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!

 

[HoofHarted]

If one was to intercool the air after the turbo before entering the supercharger would a liquid charger cooler after the supercharger suffice.

 

[patprimmer]

What Warpspeed says about response is true, but if you close a throttle after a positive displacement blower, you need a pressure relief valve in the manifold between the blower and the throttle. If this valve sticks closed, pressure will build up against the throttle plates an bend things and can result in a runaway engine. In my opinion not worth therisk.

If the blower has a low boost multiplication rate and you add liquid coolant (alcohol water blend) just before the blower, you can retain good if not perfect response and eliminate engine runaway risk.

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.

 

[Warpspeed]

If you use a turbo wastegate for the supercharger air bypass, any pressure spike will force the poppet valve off its seat (against the internal spring), and act as a fail safe pressure relief valve, up stream of the throttle.

This will still work, even if one of the two pressure control hoses to the wastegate diaphragm fall off.

All production cars must be fitted with dual throttle return springs these days, and that should go a long way to avoiding a stuck open throttle.

A stuck open throttle can be a frightening experience in any car. But on a really high powered forced induction car it can be absolutely terrifying.

 

[Scirockett]

yes. it's wise to have a kill switch (coil or fuel) within reach so you don't chance losing steering if she's bouncing off over redline. anyone ever have their gas pedal catch on the floormat?

A throttle body in front of a twin screw type charger is not such a bad thing. These chargers are quite loud, even at an idle. this type of setup also requires plumbing to be rigid because everything downstream of the charger will now also be under vacuum. referencing my setup, throttle response is immediate at any rpm, in any gear. it's great because there's always immediate boost available. this is the exact opposite of my other car, who's turbo doesn't spool until 4k, but will blow 20psi all day.

I'm going to take the TSI's side of the table for arguement's sake. please tear me apart so we can find it's shortcomings.

I think control of the SC clutch and bypass valve are easier than we may thing. in this scenerio, the TB would have to stay on the manifold because it can't be upstream of the turbo, so it may be a little noisey (maybe baffle the intake?). any standalone management system could handle it and even manage duty cycle if you wanted partial openings. I simply see the SC bypass being closed unless turbo is making boost, then it would need to open fully. can't see when it would want to be partially open, you would lose boost from SC. (off beat, supposedely these clutched eatons can freewheel? ). for oem mangement, you'll likely be restricted to electro or mechanical vacuum switches.. hmm.

now for the bypass, a throttle body with TPS could provide position feedback to a servo of some type. or, a drive by wire TB could be rigged to work. anyone have any other thoughts for other off the shelf valves or even check valves that would work?

knowing the sc makes X psi and the turbo makes XX psi, management would disengage the SC 'circuit' (pulley and bypass) when boost levels reach desired turbo levels. without management, this could be done with a pressure switch.

you'd also have the ability to turn the SC off, and you'd drive no problem. great for being quiet when you don't need it. in cabin SC on/off switch.. sweet..

back to the thread starter, if clutched lysholm style chargers were available, if that's even possible because it does take alot to spin it, my opinion is it would be superior to the eaton. but the clutched m45 or m62 off the MB is very attractive for this application.

great reading this thread, nice to meet everyone, and I'm curious to see what we all build!

 

[Warpspeed]

o/k so let's assume your supercharger is making 15 psi of boost, and you decide to suddenly de-clutch the supercharger "to reduce losses" once the turbo winds up.

What do you think will happen to the engine induction air pressure, if the supercharger that is contributing significant (probably most of) the flow and pressure to the engine, is suddenly switched off?

Why not completely bypass the turbo exhaust turbine as well "to reduce losses" even further ?

But seriously, once the supercharger is creating positive boost pressure, there is NO WAY you can just switch it off. It would feel like disconnecting half your spark plugs.

The only time you can smoothly clutch, or de-clutch a supercharger is when it is already completely bypassed, and producing zero boost pressure. That would only be the case at engine idle and very light throttle during constant speed highway operation.

If it is already totally unloaded, why de-clutch it ?

A throttle body in front of any positive displacement supercharger just turns the supercharger into a giant vacuum pump. If you think large vacuum pumps consume zero drive power, you would be mistaken. Throttling the air into a supercharger does not unload it, far from it.

It will run hot and noisy throttled at the intake. My own experience trying this, is it will cost around ten percent in wasted fuel creating all that unwanted heat and noise.

Every production supercharged engine I am aware of uses an air bypass around the supercharger that opens on closed at light throttle.

