New project @ Supertec

[PeteKz]

Reflecting on this project some more: I've posted questions and comments before about combustion chamber shape and making it more efficient.

Like Dannobee posted above, the old hemispherical open chamber has significant drawbacks which have only gotten worse as bore diameters increased. More modern designs have tightened those chambers and flattened the piston tops. I would copy that work. Other manufacturers have done similar things, notably Harley in their Twin Cam engine from 2001-2016. The GM LS series are a different animal, with wedge-shaped chamber, due to the valves being oriented side-by-side, but they are also highly developed, very efficient, and have flat top pistons. A flat top piston not only minimizes obstructions in the chamber, but also minimizes the surface area exposed to combustion, and thus the transfer of heat from combustion to the piston. Keeping the heat in the combustion chamber keeps the energy to drive the piston down, and results in less waste heat into the piston, the oil, and the rest of the engine.

Thus, the first thing I would focus on would be heads and chamber shape. Build those by welding up the stock heads and then machining them to get the chamber shape. After going to that trouble, I would want to design new head castings that incorporate that shape. It might be possible or desirable to extend the valves further into the chamber to make it shallower, and allow flat top pistons and related dimensional changes. Maximize squish by holding the clearance to .025". However, I would make it compatible with the standard cam housings, rockers, and cams in order to sell them as "drop-in" replacement heads. Finish CNC the chambers and ports to exact shapes. Consider drilling oil cooling holes to the exhaust guides like some Porsche racing engines used. I would lean towards using the 993 heads with the rockers that bolt to the heads under the valve covers in order to eliminate pesky leaks around the rocker shafts. I like the idea of hydraulic valve adjusters too.

Displacement: "There's no replacement for displacement." Whatever case and crank you choose, maximize it, consistent with reliability. Given cost and availability, I would probably choose to start with the 3.2 case and 74.4 crank, and build it to 3.6 with larger bore cylinders. I also like the engine mounted oil cooler. But if cost is not a constraint, start with the 993 platform and go to 3.8 or 4.0. Open up and radius the spaces between the crank webs and cylinder bottoms to reduce air pumping losses inside the case. It might also help reduce oil aeration inside the case. Cross drill the crank too. Shoot for .0025" as the crank oil clearance and .0025" on the rods. Slightly more clearance will allow more oil flow through the bearings, to provide more bearing cooling.

Oil pump: Go big here too. More oil flow can be used to provide additional cooling. After all, these are really "oil cooled" engines. Use the larger Turbo squirters. Use larger diameter restrictors on the heads too. Radius the return holes at the bottom of the cam housings to allow the extra oil to flow more easily back to the sump.

TBC coatings: Use on the piston tops, combustion chamber, exhaust valve, and exhaust port. My sense is that the intake valve should not be TBC coated, to keep it hot to help with intake mixture vaporization.

Valves: Use light valves made of titanium and/or with narrower stems to save weight. Use what your current experience indicates are the best sizes for the matching ports and overall displacement.

Pistons/cylinders: Use the largest bore cylinders consistent with reliability. Use flat top pistons with valve cutouts if required, but only enough to clear the valves, to prevent weakening the edge of the crown and top ring land. I would consider using hypereutectic cast pistons instead of forged pistons to minimize thermal expansion and piston rocking. Forged pistons work for racing engines, but cast pistons work better in street engines. Hypereutectic cast pistons also have lower heat transfer than the alloys used in forged pistons, which keeps more heat in the combustion chamber.

Induction: I would want to use EFI for fuel/load/ignition mapping and to get knock sensing and dynamic timing retard. Carbs may be cool, but I would want to use the best current technology to make my anachronistic last-century air-cooled engine as good as possible. To reduce weight, I would consider 3D printing the intake manifold out of a high-temp polymer.

Overall, try to keep the engine weight down by using lighter components.

I'll add other stuff as it occurs to me.

[Neil Harvey]

Henry,

It never stops amazing me how others will build their engine with your money. Keeping it real will always come back in your favor.

Use your knowledge and experience to decide. You know in this business where advantages can be made.

[PeteKz]

He wanted out-of-the-box ideas!

[Henry Schmidt]

Originally Posted by Neil Harvey

Henry,

It never stops amazing me how others will build their engine with your money. Keeping it real will always come back in your favor.

Use your knowledge and experience to decide. You know in this business where advantages can be made.

I'm confused. I don't understand this response to my inquiry.
I just asked you how you know that the "new " 4 valve heads didn't have temperature control issue.
To be specific, what were the head temps @ a given horse power output.

[Henry Schmidt]

Originally Posted by PeteKz

He wanted out-of-the-box ideas!

"Out of the box" is a term I never used.
Actually I was looking for what build formula might be an interesting project for the board members to watch.
As for your suggestions, actually, there isn't anything on your list that I/we haven"t tried in one form or another but thanks for your thoughts.

