Piston Weights?

alpa

Member
Messages
241
Looking again at the pictures I must admit ABF heads are not identical. Now I wonder if VW are Coscast, I see VW stamps. A year ago I spent so much time studying posts and pictures that I probably convinced myself. My goal was to find the same kind of chambers to know how much horses they can potentially produce.

However there are striking similarities with the VW heads in details that are uncommon on car engines.
The 2v heads can't be compared with 4v ones because they can't have central spark plugs. That's an important detail about the combustion and maximum effective compression.

Sorry A.S. Motorsport I could not find any Lambo or Ferrari 4v head with the same head architecture.
TC heads are a very traditional hemi design except for separated intake ports.

There was a post on a VW forum (can't find it) where there was an excerpt from a book explaining the VW intake valve is set at the 24 degrees angle. This is about what I found on the Mase 24v heads.
I'm copying this picture of a cutaway, couldn't insert a link, sorry. Comes from :

https://clubgti.com/forums/index.ph...and-maximum-porting-dimensions-thread.193826/


95457


Few other pictures from the web:

95458


95460

Now compare with Maserati v6 24v head:
On this picture the chain is missing, it'd be on the left side of the cams.

95461


95462


95463


Of course for every time period there are dominant ideas. So it's not surprising to find similar concepts on many engines. For example the v6 18v cambox oiling is about the same as on the Porsche 944 head, the crank case looks similar to the Porsche 928, and many other.
But the 24v head is really special, that's my point. It goes against the main idea of hemispheric heads that are supposed to create compact almost symetrical chambers with one central spark plug and the biggest possible valves that would not be shrouded inside the head. The valve inclination should decrease the amount of turns in the ports and optimize the flow to and from the cylinders.
 
Last edited:
Messages
211
9546695469
348
95467
456

What's all very specific especially to Ferrari and Maserati is the bored intake runner port at very shallow angle to the valves (very visible on a 3.2)
And that they're bored as two round machining passes, even F136 looks like this

Also the coolant gallery on the intake port side and how there are ports, (although Ferrari's port layout is often per cylinder versus split exit on Maserati)

Detail shot on a 3200GT head I did a few months ago
95468



Again, these engines aren't related at all, that happend as we all know 4200GT onward.
What i'm trying to illustrate is they share a common set of deign features on the port and cooling side, heavily race car derived valve angles and steep inlet ports.
 

alpa

Member
Messages
241
What i'm trying to illustrate is they share a common set of deign features on the port and cooling side, heavily race car derived valve angles and steep inlet ports.

Yes these points are definitely important. That was the first aspect I mentioned on 8ht of December.

About the 24v heads
They are the same on all 24v 2.0 and 2.8 engines. There are slight differences between GT and not GT heads and may be the other versions, but they are all compatible.
24v heads are from Cosworth: one of the best head foundry and design company.
These heads are very good:
  • They had a novel (probably inspired by F1) cooling flow: on the exit side (in the V) there is a collecting gallery with two exits to the pump, so that the flow inside the head can be balanced with external restrictors on the exit path.
  • The intake seats are round-shaped like in F1 engines (not a 3-5 cut shape)
  • Intake ports are straight, almost vertical, with a long separation wall between two ports

Few years ago there was an article in the Race Tech Engines magazine about a novel cooling design in a WRC (or may be Time attack) engine and they were presenting the cooling architecture of the 24v head :) . Which was already present in the 18v heads.
 
Last edited:

BiTurbo228

New Member
Messages
27
Sorry for dredging a thread up from the depths, but it's a fascinating one and I've been mulling over some desktop modification options for these engines.

As explained earlier, one of the challenges of getting power out of these engines is some of the fundamental aspects of their architecture. Super-short stroke coupled with a tall deck height necessitates a tall piston and an unfavourable rod-stroke ratio*. Increasing the rod length to reduce piston weight makes the rod/stroke ratio even worse.

One solution to this would be to increase the stroke. We already know these cranks are perfectly happy with 47.965mm nominal journal dimensions as the Ghibli engines use this diameter to run with V8 conrods. Offset-grinding the 2.8l 49.965mm crankshaft by 2mm (as is done with many an American V8) and you get the triple bonus of an improved rod/stroke ratio (1.94 instead of 2.05), a bump in displacement to 3.0l (2956cc from the now-71mm stroke crank), and the opportunity to drop ~2mm from the piston compression height meaning lighter reciprocating mass.

