Roman's 4GR V6 Carina discussion thread

[Roman]

17 minutes ago, Nominal said:

Why not run six pipes out the back?

Watch that video about the v12 exhausts.

6-1 would be way to go for epic soundtrack. But too hard. 

[ajg193]

Shirley a man of your calibre could do something funky and make the intakes be on the outside of the V so the exhausts all go into one spot

[BiTurbo228]

Just catching up on the first posts of your thread, and the 4GR does seem like quite a good base design for high revs (bar the smallish valves). I've plugged the basic dimensions in to my little engine internal forces database to see how different rpm ranges compare to other engines/setups. Mean piston speed is a halfway decent way of working out how high an engine might rev, but the thing that kills rods and pistons is maximum acceleration at TDC (the piston screaming up to the top of the stroke then doing a full 180 in a fraction of a second to come back down the bore).

At 10k rpm the 4GR is pulling 5449g which...is up there. No production engines have that high forces, and only a tiny handful of tuned ones. Your 1NZ is one of them at 5016g at 9000rpm. There was a bloke with a billet-cranked stroker S54 that was 5975g at his 9000rpm redline. K20a's at 9000rpm are 5131g. If you take F20a's to 9200rpm which they can do that's 5088g. That's it that I've come across, though I started it to look at long stroke I6s and torsional resonance so it's probably missing a fair few high-revving V6s.

If you take the 4GR down to 9000rpm then that's 4413g which is far more common for a lot of high revving stuff. High revving big-bore Alfa V6s hit 4608g at 9k rpm and they're far less built for revs than the 4GR. The factory redline for an S52B32 is 4137g, and are happy up to 4352g. Stock K20a limiter is 4685g, and F20C is 4655g. @kpr's 9k rpm short-rod 4AGE is 4616g.

All depends a bit on the piston and rod weight though, which I can't find online for the 4GR. 5000g with a 300g piston is going to be a lot less load than even 3000g with a 600g piston.

Journal/bearing surface speeds are up there as well. At 10k rpm they're 10.17m/s for the mains and 8.0 m/s for the big ends. I've only come across folks racing Jag V12s that are higher. 9k rpm again is more acceptable (though still high) at 9.15 and 7.2m/s respectively. The 993 GT3 3.8 is about there at 9.0 and 7.8 m/s, but that's pretty much it that I've come across. Not sure if they like to eat bearings like S54s or if they're ok. Big bearings and small stroke does make for a stiff crank though, so hopefully torsional vibrations would be less of an issue.

[Roman]

20 hours ago, BiTurbo228 said:

 I've plugged the basic dimensions in to my little engine internal forces database to see how different rpm ranges compare to other engines/setups.

This is possibly the nerdiest thing I've ever seen. 
I guess what I'm trying to say is... 
Will you marry me? 

That's really cool data you've put together, well done.
Will have a good look through this! 

I've not really considered bearing surfaces etc. Mainly because I'm completely ignorant on the topic.
I guess it's unlikely that bearing area needs to squeeze up smaller than ideal, thanks to 4GR being dimensionally the same as a 2GR. With the 3500cc sized spacing between bores.
It's got more room for bigger bearing surfaces and thicker crank in between, compared to a more regular 2500cc block of a straight six or whatever that's trying to be compact.
So the crank weight is probably somewhere in between a compact 2500cc V6, and a 2500 compact straight six. 

Some interesting data points using your sheet.
I think KPR has run both his long and short rod engines to 10k rpm, and they have identical stroke to my motor.
Cant remember what pistons. But at 10k:
KPR long rod motor is 5540 G,
short rod motor is 5699 G. 

The VHT Racing Starlet's 4AGE runs to 11,000rpm with what I think is standard rod length, maybe a little longer.
It's in a 4AGE block anyway, not 7A block. So even if they moved the piston pin up a bit, still wont be as good as KPR 7A block motor.
A pretty extreme build with roller barrel throttles, liiiiiggghhhttt pistons and custom rods etc. 
But that's 6700 G. sheesh! Gets rebuilt a lot though I think.

The 2500cc Alfa 155 DTM engine runs to 12,000rpm with a 61.3mm stroke, so that's around 6150 G assuming a 2:1 rod ratio. 
Obviously its a full blown race motor that' probably has a lifespan not much longer than a single race before everything gets rebuilt though.

