View Full Version : 3.5/6G74 and 3.8/6G75 bore size potential

Alan J
16-08-2009, 02:19 PM
After getting some details Dave passed on Graham has got into measuring up both blocks and just sent me some info regarding the differences and the potential of the 3.5 and 3.8 engine. He hasn't sent pics yet to show the differences but I'll get them up when he does.

3.5/6G74 block and crank

Weight = 48kg

Main bearing saddle weight = 5kg

Crank weight = 20 kg (crank has cross drilled pins and mains)

Cylinder wall thickness at bottom of No 1 and No 6 = 7 - 8mm (rough measure, not ultra sonic)

Cylinder wall thickness through No.1 water inlet = 7.5mm

Thickness at bottom of all cylinders above main bearing webs = 7 - 9 mm

Depth from main bearing tunnel to top of crankcase "vee" = 65mm

3.8/6G75 block and crank

Weight = 46kg

Main bearing saddle weight = 6kg

Crank weight = 20 kg (crank has cross drilled mains, through drilled pins)

Cylinder wall thickness at bottom of No 1 and No 6 = 5.5 - 5.7mm (rough measure, not ultra sonic)

Cylinder wall thickness through No.1 water inlet = 6.4mm

Thickness at bottom of all cylinders above main bearing webs = 6.2 mm

Depth from main bearing tunnel to top of crankcase "vee" = 67mm

So differences are:

3.5 block is heavier and has thicker walls so same maximum bore potential unless better grade iron is used in the 3.8 block. Apparently the 3.8 has a big external rib along both sides of the block, and the main bearing webs are cast full depth so there is more strength/rigidity there.

3.8 bearing saddle is 20% heavier due to substantially thicker/deeper side beams.

2mm more space in the top of the 6G75 crankcase so less relieving required for long stroke crank.

Cross drilling of 3.5 crank is as per that in most high performance and race engines. It gets good flows to both the pins and mains. The through drilling of the 3.8 may reduce main lubrication and help the pins, but it may just have been to save cost and there may have been no benefit to pin flows. Have there been any 3.8 bearing issues reported?

Cylinder bore potential

Unless the block is subjected to unusual harmonics or has especially poor grade iron it should be OK up to a 99mm bore! That's close to 4.2 ltr with the 90mm crank or 4 ltr with 85.8mm crank. Rubbish American iron regularly handles 110 HP per cylinder and 10,000 rpm on 3.5" stroke for 5,000 miles with good ring seal on 0.180" wall so a Mitsubishi block should be good for 70 HP per cylinder and 7,000 rpm without block fatigue with similar wall thickness. (Early Chev blocks regularly run down to 0.120" wall but racing block life was short and ring seal not perfect).

Rings would not be a worry due to Chev LS1 Gen III bore size but pistons would have to be made, also head gaskets.

A cheap option would be 97mm bore, 4 ltr with 90mm crank and 3.8 ltr with 85.8mm crank. Graham thinks the stock 6G75 head gasket would be OK enlarged to suit that bore size and 20 thou oversize low block series III Buick/Holden Ecotec V6 rings would fit. The forged 3.8 Supercharged piston could possibly be used (ACL #6MKRY9705).

In a 6G75 that piston is about 2mm too high but Graham thinks it could be machined. He's been told the top ring is 6mm down and the crown is thick enough for blower use and a 3.55mm deep bowl so it should be more than strong enough NA. Machined 2mm it would end up with about a 6cc bowl. It has a 23mm pin so the rod would need to be honed/rebushed.

In a 6G74 these pistons would be the correct height but the stock 14cc bowl would have to be machined a bit wider to get the compression ratio down, or the combustion chambers enlarged.

Stroking potential

Already a long stroke engine so why bother? Graham says limit would be about 96mm. Going down to a 50mm Nissan VG30 pin and rod the 3.8 crank could be offset ground for about 95mm stroke and 3.5 crank for 91mm stroke.

Graham would be interested in any info you come across on the strength and bore potential of these blocks.


