Roman's 4GR V6 Carina discussion thread

[Muncie]

How are you going on this? 

[Roman]

Not great at the moment.

Have been away for a work a lot, and then have also started putting my garage up.
So although that's setting me back a bit on having time/energy doing car stuff.
I'm hoping that It'll let me catch up once I've got some extra covered space and so can start burning the midnight oil! 

Hopefully get a few more trusses up today if it keeps clearing up.

[BiTurbo228]

Well you've certainly done that a lot quicker than I've done mine! It's taken me 8 months to do all the ground work and I'm just getting to the point of laying some concrete down.

2-bay plus a workshop area? So maybe 6x9m or something like that? Should be a nice functional working space that!

[Roman]

Yeah it's 6x9m and a 2.4m stud. 

At my last place I had a 12x9m with 3m stud, and at first I loved the idea of having a huge open big single space.
But the problem is that if you want to do painting, or grinding or whatever... you have to cover up EVERYTHING. 
So usually I just went outside anyway haha.
Then it all just ended up full of junk anyway.

So this time, on a rural section so I can have multiple things. So smaller workshop area but then have a big storage container, will build a little hut for the ride on mower, and so on.

Later on will hopefully have a little painting area or hot works area that you can just go get some work done without 100% admin time tacked on top.

But priority is to get this up, and get carina running for starters!

I've got a few issues at the moment though, all of the windows that were in it were a bit shit, and someone stood on the pile of glass and broke most of them.
I've patched up where the old side access door was (as goes out to nowhere) 

So I'm a little short on cladding, and need some better windows. I drew a decent model of it for when I submitted my plans to the council for consent, it's been handy for checking dimensions of 2nd hand doors and windows to try come up with a plan for how they might fit.

Probably need a bunch of smaller ones rather than gigantic things.

[Stu]

Thats only a short list... go DaveScience go!

 

[lowlancer]

Apologies if this has been covered, but any idea what these things sound like at full hoon?

[Roman]

Honestly it could be anything ranging from really nice, to absolute dogshit depending on exhaust and so on. 
Might take a few iterations of the exhaust to get it sounding nice.

It's hard to say because none of the (3500cc) 2GR people seem to run ITB or rev out very high, and I think thats what will define the sound on this.
Virtually everyone with the 2500cc version leaves them unmodified, because it's too hard to mod the direct injection engine.
Seems that equal length pipes is the key to nice exhaust sound though, so hopefully manifolds will help.

If the sound was hilariously out of place for a Carina, somewhat like below. That would be ideal.
(The Alfa 155 DTM has a 2500cc V6  that they run to 12k rpm)
However it might end up sounding like lil johhnys straight piped RB20DE swap 
INB4 @xsspeed tells me that's a good thing
 

 

[xsspeed]

So just like that alfa then

Puhuhu

[yoeddynz]

Your manifold looks fine to me.  It could certainly be worse...

[shrike]

5 hours ago, yoeddynz said:

Your manifold looks fine to me.  It could certainly be worse...

Clearly he just needs to run exhaust pipes out the bonnet 

[BiTurbo228]

Garage looks good. I feel you about grit and crap getting everywhere. My old place was horrendous for that (only one real garage space, horribly cluttered so impossible to clean). Doing anything that required cleanliness like engine or diff builds was a nightmare. In my current garage build I'll end up with a clean garage and a dirty garage I can keep (mostly) separate which I'm very much looking forward to!

To-do list doesn't look too bad car-wise (though I didn't spot anything about wiring on it, is that all done?). 'Finish building garage' feels like one of those single list points that expand into a whole second list with lots of bits and pieces in it. I suppose it depends on how 'finished' you need to get it before it's functional enough to 'finish' the Carina. Technically I suppose you don't even need doors or interior cladding. Just a roof, walls and floor.

11,000rpm Cosworth KF 2.5l V6 on bodies as some motivation  

 

[BiTurbo228]

Sounds like the 2GR is more in the 'seamless howl' vein like the Alfa DTM and Cosworth than the 'V6 warble' of, say, a Busso or a VQ.

Though as you say, it's mostly in the manifolding, and a bit in the rest of the exhaust.

[Roman]

5 hours ago, BiTurbo228 said:

Garage looks good. I feel you about grit and crap getting everywhere. My old place was horrendous for that (only one real garage space, horribly cluttered so impossible to clean). Doing anything that required cleanliness like engine or diff builds was a nightmare. In my current garage build I'll end up with a clean garage and a dirty garage I can keep (mostly) separate which I'm very much looking forward to!

To-do list doesn't look too bad car-wise (though I didn't spot anything about wiring on it, is that all done?). 'Finish building garage' feels like one of those single list points that expand into a whole second list with lots of bits and pieces in it. I suppose it depends on how 'finished' you need to get it before it's functional enough to 'finish' the Carina. Technically I suppose you don't even need doors or interior cladding. Just a roof, walls and floor.