Extreme example. Ducati Panigale V4 R is a 4 cyl 1 litre engine that runs to 8500 G with a 48.4mm stroke and 16,500 rpm redline. 
But 250cc per cyl, obviously tiny pistons! So like you say, it's more about the G x the mass of the piston which tests the strength.

Something else to consider about this 4GR project. Is that the valvetrain might absolutely 100% shit the bed at a much lower rpm than any of this carry on.
And it might be an incurable problem. Stupid rockers! 

Either way, will be interesting. Haha.

Buying a few blocks and making some parts to fix things that go boom is in the long term development budget. 

I really hope I dont fire anything out of the bonnet though, they're impossible to find these days. 
 

[BiTurbo228]

Thought you'd never ask 

Yeah I know practically nothing about bearing speeds and oil shear, other than that it happens and this thread mentioned a chappy trying to rev a Mazda K8-DE to 10,000rpm and killing bearings due to oil shear (he'd tapped the block to measure pressure at each main bearing which never dropped, but was still wiping bearings). I'm assuming they meant he killed main bearings, but it's possible he was killing big ends. There's also a thing when you're spinning high rpm that the oil pressure isn't great enough to force itself from the outside of the main bearings 'upwards' towards the centre of the crank, starving the big ends. That's why you get things like nose-fed cranks in high end motorsport. Either way it's caused by the diameter of the main bearings and how fast the crank's spinning.

Yeah the trade-off of having big bearings from it being a 2.5l version of an engine designed as 3.5l is that the crank should be pretty strong. Even with the smaller big ends of the 4GR it still has generous crankpin overlap which is a good indicator of rigidity. It might be a big V6, but V6 cranks are inherently short anyway. All bodes well for avoiding torsional resonance and other nastyness.

I think I'd got some numbers for kpr's pistons and rods. Not sure if it was from here or just online, but plugging those in to 10k rpm gives roughly 3.2 tons of peak force on the crankpin (fudged a little as I don't actually know the weight of the little end so it's an approximation). That's definitely going some! Although I've just done a cursory look into bike engines and they're something else entirely. Went for a Ducati Panigale V-twin as there's more info out there than the V4 and they've actually got car-sized cylinder volumes. 11k rpm on a 60.8mm stroke and 116mm bore  5272g and 4.2 tons. Highest I've come across for a production engine by some 13%.

Very true on the valvetrain! Might be worth running with no bonnet for the first few high rpm pulls 

[Nominal]

9 hours ago, Roman said:

I really hope I dont fire anything out of the bonnet though, they're impossible to find these days. 
 

Make a mould before you get started with the high rpm runs?

[Roman]

@BiTurbo228 the 4 cyl ducati that does 16500 rpm is something like 8000G.... but tiny tiny pistons so not really comparable

[BiTurbo228]

Yeah definitely goes to show how the actual thing that matters is weight. Can't find rod length for the short-stroke R, but if it's the same as the non-R then it hits 9000g at 16500rpm 

Found some piston weights for it though. It's all marketing gumph rather than a bloke with some scales and oily bits so I'm a little skeptical, but they mentioned 'saving 5g/2% on the R piston' which would put it at 250g for the non-R and 245g for the R. No idea on conrods, but I know they're Ti so if they're anything like the 375g rods of the 1299R then that puts it at about 4.4 tons. Still very much up there, but comparable to the 1299R at 12000rpm.

Amazing what lightweight spinny bits and a short stroke can do (though there's a hell of a lot of guesswork in those numbers).

[kpr]

interesting stuff.  

The pistons in my short rod engine  are 404 grams, inc pin and rings.   rods  around 465 grams.  

Long rod engine was stock pistons- 443 grams ,   rods  487 grams

guessing the long rod, even though pulling a few less g's would more likely to fly to bits due to the piston weight.

[Roman]

It was definitely interesting when I pulled the 1NZ apart and saw that the piston had ripped in half. 
Rather than the rods breaking which people usually do. (At much lower rpm, loading motor up with forced induction) 

I cant find any data on factory 1NZ piston weight. However JUN 1NZ pistons are 232 grams with the pin included! 
Not sure how much OEM ones are, but they also felt incredibly light.
Well, at least the half of one that I was holding...
Then the rods are tiny toothpicks as well. 
Looking through some logs it would bounce to 9200rpm when doing burnouts at the drags. 

 

[yetchh]

What sort of lifter arrangement does the 1nz have? valve buckets or rockers? 

[Roman]

Shimless buckets in the 1NZ. 

I think (?) rockers means you can have a more aggressive ramp rate as you're not limited by diameter of the bucket. 

