Popular Post Roman Posted January 23, 2020 Author Popular Post Share Posted January 23, 2020 Ok been feeling a bit better, feels like making some meaningful progress again. I've been working on the fuse box situation. Working towards the milestone of getting power to ECU and sensors etc - Think that'll help push along towards first engine start. cool. Been looking at a bunch of different places to mount the fusebox, but I reckon in the center console with a cover over it is going to be best option that's easiest to access. So printed a little holder thingy (will have a cover hiding the bolts and edges etc) Then I've been working out on an Excel spreadsheet where all of the ins and outs of the fusebox will need to go. I've decided I'll run an SSR for the fuel pump and engine fan, so these will need to be seperate. As you cant get SSRs in the form factor of those micro relays. But it looks like I'll need at least 3 12 pin plugs for in/out, so will go for four to leave some space for future stuff / important things I've forgotten about. So for sake of tidiness and making it easy to wire up on the bench / unplug in the car. I am printing a 4 way backshell type thing for Deutsch DT 12 way plugs, that will bolt to back of the fuse box. As per usual with 3d printing it took a few iterations to get the clearances nice. But consistent once you get it right so worth the effort. Fingers crossed works out okay and wiring it up is straight forward. Documentation definitely helps, even just looking at Excel sheet you can see how to group things better so the wiring will be tidier. 17 3 Quote Link to comment Share on other sites More sharing options...
Popular Post Roman Posted February 16, 2020 Author Popular Post Share Posted February 16, 2020 Fusebox all working. But not before I realized I'd wired 9 out of 10 relays backwards. Woops. Lucky was easy to fix. Also, I've always hated the idea of an external FPR, its just so big and shit looking compared to the OEM ones direct on the rail. Since I'm making an effort to keep things tidier though, and this will sneak in behind the cam cover and look discrete. Pinched from the Redtop engine's factory setup. Think I'll give it a hoon. Has the secondary benefit of that I can just run a smaller EFI spec rubber return line hose back to the firewall tucked under the inlet. Rather than a full sized AN-6 line from the rail to the FPR. These FPRs all only regulate to 300kpa (43psi) down from the 60psi I've been using. I havent managed to find a 60psi factory FPR, funnily enough because the only cars that run 60psi or higher are returnless systems. So let me tell you a punishing story about why that is. Traditionally, EFI systems have run at around 43psi. If you're wondering why the weird number, 43.5 psi is 300kpa, and no one in their right mind uses imperial units to design things. Anyway, this pressure range is a good trade off between how hard the pump has to work/reliability, how much fuel it can deliver, and good atomization. As you raise the fuel pressure the l/min rating of the pump decreases. So keeping the base pressure lower effectively gives you a higher flowing pump for same hardware = cheap. Look how much flow rate a Bosch 044 loses when pressure goes up. It also means that as you add boost, and the fuel pressure needs to go up to suit. Spec sheets for electric fuel pumps show that most start to creak and lose heaps of flow after 65psi or so, and with reduced lifespan. With 43psi base pressure and 20 pounds boost only means 63psi required at the pump. Which is considerably better than trying to achieve 80psi. So, why do modern cars run 60psi base fuel pressure instead? When emissions rules are really tight, Fuel vapour emissions are a big problem. When you heat up fuel at the fuel rail and send it back to the tank, it creates heaps of vapour. As well as that when you open the gas cap, this pressurized hot fuel vapour rushes out which is also an issue they need to minimize. A secondary problem here is that fuel isnt just made of 1 thing, it's made of lots of different chemicals which have different boiling points. My understanding is that the additives that turn 95 octane into 98 octane, are the first to flash off into vapour. So minimizing the fuel heating reduces your octane level loss over time. The fix for these problems is to make sure fuel that goes to the rail, never gets back to the tank. Hence returnless. But now the problem is that at 43psi, if it sits long enough in a hot fuel rail it reaches its boiling point. So 60+ psi raises the boiling point above where this is an issue. Which is also why they dont vacuum reference the FPR in the gask tank on returnless, because you never want to drop the fuel pressure below 60psi absolute. And a modern ECU can just deal with a varying differential fuel pressure a lot better than old shitters can. So in my case, going back to 43psi will be fine but with lower pressure the injector duty cycle will need to creep up a bit. As at 60psi Altezza injectors are 450cc/min Injector, at 43psi it's only 370cc. So I'll need 21% more injector opening time. Currently I only get to around 63% duty cycle at lots of RPM so that'll end up around 76% duty cycle instead. No drama. I will just need to be mindful of when controlling the fuel pump with PWM, that I'm not setting the flow rate too low that it'll have time to boil the fuel. Although these engines have a fairly well insulated fuel rail compared to some other engines where they are stuck in a valley or whatever. The other problem is that my fuel pressure sensor now has nowhere to go, as previously went into the FPR. I managed to find an AN fitting that has a pressure port on the side, so I'll mount this with the sensor whacked in somewhere after the fuel filter on the inlet hose when it turns up. I got the Bosch E-throttle all working nicely, it's very snappy testing via ECU. It looks like I'll need to ramp down the PID settings a bit compared to Altezza one. Will take some 100hz logs and try fine tune it. Speaking of which, this PID tuning explanation is bloody amazing So now I've gotta sort an E-throttle pedal. I've got two options so far to try, but isnt looking as easy as I'd hoped. First is RX-8 pedal, on the left. Convenient option but it doesnt really fit nicely in the available space. The other issue is that the APS sensors are built into the entire pedal assembly. So if it shits itself, you have to rebuy the whole thing which seems stupid. The other option I've been looking at is a 350Z APS sensor, which looks like a normal TPS but has a little bar on it to be moved by a pedal somehow. This doesnt particularly look like it wants to cooperate with the factory pedal, so need to figure out something else there as well. However it does make it a lot easier to keep a spare and replace if it packs up. But I like the idea of keeping the factory pedal, so if I end up using a cable again to test something I've still got the option. Although I guess I could just refit the factory pedal anyway in that case haha. 14 Quote Link to comment Share on other sites More sharing options...