What you need to do is relieve the supercharger of all back pressure. That will unload it, unless it is a screw blower with internal compression. Unfortunately there is no obvious way to completely unload a screw blower. But a roots blower can be completely unloaded. Imagine just the rotors spinning around in open air with no outer casing, to get the general idea. This ability to easily and completely unload a roots blower without de-clutching is a considerable advantage for a road car, and should not be dismissed lightly.

Forget about unloading the supercharger under boost. It will work about the same as completely unloading a turbo by fully opening the wastegate when up to full turbo boost.

And another potential problem that may be less than obvious. If you de-clutch a positive displacement supercharger when it has positive boost pressure at the discharge, it will run backwards at a million explosive rpm, venting all the up stream boost backwards through the supercharger. That not only makes a very nasty noise, it can break things.

 

[BrianPetersen]

Supercharger first, turbo second arrangement being discussed here.

If the supercharger is making boost from idle speed, the turbo will not be. As the engine accelerates and the turbo spools up, the turbo compressor starts accelerating and developing a pressure ratio. This automatically draws down the discharge pressure of the supercharger (Not to the point of being a vacuum, just to the point of the pressure ratio becoming lower.) Eventually it will draw down the supercharger outlet pressure to atmospheric. Under load, that's when the supercharger can be bypassed (and de-clutched at your option).

The tricky bit is what to do at part load and idle. The supercharger must be bypassed in these conditions.

With OEM fully-mapped engine controls it's not a problem to duty-cycle or modulate the bypass valve depending on RPM and requested engine load.

My comment about supercharged applications normally having the throttle before the supercharger is not based on OEM, but rather based on the traditional V8 with a supercharger sticking through the hood and a set of carbs on top ... The other way I've seen it has the throttle after the supercharger, but the supercharger is centrifugal (non positive displacement); lots of kits for Mustangs etc are like this. I don't know if they do any bypassing, but I doubt if efficiency was a prime consideration in either case.

 

[Warpspeed]

Yes indeed. The typical hot rodder and drag racer does not fit a roots blower to his engine to increase fuel mileage (smile).

So to him, the only matter of any importance is the standing quarter mile acceleration time at wide open throttle. What it does at part throttle down the highway is of absolutely no interest whatsoever.

But practical every day street engines are a whole different ball game, especially now gasoline is becoming expensive, and soon perhaps scarce too. Throttle response, drivability and economy make the whole game more difficult.

While there are many weird and wonderful configurations possible, some are going to be fairly difficult or maybe impossible to get the bugs out of.

 

[Scirockett]

so it looks like switching off the SC is a delicate task. so yes, any pressure between the SC and turbo must be bled off before disengaging to 1) prevent it from spinning backwards and 2) enable a smooth disengagement. this could be done by opening the SC bypass before the SC is disengaged once the turbo has spooled. this is very tricky as we don't want to combine boost that would result in a boost spike. with my Wintec II, I could start opening the SC bypass when boost levels exceed the turbo's 15psi (like 16psi), then immediately after atmophere is met between the two, the SC could be safely disengaged without notice. RPM would be around 4k.

I think we need to make an assumption that stand alone management is required for this. it's just too tricky to do this with vaccuum switching..

As for number's i'm thinking, expecting no more than 8-10 psi from the charger, and the sky is the limit for this turbo (running 15psi street).

at low RPM/idle or when boost is not needed, a diverter valve would open by vac. in the intake creating a loop for the SC's flow. and this would be in addition to (if it's even necessary) to a BOV between the turbo and TB. SC doesn't necessarily need to be bypassed at idle or low RPM, it just needs a way to blowoff/divert pressure.

in regards to bypassing the turbo, I would expect it to spool a little faster/earlier with 8psi coming into it. thoughts?

back to TB location, no matter where the TB is, you're not going to unload a screw type compression (I think you confirmed my suspesion). No matter where the TB is (pre or post charger), you still need a diverter or boost return. It's not a vaccuum pump even though it may appear to be with an upstream TB closed. it may be under vacuum, but it's still moving air around unrestricted in a loop and not drawing any more power than if the TB was downstream.

definately not an easy thing to do, but not impossible as VW has it currently in production. but definately alot better than pushing a turbo THRU an eaton as most VW tuners are doing today.

 

[Warpspeed]

All the problems have been solved, some very nice twin charging systems have now been built all over the world.
Yet people still try to complicate things unnecessarily, and why that is so, baffles me.

Aftermarket engine management will definitely be required, because the fuel, and especially the ignition timing requirements will differ hugely from what came fitted to the original normally aspirated engine.

Here is how to do it.

1/ Leave the throttle on the plenum in the original position !!!

2/ Mount the roots type supercharger, and the drive system to the engine. The car will still be drivable while you do this. It is a big job.

