John Sayers' Recording Studio Design Forum Archive

So I've been researching HVAC design more and like this forum, the more you research, the more questions you have.

Here are some thoughts and concerns that I'm hoping some mega pro designers can answer!

Turning vanes or smooth 90's in rectangular duct work. Specifically in silencer boxes. Check this picture: The "C" value is pretty insane without vanes. Here's a quote from Rod Gervais regarding vanes:

Just make certain to turn a few 90's along the way - and to use turning vanes in corners so you do not have a problem with sound buffeting in the process.

I realize that silencer boxes weren't really a thing when Rod published version 1 of his book (which I have) as they weren't mentioned in it. However, the total pressure of the HVAC system must remain low and every HVAC book I've read says to never use a hard 90 degree as they introduce so much extra in.wg

I've only ever seen one silencer box on this forum where the builder used curves around the baffle ends. In order to lower the total pressure of the system, it would make a lot of sense to curve it where possible.

Furthermore, I've read countless times that without vanes or smooth corners, a lot of low frequency noise is introduced into the system. This seems to be the opposite of what we want for our silencer boxes!

Now, I do realize that sound impinging the duct liner at normal incidence will be damped much more than at a shallow angle of incidence, so in this case, no curves or vanes would in fact improve the silencers ability to "silence".

When calculating the in.wg for a system, what "C" value should we use for these sharp 90 degree elbows in the silencer boxes? Also, what correction factor should we use for the duct liner? These are two things I've researched and cannot find a clear answer for.

Thanks! My head is ready to explode haha

Greg

So I decided to spend the $14 and buy an app called:

HVAC ASHRAE DFDB

It seems to do what I need it to in that I can put in any ductwork and it will tell me all of the info about it.

I will still need opinions and hopefully definitive answers regarding whether or not the silencers should have radius type heels.

Also, this app doesn't seem to have any place to enter duct liner so I'll need an answer on that too if possible please!

If anyone needs any values for your fittings and doesn't want to spend the $14 on the app, please feel free to PM me and I'll gladly punch in the numbers for you!

Thanks as usual folks!

Greg

I’ve been doing similar research regarding the design of silencer boxes and their effects on pressure loss. Let me jump right to the conclusion and start with the short version of my reply…

It’s very difficult to calculate how much pressure loss is caused by silencer boxes, however, as long as the volume inside your silencer box is large enough to keep air velocity low, then problems with excessive pressure loss are unlikely.

Now the more in-depth reply…

I’m not a HVAC expert. I present the following information in the hope that it will help you, but if I’ve come to any incorrect conclusions then I also hope that someone with more knowledge will correct me.

As you may have discovered, it’s basically impossible to find any information about the effects on pressure loss of a duct making a hard 180 degree turn, which is what we usually see inside silencer boxes. This is because sharp 180 degree turns are a major no-no as far as HVAC specialists are concerned. No one uses them so no-one publishes information about them.

For this reason, I imagine that the HVAC app you mentioned does not calculate 180 degree turns either.

The basic method I use for calculating the effects of duct bends on pressure loss is to use the “equivalent duct length” method. This method is something I first learned about in this forum. It allows you to convert the pressure loss of a bend into the same pressure loss caused by an equivalent amount of straight duct. The following webpage features calculations that describe this well:

https://neutrium.net/fluid_flow/pressur ... th-method/

As you will see from the chart on the above mentioned page, 45 degree bends cause less pressure loss than 90 degree bends, and smooth bends cause less pressure loss than sharp bends. This meets with common sense expectations. However there are many other factors that we can’t really account for accurately. For example:

1. The effect of duct liner - the surface of which may cause more pressure loss than a smooth metal duct.
2. How much the baffles inside the silencer box overlap each other. Some overlap very deeply, causing a definite 180 degree turn, whereas in other designs the baffles barely overlap at all, giving the air a more relaxed zig-zag path through the box.
3. The fact that, once again, 180 degree turns are not catered for.

All this means that a certain amount of “best guess” is going to factor in to any equations we do.

According to this above mentioned source, the equation for converting a sharp 90 degree bend into an equivalent length of straight duct is:

Duct Diameter x 60

For example, the pressure loss caused by a sharp 90 degree turn in a 250mm (10 inch) duct is equivalent to 15 metres of straight duct.

