Porsche 993 twin turbo

[hoffman900]

I love the way the fuel flow rules limited engines, like the Audi DTM engine and Formula 1 engines go with constant boost or near constant boost, and varying lamba. It also makings for a faster engine on the race track / dynamically as well.

For air cooled Porsches, it’s the heat shedding capacity that limits the power. I think the 993 Turbo runs at 10.5 AFR under sustained full load from the factory. When anything gets heat soaked above a limit, it turns off the power.

Oh for sure.

I mean, a point to consider is F1 engines are using separate oil feeds for the piston oil jets, which introduces even cooler oil to the pistons than the rest of the system. Complicated, but there are ways to work within the boundaries set.

[ptuomov]

I mean, a point to consider is F1 engines are using separate oil feeds for the piston oil jets, which introduces even cooler oil to the pistons than the rest of the system. Complicated, but there are ways to work within the boundaries set.

I am installing a second factory oil cooler in series with the existing cooler. The priority of piston oil cooling jets is relatively high as they feed off the galley that feeds the mains.

Overall, I'd guess that this engine is 60% cooled directly by air flow and 40% cooled indirectly by air flow via the oil coolers.

[ptuomov]

What's a good exhaust header primary inside diameter for this twin turbo engine?

Parameters:
3.6 L displacement from 6 cylinders, so 600 cc per cylinder
6800 rpm redline
Stock 408 hp at 0.7 bar peak power boost, target 500 crank hp at 1.1 bar boost
Exhaust valve diameter 42.5mm
Exhaust port diameter at the flange 38mm
Twin turbo K16/24 hybrids with each getting three 240 degree separated pulses
Turbine housing inlet 40mm being fed from three cylinders
Exhaust valve peak lift 10.9mm
Exhaust valve events at 1 mm: EVO 44 degrees BBDC, EVC 4 degrees ATDC
Exhaust camshaft duration: 228 degrees at 1 mm, 218 degrees at 0.05"

[hoffman900]

What's a good exhaust header primary inside diameter for this twin turbo engine?

Parameters:
3.6 L displacement from 6 cylinders, so 600 cc per cylinder
6800 rpm redline
Stock 408 hp at 0.7 bar peak power boost, target 500 crank hp at 1.1 bar boost
Exhaust valve diameter 42.5mm
Exhaust port diameter at the flange 38mm
Twin turbo K16/24 hybrids with each getting three 240 degree separated pulses
Turbine housing inlet 40mm being fed from three cylinders
Exhaust valve peak lift 10.9mm
Exhaust valve events at 1 mm: EVO 44 degrees BBDC, EVC 4 degrees ATDC
Exhaust camshaft duration: 228 degrees at 1 mm, 218 degrees at 0.05"

38mm to start and step up from there.

[ptuomov]

What's a good exhaust header primary inside diameter for this twin turbo engine?

Parameters:
3.6 L displacement from 6 cylinders, so 600 cc per cylinder
6800 rpm redline
Stock 408 hp at 0.7 bar peak power boost, target 500 crank hp at 1.1 bar boost
Exhaust valve diameter 42.5mm
Exhaust port diameter at the flange 38mm
Twin turbo K16/24 hybrids with each getting three 240 degree separated pulses
Turbine housing inlet 40mm being fed from three cylinders
Exhaust valve peak lift 10.9mm
Exhaust valve events at 1 mm: EVO 44 degrees BBDC, EVC 4 degrees ATDC
Exhaust camshaft duration: 228 degrees at 1 mm, 218 degrees at 0.05"

38mm to start and step up from there.

That would work.

I was thinking about 36mm id pipe from the head flange transition to the point where the flow is directed towards the turbo in the collector. It's a three to one collector merging at a gentle angle so the primary flow can literally see the turbine wheel tips and the turbine inlet at 40mm isn't gigantic. Should keep the spool early despite the bigger compressor wheel.

[hoffman900]

Yeah, I would have to think about common sizes. 38mm od puts you about there. Based on what Calvin has said you can shrink some if you are careful. I think machined (or cast or 3D printed) spigots do this the best. If it were my car, I’d start there and probably step up around the 10” mark.

[ptuomov]

Yeah, I would have to think about common sizes. 38mm od puts you about there. Based on what Calvin has said you can shrink some if you are careful. I think machined (or cast or 3D printed) spigots do this the best. If it were my car, I’d start there and probably step up around the 10” mark.

After working on the rpm restricted race engine for the last two years, I'm a wholesale convert to the view that exhaust port size is much less important than what the engine needs when picking the primary diameter. There's no problem for the flow to go into a smaller hole with any kind of half-way decent transition, that's reasonably efficient.

[hoffman900]

Yeah, I would have to think about common sizes. 38mm od puts you about there. Based on what Calvin has said you can shrink some if you are careful. I think machined (or cast or 3D printed) spigots do this the best. If it were my car, I’d start there and probably step up around the 10” mark.