[gsxrken]

Henry, you’re asking the community for their dream build (post #1) but then you layer on a set of conditions that is:
• Wild enough to excite a seasoned engine builder and attract auction buzz
• Different from anything you’ve done in 400 prior builds
• Feasible to build, tune, and sell to a wide enough market of potential buyers

That’s a hell of a Venn diagram, and the middle is pretty damn small! If you’re not trolling us (possible, especially trying to get Neal on the record) and this thing actually happens, I suspect you’re going to have to pick any 2, as the “fast/cheap/good” saying goes.

[mikedsilva]

Interesting thread.

[konrad911]

Well I’m thinking now about my engine. 3.2 as a base, 80.4 crank, 100mm cylinders.
I’m on the fence about the Turbocharging.
Street car. I’ve built 3 conversions to VarioRam so far, great flexibility and torque everywhere. But how many VarioRam conversions you can build before you get bored? Even though each next was better (last is 320 hp at crank and 380Nm from stock 964 short block).

I just wonder how to marry ITBs, VarioRam and Twin Turbo? And maybe no IC at all?

[Helix8]

Originally Posted by gsxrken

Henry, you’re asking the community for their dream build (post #1) but then you layer on a set of conditions that is:
• Wild enough to excite a seasoned engine builder and attract auction buzz
• Different from anything you’ve done in 400 prior builds
• Feasible to build, tune, and sell to a wide enough market of potential buyers

That’s a hell of a Venn diagram, and the middle is pretty damn small! If you’re not trolling us (possible, especially trying to get Neal on the record) and this thing actually happens, I suspect you’re going to have to pick any 2, as the “fast/cheap/good” saying goes.

Agree with this and would add the underlying hostility in replies.

[Henry Schmidt]

Originally Posted by gsxrken

Henry, you’re asking the community for their dream build (post #1) but then you layer on a set of conditions that is:
• Wild enough to excite a seasoned engine builder and attract auction buzz
• Different from anything you’ve done in 400 prior builds
• Feasible to build, tune, and sell to a wide enough market of potential buyers

That’s a hell of a Venn diagram, and the middle is pretty damn small! If you’re not trolling us (possible, especially trying to get Neal on the record) and this thing actually happens, I suspect you’re going to have to pick any 2, as the “fast/cheap/good” saying goes.

Maybe I made a mistake. I am going to build another engine just for fun with the end result of sending it on. The idea was to gauge the market and the interest in old school Porsche engines.
"What would you ask me to build if your were looking for an engine?"
Except for a few responses, that's not what I got. Now I'm being accused of being "hostile". Oh well, like everything else in life, "want to make God laugh, tell him your plans".
Be well and thanks for all the input.

[gsxrken]

Originally Posted by Henry Schmidt

I am going to build another engine just for fun with the end result of sending it on...

Please do, and document it here as you originally planned to. I'm sorry no one hit your sweet spot, but I'm glad you help out on this board (and in my case on the phone and with some parts). I'll be following along.

[Bill Verburg]

another possibility that's down to earth, sortta
since there is a huge amount of interest in earlier cars w/ 915s, particularly the longhoods

a 3.6 ITB, CoP, Motec, that revs to maybe 8K but has good torque from say ~3k up( something useful in a lightish ~2400#+/- car, something streetable but also comes into it's own on a track, and has heat

I like a filter on both pressure and scavenge side so 993 block or converted 964

the torque put through the 915s, especially in the lower gears is the main limiting factor,

[Glenfield]

I think we’re moving towards a place where revs and redline are more exciting than horsepower and torque. We’ve got electric vehicles (and turbocharging to a lesser extent) to thank for that. Since HP and torque are commodities (in everyday life anyway), I’d go after the most characterful and highest revving engine you can produce and lean into the engineering story.

In that vein, 4 valve / head is the easiest box to tick (assuming you can satisfy yourself it’s proven reliable). It’s identifiable to the largest market of would-be engine purchasers as innovation, so makes your marketing easier out of the gates. It may also open up more engineering possibilities to optimize the whole setup, which you’d know best. It gives someone an opportunity to put it in their car and tell their friends at coffee on a Sunday that it’s better than a singer or Tuthill and they didn’t have to wait [x] years or pay [x] mm.

I’d then think of what’s the most characterful engine. Don’t know why I have in my head that 2.5L is a fan favorite but let’s assume it is from a displacement standpoint. Then MFI, provided it can support the revs (which I wouldn’t know a thing about), because that’s what I think of when I think of supertec (even though I’m sure you’ve done them all).

That’s my vibe based answer to your question. The engineers can shoot holes in it.

[targa72e]

So my engine thoughts which are based on price, availability and mostly stockish parts combined with ending up with something cool that makes good power.