But, on the whole the type of rockers present in GR motors seem to be troublesome / not ideal.

[shrike]

1 minute ago, Roman said:

Shimless buckets in the 1NZ. 

I think (?) rockers means you can have a more aggressive ramp rate as you're not limited by diameter of the bucket. 

But, on the whole the type of rockers present in GR motors seem to be troublesome / not ideal.

Any valvetrain conversion options if it's the limiting factor?

[Roman]

Yes, these motors are a bit like lego sets. 

There is the head, itself, then there's another layer up that holds the underside of the cams and rockers. Then cam caps bolt onto that. 
The 1GR engine, unlike the others is actually a valve on bucket engine. 
So, although it would be a bit of an engineering exercise. Not impossible to design a completely different cam/rocker arrangement that bolts on top of the head.

However there are also some intermediate options like just designing some lighter rockers, or whatever.

[shrike]

 

 

7 minutes ago, Roman said:

Yes, these motors are a bit like lego sets. 

There is the head, itself, then there's another layer up that holds the underside of the cams and rockers. Then cam caps bolt onto that. 
The 1GR engine, unlike the others is actually a valve on bucket engine. 
So, although it would be a bit of an engineering exercise. Not impossible to design a completely different cam/rocker arrangement that bolts on top of the head.

However there are also some intermediate options like just designing some lighter rockers, or whatever.

Was doing some reading (I'm in the airport waiting to board) 

Apparently solid lifter conversions would allow circa 500rpm more (if the hydraulic lifters have issue) but would reduce cam duration so would need lager cams to compensate.

And fully agree on weight reduction helping to reduce mass/friction at higher RPM 

Been interesting reading about if certain engine parts will be able to lubricate correctly at the rpm your aiming for.

Are you planning on opening up the bottom end and doing any mods or is the plan to just run it stock and see what happens? 

[BiTurbo228]

@kpr Just added those numbers in and you're right. Although the long rod pull 150g less at 10k rpm that's offset by the added weight of pistons and rods: 3.2 tons for the short rod and 3.35 tons for the long rod.

@Roman Man those pistons are superlight! Carillo reckon their rods weigh 383g, though I'm never sure if they include the rod bolts with their weights. Even so, that's about 2.2 tons at 10,000rpm which is sort of middling factory figures for a lot of engines. It takes 12,000rpm to get to the same 3.2 tons as kpr's 4AGE. If the V6 doesn't work out like you'd hoped I vote try and get a 1NZ to rev to the moon

I reckon your deconstructed piston was the piston itself tearing apart due to g forces. People think about rods a lot but pistons can be quite happy to chuck their crowns off at high speeds too!

[yetchh]

7 hours ago, Roman said:

Yes, these motors are a bit like lego sets. 

There is the head, itself, then there's another layer up that holds the underside of the cams and rockers. Then cam caps bolt onto that. 
The 1GR engine, unlike the others is actually a valve on bucket engine. 
So, although it would be a bit of an engineering exercise. Not impossible to design a completely different cam/rocker arrangement that bolts on top of the head.

However there are also some intermediate options like just designing some lighter rockers, or whatever.

 

Yeah the early 1grs are shimless buckets but they changed to rockers about 2010-11 when they went to dual vvti, the other problem with the 1gr is, even tho it has a almost identical head gasket, it has a 94mm bore so I would guess the valve spacing would be to big for 4gr.. 

[kpr]

@BiTurbo228  any chance you could do the long rod with the 404gram pistons?  sorry thread hijack.     I was also most worried about the stock pistons letting go. i think the long rod with lighter aftermarket pistons should be fairly happy.

[BiTurbo228]

No worries lighter pistons does the trick for the long rod. 3.1 tons at 10k rpm, compared to 3.2 for the short rod with the same pistons (3137kg vs 3186kg, so not quite as big as the rounding suggests). Was 3353kg with the heavier pistons and long rods.

[h4nd]

On 14/07/2023 at 21:48, Roman said:

Buying a few blocks and making some parts to fix things that go boom is in the long term development budget.

Bahaha, if you don't start a slush fund for me to contribute to these shenanigans, I shall be frowny face. If that goes stonkingly well; after reading the comments on reducing weight to reduce high RPM stress, you could get the Italian metal wizards to make you titanium rockers, etc? Ceramic pistons??

Oh and bearing speed, oil shear etc, any option to do what the 2T goons do, and throw in roller bearings?