After working on the rpm restricted race engine for the last two years, I'm a wholesale convert to the view that exhaust port size is much less important than what the engine needs when picking the primary diameter. There's no problem for the flow to go into a smaller hole with any kind of half-way decent transition, that's reasonably efficient.

For sure.

Looking at the blown diffuser Formula 1 V8’s and how small their collector diameter was, for the rpm / power they were running, it really puts into question for me why people size everything so large.

[ptuomov]

Looking at the blown diffuser Formula 1 V8’s and how small their collector diameter was, for the rpm / power they were running, it really puts into question for me why people size everything so large.

Back in the day before full control of valve events, like with the designs I’m playing with, the factory logic was as follows. First, can’t do big overlap cams for cruise stability and idle emissions. Second, if one can’t use camshaft overlap, then one gets a better fuel economy by making the exhaust large and reducing pumping losses.

The fun part is when you hot rod those engines normally aspirated with ITBs, headers, and big overlap cams you can get almost double the power without making the exhaust ports any larger. But I’m digressing away from the turbo project.

[ptuomov]

I think I am going to buy this exhaust system as the starting point and modify it to my liking:

304A1520-5FFC-4B4A-B74B-518034E12C9C.jpeg

ACDB1AF0-BF2F-4ED7-AD52-1BECA953344D.jpeg

F56F23C7-91AA-466C-99F3-05780305B287.jpeg

EE2B01CA-7C37-4DE0-983D-1008E7F80BD1.jpeg

20D79886-343B-462E-A428-06FEC80A218A.jpeg

[juuhanaa]

Looking at the blown diffuser Formula 1 V8’s and how small their collector diameter was, for the rpm / power they were running, it really puts into question for me why people size everything so large.

Back in the day before full control of valve events, like with the designs I’m playing with, the factory logic was as follows. First, can’t do big overlap cams for cruise stability and idle emissions. Second, if one can’t use camshaft overlap, then one gets a better fuel economy by making the exhaust large and reducing pumping losses.

The fun part is when you hot rod those engines normally aspirated with ITBs, headers, and big overlap cams you can get almost double the power without making the exhaust ports any larger. But I’m digressing away from the turbo project.

How much torq are you talking about without making the exhaust ports any larger? Have you look at other NA Porsches or Hondas for example?

10557778_540079226118511_3352367627024176638_o.jpg

[ptuomov]

How much torq are you talking about without making the exhaust ports any larger? Have you look at other NA Porsches or Hondas for example?

I have in a garage a seven liter 928 S4 V8 race engine (not built by me) that has long-overlap cams, ITBs, and headers. It makes over 640 rw hp at 7000-7200 rpm and 500 rw ft-lb at 5000-6100 rpm. The exhaust port is an untouched stock 928 S4 V8 5.0L exhaust port of the first casting revision head. Not a gram of aluminum removed from the exhaust port and there's a factory stock exhaust valve in there still. In the factory stock form, that engine makes about 270-275 rw hp at 6000 rpm and 270-275 rw ft-lbf at 3000 rpm.

[juuhanaa]

How much torq are you talking about without making the exhaust ports any larger? Have you look at other NA Porsches or Hondas for example?

I have in a garage a seven liter 928 S4 V8 race engine (not built by me) that has long-overlap cams, ITBs, and headers. It makes over 640 rw hp at 7000-7200 rpm and 500 rw ft-lb at 5000-6100 rpm. The exhaust port is an untouched stock 928 S4 V8 5.0L exhaust port of the first casting revision head. Not a gram of aluminum removed from the exhaust port and there's a factory stock exhaust valve in there still. In the factory stock form, that engine makes about 270-275 rw hp at 6000 rpm and 270-275 rw ft-lbf at 3000 rpm.

Your looks impressive. I was just bringing up different ports,

Screenshot_20220907-095311__01.jpg

[ptuomov]

In my opinion, if the factory turbo engine is well designed to start with (like Porsche 993 Turbo in my opinion is), then one can make a lot more than stock power with higher boost setting, higher octane fuel, low restriction turboback exhaust, low restriction ducting feeding the compressor inlet, larger turbocharger(s), and additional (inter/oil/coolant) cooling. I don't think there's much need to open up the engine.

But if one opens up a 1980's or 1990's engine for modifications, the last thing one should do is increase the exhaust port size. Fortunately, with many Porsche turbo heads (including 993 turbo) it's impossible because the factory cast a ceramic liner into the exhaust port. That liner makes power by keeping heat in the exhaust gas and keeping the griding tools away from the exhaust port!

[juuhanaa]

Speaking of NA hemi chamber or 4 valve 4 -bangers, I havent experienced, or seen a case where the exhaust port size has made that big difference, however timing may have been sensitive (Lancia/Fiat TC engine for example).

Friends tell me all the time wedge chamber is a different animal.