Popular Post Roman Posted May 10, 2020 Author Popular Post Share Posted May 10, 2020 "Hey, I bet I could find an old key that describes your entire life" "What are you talking about, thats crazy" "Oh yeah, check this out" "....weelll, shit" 21 2 Quote Link to comment Share on other sites More sharing options...
Popular Post Roman Posted June 6, 2020 Author Popular Post Share Posted June 6, 2020 Some reckless son of a bitch made a thread about sites which make it easy to purchase from Yahoo Auctions. So here we are: This is the factory AA63 dash cluster... The only difference apart from being impossible to find. Is that this one has the tacho goes to 9k instead of 8. Which doesnt sound like much, but spending half your time like this is annoying: I'm still developing a digidash, and that's likely what I'll keep in the car. Especially for trackdays. But these AA63 dashes (or any AA63 parts for that matter) are hard to find and I'm a hoarder. So there you go. It might be fun to make a little box that converts all of the factory dash items to run from canbus. so I can still run it all. Or maybe it can go in the Yellow car. Since my car isnt likely to be going any time soon, and even when it was, it was a pain to do any dash testing. I'm working on an engine simulator that can talk to the dash back and forth as the ECU would. So this has a fairly simplified air/fuel/ignition/power simulation in it, so I can test the cruise control ignition timing logic thing without needing to go driving each time. So my dashes goal, will be to find out the ideal ignition timing for this "engine" But I dont actually know what it is myself - I've modelled in things like rpm based friction, pumping losses, ideal ignition advance at WOT based on a fixed milisecond value, ignition advance increase as load decreases, and shit like that. So the ideal ignition timing is generated in realtime I guess you could say. then I'm also starting on a revised screen setup which should run a bit faster, but this time rather than using graphic functions in the screen GUI editor, I'll generate as much as I can from the arduino code instead. The benefit of this is that I can resize, rescale, move, replace things just from the code. Rather than having to reprogram the screen, and then adjust the code to suit which was very clumsy. It's a lot more code work, but it means I should be able to setup screen layouts and save them, and maybe even set the conditions that cause it to switch between screens, all from the dash screen in real time without any programming etc. I havent gone too far down this road yet, but I've just started on making some rescalable graphs. So each of these is generated just from telling it the size, position, and what X Y variables it's tracking. You only need to watch about 1.5 seconds of this video to get the idea In other news, after having had a hoon through Markku's AE86 thread. I got a severe case of envy about his engine position, and, well, other stuff about his car that's rad. I realize this is car just a street car that puts on trackday pants. As opposed to a full blown race car. But there are some areas of easy improvement. I think engine position is one of them. So although I've pushed the motor back a bunch already compared to previous. My firewall is as good as swiss cheese from various wiring looms and so on that have passed through it various points. I've just looked at how much work it's going to be to patch it up, and it looks like a ball buster / just as much work to replace some of it completely. All of my next tasks to do, require the engine position to be finalized. Get the exhaust connected, driveshaft to correct length, and so on. I figure I'll regret not pushing it back further if I go make all this stuff for its current position. I'm also not liking my current engine mount scheme, because having the motor sit on threaded pins causes a few problems. Firstly there's a great deal of opportunity for cross threading the threads on the rubber mounts. Then the worst part is that I have to lift the motor high enough to get up over the threads, which sucks and will suck even more if motor is further back. Also I'm winding the threaded rubber mounts into my crossmember upstands, so if I cross thread the top of my engine mounts on the crossmember, it'll be shit. So I'll go back to something that uses nuts and bolts, and lets the motor slide in flatter from the front. I'm planning to push the motor back as far as I can, so long as it satisfies these requirements: -Must be able to remove rear coilpack with the factory bulkhead drip tray thing in place. Cutting this, is a bridge too far in terms of losing street car feel. -Must be able to take off the cam cover and be able to do any reasonable maintenance tasks, with engine in place. -Easy to remove/install engine and gearbox as mentioned above -Sump pan needs to clear the crossmember nicely without any cuts etc. The position of the motor will be fine tuned by accessability rather than going full blown amazingness like Markku or Hyperblade in the 86/Starlet. Unfortunately though, progress is as good as on hold for a while. As I need to keep my car easy to move around until my new shed is built. The shed should have been finished by now, but Covid threw a spanner in the works just while we were negotiating consent, then a few other hurdles turned up. It's dumb, but if that's the worst of my problems in the world then it's fair to say things arent all that bad. Will keep plodding along as I can! 14 Quote Link to comment Share on other sites More sharing options...
