Roman

Yep so the crank could swap. Ideal starting point is the 5GR motor. Its 2500cc with 2GR sized bore and a crazy tiny stroke. But this motor was only available in China for whatever reason.

Not much progress on this as I've been full steam on house stuff, and just too exhausted to do anything car related. However I had a brief respite the other day so started stripping the motor down to try get the heads off. As the next tasks will be fitting stiffer valve springs, and pressing in some plugs for the direct injection ports. It's amazing how much smaller it gets with the stuff off - all of the covers etc dont really contribute any meaningful weight. So this picture here is a reasonably good representation of where the weight is distributed. Definitely a bit lower and further back than beams motor. So, although I'd prefer the car lost some weight from the motor swap instead. This is a reasonable consolation. Something that is on my mind at the moment, that might screw my plans a bit. Is that all of the good things about this motor, are that its a big motor that's been downsized. But there's one major disadvantage to how they've retained the big motor's geometry. The bore spacing is the same as the 2GR, and the valve spacing is the same - this means the 2GR cams will fit. This is great as without beefy cams, this project is dead in the water. But it makes for an interesting design limitation with the head. If we take a look at a 2GR head, with its 94mm bore. You can see that the distance between each pair of intake valves is reasonably spaced, to not be shrouded by the bore, and also not too close together either. So, if you had a blank sheet of paper engine design this is what you'd do. So that's great. But the 4GR motor is keeping that same centers between valves, with a bore that is 11mm smaller. So they have no choice but to shrink the valve diameter to fit inside the bore, as they cant move the valves closer together. So there are these pissy little exhaust valves, and intake valves which are much more shrouded by the bore. With a massive gap in the middle. So, this might be a factor that limits its potential to make power at high rpm, even if the rest of the motor is technically capable. Will give everything a measure up once I've got the heads off. This isnt a deal breaker which will make me choose another engine, but its certainly not ideal either. One potential upside is that these smaller valves will be a bit lighter, which might be helpful. But again, ideally would trade for larger valves that are closer together.

Anyone done this before? I realize it's not cost of time effective compared to welding pipes together. But it's also bloody cool / oldschool method. Not planning on doing it myself, just wondering if anyone out there has had a play with the idea. Basically the concept is that you pack a pipe full of dry (important to be dry...) sand, then heat the area you want to bend with a blowtorch. The sand ensures the pipe doesnt fold or crush. Here are some cool looking sand bent headers from an old timey Grand Prix car popular with bike guys

"Daves weird hobby" could describe literally anything on this forum

Shit yeah! That piston looks rad. How long is it going to take them to do the other 5 though? I'm excited to see how this all comes together, gonna be sweet!

I've bought a Corvette fuel filter, someone mentioned somewhere ages ago that it has an internal FPR, and runs at 60psi. So its an easy way to make a "returnless" setup when there's no provision for an FPR on the rail. After the filter I can just tee off the fuel lines from there to each bank. Will keep the quick release fittings on the fuel rails and make up some AN-6 lines to them. I will need to reroute my fuel hard line a bit though, as I'll have exhaust pretty close to it now. I bought some of the cheap/common dual H bridge boards (L298n) to run the e-throttle motors, but reading further people say that these are quite inefficient (so run hot, and have a big voltage drop to the motors) So might not be as snappy on throttle response as I'd like. However, they're here so I can at least put together a prototype of sorts. Doing some further digging around and people reccomend the mosfet based ones instead, my ECU uses a mosfet based setup. So got these. BTS7960 43A H-bridge Will wire everything up to my ECU and then start writing some code. I will hopefully get everything that the e-throttle module needs into a single CAN frame, maybe two. Then run it on its own canbus network with nothing else on it. Will run the widebands and dash on the other canbus line. Other stuff on to-do list currently: slot flywheel holes This is with Dad, hopefully finished on CNC machine this weekend. Then will bolt it up and take a look at bellhousing situation once I can center the gearbox on the spigot bearing. Sump Pretty close! This version would be good to go, except for that I'd prefer the pan to be a little further back from the swaybar. Will get there. Starter motor This is going to be interesting. As although the 3S flywheel works with the gearbox and crank okay once the PCD is embiggened. The ring gear diameter is a little smaller than the 4GR one. So this means the Mark X starter motor in its normal position wont work. Then the 3S starter position wont work either, because it's normally tucked in on the side of the block up high. But now there's a vee in the way. So I still might need to look at rear mounting a starter. Lots of problems to figure out. Sort of getting there but also feeling like I've hit a bunch of roadblocks at the moment.

