Twincharging with Screw-Type SC instead of Roots

[jbond]

Twincharging an engine that features a screw-type supercharger (ie it already has internal compression) in my mind has the following advantages and disadvantages:

Disadvantages:
It would still feature parasitic loss when the supercharger is "bypassed" and the turbo is powering away at higher rpm

Advantage:
The efficiency being better than a roots blower, would mean lower temps and more power

Can one of the gurus post their thoughts? PatPrimmer and Warpspeed had a lot to say on previous TwinCharge topics.

In short: what is the suitability of screw-type SC for twincharging project?

PS: I really have the twincharge bug! help!

 

[BrianPetersen]

You've probably got it ... the screw type would give (theoretically) an efficiency loss during cruise, but a minor power improvement during the short period of acceleration until the turbo takes over.

The twin-charging application that I'm aware of (VW) only uses the supercharging for the first moment of acceleration in which the turbo is running "off design". As soon as the engine builds enough revs for the turbo to take over, it does. This leads me to suspect that there is little purpose in using a higher efficiency supercharger, because it is really only used for maybe a second at a time. Also, those engines are intercooled, so whatever minor extra heat comes from using a supercharger that is less efficient than it could be, gets taken away by the intercooler anyway.

At least in that application, the need to minimize parasitic losses during cruise and acceleration when the turbo is active takes priority.

'course, not every application is the same, I'm speaking only of that one, where VW's intention was to make a downsized and more-efficient gasoline engine; it's best not to throw away a few points of efficiency driving an inactive supercharger.

 

[SonyAD]

Also, I'm thinking the lower heat capacity of the oil would make for more of a temperature differential when it hits the radiator, improving cooling of the oil there. Would help any bit? Also, engine oil could go higher in temp with no risk of it boiling, even in an unpressurized system.

Anyone done this or know of this having been done?

 

[j79guy]

Have you considered electric clutching the supercharger? Mercedes, on the original SLK230 had an electric clutch on their 2.3l four cylinder, Eaton M45 supercharged, which in my mind could potentially open many options for a twin charged application. Mazda, with the twin turbo RX7 of 1993 through 1995, had a very complex system of making the turbos sequential. I'm not saying it was especially good, but when in a good state of tune, that system gave a nice flat torque curve, coming on low, which the little rotary really needed. If designed properly, twin charging should be able to eliminate the parasitic loss of the supercharger, when no longer needed, or overtaken by the turbocharger.

j79 guy

 

[patprimmer]

This has all been done before. At this stage I have nothing further to add other than a small Roots at low boost vs a small screw at low boost does not impact all that much on total efficiency compared to the alterntive ways to get the same power and response.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[jbond]

Thanks for the responses guys, the question is redundant now. I was contemplating buying the mazda Miller cycle engine 2.3L that comes with a lysholm type supercharger, then just adding the turbo - hence the question about whether or not it'll be a problem using the lysholm!

I'm now leaning towards the subaru EJ20T as used in the STi as a candidate. By choosing an engine that's already fed force induction, then compression etc is already sorted.

I was also looking at the SR20DET.

In my location, there are displacement rules that coincide with vehicle weight. It works out that a 2L turbocharged engine can be put in a 800kg car, provided the engine will meet the emissions requirements of the car. So if I choose a mazda mx-5 (miata) between '89 and '97 then it weighs 940kg and all should be good. Although I'm not sure if the boxer will fit between the miata chassis rails...

 

[j79guy]

jbond: As you know, the EJ20T has terrible turbo lag. Once on the boost, they are stellar, but waiting for the turbo to spool can be irritating. An Eaton 45 supercharger, or equivalent would snap up the throttle response quite nicely. However, I doubt you will find a satisfactory fitment to the MX5 chassis. I've contemplated this swap as well, and was leaning to the Oldmobile Quad-4, mid 1990's vintage. The new Ecotech 2.0 engine, which is an evolutionary growth of the Quad-4, would potentially be even better. The new Ecotech turbo engines will start showing up in the wreckers shortly.

j79 guy.

 

[jbond]

j79 guy, thanks for the tip. I'll look into the Ecotech turbo engines too.

I quite like the novelty of the boxer, the look, the sound, etc Plus I think they're already designed for around 18psi and 8k rpm which is plenty in a sub 1000kg car

 

[izzmus]

The Subaru engine looks just right for an upside down Eaton supercharger, and it would not interfere with factory style turbo location whatsoever. It looks like it would complement it, actually. (As such, I am acquiring the bits and pieces to take on this project!)

I realize VW does things their way, but I thought the main advantage to twincharging involved the pressure differential/heat characteristics when you ran the turbo and supercharger in series rather than in parallel sequence. Bypassing the blower would seem to revert it to a standard, if not now highly complex, turbo engine.

 

[patprimmer]

In series keeps a system that i complex by definition as simple as possible, or is that as least extremely complex as possible.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[jbond]

izzmus, yes great idea, to upside-down mount the eaton below the intake manifold. Is there enough clearance room there?
Additionally, I had always hoped of fitting an A/W intercooler after the eaton, that would take at least an additional inch of space for a highly compact solution. Perhaps spacers where the intake manifolds meet the heads (if they meet orthogonally)

twincharged AWD sounds like the ducks nuts as you wouldn't be spinning the wheels like rwd may do if you have too much torque off idle

 

[izzmus]

A custom intake manifold is assumed. Fortunately, Subaru saw fit to make the turbo engines (for the North American market at least) with three piece intake manifolds, with the fuel injectors and rails attaching to the lower stubs. A custom intake manifold is not therefore going to be nearly as time consuming to fabricate as it could be.

