|
Actually, you seem like the exact person I should ask about this... Say I want to turn an EJ25D from a 4 stroke engine to a 2 stroke air compressor. Instead of intake, compression, power, exhaust, it'll be intake, compression (pushes air out through 1 way valve replacing spark plug), intake through exhaust valve, compression (pushes more air out through 1 way valve). So all I should have to do is advance the timing on both exhaust cams by 90 degrees, ish. What do you think my chances of making this work with stock cams are? I think it should be way cheap to build a crazy powerful air compressor that way, I'm just not sure how to figure out that angle to put the exhaust cams in at without cutting a head up and observing as I carefully crank it over checking for p-v and v-v interference. The crank sprocket is 24 tooth and cams are 48 so my wild rear end guess would be 12 tooth advance. Thoughts? Since you like, actually design engines for a living and I'm just a tinkerer.
|
#
?
Mar 15, 2018 01:53
|
|
|
|
| # ? May 11, 2024 07:18 |
|
|
Oh that's an interesting one. I haven't measured an EJ25D's cams, I'll assume they're similar to the 205 though. So, in the car application, the cam timing is going to take advantage of the fact that there's air inertia ramming through the ports. You're not going to get that as much so you want the sucker closed at bottom dead center, about. And there's no problem with hanging everything open before top dead center if the valves don't hit the piston. Maybe be safest to just leave it. You're going to want the exhaust cam to open at the same time the intake was on the other stroke. *edit* Wait, duh, don't advance the cam, you'll lose compression. I'm not sure what this works out to offhand in cam teeth... easiest way is to use a dial indicator (or eyeball, whatever) in line with the valve, rotate to when the intake lobe is pointing away from the valve (just because it's an easy benchmark, then rotate the crank 360 degrees and slip the exhaust cam until the lobe is pointing straight away from the valve too. I'll sit down with Excel and figure out a guess for what that works out to in teeth tomorrow. Definitely should at least kind of work; regrinding the cams would work a little better but that costs money. What strikes me as fun would be to leave one bank as a gas engine, slip the cams / do one way valve on the other and have a self-powered air compressor. mekilljoydammit fucked around with this message at 02:46 on Mar 15, 2018 |
|
#
?
Mar 15, 2018 02:42
|
|
|
Sounds like you two need to get in touch with this guy: https://thekneeslider.com/ducati-v-one-twin-to-supercharged-single-conversion/ poo poo is bananas
|
|
#
?
Mar 16, 2018 03:39
|
|
|
mekilljoydammit posted:What strikes me as fun would be to leave one bank as a gas engine, slip the cams / do one way valve on the other and have a self-powered air compressor. I saw this on an episode of Junkyard Wars. I think it might have been water pumps on an inline-4 though. Two cylinders running fuel, two pumping.
|
|
#
?
Mar 17, 2018 03:28
|
|
|
mekilljoydammit posted:
We had one yeeeeeeeeeeeeeeears ago with a VW motor. One bank firing, the other bank compressing. Vibrated like mad but it was pretty goddamn impressive how fast it filled the tank it was attached to.
|
|
#
?
Mar 18, 2018 00:28
|
|
|
I didn't imagine it was an entirety new idea. Today's fun: transmission work on the Subaru! Putting a DCCD in a 5 speed that didn't cone with it in TYOOL 2018 is not o e of my wiser ideas.
|
|
#
?
Mar 18, 2018 01:02
|
|
|
Yeah, that's not a new idea at all for sure. The only new part was me theorycrafting about doing a true 2 stroke compressor setup without using a custom head (aside from hogging out the plug hole from M14x1.25 to 3/8-NPT that is) or custom cam. Can't you swap DCCD into the 90s phase 1 5mts easily? I can't remember, I looked into it briefly a while ago.
|
|
#
?
