Sound suppression for Lysholm compressor with blow-thru throttles 2

[stimpee]

I am considering putting a twin-screw supercharger on an engine which has individual throttle bodies mounted directly to the cylinder head. I plan to attempt to make this work in a blow-thru configuration, so I am not looking for people to dissuade me from that. I have done enough research, and seen enough setups to be confident that it can be made to work, and work quite well.

However I do have concern over the noise output from the compressor. Blow thru setups are typically quite noisy. I don't mind the sound under heavy acceleration, but I don't really want it to be screaming under normal cruise conditions. The removal of the throttle plate in front of the compressor obviously has a negative impact on the noise transmitted when running with the bypass open.

So, anyone have any good pointers to inlet noise suppression on a twin-screw Lysholm type compressor? I have looked a little bit at resonators (tuned), but the tuned resonator might be limited to being effective in a relatively narrow rpm range.

I would be interested in any good ideas, links, pointers, etc!!

Thanks,
Steve

 

[Warpspeed]

Perhaps you could describe the size and general construction of these acoustic dampers. No doubt MB have put quite a bit of time and effort into developing this system.

Some pictures would be great !

Blow through has many advantages, and if you are planning to run a decent sized front mounted air/air intercooler, there is really no alternative to fitting a down stream throttle.

 

[Tmoose]

Warmington wrote - "Yes, if it was a rootes type. This is a screw compressor. Screw compressors have little to NO intake pulsing. Sorry Pat, there is no exitiation frequency to speak of with these SC's


jcd06 wrote that the Mercedes M271 "has 3 noise dampers, first between air filter and charger......"

So, the Mercedes device is designed for broadband type flow noise, rather than anything related to noise at frequencies related to SC rotor lobe/blade pass?

 

[Warpspeed]

Even a roots style blower has almost no pulsing at the rotor inlet. But both types can pulse significantly at the discharge.

There is rarely an induction noise problem if the throttle is placed upstream of the supercharger. But a downstream throttle always requires some type of an air bypass system.

It is when this bypass is open at light throttle and engine idle that pulsing supercharger discharge air can find it's way back around to the main air intake.

 

[jcd06]

So here are the pics finally.

http://users.skynet.be/fb924332/Folder1/P1260053a.jpg

The complete thing, this is all I have...

http://users.skynet.be/fb924332/Folder1/P1260054a.jpg

That's the first damper after the charger.
The plastic insert is held in place with those metal clips.
Not a big deal huh.
The damper upstream of the charger looks more or less the same type as I can see on the
cut-away drawings in the boolet. The second damper downstream of the charger I would rather describe
as half a meter of perforated tube surrounded by a plain one. It's not clear on the drawings if the space between
both tubes contains somekind of a damping material but it's definitely an absorption system.
In the text they call it a broadband damper. No idea what gives it that property,
perhaps different perforation diameter or distance between the holes.

http://users.skynet.be/fb924332/Folder1/P1260056a.jpg

This one is a view without that bypass manifold.
The bypass valve is a DC motor opening the butterfly against a spring.
It has potmeters for positioning feedback; I think it can be called a "servo" then.
Each rotor has three lobes and is covered with a black, rather rough coating.
Not at all the smooth feeling Teflon coating like I saw on another Eatons.
Perhaps someone can tell me more about this kind of coating cos I never saw it before.

In my application noise will not be an issue unless it would be louder than the exhaust.

 

[Warpspeed]

Really interesting Jcd06.

The servo driven air bypass idea is o/k for the vehicle manufacturer, because they can fully map it's operation in the ECU. But for a project car, getting it to work properly would be quite a challenge. An entirely pneumatic system using an adapted turbo wastegate wins hands down for simplicity, and it will work perfectly well, onece a suitably light wastegate spring has been fitted.

Your best bet might be to just install everything on the engine and first get it drivable and properly tuned. That should keep you fully occupied for quite a while.

The induction noise problem is difficult to anticipate. The space available, and pipework layout will then dictate what is possible, or practical, along with some experimentation.

Fortunately changes and additions to the atmospheric side of the induction system should be possible without too much difficulty once the car is drivable.

All the noise (and any vibration) will come either directly from the mechanical blower drive system, or out through the open air filter. Minimal noise should radiate from pipework or intercooler surfaces.

 

[stimpee]

Warpspeed,

The problem with your "approach" of get it running and then deal with it, is depending on the severity of the "problem", and dealing with the problem, may completely dictate how I ultimately decide to lay out the system!

If I can get the blow-thru setup to work, and work well and safely, as well as manage the noise, then I would be inclined to use an air-air intercooler setup on the car. I would do this for both simplicity (less failure points vs water pump, etc), and also since it would (in my opinion anyway) slightly damp the throttle response a bit. The car's throttle response is already pretty amazing, and bordering on abrupt. I fear with a twin screw on top of that, that it would get to be difficult to drive.

However if I KNEW that I would end up punting, and either adding a pre-compressor throttle, or even further, end up removing the ITBs or blocking the original throttles open, then I would probably build an integrated air-water cooled intake manifold like I have in my previous system developed for the earlier BMW 6 cylinder models. The main problem with this approach is that the timing of the delivery of fuel to the cylinders could be hard to deal with, since there could be some lag in the airflow relative to the stock setup, and it could result in a stumble upon throttle application.

Those 2 approaches would be very different from a fundamental standpoint, so it is hard for me to figure out how to proceed without knowing the answers!

 

[Warpspeed]

It is a tough decision that only you can make, and the intended purpose of the car may influence your thinking.

Individual throttles are going to give by far the best throttle response, especially on closing throttle. That is something few people realize. But throttle lag can severely impact gear changing if you have a manual gearbox. With an auto is hardly matters how bad throttle response is.