16-08-2009, 03:26 PM
thanks for the info mate

17-08-2009, 12:20 PM
Alan, sorry for the noob question but what is the practical minimum thickness of the walls before you encounter serious problems?

17-08-2009, 03:56 PM
Not sure if this is of any use, if it isnt let me know and I'll delete it. From the 3000GT forums in the US regarding the same topic.

The largest 6G motor I can think of fitting in our car with readily available parts is a 6G74 Block with a 6G75 crank and a 95.5 mm Bore. 3868cc's

Just to break it down:
91.1mm - Stock 6G72 Bore
92.1mm - Max service manual reccomended 6G72 Overbore
93.0mm - Stock 6G74 Bore
94.0mm - Max service manual reccomended 6G74 Overbore
95.0mm - Stock 6G75 Bore
95.5mm - Max service manual reccomended 6G75 Overbore

76.0mm - 6G72 Stroke
84.0mm - Brian Crower Stroke
85.6mm - Pampena/Merkel 6G74 Based Stroke
85.8mm - 6G74 Stroke
88.9mm - Pampena/Merkel EN30B Stroke
90.0mm - 6G75 Stroke

(Correct me if I'm wrong Ray)

I'm thinking that there's a reason that Mitsu is doing 0.25 mm & 0.5mm over pistons on the 6G75 and not 0.5mm & 1.0mm over like they've done previously on the 6G72 & 6G74. Just a hunch. I would think wall thickness would be the main concern but the numbers I've seen out there for 6G72 wall thickness (0.288-0.300" average thickness @ 91.1mm) wouldn't make me sweat 96.0mm on an NA 6G75 so I don't know.

I don't think a 95.5 mm bore is the best idea for engines chasing 800+HP. The deck area between the bore and the coolant passages would shrink (and it's already tiny at 93mm) and I can only guess that keeping the cylinders sealed up would become even more difficult. Maybe o-ringing would push out the limits but who knows how much. Kepping the bore small(er) is always the safer bet.

Alan J
17-08-2009, 05:10 PM
Alan, sorry for the noob question but what is the practical minimum thickness of the walls before you encounter serious problems?

Practical minimum wall thickness isn't an easy one to answer. There are so many variables. Some blocks have very cheap rubbish iron, typically in USA, and yet they get away with Chev SB down around 0.100"! Before the heavy duty "Bow Tie" competition blocks we were stuck with the production block that were supposed to have a minimum 0.135" wall, but with core shift you had to go through a stack of blocks to find a good block OK as a NASCAR race block with 0.110" wall on finished 4" bore. Thats about 0.125-0.130" on sonic test. We squeezed about 75-80HP per cylinder out of them and they would hold together for a 400-600 mile race, never running below 7,300rpm for 3-3 1/2 hours. They ran 14:1 comp ratio, some teams went higher, and if they detonated you'd lose No 6 cylinder wall usually and lunch the engine. Then Chev produced the high tin content "Bow Tie" block for race teams. It was supposed to have 0.200" walls but core shift knocked that down to 0.170-0.180" on a finished 4" bore. These were V8s with horrible harmonics from the crank and the 16 lobe cam. Remember too that some were overboring these already thin blocks up to 0.060", knocking another 0.030" off wall thickness.

On the V6 Buick the race blocks finished at about 0.180" and on a 3.625" stroke crank made 90-95HP per cylinder at 8,000rpm. These were an even-fire engine but not as smooth as the Mitsubishi 6G. The engines lasted forever and ring seal was excellent.

In Europe the blocks tended to be thick due to antiquated casting techniques so the Mini would rebore 3.4mm oversize easily and the Ford 4 cylinder block normally went 4mm over without issues. When Cosworth did the GAA based on the V6 Capri they ran into main bearing issues so there were some stranges forces at work but never had ring seal/wall bowing problems even in 1000km and 24 hr enduros

The new Chev LS1 GenIII has 3mm thick steel sleeves and they will hold up pretty well at high HP/rpm. I don't know what the big bore LS2 Gen VI is but I know the current 7ltr Le Mans LS7 engines ran without liners on the aluminium block and the siamesed wall between pairs of cylinders was only 0.215" maximum, so the jacket wall would have been considerably less, probably not more than 0.150" I'd suggest.