Mar 20, 2018 02:30
|
|
|
DCCD center goes in a phase 1 extension housing, so, yeah it's much easier then. It uses its own transfer gear and not the normal phase 1 transfer gear though, so still relies on finding some of the right bits.
|
|
#
?
Mar 20, 2018 03:26
|
|
|
OK, first half day back at work after having my daughter stillborn last week - 10 weeks before she was due. Decided to measure a pile of Subaru cams! I'm copypasting from a post on NASIOC I did because gently caress off. OK a few more things than last time! Note again that some squiggles are unavoidable as I'm measuring real things with imperfect measuring equipment.   V7 EJ207 - notice how the flank accelerations are lower than the EJ205? More duration and less lift, so it will be even more gentle on the valvetrain than the 205 but not the best performer. Subaru probably figured the larger heads compensate.  EJ257 Intake. Notice it holds the opening acceleration a little longer than the EJ205 cam so gets to a bit more velocity. About 0.020" more lift than the 205 and about a degree more duration at 1mm lift. (I'll harmonize measurements at some point, bite me) Now onto the fun stuff.   Non-AVCS Kelford 199-E. Look at the velocity plot - the flat section at the opening and closing is much shorter. Those sections are constant velocity sections to open and seat the valves relatively gently, and the longer they are the more tolerant the cam will be to loose lash, but the longer they have the valve cracked at lifts where it can bleed pressure, create vacuum issues, and not have enough lift to flow anything. Tradeoffs. Now for some big muthas. But first an oddball.  Kelford R199-C Exhaust... wasn't sent the full set of these. Interestingly the flank accelerations were higher but for shorter duration than the 199-Es.   GSC S3s! The older cast core version - you might ask if the billet version changed the profile, well, you'll have to send me a set to measure.    3MI Racing custom set. What's kind of interesting is on the lift graph the intake overlays very closely with the GSC S3 intake, but looking at the derivatives, it's somewhat asymmetric - the acceleration peak on valve opening is lower but held for longer. This may be a bit kinder to the valvetrain.   Some metrics for comparison. "Area" is the integrated valve lift over the 1mm duration - basically a combination of both how far and how long the valve gets opened. If you're curious it's in inch-degrees. "1mm Duration" is pretty self explanatory. "Area/Dur * 100" is a completely arbitrary figure to look at on average how far the valve is open. "Max Lift" in this case is in MM. Now, here's the interesting (to me) things. Cams are built from a series of equations defined by fixing various derivatives (and radius of curvature on the nose) to various values for various reasons. If you look at the cams measured here, there's not much difference in those derivatives. There's a reason; there's physical limits on these things. So, lift is easy enough - too much lift and you get to coil bind. Velocity is the next limiting factor, in inches/degree. The higher velocity you reach, the farther away from the bucket centerline the cam is touching, so the bigger bucket you need - since Subaru buckets are a fixed size, that's a limit no matter how aggressive you want to be. The next derivative is acceleration - you need to turn this to actual inches per second (or whatever you prefer) to determine forces acting on the bucket to open it (too much force and you break through the oil film and wipe cam lobes) and how much spring force is required to keep the bucket in contact with the cam. Finally, there's jerk, the rate of change of acceleration... which is hard to define theoretically. A separate calculation is radius of curvature, mostly at peak lift - a smaller radius means less contact area and more stress on the cam for a given force. I'm not going to handhold you through everything I'm seeing because frankly I'm not an actual expert anyway - suffice to say that the OEM profiles are pretty conservative, and the aftermarket ones less so.
|
|
#
?
Apr 12, 2018 02:06
|
|
|
drat, what a way to start a post. I'm so sorry man. 
|
|
#
?
Apr 12, 2018 03:08
|
|
|
... I can't even read the rest of the post. So sorry man. If you ever need anything, you know how to find me.
|
|
#
?
Apr 12, 2018 03:33
|
|
|
Thanks guys. It's... it really really sucks, but it's also a giant kick in the rear end to fix a lot of things in my life before the next try. So that's something I guess.
|
|
#
?
