Dyno room exhaust setup question 2

[thrashercharged]

I'm not sure if this is the right forum for this, but I'm setting up an exhaust system for an engine dyno room and have questions. Here's what I have:

1. Room purpose - dynoing 4-cycle gas car engines, anything from 4 cyl to V8s, up to 1000HP or so, both blown and NA. No sustained durability tests, just quick ramp to redline pulls and some brief steady state at various rpms/loads for calibrating the engine.
2. We have residential neighbors within oh, 100 yards
3. The dyno room walls are poured 8" concrete
4. Room has conventional 10' drywall ceiling (hanging on resilant channel to avoid transmitting sound directly to trusses)with asphalt shingles outside.
5. We have (2) Nelson 300 dyno silencers (mufflers) 12" opening, side in, end out, about 30" in diameter, 112" tall rated 25-35 dB of silencing depending on frequency.

http://westernfilterco.com/critical_300.html

Why 2 and why these particular mufflers? We found them on eBay and they were a good deal.

Our main goal is to operate the dyno as quietly as possible to keep the neighbors happy. Preferably, we'd rather that they not even know we're running a engine. I'll point the chimneys straight up.

My plan is to keep the mufflers inside to keep even more of the noise inside the concrete walls, stand the muffler on end and run a chimney through the ceiling/roof. The mufflers themselves should get plenty warm, but I think the room fans circulating the air 10x/min should be able to suck the heat out of the room sufficiently. How warm they'll get I don't know - anyone have any input on this?

Questions:

1. Since I have 2 of these mufflers, I might as well use both. Should I use them independently and run 2 chimneys? That is, give each bank of the engine it's own muffler?

What are the pros & cons of this setup? Will the muffler be too big for just a single bank (4 cyl at most) and not provide enough silencing? I ask this because of a statement from another dyno muffler's website:

"If the silencer is much too large, then the exhaust noise simply passes through using only the initial large expansion for attenuation."
- quote from

http://www.eiwilliams.com/steel/index.php?p=EngineSilencers

My understanding is that if the muffler is too big for the volume of exhaust input, the baffling inside isn't very effective and the muffler essentially becomes a large unbaffled chamber - is this correct?

2. Or should I put the 2 mufflers in series (I can easily do this by standing one up and laying one on its side and running its end output into the side input of the other.)

Will this provide more silencing? I realize this may increase the backpressure, but these mufflers are so large already, I'm thinking the increase in backpressure will be pretty small, but I'm no expert on this (which is why I'm asking these questions here!)

3. Question about exhaust temps. At the header, temps will be in the 1400-1600 *F range, but the muffler is a huge heat sink, so what are realistic temps at the chimney? How much lower?

4. Outlet sizing: My understanding is that the outlet should be smaller than the inlet for good sound attenuation. Both openings on my mufflers are 12". This probably already is overkill for the engines I'll be typically running, but just in case I get a blown big block or diesel I guess I'll be ready. I assume with the exhaust gases cooling, they'll have less volume and a smaller outlet pipe/chimney is ok and won't increase backpressure? Ideally, what size should my outlet & chimney be?

5. Another reason for wanting an 8" chimney is cost. A 12" double wall chimney stack is pretty expensive and custom while 8" double wall chimneys made for wood stoves and such are readily available. Would a double wall chimney rated for wood stoves and furnaces be ok to use?

After I get through the roof, the exhaust gases should have cooled even more, could I even adapt down to 6" for more attenuation or will this create too much backpressure? The reason for this is I'll need a raincap, and 6" is about the largest that are readily available, anything larger is custom again.

Thanks for any help. I think I've given all the info I can, but ask if I've left anything out.

 

[thrashercharged]

Whitevette, can you (or anyone else) elaborate on your post: "Two mufflers? In series, they will work well together. In parallel, you'll have half the volume of gases (at the same noise level) going thru each one...so you'll end up with two noisy mufflers! Nothing much accomplished!"

Why does half the volume make the mufflers noisy? Does this imply that when I run a 4 cyl into my exhaust system (assume I use 1 muffler) it'll be noiser than a V8 at double the displacement?

