That is a pretty slick design. Variable valve lift designs are fun to play with, it would take a lot of engineering to make a new head that holds a stock camshaft and keeps the lubrication working adequately, an issue we didn't have to deal with in the rotary valve due to us adapting from a non lubricated air cooled engine.
My next project after the rotary valve is finished is going to be a ratchet drive CVT of my own design. I have always had a fascination with continuously variable transmissions and find an odd satisfaction in driving them. It would be ever more fun to have manual control over such a transmission.
Anyway... The current status of the rotary valve engine is fixing up a really poorly done cam position wire and replacing a coolant fitting that fatigued off in our head causing a serious leak.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
Yea I just have not had enough spare cash laying about to pursue it. Although I think I should at some point, it's the simplest and most robust variable timing and lift system I have seen. Each one of the adjustable rockers also has a secondary return spring that is not shown. With this system it is possible to increase or decrease valve lift during the valve lift/closing event which I would imagine would make for one hell of a throttle response.
In regards to the transmission, I have been looking at a shiftmaster paddle shifter conversion. You keep the clutch but get paddles instead of the shift knob which I think would make a hell of a difference in these cars.
An update to this build. We have found that we are very quickly fowling our spark plugs up and that quickly causes the engine to stop running well. It also much would rather run with no intake at all, it will idle fine controlling just the fuel but not having any air metering via a throttle body.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
do you think adding velocity stacks or a long-runner, small volume plenum style intake without a TB will help scavenging and subsequently help with the backfires?
always trying to make it lighter and faster
^^former build: http://www.bimmerforums.com/forum/sh...-neglected-M3/
current build: http://forums.pelicanparts.com/porsc...car-build.html
instant grams: doktor_b
We are considering machining ourselves a set of stub stacks for the intakes to see what will happen as that is a very easy thing to do. Yes the thought it to help a bit with scavenging and prevent the backfires. The timing is currently a bit advanced but we don't have great control over it, only 10 degree valve angle increments which is 20 degrees at the crank. This just happens to be where it runs best though it is likely the best valve timing is somewhere between where it is now and the more conventional timing locations that work for engines.
We are considering purchasing an AFR gauge to see if we are close to correct on that front. It seems like we are drowning the spark plugs with fuel.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
I remain in stunned awe / suspense. Excellent read and the best project I've seen in years. Thanks for sharing!
"Captain" 1995 Cosmos M3 5sp, s52 Swap, Conforti chip, CAI, Borla exhaust
"Hellrot" 1997 328i, 5sp ZF Swap, Vogtland springs
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
Update time!
After some recent runs of the engine we realized that we chose a very poor location to put the pick up for the OSV/CCV. The position of the pickup was in a place where oil was pooling and the crank case pressure would push oil into the oil separator. The result would be an excessive amount of oil would get pumped into the intake which would then be pulled into the engine and burned.
We are currently considering where we can re locate the pickup point and potentially removing the OSV/CCV from the system and just ventilating directly to the intake with a vacuum line.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
It's just for emissions anyway, the OBD1 cars just vent into the intake, the larger hose from the valve cover goes to the intake elbow before the TB and the small hose goes to the manifold after the TB. Not sure how that adapts to your system. Crazy awesome build by the way.
'94 325i Sedan, Arctic Gray: UUC LTW FW, EVO 3 and DSSR, +.020 Maxsil pistons, ASC delete, Eibach shocks/springs, 16" contour reps 238k
'93 325is Coupe, Schwarz, work beater 299k
'89 325i Vert, Alpine White: 5spd swapped. Sold
'04 Toyota Sienna XLE Limited AWD, In progress swapping to M50/G250, http://www.wibimmers.com/board/index...nna-25i-build/
'05 Volvo V70 R, 6mt, mostly stock, kid hauler 200k Sold
'85 Toyota LandCruiser: Lifted, gas hog. 205k
Update time! It is been a while since I updated this thread.
Progress has been good, we have pretty much solved the smoke issues with pulling oil in through the CCV. We ended up just attaching a tube from the timing chain area to the intake where the cyclone separator originally connected. An updated oil splash guard is still needed in the timing chain area but it is working much better.
The biggest development that made the biggest improvement to performance is actually spark related. It turns out the Honda motorcycle spark plugs we used for this were several heat ranges colder than the stock BMW plugs. Since the engine is roughly the same compression ratio it meant that we were fowling the plugs by not burning the carbon off of them. Going up one heat range and switching to an iridium plug solved that problem. Now the engine is able to run for more than an hour without the plugs fowling.
We made an attempt to go get the car state inspected. We made it one mile down the road before destroying our timing belt. The belt lead a slow death by means of too much oil getting on it and destroying the rubber. Thankfully the rotary valve design is non interference by nature so the solution is to clean up all of the excess oil that was laying around the engine bay and replace the timing belt with yet another $100 Gates double sided belt...