At least ninety percent of the planning, engineering, and fabrication will be involved in mounting and driving the supercharger, and moving other engine accessories out of the way.

The induction pipework is simple and fairly readily changed in comparison. In fact, until the supercharger is bolted into its final position, it may be rather difficult to plan the exact details of the pipework.

I would get that part of the project out of the way first, there will be enough problems to keep you awake at night just getting that far. Then tackle the pipework as stage two.

 

[stimpee]

Again, the issue arises, since the fundamental position and the orientation of the compressor will be different (or at least in my mind, I would intend it to be different) depending on the approach. If I do an air-water setup, the charge cooler is located internally in the intake plenum, and the compressor attaches directly to the plenum with the discharge to the side similar to my current setup on the older BMWs.

If I go air-air, I would very likely (assuming it will fit) orient the compressor differently, in a manner that is more conducive to piping/ductwork for a front mount intercooler.

If I end up doing the air-air setup, and then later decide on the draw-thru throttle, then throttle response will likely suck.

If I go to the air-water setup with the integrated manifold, that is probably a more straightforward design approach for me, but I fear that the throttle response will be too abrupt.

The good thing with the air-water integrated manifold is, that if I desire to put together a twin-charged setup down the road, it would be "easy" to make a turbo feeding a front mount intercooler and then feed that into the compressor with the second stage of charge cooling after the compressor in the manifold.

Ugh...

 

[Warpspeed]

I feel your pain stimpee, been there done that.

Throttle response with an air/air cooler and the throttle on the blower intake will not just be bad, it will be completely undrivable.

With perhaps eight to twelve litres of total contained air volume after the throttle, close the throttle and it can take several seconds for the engine speed to drop back to steady idle rpm from flat out full boost. I tried this once and was quite shocked by how bad it actually turned out. Forget that idea.

For best performance and drivability use a front mounted air/air intercooler with the individual throttles. It will be a dream to drive, and have no intercooler or supercharger heat soak problems in slow traffic. Especially if you bypass the air back to the blower intake from after the intercooler. But the small throttle transmitted noise problem will require some acoustic ingenuity.

Next best would be throttle on the blower intake and an air/water cooler with absolute minimal dead pipe volume. No nasty noises, but it will not be so responsive. The air water intercooler system will be a real pain to develop, and it will badly heat soak in the engine compartment, and be nowhere nearly as effective at intercooling in typical stop go street driving as you expect.

Either of these systems requires an effective air bypass around the blower.

Twincharging does not really change anything, the air from the turbo just goes in where the original air filter was, regardless of what else there is. Twincharging is a fantastic idea, but keep that for stage three. Solve all your supercharging problems first, just leave sufficient space for the turbo for later.

If I was doing this myself, I would keep the individual throttles, fit the largest front mount air/air cooler that would fit, and fit a second slave throttle body to the blower intake. I know that ain't particularly easy in your case, but it is only one problem to solve.

Doing it any other way is going to have heat soak and response disadvantages, as well as more potential problems.

 

[stimpee]

It does appear to be obvious that the design of the bypass setup will be critical. I also agree that a throttle in front of the blower and then an air-air intercooler setup (with the requisite cooler and piping volume) would never work without keeping the ITBs in place and functional.

I don't necessarily agree that a water-air setup will be as bad as you state, particularly since I have done it before, and achieved very excellent results. However I will concede that in slow stop-n-go it will get hot, and takes a minute or so of driving at normal (30mph+) speeds to cool the liquid back down.

So in your suggestion, is the purpose of the slave throttle body at the blower intake basically for noise (and adding the secondary benefit of another fail-safe if something went wrong with the primary ITBs)? I am not sure how difficult it will be to do that, but given the location I am thinking for the compressor, getting access to the existing mechanical throttle linkage may not be that difficult. The big question is whether the existing linkage and drive setup for the throttle can handle the load of an additional large throttle body. I would imagine that I can get away with a pretty light spring on the throttle body though.

I am partial to the simplicity (functionally) of the air-air setup. Hopefully the piping won't be too much of a bear...

Thanks for your continued feedback!

Steve

 

[Warpspeed]

The problems I had with my own water air setup was that everything in the engine bay heats up to perhaps 85C (185F) including the intercooler core. That rapidly heats the water reservoir, even if that is located in a cool remote location. Even a very large front mounted water/air heat exchanger will take a couple of minutes to cool down all this stored water. My own experience was that this was just not up to the job in mid summer heat, and heavy traffic conditions. All the extra complexity did not justify the rather disappointing results.

With an air/air cooler mounted way out front, well away from engine heat, all the engine bay pipework and induction system will still heat soak, that is unavoidable. But when you plant your foot, the rush of cold air through the hot pipework has a surprisingly low actual temperature increase.

This is where a sudden rush of really cold air is most welcome, with a quick stab of acceleration. If it takes ten seconds for the induction temperature to fall after full heat soak, it is pretty useless.

If the intercooler is properly ducted, the radiator air will first pass through the intercooler, and the radiator fan will cool the intercooler core even when fully stationary and idling in traffic. In my experience air/air works vastly better for a city car than air/water/air.

The idea of having an extra throttle body up stream, is to reduce the open flow area of the air inlet, to reduce the escape of noise. This this will be far more effective than you might imagine. There should not be any mechanical loads on this extra throttle beyond the return spring.

I believe it is now a mandatory design rule in most countries that dual throttle return springs must be fitted. These are usually one inside the other fitted directly to the throttle shaft. Where two throttles are to be fitted, one spring can be removed from each, resulting in about the same throttle pedal effort as original.

This is still a fail safe system, because both throttles are connected in series flow, and only one need close. Strictly speaking there are still two throttle return springs, so it should still comply. But regardless, it is still a safe modification to do.