Apr 12, 2018 11:25
|
|
|
loving hell, I'm truly sorry to hear of your loss. Hit me up on Slack if you need anything
|
|
#
?
Apr 12, 2018 12:21
|
|
|
Holy poo poo dude, that loving sucks. I hope you and your wife can both recover well from that, and grow stronger together because that's some crazy poo poo to deal with.
|
|
#
?
Apr 12, 2018 12:26
|
|
|
Christ, I'm sorry dude. 
|
|
#
?
Apr 12, 2018 14:32
|
|
|
There's not really anything that can be done at this point, you know? Got a lot of family and friends in the area, and even they don't have anything they can really do. We're bound and determined to work through all of this though... first counselling appointment was less than a week after all of this, and we're bound and determined to do everything we can to recover and improve ourselves before the next try. Part of that is just... saying to hell with it and getting things done. Cams included.
|
|
#
?
Apr 12, 2018 17:05
|
|
|
Changing topics: Have you read up about the Monsterbox 4 rotor drama? PPRE apparently mis-machined the eshaft and threw the lobes several degrees off. Tossed the rotors into the housings and plates. Let me see if I can find the thread.
|
|
#
?
Apr 12, 2018 19:48
|
|
|
I don't even know what a Monsterbox is! I swear, it's like the FD people are a completely different community from the rest of the RX-7 people.
|
|
#
?
Apr 12, 2018 22:53
|
|
|
Take a read of the last few photos. It's an excellent example of why I stay out of the "main stream". Too many bad shops, too much drama and BS. https://www.rx7club.com/build-threads-292/monsterbox-4-rotor-1095628/page59/
|
|
#
?
Apr 13, 2018 21:52
|
|
|
Yikes. I'm just going to stay over here with my weird non-turbo roadrace poo poo. Is it weird that I'm not even really sure what this car would be for? Also jesus I hate everyone's insistence on fancy billet electric water pumps, but I find it hilarious I could probably do a more accurate eccentric shaft on my home machinery.
|
|
#
?
Apr 14, 2018 00:56
|
|
|
the spyder posted:Take a read of the last few photos. It's an excellent example of why I stay out of the "main stream". Too many bad shops, too much drama and BS. Is there any BS there, though? This is just straight up bad, isn't it?: 
|
|
#
?
Apr 14, 2018 03:45
|
|
|
The drama/BS is organizers for an event taking the shop that poorly made that eccentric shaft to take a dig at the guy building that car. I just find it... well, I guess I shouldn't say astonishing that the engine was so badly slapped together (the equivalent is not checking piston to wall clearance before putting parts together) but ... disappointing. I'll just design and have made my own eccentric shafts for 3-rotor use I guess.
|
|
#
?
Apr 14, 2018 14:26
|
|
|
Minor bump - decided I'm going to open up my old peripheral port 13B and re-port it. I never got stuff to seal as well as I'd like so I'm going to do it... and as long as I'm at it, go to a larger bonded in tube so I can go to bigger ports - maybe 52 instead of 47mm. (aaaaaaand I realized I now actually have the resources to build the slide valve throttle body I designed 15 years ago) I mean just little stuff to keep busy and distracted.
|
|
#
?
Apr 20, 2018 18:25
|
|
|
What's your plan for sealing the P-Port inserts? Years ago I started on a design with my machinist buddy, but it's going to take $$$$ in tooling and fixtures. I've never liked the devcon/RTV option- as they always seem to leak after x number of cycles. I originally drew up a fine-threaded insert with orings and using a sealed blocking plate in the coolant passage on each side. Turns out this is pretty common. Chip Motorsports has made a nice o-ring'd side insert that requires additional machining of the housings. Several others have made nice threaded inserts and Rob's working on a set right now with another machinist we know. I kinda want to just make a 13b P-Port using our lathe and devcon'd inserts just to see how long it takes before it cracks. Maybe in a Miata.... hummmmm. Side note: Stock up on ANYTHING FD related you might need for engine builds now. Mazda is dropping a ton of stock and we're placing a order to hoard known consumables.
|
|
#
?