11,000rpm Cosworth KF 2.5l V6 on bodies as some motivation  

Yeah I mainly just need a weather proof space. I can live without power lights etc for now. 
As I just need to be able to take my motor out to work on the firewall etc without everything being exposed to the elements. (and animals)
So hopefully not too long to go.

All of the wiring is done, I need to test some things like lights and indicators and so on.
But the stuff needed for drags at least is pretty easy. 
All of the engine wiring is sorted, all of the sensors have been tested and calibrated.
Alternator works, and all that.

That video you posted is interesting, because to me that KF motor sounds more like a high strung 4cyl. 
I guess the thing with a V6 is that if you have even firing order separation on each bank. (So 60 deg V6, not 90 deg like DTM motor) 
Then you have 240 deg of separation between each exhaust event on a bank. Compared to 180 degrees with a 4 cyl.

So the exhaust can sound a lot different because if you had a 6-1 (or effectively 6-1) you've got 120 degrees of separation so it will sound higher pitched.
But then if you ran totally isolated exhausts on each bank, or the pipes are really uneven.
It could also keep its "240deg" sound, and sound like a pair of 3 cyl motors driving in unison instead.
And then there's the effect of having an unequal length between each 3 pipes too on each side.

I guess a good comparison, is that a Subaru "boxer rumble" comes from only the pipe layout, not uneven firing order or anything. 
Yet sounds wildly different to most other 4cyl motors. If you put an equal length manifold on it, they sound just like a normal 4 cyl. 
Compared to a 4 cyl, there are lots of different ways you can affect the pipe interactions on a V6. 
Can have a 6-1 manifold,
long primary, short secondary 6-2-1,
Short primary, long secondary 6-2-1,
Separate 3-1 with long or short primaries.

Then all of the ways that different mufflers can affect the sound... and what cams you have.
Then how much noise comes from the intake. 
So it's probably a good thing that nothing is set in stone in terms of how the motor sounds.
As I'm not expecting it to sound particularly pleasant, or even go particularly well on first iteration.
But at least when the motor is going, and the car is driving. The fizz level ramps up, and doing the iterative changes and testing is the fun part for me.

[BiTurbo228]

Yeah I remember watching a video a while back about someone who had made an unequal length manifold for an S2000 to get it to sound exactly like a Subaru boxer.

IIRC the Alfa Busso is of the 'short not-quite-equal primaries, long but unequal length secondaries (can't remember if they're unequal but divisible, or fully unequal). They're of the 'warbly' variety. My 156 was a 24v though, which aren't as warbly as the 12v versions. Given enough time to fish the car out of the weeds I might be able to get some measurements.

Interesting you mention the KL sounding like a 4-cylinder as that's what I thought about that 2GR in the MR2, though I did get the hint of a V6 warble right at the top of the rev range in one of the clips.

Definitely in for the continuous development on this thing.

[shrike]

Not sure if you've seen the latest update good info on factory injectors ie 4bar gives 15% more flow (good to confirm)

Also talks about his recommendation on aftermarket injectors.

He's developing a billet oil pump gear 

He thinks the 2GR will do 8500rpm with rods/pistons/headwork. So hopefully the 4grfe will do more

 

[shrike]

Can't remember your cam selection but they do a recommended cam profile

https://frankensteinmotorworks.squarespace.com/shop/p/2gr-fe-frankenstein-motorworks-camshafts

Looks more aggressive then the kelford options 

[Roman]

Kelford biggest are 242mm @ 1mm, 12.3mm lift 
Vs 225mm and 11mm lift for the Frankenstein ones. 
So Kelford still the best option for biggest possible cam at the moment. 
Also considerably cheaper.

Quote

 

Not sure if you've seen the latest update good info on factory injectors ie 4bar gives 15% more flow (good to confirm)

Also talks about his recommendation on aftermarket injectors.

 

I'm using the direct injection only motor, so factory injector information isnt relevant here as I'm using BMW ones mounted in the throttles. 
I'm anticipating it'll have some fairly gross transient throttle response, as you always do when they're mounted high up. 

When I do a version 2.0 motor I will drill the underside of the ports and run 2x injectors per cylinder, one into each port. 
So then I can run some nicely small injectors and get a really good spray pattern from 32 nozzles total (or whatever)
and then less ugly shit up on top of my motor too.
 

Quote

He's developing a billet oil pump gear 

That will be cool! I mean it's not a known weakness at this stage, but probably a good idea. 

I dont know why everyone with a 2GR is so shy about revving them. 86mm stroke engines (basically every 2 litre 4 cyl motor) have no problem doing 8500rpm all day long. Beams motor was happy doing that. 

Only an 83mm stroke on a 2GR and I think they have a better rod ratio too.


To further throw a spanner into the works of my already stupidly compressed timeline to try get to the drags. I've been absolutely dead with the flu for about 2 weeks, only just getting better. 
So I'm back on the mission to get my garage finished but its blowing a gale today so cant put any more cladding up for now. 
Hopefully by next weekend the garage is all finished, which leaves me.... fuck all time to work on carina haha.