3/ Fit an air bypass system directly around the supercharger using an external turbo wategate. Connect up the wastegate actuator to work from the differential pressure drop across the throttle body.

4/ Connect the supercharger discharge to the throttle body via the largest air to air intercooler that will fit in the available space in front of the radiator.

5/ You now have a very nice, really tractable supercharged engine with excellent drivability and fuel economy. It will be very well behaved, but will probably lack extreme top end power. Nevertheless it is an excellent start.

6/ To twincharge, just install a turbo to that engine in the usual way. Turbo compressor discharge goes straight into the supercharger intake. Exhaust wastegate sensing pressure comes from total combined boost after the supercharger.

This is a PROVEN system.
It is SIMPLE as it can possibly be.
And it works extremely well, as many people that have built it this way all agree.

If you want to try doing it some other completely different way, I wish you luck.

 

[Scirockett]

Holy defensive.. the system you just described is common, and proven, however I'm trying to discuss alternatives to it. On the table here is basically replicating the VW TSI design, which is in production, so spare me the over-complication nonsense.

first of all, TB positioning is getting confused here. for the TSI replication of a twin charged design, the TB stays on the plenum downstream of the chargers. For a twinscrew ALONE, I'm not aware of any losses associated with a TB positioned upstream of the charger. The charger is not under any more continuous load than with the TB downstream. there is still a recirc. this is extremely common with lysholms due to their noise. I'm doing it too and have no temperature or efficiency issues.

for the record, I already have a Lysholm twin screw blowing a 2l 16v motor. as you stated, drivibility and fuel economy is excellent. I'm not making any changes to this car. It's running OEM digi-1 boost aware management with a custom chip, so aftermarket management is not necessarily required for boosted motors; however, it's I/O's will help with the twincharged specific issues of the TSI design.

my other '87 is the same 2l 16v motor, but with a big turbo. it's an all or nothing kind of setup spooling at 4k. 230whp in 2200lbs. this is the setup that needs a supercharger for low end which is lacks now.

Also have a supercharged cabriolet, a small turbo type I scirocco, and completed countless other boosted swaps. not looking for you to teach me how to build a setup we are not even talking about.

now as far as being constructive, one of your comments caught my attention.

If it is already totally unloaded, why de-clutch it ?

agreeing that the charger must be unloaded to de-clutch it, this brings up another issue. when to re-engage it. also agreeing our target is street driving, right? All of my cars are daily drivers and I've never been to the track. so this would mean re-engaging it between shifts if and only if the next gear RPM will be lower than the turbo spool RPM.. now here's a tricky area!!

complicated, yes.. impossible, no. I know we could build the system you described with a turbo blowing a eaton (didn't I mention I didn't want to do that in the last sentence of the previous post? ? ), but I found this thread researching the TSI and thought it looked open for discussion. believe the first thing I said was I'm taking the TSI side of the table... maybe you need to take a closer look at the TSI setup..

 

[Scirockett]

one more quicky -
why disengage the charger under light load/low RPM? I would want it spinning and ready to boost. it would be recirculating because of the diverter. once the TB is open for accelleration to the point where there is no vac, the diverter/bov closes, the charger is already spinning, and you get immediate boost. you're just closing a diverter to get boost.

 

[HoofHarted]

I'd have say that this thread has been very fruitful.

But the way I've decided to do it is way Warpspeed suggests, Sc then Turbo with two wastegates controlling the boost. And Possibly Using Two Intercoolers, the one might be a charger cooler.

I'll use the Sc first and how it behaves then i'll add the turbo.

 

[Scirockett]

no agruement that setup will work and has been done before, but it suffers from the same fundimental problems where all intake air passes thru both chargers. this means compressing and/or displacing air twice throughout the entire RPM band. needless to say, heatons are best for their mid-range performance, yet it must run past its efficiency range in this setup. Again, this is an option, and you'll make gobs of HP thoughout your nice TQ curve.. TC FTW.

the point of the TSI engine is to allow both compressors to work within their efficiency range, and for one charger to not effect the other charger, almost like it's not even there, eliminating almost all paracitic and flow related losses. This is a breakthru in twin charging and I believe VW is the only one doing now. they get 168bhp from a 1.4l motor! from the factory! that alot of power from a little package, efficiency at it's best.

Just looking for a healthy discussion on understanding the concepts and workings behind the latest, production automotive twincharged setups made by VW or whomever.

 

[Warpspeed]

Yes all the intake air passes through both chargers in series, all of the time. And trying to get around that somehow, will not gain you much. But it will surely increase complexity, and introduce some other rather interesting little problems that will likely keep you occupied.