Apr 20, 2018 19:24
|
|
|
General plan is machine for an interference fit, coat with RTV before sliding in, fill with devcon and put it under pressure to force it through any cracks. If that doesn't work, something like the Chips setup. What kind of FD specific stuff is going away? I'm assuming seals and stuff are sticking around, corner seal springs likewise... housings going away? Sucks to be on this part of the product life cycle.
|
|
#
?
Apr 20, 2018 19:49
|
|
|
Decided to horribly abuse Imgur by uploading a bunch of albums of pictures I took at races over the last couple decades. I'll be more picky once I get to later years where I have a long lens and was taking a lot of shots on track simply because holy poo poo that'll take forever to upload. So first, albums that had relatively few pictures. Note I'm doing very little culling here and periodically I'm taking pictures of spergy poo poo like suspension, chassis, engine or whatnot details. Also in earlier years I was using shittier cameras and didn't have a clue what I was doing. Stuff's completely out of order and I'm not inclined to explain stuff ahead of time because SO MUCH. That said, have some links. https://imgur.com/a/pw4DwRn https://imgur.com/a/zOfF3tl https://imgur.com/a/qTLUd42 (big one) https://imgur.com/a/cuHX1pJ https://imgur.com/a/X01uIcl https://imgur.com/a/YAcVR56 https://imgur.com/a/n1ZLMBu https://imgur.com/a/exr6epf https://imgur.com/a/nSARR2K https://imgur.com/a/lIWCG2u https://imgur.com/a/ZmY2v7k https://imgur.com/a/H6KGcVE Oh and Sevenstock 9 https://imgur.com/a/VFSGPZK
|
|
#
?
Apr 29, 2018 02:31
|
|
|
Thinking of doing a nice accomplishable project this summer for a street terror. One of my FBs I had titled not too long ago... quarter panel rust and so on though. I was thinking doing fender flares, converting to 4x100 in the rear, making new suspension based around MR2 Spyder knuckles, Miata brakes on all 4 corners, and maybe a front subframe to take a circle track rack and pinion. Nothing major. I've been wanting to refresh the peripheral port engine anyway, and nice healthy braps for a weekend toy sounds like a hoot. Sunday was one month since my daughter was stillborn. Even without that, work stress would leave me in a state where I'm not doing good. I'm trying to get to setting goals I can tinker on so I don't just sit and stagnate into (more) depression.
|
|
#
?
May 8, 2018 17:08
|
|
|
You should! Better yet, go racing for a weekend. I'm sure between you and your dad there's a car ready now. I need motivation to work on our 99 Miata. It needs a solid weekend of work to fix the crash damage and a front bumper. A hard top would be awesome, but $$$$.
|
|
#
?
May 8, 2018 19:55
|
|
|
Oh yeah, already planning on taking the NC MX-5 out memorial day weekend - double race weekend so we were going to trade off on the crew/drive. Reminds me I need to bug someone for the new copy of my competition license. Really though I'm thinking it would be nice to have a nice uncivilized monster to go hoon around backroads on weekends. The RX-8 would work too except emissions (I'm in a "scan the OBD2" county) are a bit more tricky, and everything is more expensive because I can buy parts rather than fab poo poo. That sounds weird, but I can fix all the 1st gen problems with fab work, while the RX-8 problems are more complicated. Or probably I should get the WRX off the lift, lol. *quickedit*   Decided to link a couple of my favorite shots from that big pile of albums. 2011 SCCA Runoffs at Elkhart, and a bit wet... white #95 is a friend of myself and my dad. Imgur, so remove the last l in the filename for big. Or go poke through the album. 
mekilljoydammit fucked around with this message at 20:59 on May 8, 2018 |
|
#
?