Popular Post Roman Posted September 7, 2020 Author Popular Post Share Posted September 7, 2020 Some minor progress on making things a bit lighter and a bit fancier. Made a new shifter with a 3d printed dissolvable inner and some carbon sleeve. Turns out nice this way because there are no seam lines like you get with flat sheets. I was grizzling about how that bottom bracket seems heavier than it needs to be too, but then I remembered I've got the SQ Engineering relocation kit thing so that gets replaced too. And then to the most important part of anything I'm involved with, max effort for negligable differences and pointless graphs: I couldnt find the old gearknob but I think I remember Nic saying it was about 60 grams. and missing one of the bolts which are 20 grams so around 1400 all up. "That's fine dave, thanks, I can see difference from the pics" No, no, I insist Also bought a brand new OEM Silvia 6th gear kit finally. Looking forward to those sweet sweet sub 3000rpm cruising speeds... What fuel economy dreams are made of. Still no progress on getting a new shed built before I can do any major stuff needed, so probably a few other random additions like this in the meantime. 24 1 4 Quote Link to comment Share on other sites More sharing options...
Popular Post Roman Posted February 9, 2023 Author Popular Post Share Posted February 9, 2023 Long time no post. Two years doing sweet fuck all on the Carina haha. I've spent at least the last year mulling over new motor options. Probably should have just finished getting beams motor going, but oh well. However I've got some renewed enthusiasm at the moment. New motor Criteria: -Capable of big rpm -less weight than beams motor -newer motor / alloy block -Be nice if it bolts to a J160 without adapter These are some of the engine options I've been going back and forth between: 1NZFXE NA Probably a bit too slow for a ~950kg rwd car... But maybe alright with a diff ratio change. Would definitely be the lightest option. Lots of room for activities with intake etc! 1NZFE+turbo This would be great because it would achieve nearly everything I want (lots of rpm, a power increase, less weight) However I do prefer NA motors for ease of working on but also tuning is more fun I think. This wont bolt up to a J160 directly, so that's a headache not present with some other options. Also the exhaust runs down the drivers side which is a bit annoying. Bike motor I am keen to do a bike motor project, but, I think this isnt the right car for it. Too big/heavy. But how cool are bike motors? Built with amazing ports, mega rpm, and sequential box that can shift near instantly. Bloody good. It just really excites me to buy a 10 second bike and turn it into a 14 second car haha. Have you considered a 2ZZ Yes, yes I have. Pluses to this are that it's lighter than a beams motor, VTEC! and built from the ground up for lots of rpm. Good power potential. Downsides - motors are hard to find, that arent either clapped out from high km, or needing to buy a whole car to get an engine. Needs a bellhousing adapter, exhaust runs down the drivers side (steering column/brake master cyl annoyances) Niteparts is currently working on fitting a 2ZZ to AE86, really excited to see how it turns out. 2ARFXE or 2ARFE This is a big late model 2.5 litre 4cyl motor that's gaining popularity for modding. The FXE version uses the same tricks as the Echo motor, however, the factory cams have some pretty decent lift and duration unlike the 1NZ. So it looks like the popular cheapish option is to use a second intake cam in the exhaust side, then you've got 270deg cams with 11mm lift. Pretty sweet. Downsides? It's a tall bugger. I'm definitely not wanting to do anything that requires cutting the bonnet. Might not fit. Also, it would be weight neutral rather than losing any weight from front of the car which was one of the goals. Since it has quite a long stroke, it probably cant rev as high as I'd like. I'm more concerned about engine sound/feel/revvyness rather than outright power. It's a good option, especially if it can bolt to a 3S bellhousing - but being rpm limited to 7500 or something. Blargh! So, yeah. Those were the options I've been considering and trying to decide between. However while browsing through Dodson's trademe site for cheap big block 4 cyl FXE engines I spotted something interesting that I'd not considered... A new challenger appears! 