Twin Turbo Ideas

[michaelmurray]

I have a few questions and ideas to pass along for those interested...



1. Is it possible to mate a Holset H2E 17cm2 exhaust turbine to a Holset HX65 turbo? I suppose machining the turbine housing may be necessary, but has anyone done this? I am considering using this for a secondary turbo.



2. Can you buy the Holset HY40 VGT turbo aftermarket yet? I would like to use this as a primary turbo. The actuator for the HY40 is a spring/diaphram configuration, and the adjustment comes from varying the air pressure to the diaphram using a servo regulator. The command signal to the servo regulator would come from a custom controller. I would use the throttle position sensor (TPS) and boost sensor as inputs as well as a tuning pot for calibrating the boost-TPS relationship. The controller would use a PID loop to control the acutator's position to maintain the desired boost. The turbocharger's RPM sensor would also be used as an input to prevent over-spinning the charger.



3. I would like to use a liquid-air intercooler between the two stages of compression. I have never seen a twin setup in a Dodge using an intercooler like I have suggested. I would like to mount the intercooler in the stock air filter location. I would use an AFE Mega-Cannon air filter mounted in a custom fiberglass box that is ported to an induction hood. The air box would be mounted in the stock passenger's-side battery location.



Please feel free to share your thoughts...



Michael

 

[drees1]

Intercooling

The benefit of intercooling between stages is to provide denser air to the 2nd stage. That results in better efficiency of compression for the 2nd stage of compression, and possibly wider boost range on the 2nd compressor map. For ultimate boost, you would have to take that step. The current leaders in horsepower development have not figured out how to get or use the full 150psi that a twin set up could be capable of in our engine... Yet!. The current usage of twins on the Cummins B is to resolve spooling problems with a large single, and to get in a better position on efficiency in the 35 - 70 psi boost range. So, unless you are going to take the next step in performance develpoment, and go beyond the current limits, intercooling is probably more effort then it is worth. A super big aftercooler feeding the engine can achieve the same inlet air temp as intercooling and aftercooling combined, so the real gain to intercooling is efficiency at the 2nd stage compressor, which would only be needed if you approached the limits using twins.



This is the beauty of the diesel. The horsepower limit is determined by structural design of the engine, and nothing else. Boost can be as high as you want it, as long as you have the fuel to match it and keep egt in check. But then, thermal barrier coatings on the pistons, and stainless manifolds and a water cooled 1st stage turbine housing would allow much higher EGT safely. The hyperbar engines we made for the military had very low compression, would not start with out a flame heater, had twin and quad turbos, and would scream when they came up on boost.



The fact that little fails structurally on our engines, (we have driveline problems), indicates that there is still much more potential left in the engine, which is exciting.



Doug Rees

 

[nickleinonen]

Holset HY40 VGT turbo



i'll take that please when i [hopefully] get my 04 cummins HO ram, i'd like that much more than the little hy35, or maybe a variable geometry holset hv 50... oh no... don't even own one yet, and am thinking of what to bomb it with. .

 

[JHansen]

The VGT turbo is on my list of future improvements... ... . to my mind it would be the best of all worlds-quick spoolup on the bottom, minimized drive pressure up high, plus you gain that nice water-cooled bearing housing for durability. A few of us have been kicking around the control circuitry as well-I was thinking an RPM signal as well to assist the controller in calculating engine load in concert with the boost and TPS readings. My thoughts weren't to use it in a twin setup though (although I'm sure that would be a sweet setup!)-more along the lines of the HX55 sized monster-that is the description at least to me of the term "Super Single"!



Jason

 

[Devildog]

Single turbo thoughts

A PID loop is fine, but why so many parts? Would the right spring configuration do the same? Wouldn't the exhaust pressure be linear? Start with a 20cm housing diaphrammed down to 9cm And as the exhaust pressure increases the diaphram opens to the final size of 20cm. Anyway that what I've been thinking about for the last couple of months

 

[JHansen]

The only reason I was thinking of such a sophisticated controller was to maximize the performance of the turbo (ie. quickest spoolup possible). For my type of driving spoolup is the key. Figuring that by monitoring as many variables as possible you could keep the turbo operating at its peak efficiency for a given engine state at any time..... hey if you're gonna dream things it may as well be in Technicolor right?



Cool to get a few guys' input on this deal.....



Jason

 

[michaelmurray]

An open loop control system, as Devildog has described, would work. In fact, that is what Cummins currently uses on its wastegated turbos. However, a controller-based system is much more configurable. You can taylor the boost-tps relationship easier (tuning pot) than changing springs, and in some cases you would not want the boost-tps relationship linear. One such case would be sequential turbos. With an open loop system you would get a surge when the second turbo kicks in, but with a closed loop system you can raise the pressure ratio of the primary turbo until the second turbo kicks in. At that point, the primary turbo reduces the pressure ratio so as to maintain a linear boost-tps relationship. This makes driving a twin sequential turbocharged engine much easier in slick conditions. I believe the VGT comes with the pneumatic/spring actuator and servo regulator so only the command signal [from the controller] is required.



To overcome the problems associated with high manifold pressures (80 psi and up) and still maintain drivability at low manifold pressures, I think a variable comression ratio system is necessary. I find the concepts from www.alvar-engine.se very interesting since it does not involve changing the block, pistons, crank, etc. I can see modifying a 12V head to incorporate the pistons, design an upper crank and rods, timing gears, and use a rotary hyraulic actuator to advance or retard the timing (like BMW). I understand this is getting complicated and expensive, but it is fun to think of the possibilities.



Later