You can then use the following calculator from Engineering Toolbox to calculate the friction loss of your equivalent amount of straight duct:
https://www.engineeringtoolbox.com/duct ... d_444.html

Here’s an example:

Imagine a silencer box with 4 internal baffles. Let’s make the passage inside of the box to be 400mm x 200mm in cross sectional area (16 inch x 8 inch) all the way through.

The area of the passage is 0.08 square metres, which is equal to a round duct of 320mm diameter.

If we treat each baffle as a 90 degree turn (I know, it’s 180 degrees, but just go with me for now) then we have an equivalent straight duct of 4 x (60 x 0.32m) = 76.8 metres.

Now let’s push 150 litres of air per second through it. That’s enough air to provide 6 changes per hour to a sizeable room of 6m x 5m x 3m (90 cubic metres).

According to the Engineering Toolbox’s equation, air would be moving through the box at just 1.87 m/s and would cause just 11.6 pascals of pressure loss. This is not going to cause any significant noise, and it's not enough pressure loss to cause any issues for a 150mm centrifugal fan pumping 150 litres per second, such as this one:
https://www.fanco.com.au/product/vkm-ce ... fan-150mm/
You could have 4 silencer boxes in use, resulting in a total of 46.4 pa (11.6 x 4) of pressure loss. Still not a problem for this fan. Even if we then doubled the expected pressure loss to make up for any other factors that we can't account for, 92.8 pa of pressure loss will still only cause about 15-20% reduction in fan performance.

The smaller you make the passageway through your silencer box, the faster the air flows and the higher the pressure loss. It can climb very high if you’re not careful. But with the right design it can be kept so low that it’s almost inconsequential, regardless of whether you round out the corners inside the box or not.

Excellent post! Makes it very clear. Very useful information, and method.

Just one thing I would add: A highlight from your first link:
I'd take that with a grain of salt, though: It would probably be higher.


I'm making this into a Sticky.

- Stuart -

However there are many other factors that we can’t really account for accurately. For example:

1. The effect of duct liner - the surface of which may cause more pressure loss than a smooth metal duct.

In my calculator, I use 0.003 for duct liner Absolute Roughness e, ft

For this reason, I imagine that the HVAC app you mentioned does not calculate 180 degree turns either.

It actually does!

I will still need opinions and hopefully definitive answers regarding whether or not the silencers should have radius type heels.

After some experimentation, I've determined that radius type heels do not help.

If I get some time I'll post my homemade total pressure template (done in Google Sheets) and a description how I calculate it. It's definitely a chore!

Thanks for making this a sticky Stuart. I know it has been a long grueling road for me, so if this post will help anyone figure it out faster than I did (am), hooray!

Greg

Soundman2020 wrote:Excellent post! Makes it very clear. Very useful information, and method.

Just one thing I would add: A highlight from your first link:

HVAC-180°-bend-equivalent-loss.jpg

I'd take that with a grain of salt, though: It would probably be higher.


I'm making this into a Sticky.

- Stuart -

Thanks Stuart. It's funny - from memory I referred to that chart because it was one of the few that incorporated 180 degree bends, but then I decided to ignore the figure for 180 degree bends because it was LESS than the one provided for 90 degree bends (presumably because the 180 degree bend is rounded, whereas the 90 degree bend is mitered). I guess we can take away one useful thing from this (assuming that the chart is correct...hmmm) which is that rounded 180 degree duct bends cause less pressure loss than mitered 90 degree bends.

Okay here is my homemade, maybe crappy static pressure calculator. I will admit that I've thrown this together with as much information as I was able to gather. There is so much confusing and contradicting information out there. I asked several friends who own their own HVAC contracting companies for help on this topic and no one could help. So, again, I must say that this calculator might not be the best or be correct but I hope that it allows us to get in the ballpark for purchasing correct fans/blowers as well as to see where our designs need to be modified.

I welcome suggestions to improve the calculator accuracy as I'm offering it here to help people in their studio designs!

