are not allowed to view links. Register or Login offers some kits, and dimensions for all the ICs and heat exchangers that they have, which made the selection much nicer. Essentially, I wanted something to sit in the same spot as the OEM TMIC and allow the hood to close with minimal modification, which really cut the options down to just the You are not allowed to view links. Register or Login (3.5" thick core), which has offset air inlet & outlets. This IC was only spec'ed for about 350 hp on the website, which should be OK for my plans with this car (about 350-400 hp max, I think I'm around 300 now)

 

By using reducing couplers on the inlet and outlet, 2" tubing straight off the BOV tube, and 2.25" tubing and 135 degree coupler cut down a bit, it was possible to piece together a fairly smooth and much lower volume air path to get to the original throttle body elbow. All I need on this side are a few pieces of bent aluminum bar-stock to hold the IC in place and mount to the engine & UIM.

Moving on to the heat exchanger, this would need to fit in the front of the car, before the radiator and oil cooler so that it will get the coolest air. The AC condenser has been removed from the car many years ago, so I don't need to worry about competing with that either. FrozenBoost had options that were more square, similar to a typical radiator, or much more rectangular like our oil coolers. I wanted to do minimal cutting of the radiator support (none if possible), and absolutely no structural cutting to get it to fit, and I wanted to use this heat exchanger basically as a large capacity tank to prevent water temps from changing too quickly, and to provide a constant head of water to the pump, which should ideally have its inlet lower than the heat exchanger outlet. This way I could avoid having to get another tank, as I don't intend on using it to dump ice into. Basically, these set of criteria meant that I was looking at something similar to the oil cooler, sitting above it (and above the round tube that braces the front chassis legs in front of the radiator). The two options were the You are not allowed to view links. Register or Login with a 3.5" thick core and the You are not allowed to view links. Register or Login with a 2" core. Since I anticipate being marginal on the intercooler capacity, and don't mind the additional water capacity in the heat exchanger to avoid a tank, the thicker Type 101 looked ideal. 26" wide would be very tight between the sides of the radiator support, especially with fittings, but essentially it looked like the only option.

I ordered the 600 hp kit, with the specific IC and heat exchanger (the 350 hp kit didn't offer the bigger heat exchanger), and omitted the fan for the heat exchanger since I should be able to trigger & use the radiator fan with proper ducting. I also got a variety of tubing bends and couplers to piece together the rest of the intake system. Everything arrived in late November, and I've been slowly working on fitting things since then. Between Christmas & New Years, I got a bit of time where I could use the angle grinder & not wake anyone up... no going back now!


Basically, most of the structure for the battery vent on the drivers side needed to be cut out to make room for the fittings, and the bungs for the fittings themselves got cut down and re-tapped so that the fittings won't protrude as far.


The final cut looks like this, from the front first, then from the engine bay:



And with the heat exchanger test-fit:


This arrangement will have a 90 deg elbow in the bottom bung pointing forward as the outlet, straight into the pump mounted next to the driver's side headlight, and a 90 deg elbow in the top bung pointing backwards into the engine bay to connect up directly to the intercooler outlet. There is also room to run the pump outlet hose back through the same hole to go back towards the engine. I'll need to make some block-off panels to close up the rest of the hole and help keep the cold air up in the front of the car, as well as the brackets for holding things in place, but conceptually most of the work for the components and routing is done.

The electrical and switching side of things is still a bit unknown. I've got bits and pieces done, and tentative wiring diagrams drawn up, but I want to do some testing and playing around with an arduino first before writing things up.

[ITSWILL]

On the air to water setup I did, I cut the radiator cap off the heat exchanger, welded that shut and put ithe radiator cap on the top of the intercooler so it could be filled at the highest point.

[toplessFC3Sman]

The long, thin heat exchanger didn't come with a radiator cap so I got one of the external ones, which I'm going to mount on the firewall near the brake booster so its at the highest point, for filling and burping air. Thats a good point to make - filling and air venting needs to be at the highest point, and the pump inlet should be at the lowest point, with fluid that is cooling moving from high inlet to low outlet and fluid that is heating ideally moving from low inlet to high inlet.