So the bottom line is generally about 3mm wall is fine in a normal road engine unless there are some unusual forces at work or the casting/casting iron is very poor. For a race engine 4-4.5mm wall is what you want.


Alan J
17-08-2009, 05:17 PM
;1103036']Not sure if this is of any use, if it isnt let me know and I'll delete it. From the 3000GT forums in the US regarding the same topic.

Interesting that they are saying the 3.0 ltr has about the same wall thickness Graham measured in the 3.5 block. So seems Mitsubishi have been consistent in keeping with thick wall. They probably had to go thinner in 3.8 to get water space between the cylinders.

Has anyone seen what size bore Mitsubishi used in later 3.5 based rally raid (aka Dakar)engines?


Alan J
23-08-2009, 08:19 AM
Graham has finally sent me those pics of the 3.8 block and an AWD block(apparently same as 3.5 block but has a bit sliced off the rear to clear the transfer case).

1) and 2) Here you can see the rib added along both sides of the 3.8 block to stiffen it. The 3.5 has some vertical ribs that aren't in the later model block.

3) and 4) The 3.8 has full thickness main bearing webs but the 3.5 is a bit lighter with webs dished in.

5) 3.8 main bearing saddle has about 5mm taller side rails than 3.5. Otherwise they seem the same with same thickness and depth main caps etc.


23-08-2009, 08:58 AM
Always great to get photo's with the explanation! Thanks fella's!

26-11-2011, 11:18 AM
Just a heads up, it looks like my Pajero block is considerably different to the 380 one (just in case)

the V77W is on top in both pictures...

for more info see here: http://3000gt.com.au/car/74%20vs%2075/6g75/My%20V77W6g75/

26-11-2011, 11:22 AM
Can someone please post up bare block shots of a 380, all around though (front and back as well)

26-11-2011, 11:55 AM
My uneducated eyes can't see a huge amount of difference in the two blocks 4000GT. One looks a little 'heavier' and thicker around the cylinders? were you suggesting that one could be bored further? I'm a fan of 4L, its alot easier and quicker to say than 3.5L or 3.8L :P

26-11-2011, 05:29 PM
let me help :)


the one the guys on your site measured it to be 46kg, mine is 43
you guys have some extra bracing around the block, mine does not (pretty simply a 74 bored)
you guys also have the mount for the AC bracket which mine does not

let me know if you want more detail

22-02-2012, 08:07 AM
I just realized I did not post in here..

I sent both a 6g74(Debonair) and a 6g75(2004 pajero) block into Duncan & Foster Engineering to get the walls tested( to see what I can do by way of Boring it out), the results came back a week ago. For boost you want a minimum wall thickness of 6mm, so for each block machined at it's thinnest point down to this point you can get the 6g74 up to a 95mm bore, the 6g75 can go up to a 97mm bore.

the blocks are not the same, the 6g75 has a 1mm greater outer radius, keep in mind there seem to be slight differences with the 75 blocks so I cannot speak about other years (you'll have to send your block in for testing...it's only $150.)

for those running NA engines they were suggesting that another 2mm could be added (so a 99mm bore on the 75)

11-07-2012, 01:14 PM
So can you stroke a 6g74 engine with me 6g75 crank?

11-07-2012, 08:10 PM
So can you stroke a 6g74 engine with me 6g75 crank?

Yes, but you need to check the clearances of the rod big ends (some blocks are fine, others need a bit of cleanup).

19-07-2012, 07:46 PM
4000GT Have you done anything with the two blocks tested?

I am about to build a '74/'75 engine for mine and am wondering about pistons, is there a commmon piston available that may suit a certain bore/stroke ?

Screamin TE
08-09-2012, 10:21 PM
Can you safely fit 380 pistons to the '74 block or will there be clearance isscues with the valves?

10-09-2012, 04:09 PM
My understanding is the crank and heads from the 380 will be fine, no issues as the '74-'75 deck heights are the same.

Would pay to check if you are using the '74 heads with the '75 cams...........