I was also concerned about exhaust leaks and heat, but I'm assuming my ventilation fans will be able to cycle enough air to eliminate those concerns. I've worked in commercial dyno cells that had mufflers inside, but they fed into some elaborate ventilation system that I believe had additional fans helping to draw exhaust out.

This brings up turbocohen's response about using fans and no mufflers, as the fans will attenuate noise via dilution and breaking up standing waves and pulsations. I see how that can work, but I wonder if the dB of noise reduction is enough for my purposes of keeping the neighbors happy. I can see it being enough for a lab or pure industial environment.

You state: "They run GM LS2 and other stuff at full load for extended intervals with one of these fans with a single stage large dia attenuator inline."

Is this single state large diameter attenuator essentially a muffler like I have? Are you suggesting combining a fan with this muffler? Is the fan before or after the muffler?

 

[Warpspeed]

Thrasher, this is sure becoming a many faceted discussion.

Those commercially designed mufflers look excellent because they have maximum attenuation located around 100Hz which is in the general region of the very strong fundamental firing frequency of many stationary engines. They should work extremely well on a dyno, and as you already have a pair, I would connect them in series.

Temperature rise will have a lot to do with dyno running time. A stationary diesel running near full load for may hours on end will be a rather different proposition to short bursts on a typical engine tuning dyno. Diluting the exhaust with plenty of fresh air by using a suitably powerful exhaust extraction blower will make a huge difference to muffler temperatures. The system will cool down rapidly between dyno bursts.

Expect an exaust flow from the engine of around 2.2 CFM per horsepower. From that you can size the blower to flow sufficient extra air against the back pressure of the total exaust system. In theory, this blower can be placed anywhere along the exhaust system. More on that later.

Rubber tyres can stand a fair bit of heat, and short of actually catching fire, should work fine. Even if the tyres do deteriorate over time, they are never going to see the road again, so it probably does not matter. They should also be fairly resistant to acidic condensation in the exhaust.

Most standard rectangular metal airconditioning ductwork installed in commercial buildings has an acoustic lining. This is usually fibreglass wool packed behind a perforated metal liner. The whole air duct is constructed in the manner of a large rectangular fibreglass absorbtion muffler. It may be possible to salvage some straight lengths with bolted and flanged ends from a building demolition site. A box of beer to the site foreman should do it! If fresh entering air into the dyno room, and dyno room exhaust air go through suitably long sections of lined ductwork, it will be very effective at noise attenuation outside the building.

Direct higher frequency mechanical noise from the tyres and the general machinery will be less trouble to attenuate than the powerful low frequency fundamental exhaust noise. The only other expected problem might be rumble transmitted through the ground. Rubber mounting of the dyno frame might be worth a thought.

The only "quiet room" technology that I am familiar with are radio and TV studios. These have exactly the opposite problem of keeping traffic noise out. One TV station I worked at for many years (ABV2 in Melbourne) has a suburban rail line going right past one wall of a TV studio, just over the fence. A tain goes past every few minutes during the day. The whole studio is a building within another building, and the inner building is isolated on rubber mounts.

Thanks for the advice on the sound proof door construction, that is extremely helpful. I will go back to that TV studio and take a much closer look at how they built the larger doors. They have dual massive barn sized sliding sound proof doors at the rear for moving studio sets and scenery, but I cannot now remember how they were constructed.

I agree that the most effective low cost sound absorbent treatment for walls and ceilings would be ordinary fibreglass wool batts held in place with fine steel mesh (chicken wire). Sound studios are built exactly like that, but the ugly fibreglass is hidden behind very expensive looking spaced slim polished wooden battens.

I know exactly what you mean about these quiet rooms being absolutely silent and it being difficult to hear people talk. It is a really strange experience. There is something really odd about total silence that most people find rather disturbing.

Now we come back to exhaust systems, and the theory about back pressure and reduced exit area.

Individual exhaust pulses can have steeply rising wave fronts that contain a lot of very harsh objectionable harmonic energy. Getting rid of the worst of the higher frequency harmonics with an absorbtion muffler system is not terribly difficult. But it is the violent low frequency pulsing that is much more difficult if not impossible to eliminate, without adding back pressure.

