Sound suppression for Lysholm compressor with blow-thru throttles 2

Positive displacement supercharger + closed downstream valve = ? ? ?

+ Large Bypass = no problem.

Its been done, hundreds of times, and even on a few "production" cars or aftermarket SC kits.

Do you have any answers for my question?

No problem until a bypass valve sticks, but you asked not to be dissuaded, so I will simply ask where you intend using these so I can be somewhere else.

I would expect the bypass valve to make a lot of noise on closed throttle at high rpm.

I would expect a throttle upstream of the blower to be quiet when closed at high rpm.

I will not help in nor encourage you to do something dangerous and stupid.

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.

Secondary bypass/blowoff or pop-off valve can be used in case the primary bypass sticks. I don't care how loud it is when I close the throttle at high rpm.

I care how loud it is under normal cruise conditions.

I think i recall in past threads where your feelings against doing this is quite apparent. I recall your concern about a bypass valve failure. What other concerns do you have (if you can provide them concisely)?

It has been done on a production engine by Lancia, aftermarket supercharger setups on VWs and Miata's. I don't recall reading about any deaths or global armageddon resulting from it, so I am curious.

The engine the setup is going on has ITBs from the factory. I am quite reluctant to remove them. I COULD just remove the throttle plates and connect the DBW actuator to a large single throttle body, but I am still not certain that would be the way to go. If I do this, I have significant problems with engine management, since the timing of fuel delivery upon throttle tip-in is based on the rapid response of the ITB. Even though a draw thru TS mounted on a compact manifold with a water-air charge cooler internal to the manifold has excellent throttle response, the lag, measured in ms, is probably enough to cause a brief rich condition and slight bog upon throttle application.

I can deal with hardware issues a lot easier than I can deal with something as fundamental as the timing of the injector firing in the software code.

Anyway, interested in your opinion, particularly if you have any ideas of how to quiet the compressor, or address the concerns that you have...

My two cents....

There will be 3 major sources of noise. You didn't describe your application much, but with individual throttle bodies I assume it is a motorcycle.
1. supercharger drive - Obviously a ribbed belt would be better than a toothed belt. An enclosed gear drive MIGHT be better but that depends on how stiff the enclosure is, the gear tooth profile, and backlash.
2. air intake - Provided you have the room you can treat intake noise with various types of silencers. (ex. helmholtz, 1/4 wave, expansion chanbers) If you don't have the room you are out of luck! Also make sure that your intake tract isn't vibrating and causing noise.
3. supercharger case (and other structures secured to it) - This will be tough to manage as absorbing/dampenging materials will act as insulation and might cause the SC to overheat. The best luck here might be a cover.

ISZ

Application is BMW inline 6 cylinder.

Will use a ribbed belt for drive, as i have in past setups with non ITB engines of the same type.

The intake tract can be relatively long, and I can use non-metallics in it as well, probably in the inlet housing to the compressor as well, to help provide some acoustic damping. I also think I will have room for a decent sized airbox, and possibly a resonator/expansion chamber. Hopefully that will do it.

The compressor itself does not seem like it is a horrible source of noise thru the case in the current setups.

Stimpee,

Why do the blow through setups become more noisy than the suck through throttle types? I assume that either a) the pressure in front of the throttle plate is greater causing more turbulence at the edge and therefore noise or b) the bypass valve has to be used as a means of part throttle boost contol which creates noise that is an inherent feature of the bypass?

I am not sure I completely understand why a blow through would necesarily be MORE noisy, can you expand?

Happy new year!

During cruise with a draw thru system, the partially closed throttle plate must block a lot of the inlet noise of the compressor. Couple that with the fact that the compressor is not doing much "work" due to low density of the inlet air, and the result is that you barely hear the compressor at idle and cruise conditions.

I have seen several blow-thru setups and they are much much louder at idle and cruise. Presumably this is due to the lack of the throttle plate blocking the air (and noise), as well as perhaps due to the fact that the compressor is passing more "mass" thru it due to the inlet being at ambient pressure. However I can't imagine that this is a huge contributor, since the outlet of the compressor in this case is also at ambient due to an open bypass...

Not sure, but empirical evidence shows that blow thru twin screw setups are much louder under normal cruise (partially or significantly closed throttle) conditions. At WOT I would imagine that there is little difference.

OK, i see. So basically the extra noise is due to the fact that the SC is un-throttled at intake meaning that it is passing ambient pressure air and doing alot of work, plus the fact the the throttle plate itself would act as suppressor.

The lack of air pasing through the SC will be quite a major contributor as if there is no air then the drive system of the SC will be running almost unloaded except for the system inertia. The drive will not make alot of noise until it is under load (compressing air).

Surely any pulse or wave tuning cannot apply before the SC, i.e. Helmholtz, 1/4 wave etc as we are not dealing with any engine exited noise only the linear noise of the SC. No pulse waves will pass the SC and back out causing noise. Maybe try the open end of the system first as this is where you have the SC and BMW have probably made quite a good job of the other bit.