It looks like Toyota 3S GTE rods are ideal for this as they're very close to the original rod length (137.787mm vs 137.35mm), have 47.9935mm big ends, and are readily available in aftermarket forged varieties. The only variable I haven't been able to find information about is big end width (assuming the Maserati engine uses crank-guided rods). The 3S GTE rods are apparently 26.797mm wide at the big end. So long as that's wider than the Maserati rods they can be machined down to the correct dimensions.

I haven't been able to find a suitable piston from another manufacturer that would work with this combination (hindered by not being able to find the compression height of Maserati pistons), but custom forged pistons would be on the table anyway if you want to change the rotating assembly of a Maserati engine, so that's no real change if you spec them with 22mm pins and ~2mm lower compression height.

Also, I've got a natty little spreadsheet I've put together calculating the internal forces that engines see at different RPMs (borne out of research into I6 crankshaft harmonics). At the standard 2.8l redline of 6400rpm these engines are experiencing a paltry 1.6 tons of reciprocating weight. Might sound like a lot (and it is in absolute terms!), but compared to many other engines that's nothing. RB26s are somewhere around 2.4 tons at their 8000rpm redline, Alfa 3.2l 24v bussos are at 2.1 tons at redline, as are BMW M30B28s, Honda C30As are at 2.2 tons. ****, you can get to 1.7 tons on the old 4-bearing 95mm stroke Triumph OHV I6s with a design rooted in the 50s before they start coming apart. Maserati cranks are far beefier than those.

Even with like-for-like weight forged pistons without the 2mm reduction in height mentioned on page 1, the lighter 3S GTE forged rods, and a longer 71mm stroke you'd be at 1.3 tons at 6400rpm which is absolutely nothing. You'd have to get to 7100rpm before your reciprocating forces equal the standard 2.8l 18v's 1.6 tons. All the while having better thermal efficiency from the improved rod/stroke. 8000rpm is still a perfectly reasonable 2 tons.

What would be amazing to know is:

  1. What is the weight of a 2.5l 18v piston is. We know Group A cars could rev between 8000-9000rpm, and I have the weight of the rods from the FIA papers. If I knew the piston weight, then I can get some kind of steer around the forces Maserati deemed the 63mm crank capable of withstanding. Even without that, we know that the rest of the architecture can support (or be made supportive of) at least 8000rpm.
  2. What sort of rpm people have reliably taken the 2.8l 24v Ghibli engine to. That way we can calibrate a little around what difference, if any, the smaller big ends have (an idea of what rpm the 2.0l 18v is happy to would also do this as I have that the 2.0l 24v is quite happy up to 8000rpm).
  3. What the longer 145.3mm V8 rods weigh. I have the piston weights, so if I know the rod weights I can estimate the forces a Ghibli 24v crank sees, again giving some form of idea what the 2mm reduction in big end diameter might do (though not a perfect one as offset-grinding will reduce the bearing overlap more than just straight grinding it down, though would still be a super-generous 23mm. For reference that's 9mm more than a 2JZ has).

Reciprocating forces aren't the only thing that limit the rpm capability of an engine, but they're one of the primary limiting factors. An 8000rpm 3.0l twin-turbo Maserati engine would something to see!

*Interestingly, having a rod-stroke ratio on the other end of unfavourable (i.e. too short rods for the stroke) has benefits for turbo cars as the speed at which the piston is pulled away from TDC helps mitigate against detonation. Possibly another reason it's tricky getting big power out of Maserati V6s.

Edit: the answers to question 3 is in the link to the French forum on page 1! One question down...
 
Last edited:

alpa

Member
Messages
241
I think you are searching in the wrong direction. Too complicated, not the low hanging fruits.
As I said in this thread to me the reason why v6 24v are not able of high power are valve and intake port sizes. I mean they are ridiculously small. To me 18v heads are better than 24v stock ones.
Big end width are specified in the workshop mans 18.9mm ghibli 2.0, 22mm (I don't trust this number) 2.8 24v, 21.4mm all older engines.
Ghibli big ends match some modern audi engines. Don't know more about audi rods.
There are many other aspects. For example the 2.5 liners are the most thick: 8mm. 2.8 are 6mm thick in some sections. And custom liners are very expensive.
Even if R/L are not perfect we are still far away from the usual numbers. And conrods look great and never break (to my knowledge).
Ghibli rev limiter is at 7300. It's way enough with these small valves.
The ghibli 2.0 piston pin diameter is only 20mm. It's not alot. But it should be ok up to 500hp.
All v6 and v8 rods are crank guides.
V8 and ghibli v6 use the same conrods. Weights are documented. Those I have go from 507 to 510gr. There were different designs, in my collection the best are from a qp4 2.0 engine.
Never had 2.5 pistons in my hands.
I'm in process of assembling my frankenstein 2.5 24v engine and I have a small stock of spare parts, I can measure them for you.
 

alpa

Member
Messages
241
And more generaly there are many other reasons that may prevent v6 from making a good level of power (in addition to the head flow problem). I've already talked about them on this forum: weak valve springs, low CR, low thermal efficiency of short stroke engines, thin liners.
 