Question ? How do you turn a violently pulsing gas flow into a perfectly steady flow ? That is what you must really try to do to acceptably reduce the low frequency sound amplitude.

The only practical way is to allow your pulses to expand into a sufficiently large open volume, and then restrict the outlet area in such a way that the outlet is then forced into almost steady flow.

Restricting the outlet can be either a simple step reduction of pipe size, or it can be an active variable flow control device, rather like a throttle butterfly or weighted flap. The design problem is that engine exhaust volume is going to vary over a wide range, and ideally so must the outlet restriction adjust to compensate. It is really the pressure drop at this point that holds back the noise while it beats itself to death inside the expansion chamber.

Some Ferrari and Lamborghini models have exhaust butterflies so their cars sound civilized at small throttle openings. This is not a new idea.

As to flow control device flutter. It can CREATE massive noise all by itself, by modulating the escape of the higher pressure exhaust gas from the expansion chamber. If the flow control device can build up to a self resonant flutter, the noise it can generate is horrific !! So the respone of any variable area flow control device needs to be slow and heavily damped to prevent any flutter from developing.

Another better approach is to keep the flow through the whole exhaust system artificially high with an air blower. Then design the outlet restrictor to suit the much higher blower volume. The blower will more than make up for the deliberate pressure drop that you are going to require in order to achieve a sufficient sound attenuation.

In an earlier post, Turbocohen suggested that injecting fresh air into the exhast stream with the aid a blower dilutes the noise, and that is a rather neat way to express it. It would potentially lower exhaust gas temperature too, and that would offer a few advantages, particularly in mid summer heat in your dyno room.

My own future plans are to install a powerful centrifugal high pressure air blower (0.5 psi?) feeding exhaust plus fresh air into a long baffled underground chamber. This chamber will be filled with old tyres stuffed with fibreglass batts. It will have a tall and deliberately restrictive exhaust stack at the exit. The blower will be run fast enough to create a significant pressure rise within the underground chamber, to allow a suitablly high pressure drop across the exit stack. The higher the design pressure drop, the greater the total sound attenuation will be.

Obviously the exhaust from the vehicle feeding the blower intake will see none of this deliberate back pressure.

As a matter of interest required blower power, (assuming a 100% efficient blower) will be roughly psi x CFM divided by 300. Typical blower efficiencies might be in the region of 30% to 50%, so you will require two to three times the drive Hp calculated from the formula.

So as a design example 0.5psi at 2,000 Cfm works out to 3.3Hp, but a practical drive motor may end up being in the 7Hp to 10Hp range. Not exactly small.

 

[thrashercharged]

Warpspeed & others - thanks for the detailed responses, I've learned a lot from this discussion. Thanks for taking the time to write such a detailed explantion. I understood the importance of attenuating low frequencies and converting into a steady flow, but did not realize the importance of restricting the outlet area to accomplish this.

It sounds like ideally I should add a large blower to the system and dilute the exhaust with fresh air. I assume it would be better to put the blower after the muffler so it's not subjected to such intense heat, but does it matter to the blower if it's sucking through the muffler restriction or blowing through it? A blower is more efficient one way vs the other isn't it? Which side do you intend on placing your blower?

With a blower added to the system, instead of directly coupling my 3" or 4" flex pipe from the exhaust headers and tightly sealing them to either the muffler or blower opening, I'd probably have a large, say 12" opening and just stick my flex pipes into this opening and let a lot of ambient room air get drawn in with it. Wouldn't this setup allow a lot of exhaust noise to spill into the room? Or with a sufficient blower, and if I stick my flexpipe deep enough into the 12" hole, no noise would escape back into the room? Is this how you intend to make your room exhaust ports?