[ITSWILL]

I would still be worried that without having the cap right on top of the intercooler water end tanks that it could get air trapped above the water inlet and outlet.

I guess you could put a vent port on top as well and close it once the air is purged.

[toplessFC3Sman]

Yea, when filling it I intend to disconnect the IC and tilt it to allow the air to escape, and have enough slack in the hoses so that the filler neck can be raised over it. I'm also going to be drilling into the end cap on the water outlet (hot) side to install a temperature sensor, so I could always unscrew that and use that hole as an air bleed too.

[toplessFC3Sman]

Got the heat exchanger mounting basically done yesterday. The heat exchanger had 4 tabs welded onto it, 2 on top & 2 on bottom, so essentially I just needed to pick these up and connect them to different spots on the chassis. Using some aluminum L-channel material, I made a span to go between the top two brackets, and pick up one of the existing threaded holes on the upper radiator support. The other hole I had to drill in a symmetric position. You can tell that the heat exchanger is offset to the pass side slightly to make room for the fittings on the driver's side. Similarly, some L-channel is run between two existing threaded holes in the radiator support that were unused - I don't know what they originally held, but they appeared to have never been used; the threads were still painted.


I needed to use another small section of L-channel bolted to the main one to pick up the threaded nut on the drivers side, since it was lower than the one on the pass side. The top bracket and the downward bent piece of the upper radiator support also needed to be bent out slightly to match the angle of the radiator and heat exchanger, since they're tilted with respect to the top surface of the upper radiator support.

View from the top & front:


To get the hood latch to fit again, I needed to both cut away some of the bracket, and do a little percussive manipulation of the top row of the intercooler (read: bent it down slightly with a hammer).



This should allow plenty of room for adjustment and slight flexing. I still need to remove the heat exchanger again to flush it of metal shavings from tapping it further and add some rubber bushings inside the holes in the tabs to try to provide some isolation from chassis flex. I'd also like to cut some speed-holes in that lower L-channel, because as it currently sits it's blocking about an inch of the radiator height. I'll also need to add lock-tite, anti-seize, and sealing compound to the various fasteners and water fittings, but overall I'm very happy with the final fitment. To get it in or out, it definitely requires that the radiator be removed, but thats a much easier task now with the E-fan.

Finally, the wiring diagram for all these changes that I'm planning on using. Some of this has been done already, like most of the changes in the Lexus ABS solenoid box, and mounting of most of the components, but I haven't had a chance to do much testing yet, and don't have the Arduino to play with (hopefully its coming this week).

Anyway, the BMW thermo-switch has two temperature switches inside, one at 90C and the other at 98C, so that directly goes to the blue solenoid for the e-fan low speed & the green solenoid for e-fan high. Since the switch is "dumb", meaning that both "low" and "high" will be switched on whenever "high" is on, and in testing the fan seemed to blow harder when only "high" was energized (instead of sending power to both "low" and "high"), i built that switching logic into the ABS solenoid box. When the "High" green solenoid is not energized (as shown in the diagram), it is sending +12V from the battery to the coil of the "low" blue solenoid. When the thermo-switch "low" gets triggered, the other side of the blue solenoid coil will see ground, and the fan will turn on "low". When "high" gets triggered, the green solenoid will switch +12V from the contact going to the blue solenoid to the contact going to the "High" wire on the e-fan, turning on the "high" speed while turning off the blue solenoid, even though the "low" coil is still grounded by the thermoswitch.  Additionally, the "high" green solenoid is taking its +12V from the switched ignition power, so "high" speed can only be engaged when the engine is running. "Low" can be on at any time, even with the car off. I don't know if this will cause a problem yet... I don't expect so since even though the water pump won't be circulating coolant when the engine is off, natural convection in the radiator will tend to keep coolant moving a little bit (as the coolant cools, it will become more dense and fall in the radiator, pushing hot coolant through the system and into the top of the radiator to keep a small amount of flow going). Also, the e-fan is blowing directly on the thermostat neck, which should help cool that part down too (which is where the thermo-switch is located). Either way, its something I'll need to keep an eye on once the car is running. Also, all of the switching (whether by the arduino, manual switches, or BMW thermo-switch) is done on the low-current signal side, not on the high-current power side, so all the components should be much more reliable.