Gregwor's Static Pressure Calculator Version 1.00

Greg

Hopefully this helps some people with designing basic single path silencer boxes. Greg

Cool.

And Z=X presumably?

Cool.

And Z=X presumably?

Z can be whatever you want it to be. It would basically just be your available height minus 4 inches. The 4 inches comes from two layers of 1" MDF and two layers of 1" duct liner. Or of course, if you have a ton of height to work with, make Z as big as your calculations say it needs to be to achieve your desired cross sectional area.

This is a basic layout that would work. However, things you could incorporate if your space allows it would be a larger "X" around the ends of the baffles to do two things:
1. lower static pressure (as per the original topic of this thread)
2. add yet another doubling of cross sectional area to create another set of impedance mismatches.

Also, you can double "X" after your last baffle to give you a really nice low air velocity. You can add another baffle. You can use 1" baffles instead of 1/2". You can use 2" or even 4" duct liner. You can mirror image this design and make a Y split type silencer box like Stuart incorporates in many of his designs. You can add a tuned stub. Many things to further improve on this design. However, this basic design is tried and true and as it's drawn will give you great results!

Greg

Here’s another take on silencer boxes that raises a few more questions. These silencer boxes have ducts inside them, with all remaining spaces filled with insulation. I saw this on the following YouTube video:
https://youtu.be/_aONYq_8vnQ

According to the fellow who made them, the ducts help to prevent the box from developing a musty smell (which is what happened to boxes he made earlier without ducting). As someone who lives in a cool climate with very high humidity, avoiding dampness and musty smells is a priority for me.

I wonder what are the other pros and cons are of using ducts inside silencer boxes?

Are they likely to be less soundproof at certain frequencies? I imagine that higher frequencies might bounce through the tube a bit, but lower frequencies would pass through the walls of the duct and into the box.

Might the curve of the ducts result in improved airflow, or would a box of similar size without ducts provide better airflow because there would be more free space for air to move (hence less static pressure)?

Might the extra insulation be beneficial for soundproofing?

Thoughts?

I saw this on the following YouTube video:

I saw a video on YouTube that "proved" that the Earth is flat, and another that "proved" that the queen of England is actually a reptile alien invader!

I would never, ever build a studio based on the videos on YouTube about building studios! Most of those are plain silly, many more show gross ignorance of acoustics and wishful-thinking, even more are pure myth and "my girl-friend's cousin's step-father's great-uncle's best friend's daughter used to date the brother of a guy who worked at a studio, and he said....". Some of those videos show dangerously unsafe practices. Some are just plain illegal. A very, very few of them show good, solid, sound advice. The one you saw is not in the latter category:)

the ducts help to prevent the box from developing a musty smell (which is what happened to boxes he made earlier without ducting).

Which indicates that he didn't know how to make the "earlier" one right either! Then he didn't learn his lesson from that, and instead of doing research on how to do it right, he came up with yet another hair-brained scheme...

If he would have built his first one correctly, building a properly sealed massive box from the right materials and using proper duct-liner for the interior, instead of products that were never made to be used inside air ducts, then he would not have had the "musty" problem.

I wonder what are the other pros and cons are of using ducts inside silencer boxes?

The pro's are that it is cheap. The cons are that it won't work, and will make things worse for the HVAC system, not better. That's the brief summary... now for the details:

Are they likely to be less soundproof at certain frequencies?

Yes. Mostly in the range 20 Hz to 20 kHz...

More seriously, there's no useful mass in that box at all, so it is NOT going to isolate, and in fact will probably resonate. You can clearly see that it is built from very thin plywood panels attached to a light frame, which seems to be MDF. So it's basically a resonant box, with membranes stretched over a frame. In reality, the box needs to have roughly the same surface density as the wall or ceiling through which it penetrates, since it is an extension of that wall or ceiling.... Which brings us to the second point of that box in the video: It isn't even on a wall penetration! It just has a duct going in, and another duct going out, and one of those ducts then goes through the wall.... Sigh! Unbelievable. What's the point of having a silencer box that is supposed to replace the missing mass in the wall, but then not do that, and run a very low mass duct though the wall instead? Pure ignorance. The guy doesn't have a clue of what he's even TRYING to do...