BiTurbo228

New Member
Messages
27
Yeah I wasn't positing the idea as a sensible or economic starting point for modification on these engines! More an exploration of just how far you could go with development, should you want to make the best Mas V6 possible. Say more of a 'maximum effort' application than a 'street tuning' one. There's not many engines out there that you could offset-grind the crank to gain stroke and displacement and theoretically end up with an engine that's more likely to tolerate high rpm than the standard engine.

I've seen a dyno graph on this site of an unopened 2.8l 24v with 3200GT turbos making 339whp, so it's also worth caveating that 'Maserati V6s don't make much power' is very much a relative term. Relative to big-boosted BMW M50s, or RB26s, or 2JZs no they don't. Compared to, say, Rover V8s they're phenomenal.

Agreed that there are many other aspects as well that would need to be overcome to make the most of a high-revving larger displacement. Valves and port sizes being a big one, as you say. Personally my interest is in the 18v as that's the one I'm closest to purchasing, but my understanding is that both are relatively well matched (if not a little undersized) for their stock power. I say relatively well matched because there's a reasonable bit of evidence out there that narrower-than-expected ports help maintain flow velocity far better, improving low rpm torque (a la Alfa 12v V6s). Doesn't help with modifying though. How much meat is there around the port walls? That will determine how much larger you can make the ports. I understand valve shrouding is also an issue, which could be improved slightly by working the bowl to unshroud them and making the compression ratio back with a domed piston. Again, more maximum effort stuff.

If the 2.5l has thicker liners, and thus is better able to tolerate abuse, then the same offset-grinding approach could be taken for one of those. A 2.5l with a 71mm crankshaft would give you 2807cc, the same 1.94 rod/stroke ratio, and an even lighter piston. And we know for certain that the 2.5l package was able to rev to 8000-9000rpm on stock conrods, crank and valve train. Both of these were specified by Group A rules, though there was certainly some cheating going on. Rover SD1s used Volvo rockers to reach their 7500rpm rev limits, and fooled the inspectors by carrying a stock around with them and giving them to local BL dealers so if they carried out a spot-check they'd match the rockers on the race engines.

Weak valve springs should be easy to rectify (he says...). Hopefully there would be another car with compatible springs, but if not they can be custom made again. I've not scoped out costs for that though, so not sure how prohibitive that would be. Easier for the 24v as well as you don't need to worry about extra strain on the weird double-rocker arm situation (though presumably the Group A cars got around that somehow, even if it was by changing the rocker arms after every race).

Low compression ratio would be improved (slightly) by stroking. 7.3:1 on a 2.8l becomes 7.7:1 on a 3.0l, provided the combustion chamber is unchanged. If you're specifying custom pistons then there's no reason you couldn't specify a dome to raise that to something more sensible like 8.5:1.

Short stroke again is improved by lengthening it via offset-grinding, though is still comparatively short. Our theoretical 3.0l long-stroke engine is still quite oversquare at 1.32:1 bore/stroke ratio, though this isn't wildly out of whack with other engines that make power fine. 3.0l Alfa V6s are 1.28:1 (with big-bore version getting up to 1.38:1, close to the 1.4:1 of the 2.8l Maserati). Fiat Dino V6 is 1.54:1 and that won rallies. Ford Essex RS3100 is 1.29:1 and that was fairly dominant in its class in road racing. 3.3l turbo Porsche flat 6s (930 turbo) are 1.3:1), and the 911 GT3 4.0l is 1.28:1 and makes 400hp NA. Our theoretical 2.8l long-stroke engine is 1.15:1, which is almost spot-on what an RB26 is at 1.17:1. Low thermal efficiency is less of an issue if you can rev to the moon!