Two stage compression does not mean the compressed air will be twice as hot as many people sometimes expect. It is compressed twice, but each stage only provides roughly half the total compression. The net effect is not going to vary by much, no matter how this compression actually takes place.

By far the more important issue is providing sufficient intercooling AFTER all this violence has been inflicted upon the air, (by whatever means).

If intercooler capacity is made generous, the efficiency of compression becomes hardly significant at all. If you can reduce induction temperatures sufficiently, the engine will be very grateful. And it will happily show it's gratitude by providing a little extra supercharger drive torque, and tolerating a slightly higher turbine intake pressure.

So to a great extent, a less than wonderful supercharger and turbo combination can be offset by fitting the worlds largest intercooler. At the end of the day, what comes out of the flywheel is what matters.

Hoof, use one monster intercooler last thing after the supercharger. Cooling the air before the supercharger is not of any real practical advantage. This is especially true if it means the final intercooler must be made much smaller, which is often the case.

The supercharger can handle hot air just fine, it will be fairly dense hot air, but that is no problem. Just alter the drive ratio to get the required pressure increase. A few extra supercharger rpm is no big deal. Place all the charge cooling last, and pull out as much heat as possible there.

As for clutching and de-clutching, that too is something of very doubtful value. The manufacturers do it for two reasons. The Toyota blowers are rubbish, and wear out extremely fast if continuously driven. They also run very hot if driven continuously. Toyota clutch their blowers simply to make them last.

I am told Mercedes clutch their blowers for noise vibration and harshness reasons only. The clutches are a big problem, they operate under high stress, and figuring out an acceptable clutch control strategy to do this is far from simple.

Just use a solid well designed supercharger (most are) and keep it spinning. It simplifies things greatly, and there is no disadvantage except perhaps some slight continuous mechanical noise from the drive, especially if a toothed belt is used. Use a silent multirib belt if at all possible.

 

[Bribyk]

If you intercool between compression stages your max ideal power savings are on the order of ~0.5 hp (assuming a 2.0 L engine with 90% vol eff at 7000 rpm and one bar of total boost). You will get better heat transfer with only one intercooler after both stages because your temperature differential will be greater so you should easily make that 1/2 hp back with a cooler intake charge (as long as the supercharger can handle the hotter air).

 

[Warpspeed]

The supercharger will usually be mounted right behind the radiator, and often right up close to the exhaust manifold. The whole engine bay will be as hot as hell, and the supercharger will rapidly heat soak to perhaps 85C or more.

What air comes out of the turbo will also be pretty well "roasted" even under non boost conditions. Just feed the turbo straight into the supercharger intake.

It makes for a nice neat compact installation, with minimum pipe volumes, and don't worry about high interstage air temperature, there is nothing you can really do about it anyway.

 

[patprimmer]

You still have the real danger that if the blow of valve sticks closed, you have a runaway engine if it then bends the throttle shafts and sticks them part opened.

A closed throttle downstream of a positive displacement pump builds pressure until something relieves the pressure. If the BOV sticks, something else gives.

If it were mine, I would fit a burst panel as well as the BOV so the burst panel broke well before the throttle shafts bent. A single valve safety device really is jst not good enough when the consequences are considered.

Two BOVs are also not all that good re safety as if one fails the other operates with no real indication to the driver. It can stay that way undetected for years. A burst plate is very obvious when it bursts. The burst plate should be upstream of the throttle for obvious reasons

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.

 

[Bribyk]

Or just leave the IC piping clamps loose . A good point though, Pat.

 

[Warpspeed]

An external turbo wastegate is pretty fail safe. The spring holds it closed. There is nothing there that can obviously fail and jam one shut. The wastegate is also working with with clean air, not red hot exhaust gas. The bypass is also in fairly constant motion (with all throttle movement), which should discourage sticking anyway.

A bursting plate is not a bad idea. Drag racers have been known to use light gauge aluminium cut from a drink can to work as a bursting plate on big gasoline and nitro filled roots blowers. Blower explosions can be pretty expensive.

For the truly paranoid, a pair of wastegates working in parallel would offer another solution.

My own experience with extreme pressure spikes, is that either a hose blows, or the supercharger drive belt first slips and then shreds.

 

[patprimmer]

I am only paranoid when the consequence of not being so can be fatal.
I agree belt or duct failure are likely before throttle shaft, but how well you secure a pipe or how much overcapacity drive has is a fairly inconsistent method.

I have seen blowers with direct crank drives and gear drives. It would take a lot for that system to fail first. Just because belt drives are by far the most common now, I would not automatically presume it in the absence of alternative information. I know it's a stretch in this case, but others read what we post here.

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.