Believe it or not, my plans are for front drive. As it will be entirely on an experimental, beer-money type basis, I don't want to have to deal with the reputation for drivetrain breakage that Subarus have earned. Not yet, at least.

 

[jbond]

FWD in what body? An imprezza or liberty, or something entirely different?

Here's some pics:

Alternator, Power Steering Pump, AirCon Compressor - which one are you going to remove? Or, will you accomodate them with a bracket to re-mount it a little to the side?

 

[izzmus]

This is getting off-topic, so this will be my last post on the subject (a valid e-mail address will not be hard to find if you wish to go off-board) but the models here are a tad different, and all one needs to do is look into "reversed intake manifolds" (positions throttle to face forward) to see the options for relocating accessories.

Trying to steer back on topic, the screw type superchargers require less power for a given boost level, so the drop in parasitic loss alone would be advantageous.

 

[TDIMeister]

The twincharged VW engine has an electromagnetic clutch that completely disengages the supercharger when it's not needed, so the associated parasitic losses of the bypassed supercharger is not there. In fact, I don't know of any contemporary OEM supercharged application that doesn't have a clutched disengagement during part load.

Care should be taken about discussing efficiencies. While it is well known that screw compressors have internal compression, the general blanket conclusion is that this is more efficient than a positive displacement blower. This is true at increasing pressure ratios, but in fact, at low PRs the difference approaches insignificance, and a Roots blower can get better efficiencies than a screw compressor because of less aggregate losses (the work of a compressor covers more than simply the process work).

Generally, if the PRs of the compressor stage is less than about 1.5, the differences in efficiencies are pretty small, and cost becomes the overriding factor. However, the VW TSI engine has the blower delivering a PR of 2.5 barely above idle at 1250 RPM, and IMO a screw compressor would be better suited for this, but again, cost was the overriding factor.

In an SI engine, a large reduction in compressor work and charge temperatures can be achieved by injecting fuel at the compressor inlet, as Lotus has done with the Exige 265E and 270 Tri-fuel prototypes.

 

[TDIMeister]

After re-reading a paper I have on the TSI, I must amend my comment on the supercharger PR. At 1250 RPM full-load, the total PR is 2.5, but the contribution of the supercharger itself never exceeds a PR of about 1.75. At this operating point, the total PR is achieved by a series connection of the supercharger and turbocharger, the respective PRs being 1.75 and 1.43. The supercharger PR is on a downward slope, and the turbo PR is going upward, until the turbocharger takes over completely beyond about 2500 RPM to redline, again at full-load.

 

[carnage1]

Although I'm not sure if the boxer will fit between the miata chassis rails...

its a bit off topic but you can make anything fit
I once saw a crosly wagon with a 426 hemi.

http://classiccars.com/Uploads/Preview/34926.jpg

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[HoofHarted]

The Thing Is I Have Both A Mercedes Supercharger (Screw Type) Off A C180K And A Mitsubishi Turbo. It Was Not My Intention To Use Both On The Same Car (1700cc Motor)

But The Supercharger Does Not Have An Electro-Magnetic Clutch Which Would Be Ideal. Short Of Designing One Or Adapting Another I Need A Plan.

Plus How Would One Design The Plumbing. Would Both The Chargers Meet At The Intercooler.

If It's A Ball-Ache I Rather Not Have The Labour Pains

Has Anybody Done Something Like This Before?

 

[patprimmer]

The plumbing layout is easy.

Air into turbo via filter and bell mouth.

From turbo to intercooler if available.

From inter cooler to throttle body.

From throttle body to the belt driven supercharger inlet.

From the belt driven supercharger to the intercooler if available.

The use of intercoolers and where thy are is optional but desirable. One should suffice unless you are generating a lot of boost at both stages.

Typically best results are obtained with moderate boost at the belt driven but a lot of boost from the turbo, so the critical intercooling area is after the turbo, but this can either be directly after or the belt driven. I think there is a slight advantage to feed cool air into the belt driven, but there is also some intercooler efficiency gain by feeding hot air into the intercooler, but I am guessing re this aspect. One intercooler after the belt driven is the simplest and most thermally efficient so long as the belt driven can run efficiently and durably at the inlet air temperatures it would get directly from the turbo outlet.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[HoofHarted]

Yes That Is A Method That Could Work.

There Are Three Ways To Go About This I Suppose.

Turbo Before (Feeding) Super
Parallel
Super Before (Feeding) Turbo

I Was Thinking More Along The Lines Of Going Parallel With The Super Working Until Say 2000Rpm Then Disengaging The Clutch Of The Super By Then The Turbo Should Have Overcome It's Lag Stage.

But The Super Doesn't Have A Clutch.

Also Doesn't Feeding A Turbo Or SuperCharger With Compressed Air Further Increases It's Compression? Just A Thought. Sort Of A Compound Turbo-Charging Situation. If So I Would Have To Decrease My Compression Ratio To A 5.8 :1

This Is Exactly Why People Don't Often Do This. The Road Less Travelled.