Half of the issue with short strokes and thermal efficiency is them often being paired with overlong rods. If you look at a number of engines that are known for being a bit gutless they have overlong rod/stroke ratios. 2.9l Jaguar AJ6s for instance are 2.34:1. The 2.5l PRV at 2.32 and the Jag V12 at 2.18 aren't exactly known for being gutless, but they're not known as bhp/l powerhouses either. Again, this is improved by stroking the Mas crank. 1.94:1 is much more sensible (though still on the high side of the 1.8:1 'ideal', boding well for high-rpm use). Really, rod/stroke ratio isn't a huge variable in making power, so the principle benefits of stroking would be greater displacement and lighter reciprocating mass. Improved thermal efficiency and resistance to detonation are just nice bonuses to have on the side.

Thanks for the info on the rods. Looks like big end width would be fine for use with the 3S GTE rods. They'd need some machine work to thin them down a bit, but that's a lot easier than the other way around! They're 22mm pins as well, so should help at that end as well.

Is that 7300rpm rev limiter on the Ghibli 2.8l or 2.0l?
 

BiTurbo228

New Member
Messages
27
I suppose the way to look at it would be 'if you can sort out bigger valves and ports, all of the bottom end architecture is well set up to take advantage of significantly increased flow to make power. Be that through increased rpm, stroking to increase displacement, or both'.

On the thin liners bit in particular, am I right in understanding that that's primarily an issue when you're trying to cram a ton of boost into it? If you can increase the flow enough to support bigger displacement and/or increased rpm, you don't need higher boost to make power. You just need a turbo that can support more flow at the same boost level. That might help with the thin liners issue (again, if my understanding is correct, which it might not be!).

As you say, it's flow that's critical. Valves and ports primarily, unshrouding the chamber secondarily, everything outside the engine after that.
 

alpa

Member
Messages
241
To me searching to make an ideal engine and reusing existing parts is in total contradiction. Today everything is possible, just pay. Steel or titan, DLC and ceramic coating, everything. Trying to reuse existing pistons or rods won't make you save big money (compare prices) but will cost you in a lack of freedom in your choices.
No 339whp at 1.6 boost bars is not big poney. It's common for good engines. For a street engine of that period it was excellent, unfortunately it's all you can probably get. Golf gti guys get more or less the same numbers from their 2.0 4IL boosted engines at same pressures. Gt3 is a modern engine, it's totally different.
 

alpa

Member
Messages
241
Please read my post where I was explaining what's wrong with 24v heads. It's not even the size of the valves, it's the ratio between valve and port diameters. Yes of course making huge ports does not help for drivability. We are far from these issues, we just can't get enough air. Look at the FIAT 5cyl 2/2.5 engines, they are able to make reliable 500hp at 8000. Both use the same head, but it's a head with well sized valves and ports.
 

alpa

Member
Messages
241
There is enough material in 24v to increase ports by 2mm. I did it. Certainly more, but I kept stock valves.

Unshrouding 18v valves is easy, I did it, enough meat. I should go on a dyno, not enough time, still need to better map ignition. You can also use bmw 6mm stem valves, saves weight and should flow slightly better.

There are many companies selling springs and retainers, just make your combination. Custom made springs are very expensive, at least this is what I see.
 

alpa

Member
Messages
241
I believe groupA biturbo were using 2 bar boost. The turbos (oil cooled) were trashed after one race. Engines certainly fully rebuilt.
 

BiTurbo228

New Member
Messages
27
To me searching to make an ideal engine and reusing existing parts is in total contradiction. Today everything is possible, just pay. Steel or titan, DLC and ceramic coating, everything. Trying to reuse existing pistons or rods won't make you save big money (compare prices) but will cost you in a lack of freedom in your choices.
No 339whp at 1.6 boost bars is not big poney. It's common for good engines. For a street engine of that period it was excellent, unfortunately it's all you can probably get. Golf gti guys get more or less the same numbers from their 2.0 4IL boosted engines at same pressures. Gt3 is a modern engine, it's totally different.
Personally I quite like the challenge of being as ambitious as possible with an old engine, but trying to do it within a semi-achievable budget. Especially mixing in old-school hot rod techniques (like offset grinding) that don't often make it outside of American V8 circles. Because I'm not made of money (barely even a fraction of me is!), making everything custom is usually out of my league if I can't make it myself. Manifolds and intake piping, fine. Conrods are a bit tricky! However, there's many cheap high quality rods for popular cars knocking around.

To be fair, I haven't got many quotes for custom rods and pistons in my time. 2 sets of 3S GTE rods (to get 8, so 2 spare) come to £738 from Maxpeeding Rods (who despite the rather janky image have been proven to be of high quality).