4GRFSE Now this is a bit of a WTF option but hear me out. The 4GR's big brother, the 2GR. Is a 3.5 litre V6 that is commonly swapped into MR2s etc, because can bolt up to a 3S bellhousing. However most of the 3.5 litre versions suit having the motor tipped over for east/west config. I think you need to buy a lexus motor to get the RWD config motor, and uhhh yeah. Lexus tax. The 4GR is the 2.5 litre direct injection V6 from a Toyota Mark X, which is setup for FR config. It's got a rear mounted sump, which by eyeball should easily fit behind Carina crossmember. Also, they're cheap and readily available as Mark X are a fairly common car. Why not 2GR for an extra litre of displacement? Well, the 2GR makes good power but it's got a big bore, big stroke, and will be rpm limited on factory pistons to less RPM than I'm willing to settle for. The short stroke and smaller bore of the 4GR motor means pistons/rods are lighter and the mean piston speed is considerably lower at a given rpm. To be honest I've never really consided a V6 because I think a 4 cyl motor is more in keeping with the spirit of the car. But when you look at the numbers, there's no way to get an off the shelf 4cyl motor that has a big enough bore and short enough stroke for mega rpm. Even the tiny 1NZ has rpm disadvanges compared to it. Another interesting aspect, is that since this motor is DI only from factory. It doesnt have any injectors in the ports, which usually seem to result in more angle on the port. In this case the DI injectors are mounted below the intake ports, so it's got STEEP intake ports. Another good sign for high rpm. The downside is that as per every DI only engine, the intake ports clog up with shit. Solved by converting back to regular injection. Still thinking "why a V6?" (boooo) I think it's going to be possible to get a semi reliable 10k rpm from this motor. Honestly how many engines are capable of running to 10k? How good is that. If these numbers stacked up the same way with a 4cyl motor, I'd use that. But they dont. Also to remove any doubt of this motors absolute peak performance potential this son of a bitch has a modern day TVIS equivalent. shit a brick. I've come to realize from the echo that revving the absolute bejesus out of something puts a smile on your face for longer than outright power does. This motor, the bottom end at least, should be 10k rpm capable with considerably less stress on the motor than any other option from the Toyota stable that I can see. Here's some graphs, because, well, it's my thread and thats how it goes. Bore size 4GR is about middle of the range for bore size, which is a good indicator of piston weight. Keeping in mind that piston area goes up non linear to diameter, so only a few mm more can mean a considerably heavier piston. pistons will definitely be a a fair chunk lighter than 2AR or 2GR options. Stroke length 4GR has got by FAR the shortest stroke out of options. Nearly 8mm shorter stroke than a 1NZ. Considerably shorter stroke than a 2ZZ, which is a pretty good benchmark for a high rpm capable engine. God damn delicious! Compression ratio Lower than some of the options, nothing beats the 1NZ! However 12:1 is still awesome, the fact it's a non hybrid factory engine with CR that high is great. Rod length (Thats what she said) 2AR has the longest rod, but it's also got the longest stroke. Apparently the 2GR and 4GR have the same rod length, which would mean the 4GR block must be more compact. Good news for keeping the engine height down a little. Nearly 10mm longer rod than a 2ZZ. Rod/stroke ratio The highest the rod/stroke ratio number is, the straighter the rod stays when travelling up and down the bore. So less side loading on the bore. This is generally considered a good thing for high rpm, and it also changes how long the piston dwells at TDC and BDC. (higher number favorable for more rpm) An S2000 has a rod ratio of 1.82:1 Having a 1.92 rod ratio is god damn glorious for high rpm from a factory engine. Mean piston speed Rather than picking an RPM and listing the mean piston speeds, I've chosen a mean piston speed (4500 feet per second) then shown how high each motor can rev when reaching this number. The 4GR does wees over everything else. Keep in mind that I'm revving 1NZ to 9k currently, which would be same piston speed as nearly 10k with the 4GR. (4GR will have a heavier piston though to be fair) Sooooo yeah. I think a 9500-10,000rpm motor should be doable, as there are big cams available. If so, it will be friggen wild. Well, there are big cam and stiff valve spring options available for the 2GR - but I'm not yet sure if these will fit the 4GR. Looking at pictures of the head gaskets, it looks like the 4GR has the same bore spacing but just bigger gaps between cyls. I've bought a motor to test fit everything initially. It's not much of a gamble as they're reasonably cheap. I'm hoping things will all work out, but there could be some issues yet. Good thing I got some practice using a hand drill to modify engine blocks with the 1NZ. Might need these skills. Here's a few pics of the next modern Toyota FE motor I'm planning to turd polish into oblivion. I'm super excited about this! There's still a zillion reasons why it might not work out, or something doesnt fit, or blah blah blah. But if it all works how I'm hoping, it's going to be cool as hell. 41 9 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.