Then there's the issue of impedance mismatch: there isn't any on this "design"! The cross sectional area of the air path remains roughly the same throughout, so he's missing a key benefit of having a large silencer box: the impedance mismatch at the entry and exit, where there's a sudden very large change in cross-sectional area. When that happens, low-frequency sound is partially reflected back up the way it came, due to the difference in acoustic impedance. That larger the change in cross-section, the more pronounced this effect is. Minimum useful is a change of twice or half the area, better is four or five times the area (or 1/4, 1/5). In this case there's no change at all, thus no benefit.

I imagine that higher frequencies might bounce through the tube a bit

Highs tend to act in ray-like fashion, moving more or less in straight lines... and if you look at that second picture you posted, you can see that an imaginary straight line going up the center of the input duct is going to hit the curve at the far end of the "bend"m then carry on straight through, where there's almost no insulation outside of the duct, then continue on through the thin, resonant panel that is supposed to be the wall of the "silencer" box.... So at best, there's only slight attenuation of the highs.

but lower frequencies would pass through the walls of the duct and into the box.

Right! And then continue onwards, through the thin resonant panel that is supposed to be the wall....

Might the curve of the ducts result in improved airflow,

Think of it this way: Does traffic on a road move faster and more smoothly down a long straight stretch of road, or does it move faster and more smoothly through a pair of tight hair-pin bends, back-to-back? There's your answer. Any time you have a bend in an HVAC duct, you are basically making it harder for air to get through, because it has to change direction to do that. It take energy to make the air change it's route, and something is supplying that energy: the fan. So if the air is forced to go around many tight corners, the fan has to work harder to make that happen, because it feels more "resistance" in the air flow. Simple illustration: blow through a drinking straw, then pinch the straw a bit... it's harder to blow through like that. You have increased the resistance to the flow of air. Technically, this is called "static pressure". One goal of HVAC design is to keep the static pressure low, so the air can move through easily and the fan isnt overloaded. The more bends and corners you have, the higher the static pressure. The more you pinch down the air flow by narrowing the path, the more you increase static pressure. Etc.

So no, those curves do not improve airflow: they make it WORSE! In a normal silencer box, there's actually a larger cross sectional area for the air flow around the corners, because the diagonal distance from the tip of the baffle into the opposite corner, is larger than the straight-across section just before and after the corner.

or would a box of similar size without ducts provide better airflow because there would be more free space for air to move (hence less static pressure)?

Exactly! Yup.

Might the extra insulation be beneficial for soundproofing?

Nope! It does nothing useful, because the entire box does nothing useful.

So, to summarize:

1) There's no mass in this box, thus no isolation.
2) The box is resonant, thus probably amplifying at some frequencies.
3) There's no changes in cross section, thus no impedance mismatch.
4) High static pressure.
5) There's no massive "sleeve" that penetrates through the leaf it is attempting to isolate, thus no isolation.

Those are the issues that I noticed at first glance.... There's probably other issues too...

I think you can safely ignore all videos from this guy...

- Stuart -

Soundman2020 wrote:I saw a video on YouTube that "proved" that the Earth is flat, and another that "proved" that the queen of England is actually a reptile alien invader!

LOL! Tell me about it. You would not believe how many videos there are on YouTube from people claiming that the earth is flat. The flat earth movement has been gaining momentum in recent years. Fortunately there are some great channels dedicated to debunking this ridiculous idea. “Sci Man Dan” is one of my faves. Something tells me you will enjoy his brand of humour: https://www.youtube.com/channel/UCRtsZ5 ... sQLQ3XOAeA

I assure you that I don’t assume anything I see on the internet is valid, nor do I dismiss it outright without educating myself, hence my questions to you. Really glad you could chime in with such a thorough response for me and everyone else here.

Soundman2020 wrote:It isn't even on a wall penetration! It just has a duct going in, and another duct going out, and one of those ducts then goes through the wall.... Sigh

I could have sworn that the box was fitted in such a way that it is sealed to the ceiling leaf, not just raw duct passing through?