There is enough material in 24v to increase ports by 2mm. I did it. Certainly more, but I kept stock valves.

Unshrouding 18v valves is easy, I did it, enough meat. I should go on a dyno, not enough time, still need to better map ignition. You can also use bmw 6mm stem valves, saves weight and should flow slightly better.

There are many companies selling springs and retainers, just make your combination. Custom made springs are very expensive, at least this is what I see.
2mm is encouraging. Is that with similarly increased valve sizes? Give me a moment and I can do some comparisons with other big power boosted engines.

I believe groupA biturbo were using 2 bar boost. The turbos (oil cooled) were trashed after one race. Engines certainly fully rebuilt.
Knowing how other teams ran their racing efforts, I'd be surprised if they did a full rebuild after every race. I'd certainly expect rebuilds of the known weak points each race though. Things like rockers, if that's an issue. Not sure if it was though, we'd need to find someone on the team to ask!

The crank and rods certainly seem more than beefy enough to take the strain. Short stroke (so low forces) and really beefy main bearings, plus a short V6 crankshaft makes for a stout package.

Not surprised if the turbos were trashed after each race though. Those were controlled by homologation so you couldn't swap them out to bigger, more appropriate units as easily. Not something we have as much difficulty with on road cars (barring packaging issues).
 

alpa

Member
Messages
241
Stock rods get ovalized by up to 0.02mm after 100k km. Both ghibli/v8 and earlier rods. It's ok, but I don't know how long they can last at 7000 rpm.
The crank is undestructible, but it's nitrited.
 

alpa

Member
Messages
241
I don't understand. You are saying you are searching for economic solutions and you start with a mod of the crankshaft. To me the most expensive operation.
Last time I heard of a shamal crank restoration (add material, machine, nitrid, balance) it costed 6000 euros
 

BiTurbo228

New Member
Messages
27
Hmmm, all the more reason to get some aftermarket forged ones. I've noticed you can get 137.35mm stock-length forged rods now which is encouraging.

I see what you mean about the valve and port sizes. 30mm is on the small side, and 22mm port diameter especially so! Taking the port out to 24mm gets it back into the realms of reasonable for 30mm, but you're still down on the 34.5mm RB26 valves, 33.6mm 2JZ valves, and 33mm M50B25 valves.

You could fit larger 32mm valves to equal that out, but then you bump into the issue of small ports for your valve size again (even with 24mm ports). If there's no further meat in the casting to take the ports out to, say, 26mm (0.81 port to valve ratio, still not ideal but better) then I see where the real bottleneck is. Then you're looking at some of the more advanced techniques like machining out the port in its entirety and pressing in tubular sections as was done on racing XK engines for a long time.

I don't understand. You are saying you are searching for economic solutions and you start with a mod of the crankshaft. To me the most expensive operation.
Last time I heard of a shamal crank restoration (add material, machine, nitrid, balance) it costed 6000 euros
Offset grinding a crank to a smaller journal diameter would be much more affordable than that. No adding material. Just machining it offcentre. Costs are little (if at all) different from a crankshaft regrind for oversized bearings. I had my crankshaft reground (plus 6 bores honed) for £342 back in 2013. No doubt prices will have gone up by then, but not 10x higher.

Re-nitriding would add to it, but wasn't extortionate last time I looked at it for Triumph cranks. Similarly balancing wasn't extremely expensive. Last time I had a rotating assembly balanced the cost was £120 (again in 2013), though that was with an I6 crank that isn't as involved as a V6 crank (bobweights and such).

Let's say prices have tripled since then. You're still looking at somewhere around £1400 plus nitriding, which isn't a small amount of money, but isn't extortionate either.

However, I see where you're coming from. If we don't have enough valve and port to make use of our existing displacement, there's little point in increasing the displacement further (other than for the fun of it, of course). There may still be benefits to be had in increased thermal efficiency from the squarer bore/stroke and more favourable rod ratio, but those are going to be on a lower order of magnitude than anything that can be done to get more air into the engine. Unlikely to be worth the cost, unless you just want to do some experimenting.
 
Last edited:

alpa

Member
Messages
241
Get a quote for a v6 3 web crank balancing. No, better, get a list of those able to do that.

You can't increase 24v valves, unless you offset guides or make them offset (but then good like to install them properly).

Your ideas are very theoretical. You'd better open one of these engines to make the long list of existing constraints.

Grinding a crank, nitriding + balancing a v6 crank is affordable ? I'll wait until you did this, then we'll talk.