Ooohhh awesome to see this getting some love! Gonna be awesome. Anything changed in terms of plans for motor etc for the car now?

Yeah another thing in favour of 4GR is keeping max torque down a bit so a J160 wont explode behind it. However that now seems a bit of a moot point that it'll need a gearbox adapter anyway. However it'll still be nice to use the J160 as it's got good ratios, and, I've already got some to use. I'd rather not go to a bigger heavier box if I can help it.

It's just a number I pulled out of my ass to show a comparison of engine rpm at that piston speed. That was somewhere around max rpm for each of those types of motor. As it seems a more meaningful comparison than the other way around (comparing piston speed at a fixed rpm) And yes I see I labelled it wrong, should be FPM.

Not too sure, havent weighed anything. However gut feeling is that it wont be meaningfully lighter than a 3S setup. Was using this calculator. Was using feet per minute as the units. Sorry if mislabeled it somewhere. https://www.omnicalculator.com/physics/piston-speed RB20 has a nice short stroke and a pretty good rod ratio.

Here's the remix.

HAHAHA This car is the gift that keeps on giving (Go to 11:20 if it doesnt automatically)

I cant flip them the other way because the fuel hose connection onto the fuel rails will hit the bonnet. The problem with standing them up too much more is that it limits me to something like a 20-30mm trumpet on the end. Not sure if that will be enough runner length. Although, the head does have quite a deep port. Key thing here is to try keep making progress towards getting a running motor. Then play around with all the variations on things for about 55 pages later on.

Ahh look at that luxuriously spacious 90 degree Vee! 60 deg vee makes life a bit more tricky for the intake. But, it also gives better room for the exhausts which is sorely needed. Win some lose some. It would be nice to be able to cross the runners across the vee like that V8. But that only works if your throttles are spaced apart at least as far as the ports. Where as my throttles are closer together than the ports. The other pair of throttles turned up today, as well as the H bridge controllers. Looks like I'll have way more room than expected if I want to make them more vertical or push them closer together. I'm desperately trying to avoid raising the motor any more, so the sump situation is taking a fair bit of trial and error. However I've got a setup that now clears the front pulley, the pan nearly fits okay (will push it back a bit further from the swaybar) Then still dipstick hole and oil filter to figure out.

I'm not too sure, but I'll measure it when next in the garage. However I can confirm that it's bloody huge. @anglia4 my ECU actually only accepts a digital input for "clutch", so I'm actually using a general purpose output that then triggers a virtual on/off switch. Reason for doing it this way, is that matching the on/off switch to the clutch engagement point for rev matching or flat shift is super critical. So being able to adjust in software is super awesome. It's not strictly necessarily but it's an absolute life saver. I can add other analog or digital stuff over canbus as well, so not hugely stressed about running out of IO. It's mainly just a case of making sure that all of the mission critical stuff is direct to ECU where possible.

I'm pretty close to final version for throttles I think. I've pushed the pairs a little closer together towards the center of the vee, so hopefully still fits okay. I've added allowance for the injector and added 10mm of straight section to the top of the outside profile, for the silicone joiner to clamp onto. So although it looks a bit derpy on the outside shape, the inside shape means the injectors should be spraying more towards the center of the port rather than onto the walls. So that's nearly one job finished. Only about 50 other things to figure out! Also, I was talking to my Dad about ethrottle situation. He said that I should stop being a little bitch and just make something to control the motors over canbus (Actually he didnt say that, he's a great person) But, yeah, it's not actually all that complex to control a DC motor with a PID routine and an H bridge I guess? Then I could add a feed forward table so it works better than the standard setup. The MaxxECU actually has a way better e-throttle control scheme than Link does, in some good detail here: https://www.maxxecu.com/webhelp/settings-ethrottle-throttle_control.html So I might be able to make it work a bit better than standard. Will get someone smart to review my code though so I'm not gonna kamakaze myself into a wall because I divided by zero or something. However, the ECU would still be accepting all of the inputs, and still be in control of safety shut off features like killing power to the e-throttle relay if it sees a fault. However this does mean that I'm absolutely nuking my ECUs allocation of analog inputs. So if needed I'll try push some more stuff to canbus, and keep critical things direct to ECU.