Start with a 3.0/3.2 case, these generally are less expensive than 3.6 cases and more readily available.
Take a spun bearing 3.2 crank and turn it down to 3.0/GT3 rod journal sizes, center drill mains for better oil flow. These are generally available at reasonable prices and are fully balanced.
Bored and plated Mahle 98mm or use 100mm nickies for a 3.4-3.5 displacement.
Use GT3 Pankle Titanium rods 130mm. These are light and generally available with 22mm or 23mm wrist pins.
Pistons CP X forging with coating on skirts, custom designed for application 10.5 to 1 for street use with 91 octane. Minimize dome and valve pocket depth based on cam profile and head volume.
I would love to see something like this with Peanut heads (dream). As a less expensive alternative I think twin plug 3.0/3.2 heads with larger valves and 8mm stems with the largest valves that don't require seat to be replaced, with light weight retainers along with appropriate porting would be good lower cost substitute for peanut heads.
Intermediate shaft with later steel gears.
Later GT3 pump
Larger piston squirters.
Rockers using light weight forged 993 hydraulics rockers modified with mechanical adjustment (used on 993 gt2 race cars I believe) for use with standard rocker shafts (bored and bushed). Or use 993 cam housings so you don't need to bore the rockers and eliminate some potential leak points.
Modern profile high lift (based on head flow numbers) higher duration cams. Looking for peak power around 7K with 7.5-8K red line but still with good low end torque.
Combine cams with bee hive springs for lowest pressure minimum resonance valve control.
Cams dictate ITB induction set up, MFI or EFI fine (I prefer EFI with ignition control).
The obligatory tuned headers with free flow muffler.
Ultimately looking for cool engine that makes good power but more importantly has a great personality and cool talking points.

john

[KNIGHTRACE]

Glenfield. I think it is going to fit.

[Henry Schmidt]

Bill Verburg, Glenfield, targa72e have offered exactly what I was hoping for.
Cost effective, old schoo, hot rod engines that can be built without reinventing the wheel.
All three as well as Turbo Pro's short stroke motor, and Mike's turbo are doable and sound like fun to build.

My faith in humanity is restored...lol
Thank you.

One side note: all the 4 valve inquiries have me thinking. Can I convert a GT3 heads fit on an air-cooled engine.
I have a customer with 959 parts but no heads. He was talking to me about his plan to do just that. I told him he was "craaazy" and here I am....

[Henry Schmidt]

Notes on Bill build.
We have a 3.2 case bored for 100mm
We have Mahle 100mm
We have a 74.4 crank with 53mm journals and also one with 50.80mm (Chevy)
We have 10+ sets of Pankl ty rods. 127 and 130mm
We have some big valve heads
I don't use GT3 oil pumps in air-cooled engines but I do have 996 twin turbo pumps
We're developing a new 48 tooth, steel intermediate gear in 5 or more sizes so backlash becomes a new easily adjustable metric.

[dannobee]

Some of you guys are getting way out there on the cost/gain ratio.

I was going to elaborate on what Bill Verburg mentioned and go big or go home. Take a 3.6l, add 102mm cylinders, offset grind the stock crank, and have some custom rods and pistons made. But with a twist. In higher division racing a very common crank pin diameter is the 48mm Honda size. Bearings are dirt cheap (compared to Porsche) and are readily available. The difference in size from the stock Porsche crank pins would allow the crank grinder to offset grind the crank to increase the stroke to 86.4mm, likely even on a core crank that would otherwise need welding or grinding anyway. This would allow an overall smaller size big end of the rod, giving more clearance for the oil pump.

102 x 86.4 = 4236cc. Or keep the stock bore size. 100 x 86.4 = 4071cc

Match an intake and cams correctly and "modern" head porting and you'd have a torque monster street engine. Close to 400 lb ft of torque and ~500hp with a redline of 6800 if you do it right.

If one were limited to a stock 3.0/3.2/3.3l case and 98mm.

98mm x 86.4 = 3910cc.

[Turbo_pro]

Just another opinion but I would be much more likely to de-stroke the 964 crank and run long rods (close to 1.95:1 rod length to stroke ratio) than try to jam 10 lbs of poop in a five lb bag.
If I understand it, spinning the long stroke motors seems to be cracking cases more frequently than ever before.

[Bill Verburg]

Originally Posted by Henry Schmidt

Notes on Bill build.
We have a 3.2 case bored for 100mm
We have Mahle 100mm
We have a 74.4 crank with 53mm journals and also one with 50.80mm (Chevy)
We have 10+ sets of Pankl ty rods. 127 and 130mm
We have some big valve heads
I don't use GT3 oil pumps in air-cooled engines but I do have 996 twin turbo pumps
We're developing a new 48 tooth, steel intermediate gear in 5 or more sizes so backlash becomes a new easily adjustable metric.

The thing about 3.2 case vs 3.6 is the head stud spacing
86mm vs 90mm

which in turn affects the meat available for cylinder base bores

the 3.6 having more base meat allows for a bigger more stable(reliable) base bore

when I built my 3.8 I knew that I needed the 109mm bas bore not the stock 107mm

I've heard horror stories about 100mm cylinder on 3.0/3/2 cases, maybe it's like the 4 valve air cooled heads, so few of them that a issue in 1 becomes a meme.

The other thing is the the crank vibration issue which gets worse as revs go up, I'm a huge fan of Neils ATi damper for the GT3 crank,