One way to cut down this noise is to increase the level of dampening or insulation in the inlet tract before the SC. Maybe a box-inside-a-box chamber or muffler between the SC and air filter would be the easiest solution without sacrificing any response? Also if the noise is eminating from the SC housing itself, then a cover or dampening mounts may help.

You can use conventional intake tuning (ie expansion chambers, quinkies, and helmholtz) to try and reduce the noise from the s/c. The main excitation will be at number of lobes*s/c rpm.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

Greg's answer while absolutely correct to the extent that it goes, does not address the noise generated downstream of the blower and being emitted through the manifold and air duct walls. Heavy walls and curved surfaces will help reduce that noise as will wrapping the manifold and air duct with sound insulating materials. These might create a secondary problem by also insulating the heat in.

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.

"Surely any pulse or wave tuning cannot apply before the SC, i.e. Helmholtz, 1/4 wave etc as we are not dealing with any engine exited noise only the linear noise of the SC. No pulse waves will pass the SC and back out causing noise."

Unless someone changed the laws of physics all "sound" or "noise" is made up of pressure waves in air. ISZ

Pulse waves are generated by the rotors as they take a gulp of air then close of then the next one takes another gulp.

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.

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.

Even the rootes type with its three twisted rotors does not have the same problem anymore. As the lobes are twisted, only a very small section of the intake between the lobes is ever opened up at any one time. A correspondingly small exhaust gap will be beginning to open on the other side. This does not cause huge pressure fluctuations and therefore inlet tract pulses. Also the timing of these small pulses have been timed to allow cancelling. The more lobes, the less pulsing.

What I am trying to say is that the flow is quite smooth from the lysholm SC. The noise generated from the flow of air through the SC will rise in frequency as a function of engine speed and will also be too high to exite any large resonance in the inlet tract materials anyway. This reduces the usefulness of the suggested methods of inlet tract tuning.

Assuming that the inlet tract is well constructed between the SC and throttles (being BMW it will be), and that the noise is from the SC itself, the part I would attack first would be the mouth of the SC to the airfilter as this is likely to be where the noise is coming from. From some SAE papers I have read, the only way to STOP the noise that a twin-screw makes off-load is to stop the air passing through it. It has a compressive element that slightly loads the SC at all times.

I think that proper management of the bypass valve could almost cancel the noise if it is big enough. Or what about a clutch system? It depends on the application really. I assume its a road car hence why you are concerned with noise, probably an ///M engine? Why not just have the SC set to come in at say 50% throttle?

Other than that, as suggested myabe insulation on the SC body and some kind of broadband muffler on the intake will help.

Greg

While screw compressors may not have the same "pulsing" as a roots, the internal compression in the unit (pressure ratio built-in of approx 1.4) results in a case where air is being compressed, and the "re-expands" when the bypass valve is open. This expansion results in a "popping" sound as it is often characterised, and the screw compressor is typically regarded as being noisier than the newer style Eaton type roots blower with the 3 lobed twisted rotors.

I would still imagine that the number of lobes (male and female rotors have different numbers, 3 by 5 for the lysholm compressor, and 4 by 6 for the Opcon) times the operating rpm of the compressor rotors will provide the appropriate frequency range to avoid. In my case I will want to mitigate noise in the 2000rpm-3000 ( or perhaps 3500) rpm range. Outside of that range I am basically either lugging the engine, idling, or revving it out for power in which case I don't really care about noise under power.

The bypass will be left open probably up to 40%-60% throttle opening depending on how I tune it, with it closing progressively at the appropriate vacuum level to start generating boost.

It is a road car, and is the S54 engine in the 2001-2006 E46 M3.

Insulation on the SC body will probably be effectively accomplished by the general packaging of my setup, and muffling or resonation chambers in the inlet piping, which will probably also be composite and high damping from a material perspective, are all on the table.

Steve

Steve,

Interesting, I hadnt considered the expansion scenario. Maybe then a tuned-pipe solution immediately at the SC exhaust to cover the transitionary period would work in this case?

Also, the correct placement of the intercoler will go some way to providing a plenum style voulme which will dampen the popping noise you speak of. The closer the intercooler is placed to the SC output, the better the response.

Greg

On the previous setup I designed for the S52 engine (no individual throttle bodies), I had the compressor mounted directly on the plenum (cast aluminum), and a water-air charge cooler core mounted inside the plenum with some volume after the cores.

Compressor noise in that setup is barely noticable at idle and cruise. It is certainly noticable at WOT, but not unbearable.

That setup was much simpler though, due to the standard draw thru setup. I used a cast aluminum inlet housing to the compressor, steel inlet piping (3" or 3.5") and the throttle in front of the inlet about halfway down the piping. There were a few silicone couplers, and then a hot film MAF with a cone type open filter CAI just past that.

Steve

"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."

The most powerful loudspeaker I ever worked with used a sliding plate to chop a stream of compressed air.

That chopping action is the direct cause of almost all intake noise in cars.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.