Since we put the new belt on we were disappointment that the engine really refuses to idle. We had a suspicion that the air fuel ratio was off due to the engine idling better with the MAF unplugged. We decided to stuff an O2 sensor up the tailpipe and with our Innovate Motorsports AFR gauge measure the AFR. It turns out the engine was running so lean that the gauge wouldn't even register off the high end most of the time. If the MAF was unplugged and the engine in warm up mode, at full throttle the AFRs would come close to inline and then the car would rocket off with expected power. This lead us to check for intake leaks. So holding the engine at a constant RPM we started spraying the intake down with ether (starting fluid). We determined that we had several bad intake leaks. This is quite a problem in an engine that has poor vacuum from the get-go with the excessive valve overlap we are running (equivalent to 36 degrees of overlap). On a good day, at idle, the rotary valve engine makes about 5 in/Hg of vacuum. A well running engine should be making 15in/Hg. So an intake leak basically turned the vacuum down to zero.
We are currently in the process of filling in holes in our custom carbon fiber/OEM air plenum Frankenstein contraption. The glue is dry and the next step is to check it for more leaks before installing it in the car. If all goes well this will result in an engine with smooth idle, snappy throttle response, and the confidence to drive it three miles down the road to the mechanic to get it state inspected.
Last edited by pizzaman09; 05-09-2017 at 09:16 PM.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
awesome
always trying to make it lighter and faster
^^former build: http://www.bimmerforums.com/forum/sh...-neglected-M3/
current build: http://forums.pelicanparts.com/porsc...car-build.html
instant grams: doktor_b
This is so cool, hoe this continues and I get to see it in action someday.
It is getting close to being in action. Drove it up to the corner and back today to see how it is performing. My guess is that it is putting out about 50 hp at best right now.
Just finished fixing all of the intake manifold vacuum leaks. It certainly runs better however we are still having serious issues with it running very lean. It will run correct AFR at full throttle at 4000+ rpm but that is the only place.
We did determine that we are leaking air into the seals on top of our rotary valves.
There have been times over the last few months where it put out lots of power and was even able to do a burnout, not quite sure why we can't get it back to that state of tune.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
Agreement here on everyone's amazement! Nice work!.....even though I have only a dim understanding of it. One dumb question (of many): what are the Weber-carb-looking things?
Awesome build I love it. I'm curious as to what kind of an effect would blowing air into the engine have? Supercharger low boost? My thoughts were that by blowing air into the intake valve it would help counter the oil that's seeping in. And would help bring the compression up. Just a thought.
Sent from my VS810PP using Tapatalk
Did you figure out how to get around the VaNos control and errors in the ECU? I think the ECU goes into limp mode with multiple VaNos errors.
Id imagine you would need much thicker oil to really combat pressure on the rotary valves.
Did you figure out how to get around the VaNos control and errors in the ECU? I think the ECU goes into limp mode with multiple VaNos errors.
I believe your ECU is in limp mode, timing and fuel is limited, causing rough running condition
Sent from my iPhone using Tapatalk
Thanks! I would assume the Weber-carb-looking things you are referring too are just the intake trumpets. They are the entrance for the air to go down into each valve from the plenum. I am open to answering all dumb questions, let me know!
The advantage to blowing air into the engine would be a supercharger bost type effect just like a normal poppet valve engine. The other unique advantage in this design is it would solve our intake leak problem by blowing boost out of the leak locations, I know this is a terrible idea but it would help.
Pretty much. By plugging the Vanos solenoid in the only error left is an occasional cam shaft position error due to the cam shaft not getting phased by the vanos. Frequently there are no OBD2 codes surprisingly. We are looking into getting a WAR flash tuner and a tune to get the Vanos turned off and have them do a baseline tune for us.
Not particularly. The contact surface area on the valve is much larger than the main crank journals so the pressure on the valves is much lower, hence by deduction the oil doesn't need to be a higher viscosity for this application. The bearing surface is also high lead bronze so it can go with oil temporarily with little issue.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
Big news! The rotary valve e36 has made it to the state inspection facility for inspection!
After a few months of fiddling we really ironed out a few bugs. This includes stopping some reoccurring vacuum leaks once and for all. We also solved the overheating problem by simply modifying the original fan to fit the car and reinstalling it. Each fan blade is now 1/2" shorter. We added a catch can to the crank case ventilation to reduce oil ending up in the intake. The fuel injectors received their original clips to hold them to the fuel rail which solved fuel leak issues.
Right now it would seam that the engine makes great power between 3k and 4.5k rpm, everywhere else it is currently a dog.
I will update later this week with the state inspection results. It only has to pass PA safety inspection as it hasn't been driven enough since last inspection to need emissions inspected.
Current fleet:
1999 BMW e36 M3
1999 BMW e36 328is with rotary valve engine head
1999 Oldsmobile Eighty Eight
1990 Jeep Comanche Eliminator
1962 Austin Healey Sprite
"So we've come to the conclusion that BMW just has parts laying around they decide to throw on cars for no reason."
Interest on a Very Budget Supercharger Build??
Bookmarks