Soundman2020 wrote: So, to summarize:

1) There's no mass in this box, thus no isolation.
2) The box is resonant, thus probably amplifying at some frequencies.
3) There's no changes in cross section, thus no impedance mismatch.
4) High static pressure.
5) There's no massive "sleeve" that penetrates through the leaf it is attempting to isolate, thus no isolation.

Those are the issues that I noticed at first glance.... There's probably other issues too...

I can add a 6th. There are only 2 baffles in this box, which is not ideal. I believe that 3 or more is recommended.

Final thought. The world could use a YouTube channel dedicated to debunking studio construction myths and misinformation.

Gregwor wrote:Hopefully this helps some people with designing basic single path silencer boxes.

Gregwor's Silencer Box.png

Greg

Greg, I've read this thread a few times and have a couple dumb questions. I get the sense that there is a prerequisite to this thread, probably contained in dozens of other threads. This forum does not do Google searches very well.

In any case, per your drawing, where is the inlet and exit? I assume inlet is in the y or z dimension on the left and outlet is on the right where the walls have widened.

If one were to do the mirror image idea, does that mean the left would become the center? This is essentially what I'm looking into for a fairly simple one room design.

It looks like X is derived from Y. But shouldn't it be the other way around? For example if you need to match up with a 10x10 duct, assuming X on the left is the full inlet, then you'd have to derive Y from X (10in). So 10 = (Y-7)/2 or, after some algebra, Y = 27in. Or, am I just completely out to lunch here?

I get the sense that there is a prerequisite to this thread, probably contained in dozens of other threads. This forum does not do Google searches very well.

There is no prerequisite. HVAC is a crazy topic for studios. I'm friends with a few guys who own their own HVAC companies and they have no clue about the workings of our silencer boxes. To them, we are breaking all the rules and for efficient HVAC, we certainly are.

In any case, per your drawing, where is the inlet and exit?

They are at either end of the box. Ideally you want to have the inlet and outlet heading in different directions. To try and clarify that point, as an example, you would have the inlet coming from the left and the outlet shooting out the top.

Here is a picture to represent the idea. You can see that the inlet is coming in on a horizontal plane and the outlets are shooting out the bottom. The same design principles can be applied to this Y (or split if you want to call it that) design. The idea behind this design is that you will be quadrupling your cross sectional area as each side of the box is double the cross sectional area as the round duct.

If one were to do the mirror image idea, does that mean the left would become the center? This is essentially what I'm looking into for a fairly simple one room design.

I'm not sure what you mean by the mirror image idea. Are you referring to Y/split design like I posted above?

It looks like X is derived from Y. But shouldn't it be the other way around?

There are several ways to figure out the dimensions of your box. In my simple design that you are referring to, the formula will help you work backwards. This is because in most cases, people are building boxes to fit into already built spaces. Knowing the size of the available space, you can enter that measurement as the value of Y. This could be a space such as the distance between floor joists. With these values, you can then easily determine the value of X. With the value of X, you can figure out your maximum cross sectional area. And with that information, you can determine whether or not you can get a slow enough air velocity and whether or not the cross sectional area is at least twice that of the duct work feeding the box. Finally, the formula at the bottom of the image will make figuring out the overall length of a box containing 3 baffles easy. There is no Z value in my simple design image. The Z value would be the depth of the box. If you have the space to have a deep box, you could adjust the value Z so that it along with (multiplied by) the value of X, would result in your desired cross sectional area. There is one thing to consider when determining the size of Z - the aspect ratio. Follow these rules:

For example if you need to match up with a 10x10 duct, assuming X on the left is the full inlet, then you'd have to derive Y from X (10in). So 10 = (Y-7)/2 or, after some algebra, Y = 27in. Or, am I just completely out to lunch here?

So instead of entirely following my design (which determines the values by working backwards -- like mixing music top down), you can start like this:

10"x10" rectangular duct = 100 square inches of inlet cross sectional area. That means you need at LEAST 200 square inches throughout the inside of the box. After that, figure out where you can put it and if there are size limitations in that space. If you have all the space in the world, then build the box however and wherever you'd like. If you have a certain spot you need to fit it, then use my formula to figure out if you can get that 200 square inches to work for you. We can work out your design in more detail on your design thread but I hope this helps you out to get started!

Greg