Plus sized SSIs

[jpnovak]

I have considered these for my 3.0 hotrod build. Larger might be a good thing. I calculated 40.5mm would be ideal for my airflow and peak torque rpm values.

[Jonny042]

Originally Posted by jpnovak

I have considered these for my 3.0 hotrod build. Larger might be a good thing. I calculated 40.5mm would be ideal for my airflow and peak torque rpm values.

Curious if that would be inner or outer diameter? Also IIRC correctly you have pretty torquey cams?

[jpnovak]

Yes. I have basically about a 964 exhaust side cam. I wanted area under the whole curve for my build.

My calculations are based on 5800 rpm peak torque. That is about right based on fuel delivery curves and seat of pants. Also consistent with using these on a small engine vs 3.6 and not having the same intake resonance and restrictions of a typical 3.6 installation.

I calculate 43mm with the same cam and your 3.4 displacement.

I do admit that sometimes these "numbers" are far from perfect. Lots of other considerations of cam timing, overlap, etc that can shift the requirements. Plus there is no consensus for ideal EGT and Exhaust gas velocity. Basically, Pick the harmonic you want, size them and weld them. If they are near equal length they will be better than any stock system.

[Jonny042]

Thanks, Jamie! So these calcs would be ID, then?

I figured the 41mm (OD) are a little on the small side but it's probably better than being too large. I chose the cams (DC43-106) to provide good mid to high range torque, 4 degrees more lobe separation than the project car. Just to push the peak power down a bit for Auto-X and drivability.

[jpnovak]

Yes. I am calculating ID.

In general, The tubing ID will increase with larger engine capacity and higher rpm volumetric airflow driven by higher power requirements. The tubing lengths are adjusted based on the resonance peak rpm requirements and matching harmonics. I am modeling a half wave initial primary and a quarter wave stepped secondary.

I really wish I had a dyno I could regularly use so I could test and refine the calculations.

The challenge is that you may calculate for a peak number but that's not what you want because it is too large for normal driving where you are not at peak torque. These are not all track cars that spend time in the rpm range that matches the exhaust build.

Many years ago (15?) I did an EFI 3.2SS with SSI and ITBs. I used a DC40-108. Loved the power band. I am considering swapping my cams in the current engine back to this cam but my intake/exhaust ports are not large enough to support. Right now, my engine is very well matched all the way through.

I am sure your 3.4 will end up the same. The difference in a 43mm and 41.5mm ID is only 400 rpm in a calculated peak torque. The advantage of going smaller is that the torque curve levels off and flattens as it limits airflow. This is way better than going too large and not having any velocity at low rpm to provide torque. The engine ends up being peaky instead of drivable. Fine for a race car. Not as much fun on the street.

Of course, YMMV.

[Jonny042]

If anything I'm worried that the Carrera heads with the 41mm (more like 42mm in actuality) ports are a little bigger than the motor really calls for - but sometimes you need to work with what you have. The 46mm PMOs have a 42mm manifold outlet so in all it should be a reasonably balanced system.

[jpnovak]

You will be fine with 3.2 heads. No one ever complained about that engine not having torque across the rev range. Just have comfort knowing that they will flow enough when it needs to breath.

Only thing I would do is to smooth the casting around the valve guide boss and then some coatings.

[Bill Verburg]

Originally Posted by jpnovak

Yes. I am calculating ID.

In general, The tubing ID will increase with larger engine capacity and higher rpm volumetric airflow driven by higher power requirements. The tubing lengths are adjusted based on the resonance peak rpm requirements and matching harmonics. I am modeling a half wave initial primary and a quarter wave stepped secondary.

I really wish I had a dyno I could regularly use so I could test and refine the calculations.

The challenge is that you may calculate for a peak number but that's not what you want because it is too large for normal driving where you are not at peak torque. These are not all track cars that spend time in the rpm range that matches the exhaust build.

Many years ago (15?) I did an EFI 3.2SS with SSI and ITBs. I used a DC40-108. Loved the power band. I am considering swapping my cams in the current engine back to this cam but my intake/exhaust ports are not large enough to support. Right now, my engine is very well matched all the way through.

I am sure your 3.4 will end up the same. The difference in a 43mm and 41.5mm ID is only 400 rpm in a calculated peak torque. The advantage of going smaller is that the torque curve levels off and flattens as it limits airflow. This is way better than going too large and not having any velocity at low rpm to provide torque. The engine ends up being peaky instead of drivable. Fine for a race car. Not as much fun on the street.

Of course, YMMV.

While length and width are factors the more critical one is muffled or non

The effect of the desired reflected acoustic signal depends on the transitions from one cross sectional area to another, By far the largest such transition, which generates the strongest reflection, is the the one to atmosphere, which is totally muted by a muffler but would be enhanced by a long open expanding taper from the collector. A reverse cone added to the end of the taper smears the signal a bit and broadens the torque curve. The valve overlap the more effective the signal will be.

for an SC w/ stock smog cams there is little to no TDC overlap
an RSR w/ open header profits greatly,

Here's an RSR the pulses are shown side by side due to pulse duration


an SC not so much

here's an SC/Carrera w/ inline pulse because little to no tdc overlap and little to no exhaust gas pulse duration overlap



The gases leave the exhaust port at ~120m/s but slow to ~100mps @ the collector and then into the 40-50mps range or lower depending on the design in a muffler

[Peter M]

Thanks for the update Jonny.

It's good that we have another option for our 3.4's if we want heat exchangers and 1 5/8" primaries.

The price of the 42mm Eisenmann's scared me off too along with having to cobble up the hot air ducting.

Consequently I went with the generic 911 headers from BBE with 1 5/8" primaries specified (FPOR - 0401) and also allows for the use of the generic 2 in muffler. These were good as the standard 3.2 hot air duct was easy to modify to make it work without making it look Frankenstein!





I look forward to seeing how you rig up your hot air ducting.

[Jonny042]

Those look like a great solution - especially if you want to keep that crossover duct for the air! I had already backdated the heat ducting to eliminate the top mounted blower motor and add back the second air duct, but I also wanted to ditch the AC compressor.

Originally Posted by Peter M

Thanks for the update Jonny.

It's good that we have another option for our 3.4's if we want heat exchangers and 1 5/8" primaries.

The price of the 42mm Eisenmann's scared me off too along with having to cobble up the hot air ducting.

Consequently I went with the generic 911 headers from BBE with 1 5/8" primaries specified (FPOR - 0401) and also allows for the use of the generic 2 in muffler. These were good as the standard 3.2 hot air duct was easy to modify to make it work without making it look Frankenstein!





I look forward to seeing how you rig up your hot air ducting.

[Redliner911]

Does anyone know if these new 41mm SSIs will work on an '84-'89 ?

[cmpski]

https://www.youtube.com/watch?v=r-Mb5DsSUv8

The engine will have RSR sized MFI. The S cams and custom MFI fuel pump space cam will probably seem more like an E in this 3.2SS engine.