I will do the same trick as before, print with Nylon PA12 and then wrap with carbon fiber. This time I'll add some crush tubes so the bolts can do up tight without having to be precious about overt tightening them.

I got both sides fitting nicely. Sweet! I guess I should try find some injector plugs and TPS plugs that suit the bike, hopefully they're fairly generic.

Yeah welding it isnt even really an option, because it's super thick there. As this replaces a dual mass flywheel. But that's also what gives me some confidence that it's not going to blow to smithereens either. Alternatively I could drill new holes in between centers of existing, but I think slotting would likely be the better option. The consensus of internet facts(tm) seems to be that so long as you're slotting outwards rather than inwards, it's okay. @shrike I think the radiator will be fine, will just mix and match hoses to make it work.

Few bits of scatterbrained progress, on everything but the sump. haha. I pick-a-parted a Mark X radiator and fan assembly, as the radiator hose sizes on this motor are real big. It's pretty big, but it clears the bonnet. Sits too high for the hoses to line up which is a bit meh. But, keeps things simple in a lot of ways I guess. Also one set of throttles turned up. Took some quick measurements then printed just one runner to test fit. Has an alriiiggghhhttt amount of clerance to the bonnet. But these throttles actually have a massive head height above the throttle plate. So I might machine them down a bunch, so I can start a curve on the runners earlier. Just printing another iteration that's about 10mm taller, and a little less angle. The throttle spacing is not quite as good as my guesstimate with the 1NZ / S1000RR throttles. Should be alright but longer term will probably try come up with something that keeps them all straight. Also, I'm glad I didnt jump the gun and buy a 2GR or 1GR or whatever flywheel. It looks like all 3 motors have different PCD on the crank. Which is good news for me, as the 4GR is closer in spacing to the 3S than the 2GR is. With a real life measure / eyeball it's probably close enough that put my chromoly flyhwheel on the CNC machine to slot the holes and it would be sweet. This would make life a bunch easier for the gearbox situation as well. As the other FWD flyhweels dont have allowance for a spigot bearing that the J160 needs. Also some random trivia, there is a 4WD Mark X. As seen at Pickapart. Unfortunately the sump is considerably less useful than my existing one, and looks like a bastard to get out. I dont know why a 4WD Mark X needs to exist, 210hp in a 1600kg+ car surely makes for a pretty sluggy driving experience. I've gone through a few more iterations of trying to simplify the sump situation but I'm not quite happy with it yet.

Well I think we can consider the Alfa 155 DTM as the real life version of maxxing it out for a 2.5 litre V6. So around 365kw... 196hp per litre!

I have had to narrow up the sides a bunch, as it's tight for clearance where the steering column is, and the brackets for the steering rack. However, getting closer towards a workable plan. One annoying thing is that the edge of the front pulley hangs down lower than the bottom of the motor, so will need to allow a notch or whatever. Will move the oil pan about 50mm forward of this point so its got heaps of clearance to the crossmember to make life nice and easy. This will also mean I will be able to weld a little upstand piece on it, to fit the 3SGE dipstick as well. Then the only hard part left is figuring out the oil filter lines.

If you have a steep valve angle, as in, the valves are more vertical. Then you need absolutely insane amount of lift and overlap in order for both of the valves to hit. At full intake cam advance, my intake valves are chasing the piston down the bore as close as they can. So hitting exhaust cam isnt the limiting factor.

I whole heartedly encourage graphs and wild specualtion

I somehow messed up my sump scaling. I printed the first half of it, and it was about twice as big as needed haha. So started again from scratch because it messes things up if you just try rescale it. So the good news now, is that I can fit whole sump on the printer at once. And the beams oilpan might fit pretty awesomely to the shape. See how it goes.