May 8, 2018 20:53
|
|
|
Well, in my amazing ability to go down weird rabbitholes, I think I figured out a workflow to section a cylinder head into a 3d model. No pictures yet but I'll take some when I'm doing because it'll be pretty amazing if it works the way I think it should. (amazing in that I'll be turning a WRX cylinder head into chips) Really though it's hard to get much of anything useful done - going to counselling and stuff just... I don't figure this is going to be a short process but good lord is it frustrating. And sad. And so on. Anyway, went to my folks to work on stuff last night, my mom commented she found something at a thrift store. 
|
|
#
?
May 18, 2018 17:23
|
|
|
Dope. Is that the starter up on top of the transmission? I can't identify which engine that is off the top of my head.
|
|
#
?
May 18, 2018 20:54
|
|
|
It looks like an older 12A to me - the RX-2 engine my dad has had kicking around in our barn as long as I've been alive looks like pretty much dead on.
|
|
#
?
May 18, 2018 21:00
|
|
|
Yeah, the model is of an early twin-dizzy 10 or 12A.
|
|
#
?
May 18, 2018 21:15
|
|
|
That rules.
|
|
#
?
May 19, 2018 00:46
|
|
|
I try to keep out of drama because my own life sucks so much, but seriously, man, don't know what the gently caress to say but I'm very sorry to read of your lost child.
|
|
#
?
May 20, 2018 13:55
|
|
|
BTW that rotary model is of the OG 10a in the mazda cosmo 110s. It's based on the NSU design with the starter at the top, and with the twin dizzies just chucked straight up on top because..whatever Later twin dizzy 10a's, for like the r100 had them on an angle towards the sparkplug side with the generator/alternator in the middle (similar to common more modern rotaries, just two of them)  By the time the 10a rx3 came out the starter was at the bottom, same dizzy anle and starter location for the rx2 twin dizzy( with tilt to the side for dizzy and alternator - I think they were alternators by then and not generators) Fo3 fucked around with this message at 17:00 on May 20, 2018 |
|
#
?
May 20, 2018 16:47
|
|
|
That makes sense, thanks. Though I do note that at least some of the twin dizzy 12As were still top starter.
|
|
#
?
May 20, 2018 16:52
|
|
|
Yeah, just edited my post because I found a top starter r100. My first car was originally a 12a twin dizzy rx3 but had the motor replaced before I got it, same for a mate's rx2. I don't think we ever paid that much close attention to them besides seeing the others with originals while in car clubs back in the 90s.
|
|
#
?
May 20, 2018 17:04
|
|
|
  Guys, there's something wrong with my EJ205 head! OK, no seriously. I took one of the spare heads I have around with cracks by the spark plug, bolted it to a steel plate and filled it with black tinted polyester resin. What comes next is turning it into metal chips 1mm at a time, and use a flatbed scanner to take an image each layer. See, it's my reasoning that nice bright aluminum is going to show up nice and contrasty compared to black polyester resin. Then it's just a matter of turning all of those images into a 3d model. There's a few automated tools I can use, but it might be advantageous for me to do it in CAD so that I can more easily edit it. Which is the point - if I can have a dimensionally accurate 3d model of the head, I can tweak locations of things like ports, then 3d print and investment cast it. Totally not ambitious at all, right?
|
|
#
?
Jun 2, 2018 12:59
|
|
|
|
| # ? May 11, 2024 07:18 |
|
|
Neat, but what's your end game here? Do you have any sort of simulation software to show that changes you make will actually be beneficial? Moving ports sound like a huge pain because then you're designing me manifolds too. Otherwise without changes, you'd probably be better off buying used heads than backyard casting them, yeah? I don't mean to constantly sound like I'm busting your balls, but you appear to like taking the complicated route vs cheap and easy (my specialty  ).
|
|
#
?
Jun 2, 2018 13:12
|
|