[shrike]

7 minutes ago, Roman said:

Kelford biggest are 242mm @ 1mm, 12.3mm lift 
Vs 225mm and 11mm lift for the Frankenstein ones. 
So Kelford still the best option for biggest possible cam at the moment. 
Also considerably cheaper.

You are correct I was looking at the 0.1mm measurement 😆

7 minutes ago, Roman said:

I'm using the direct injection only motor, so factory injector information isnt relevant here as I'm using BMW ones mounted in the throttles. 
I'm anticipating it'll have some fairly gross transient throttle response, as you always do when they're mounted high up. 

When I do a version 2.0 motor I will drill the underside of the ports and run 2x injectors per cylinder, one into each port. 
So then I can run some nicely small injectors and get a really good spray pattern from 32 nozzles total (or whatever)
and then less ugly shit up on top of my motor too.
 

That will be cool! I mean it's not a known weakness at this stage, but probably a good idea. 

I dont know why everyone with a 2GR is so shy about revving them. 86mm stroke engines (basically every 2 litre 4 cyl motor) have no problem doing 8500rpm all day long. Beams motor was happy doing that. 

Only an 83mm stroke on a 2GR and I think they have a better rod ratio too.

To further throw a spanner into the works of my already stupidly compressed timeline to try get to the drags. I've been absolutely dead with the flu for about 2 weeks, only just getting better. 
So I'm back on the mission to get my garage finished but its blowing a gale today so cant put any more cladding up for now. 
Hopefully by next weekend the garage is all finished, which leaves me.... fuck all time to work on carina haha.

Get well soon, keen for more updates 

I'd forgotten you had the direct injection only motor, version 2.0 should be alittle more refined

I just like he's doing RND and can show real world results etc 

Plus more part options is always good

You should send him some vids see if hes keen to collaborate

[Roman]

I probably need to have an engine hold its guts in for more than 3 minutes at a time first. haha.

Hopefully I'm past the worst of it and can start getting to the good part... (drawing graphs and posting them on the internet)

He's got very different objectives to me, so I dont think a collaboration video would make much sense. 
As he's running a business and looking for the cost/benefit point of making parts and so on. 
So ultimately his videos are promotional material for his business which is cool.
I think it's awesome he's able to do that as a full time gig.
 I dont really add any value to that equation.

One thing that I'm keen to test though, that he's pointed out as a flaw with factory cams. 
Is that he says there is a bank to bank imbalance because of the way the rockers work. 
I am thinking that this imbalance likely completely goes away once you have one bank's home position advanced slightly ahead of the other. 
Rather than assuming that the zero degree "hard stop" is each cams zero point in relation to each other. 

However that's just my theory and he's done a lot more testing than I have, so I could be totally wrong on that one.

[BiTurbo228]

Thought some graphs I've produced for the Maserati folks might help speed your recovery...

I've been looking into benchmarking port and valve sizes for as many engines as I can, just trying to get a steer on what's what. As I don't have a storage container filled with various engine heads, I've been looking at the FIA Historic Database which has homologation docs for tons of cars up to the early 90s (and a few a bit later). That has specs on the valve diameters and port dimensions at the cylinder head face (would have preferred throat diameter, but they don't give that...the prudes).

Long story short I've made some graphs plotting valve and port area against cylinder volume for as many engines as I can get my hands on. Makes for some interesting reading for saddoes like me. These are all multi-valve engines.

I haven't got the 2GR and 4GR on the graphs, but I have worked out where they'd plot on them.

For valve area (orange chart), the 2GR is 3.9mm2/cc which is pretty high up. Just a smidge above the YBB Cosworth. Interestingly, while your 4GR has smaller valves overall, they're similarly proportioned in relation to cylinder volume as its bigger brother (also 3.9, but a little further down so just slotting in above the Audi 20vT Sport Quattro). Probably bodes well for getting decent power and revs out of the 4GR as it's unlikely to be significantly undervalved.

Port area (green chart) is a tricky one, and shows one of the limitations of using port area at the cylinderhead face. The super-heavy taper on the 4GR throws things off considerably. If the area is about the same as a 52mm circle as you mentioned, that would put it at 5.1mm2/cc which is off the top of the chart. Way in excess of the excessively large ports of the ST165 (which I note they progressively made smaller with the ST185 and ST205 GT-Fours). If it could comfortably lose about half that area while still being proportional to the throat area (as you guessed), that would put it closer to the mid-pack just next to the 323 GT-R. Tricky to tell what's better in this case as well as there's some high performing NAs with comparatively big ports (4A-GE, B16, B18), some with middling ports (S50, TU5, Mi16), and also a cluster with comparatively small ports (S14, 190E Cosworth, BDA).

Edit: Mistyped on the RB20 and only counted one port out of its two. That should now be way at the top of the port graph at 3.9mm2/cc, above the B16A1