Regarding the quiet room I've worked in, it was a major OEM facility (one of the US big 3). The vehicle was run on a chassis roll - we weren't interested in logging power, we were simply using them to simulate road load and using the room to isolate an engine noise. I took a close look at the sound insulation out of interest. I believe that NVH engineers typically would use the room for their work. The entry doors were a major, expensive commercial door where the panels, when open, didn't slide on a track from vertical to horizontal overhead like a typical residential garage door would. These panels stayed vertical and stacked one in front of the other at the top of the track(s) - each panel was on it's own track. I think the the thing for you to do from a cost standpoint would be to have a rolling accoustic panel of insulation wool formed from chicken wire in the design I described previously to place in front of your door and effectively make it just another wall. This was done in this room as well.

 

[Warpspeed]

Wherever I have seen exhaust extraction blowers used with dynos, the exhaust just pokes into the vicinity of immediate blower intake funnel. The general idea being that the open tuned exhaust collector end still sees free open atmospheric pressure. But your idea of feeding the exhaust in slightly downstream of a fresh air entry duct to reduce noise, sounds like a fairly worthwhile thing to try. It will also help to ventilate the general dyno space.

Another possibility would be two quite separate blowers feeding the main pressurised muffler system. One just raises the exhaust pressure to the desired muffler pressure, and the other does the same thing with fresh air. The exhaust connection to the engine could then probably be completely sealed, and the exhaust blower run from a VFD to maintain atmospheric pressure measured at the engine exhaust exit point. Many VFDs have an internal PID controller, and all it would need would be a MAP sensor hooked up to the VFD. The fresh air blower could be arranged to have some additional muffling in series to prevent noise from escaping from the fresh air intake. Lots of possibilities here, and all relatively easy to try.

As far as a large sound proof entry door goes, something like a heavy fire door, or pair of doors on rails that can be hermeticaly sealed around the edges. My first thought would be a big hollow rubber gasket (tube) that could be inflated with compressed air, after the door had been firmly clamped to the wall somehow. And then as you suggest, a sound absorption lining on the inside. Sheet lead should make a fairly effective sound barrier without being too bulky.

I would like to stress that I have not yet put any of this into practice yet, but am still at the planning, thinking, research stage.

Fortunately I already have a Bruel and Kjaer sound level meter and an audio spectrum analyser, and for my calibrated sound source (hehehe) a Briggs and Stratton lawnmower engine either muffled or unmuffled. I can kick up a fair racket while testing things without the neighbours suspecting what I am really doing. That is the grand plan...

I believe my neighbours will not complain provided the sound level is kept low enough, and test sessions are kept short and at appropriate times of the day. If all the problems are tackled one at a time in a systematic way, It should be possible to construct a relatively quiet suburban dyno.

 

[turbocohen]

Thrasher, to answer your earlier question, the extractor fan inlet is supplied from about 60 feet of 4" rigid pipe collecting gas from what is presumed to be parrallel GM truck mufflers. The dyno engine dumps into a common collector with a cone leading to the muffs with flex pipe. Nothing sophisticated, muffs presumably hacked from a salvaged test mule, quiet enough.

 

[thrashercharged]

Did we lose some of the most recents posts on this thread? I could swear a member named cone something responded saying he works at Maxim Silencers and would be glad to answer muffler questions - his post and at least one more after that are gone?

Warpspeed said:
"Wherever I have seen exhaust extraction blowers used with dynos, the exhaust just pokes into the vicinity of immediate blower intake funnel. The general idea being that the open tuned exhaust collector end still sees free open atmospheric pressure."

I see how this would be best for accurate dyno measurements as the exhaust output isn't under negative pressure. But how does this affect noise levels? Does any noise bleed into the room significantly? My idea of feeding the exhaust in slightly downstream of a fresh air entry duct to reduce noise is entirely for noise control, but I can see how it would skew the dyno results by helping to suck the exhaust out.

Turbocohen - my understanding from your description is that the 2 sides of the engine's exhaust headers dump into a common collector, and this collector has 2 parallel outputs leading to 2 production pickup truck mufflers via flex pipe, and a blower is present after these 2 mufflers. Since this setup is using 2 production truck mufflers, they would have fairly significant backpressure wouldn't they? Is any ambient air being mixed with this exhaust, being sucked in by the blower? If so, isn't the flow volume much greater than what these mufflers were designed for? Also, wouldn't the main reason this setup is fairly quiet be due to the use of these production mufflers? I would guess it'd be no louder than a production pickup truck, but the backpressure would be almost as high as well, even with the blower?

Note I'm not questioning or doubting your observations, just questioning if my understanding of sound attenuation and backpressure is correct, and trying to understand why I need large expensive dyno mufflers if production vehicle mufflers coupled with a fan are indeed sufficient.

 

[Warpspeed]

The exhaust blower I mentioned earlier in that film clip, was being used on a normaly aspirated Formula One engine, and both the engine and blower were completely enclosed within an engine test cell. As you quite rightly point out, for correct exhaust functioning, the tuned collectors must feed into a large open volume at atmospheric pressure. In fact the excess blower intake air passing over the exhaust and intakes might even simulate a fairly high road speed for such an engine.

The dyno test cell itself would have been designed to contain most of the noise. None of that is really applicable to a chassis dyno, or what we are trying to achieve here. Except for the basic concept of diluting the exhaust with fresh air, and using a blower to offset deliberate silencer back pressure.

I am still fairly certain that if you were to use a very large stationary diesel muffler on a much smaller capacity engine, there may be insufficient flow volume for the muffler to develop its full rated acoustic attenuation, particularly at the lower frequencies.

In about the sixth post here, Greg Locock mentioned using one of those monster dyno mufflers, and the requirement to adjust the back pressure to the correct figure with an exhaust butterfly. That chimes very well with my own testing and exerience.

A couple of posts in this thread have been deleted. A new poster popped up and made it very clear he was representing a commercial muffler company. A cardinal sin.

 

[patprimmer]

If you read the rules of the forum, you will see that it is a requirement that participants be engineers discussing work related subjects.

Also salesmen and direct commercial promotion are specifically banned. Maybe cone 1 crossed that boundary.

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.

 

[turbocohen]

Thrasher, The air drawn into the siamesed mufflers draws a lot more raw air than the engine can can supply exhaust. The exhaust goesoutta is loosely attached to the muffler goesinta and the blower does not reduce the pressure at the exhaust outlet to make any meaningful difference. After a run the blower is shut off since it wastes indoor air.

 

[franzh]

I just returned from a day in the dyno cell, and I can easily say that several times during the day that the walls and floors were throbbing (there were about 4 engines running in adjacent cells). Inlet and exhaust noise still withstanding, a lot of noise was transferred mechanically. None of these engines were over 3 liter, two were diesel, two were IC propane. When some of the engines synchronized RPM, the relative noise felt like it doubled.
The engines were isolated with thick rubber pads and the steel leveling plates were 6" thick, 10 feet long, 4 feet wide, but still, a lot of noise was transferred.

Franz

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.

 

[GregLocock]

We were refurbishing one of the engine dynos and somebody suggested it might be a good idea to make it a 'quiet' room. We don't actually have a engine noise test cell, so this was to be the next best thing.

By treating the project as a normal NVH project, and applying the usual philosophies, the engineer who designed it succeeded. He spaced semi porous tiles about 3 feet away from the walls, to create 1/4 wave absorbers. The engine is mounted on the cradle via normal engine mounts, and some care is taken to suspend the bits that vibrate, properly.

The guys in the labs say it is a much nicer cell to work in than the others, even when they aren't doing noise work.



Cheers

Greg Locock

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

 

[Warpspeed]

That sounds very encouraging Greg.

One concept to keep in mind.

Reflection is excellent for keeping noise OUT of a structure. For example, to keep traffic noise out of your home, smooth hard external surfaces and double glazing would effectively reflect most of the sound back into free space.

Preventing sound from escaping from within a noisy dyno cell requires internal absorbtion. Every time a sound echoes off a wall a little bit escapes through the wall. The greater the number of reverberations, the more sound escapes through the wall. Standing waves and resonances can build up in amplitude to the point where significant energy will pass through almost any practical structure. An internal sound absorbtion lining will rapidly dissipate that acoustic energy, and give the mass of the wall a fighting chance to attenuate what is left.