First off, I am an old man trying to learn some new tricks. After stumbling through WinISD and having a fairly successful build, I am moving on to embarrass myself with Hornresp. I have read many of the tapped horn build threads here and it is really amazing what you guys are doing.
So, I am trying to teach myself this program by reverse engineering some of the fine tapped horn cabs you young punks have already built. Are these following assumptions correct?
Using Hornresp, the S1-S5 values are AREA MEASUREMENTS at a particular point of the horn (the internal width of the horn multiplied by the depth of the horn at that point).
The L12-L45 are the actual measurements of the horn fold. They reflect the path the sound follows from the face of the speaker to the output of the cabinet.
If the previous assumption is correct, and I wanted to see the effect of making the cabinet slimmer, I would reduce each of these S1-S5 values proportionally to the new width of the cabinet. I would not alter the L12-L45 values at all.
EXAMPLE: If the cabinet I am studying has an internal width of 24" and I am attempting to model it with an internal width of 18", I would then reduce each S1-S5 values by 25% (18 divided by 24 = 0.75). I would then multiply the original S1-S5 values by 0.75 to see how the cab would model with an interior width of 18". I would not alter the L12-L45 values at all.
If the previous example is correct, then I could also use the same approach if I wanted to model making the cab wider.
EXAMPLE: If the cabinet I am studying has an internal width of 24" and I am attempting to model it with an internal width of 30", I would then increase each S1-S5 values by 25% (30 divided by 24 = 1.25). I would then multiply the original S1-S5 values by 1.25 to see how the cab would model with an interior width of 30". I would not alter the L12-L45 values at all.
So that is my question for today. Hopefully my understanding of this process is correct and my assumptions are true. If not, could one of you boy genius's help an old man out and try to help me understand what I am doing wrong?
So, I am trying to teach myself this program by reverse engineering some of the fine tapped horn cabs you young punks have already built. Are these following assumptions correct?
Using Hornresp, the S1-S5 values are AREA MEASUREMENTS at a particular point of the horn (the internal width of the horn multiplied by the depth of the horn at that point).
The L12-L45 are the actual measurements of the horn fold. They reflect the path the sound follows from the face of the speaker to the output of the cabinet.
If the previous assumption is correct, and I wanted to see the effect of making the cabinet slimmer, I would reduce each of these S1-S5 values proportionally to the new width of the cabinet. I would not alter the L12-L45 values at all.
EXAMPLE: If the cabinet I am studying has an internal width of 24" and I am attempting to model it with an internal width of 18", I would then reduce each S1-S5 values by 25% (18 divided by 24 = 0.75). I would then multiply the original S1-S5 values by 0.75 to see how the cab would model with an interior width of 18". I would not alter the L12-L45 values at all.
If the previous example is correct, then I could also use the same approach if I wanted to model making the cab wider.
EXAMPLE: If the cabinet I am studying has an internal width of 24" and I am attempting to model it with an internal width of 30", I would then increase each S1-S5 values by 25% (30 divided by 24 = 1.25). I would then multiply the original S1-S5 values by 1.25 to see how the cab would model with an interior width of 30". I would not alter the L12-L45 values at all.
So that is my question for today. Hopefully my understanding of this process is correct and my assumptions are true. If not, could one of you boy genius's help an old man out and try to help me understand what I am doing wrong?
Last edited:
Your understanding of the parameters are correct, Merry Christmas, Old Man!
Art (old sound man) Welter
Art (old sound man) Welter
I’m 67, so not sure if I qualify as a ‘young punk’/’boy genius’ to you……………
Regardless, HR assumes a cylindrical aspect ratio, ergo the most correct conversion is to a square cross sectional area [CSA] ratio of the same area and as it increases to a rectangle, its accuracy declines with increasing aspect ratio to the point where this now ‘duct’ eventually becomes resistive enough in reality [~aperiodic around > 9:1] to no longer match up to the sim close enough to be of any value.
Consequently, getting the folds right severely limits the usable aspect ratios and why corner reflectors often hurt performance in higher aspect ratio designs. Ditto, ‘duct’ sections that act as acoustical resistors.
Both of course can be used to tailor a speaker’s response, just that HR can’t model them accurately.
GM
Regardless, HR assumes a cylindrical aspect ratio, ergo the most correct conversion is to a square cross sectional area [CSA] ratio of the same area and as it increases to a rectangle, its accuracy declines with increasing aspect ratio to the point where this now ‘duct’ eventually becomes resistive enough in reality [~aperiodic around > 9:1] to no longer match up to the sim close enough to be of any value.
Consequently, getting the folds right severely limits the usable aspect ratios and why corner reflectors often hurt performance in higher aspect ratio designs. Ditto, ‘duct’ sections that act as acoustical resistors.
Both of course can be used to tailor a speaker’s response, just that HR can’t model them accurately.
GM
H DHAA,
Take a look here:
Hornresp for Dum... hmm... Everyone 😉 - Home Theater Forum and Systems - HomeTheaterShack.com
and here:
Simple Tapped Horn Tutorial using Hornresp
That ought to keep you busy. 🙂
As GM pointed out you have to get a feel for what may, or may not, work; and David McBean is in the process of adding additional functionality, I'm especially impressed w/ being able to see what stuffing does to the response.
Regards,
Take a look here:
Hornresp for Dum... hmm... Everyone 😉 - Home Theater Forum and Systems - HomeTheaterShack.com
and here:
Simple Tapped Horn Tutorial using Hornresp
That ought to keep you busy. 🙂
As GM pointed out you have to get a feel for what may, or may not, work; and David McBean is in the process of adding additional functionality, I'm especially impressed w/ being able to see what stuffing does to the response.
Regards,
Last edited:
Thanks Art (old sound man) Welter and GM for your quick help. I guess I am not the only old man here! Let's just say I built my first cab over 30 years ago (when you had to do the heavy math yourself) and I really am liking these software cabinet modelers. After may years of inactivity, I have been building like a madman for the last six months and am really excited about the progress I have made. Thanks for coming to my "old man's roundtable." Would you like a beer with that?
GM - your comment about "HR assumes a cylindrical aspect ratio" seems to me to mean HR models the horn like it is a french horn. So any build with plywood will therefore never truly match HR's modeled predicted response. That is the first time I have heard that, thanks.
Then you commented "as it increases to a rectangle, its accuracy declines with increasing aspect ratio to the point where this now ‘duct’ eventually becomes resistive enough in reality [~aperiodic around > 9:1] to no longer match up to the sim close enough to be of any value."
How do you know when that happens? That is the first time I can remember hearing the term "[~aperiodic around > 9:1]". Is there a way to calculate that within HR, or is that something that requires deeper thinking and further calculation outside of HR?
GM - your comment about "HR assumes a cylindrical aspect ratio" seems to me to mean HR models the horn like it is a french horn. So any build with plywood will therefore never truly match HR's modeled predicted response. That is the first time I have heard that, thanks.
Then you commented "as it increases to a rectangle, its accuracy declines with increasing aspect ratio to the point where this now ‘duct’ eventually becomes resistive enough in reality [~aperiodic around > 9:1] to no longer match up to the sim close enough to be of any value."
How do you know when that happens? That is the first time I can remember hearing the term "[~aperiodic around > 9:1]". Is there a way to calculate that within HR, or is that something that requires deeper thinking and further calculation outside of HR?
TB46 (Olliver), thanks for those links. I had actually stumbled upon the "Hornresp for Dummies" once before, but that was at the beginning of quest and I was sub-dummy at that point. I think I have gained enough knowledge now that if I reread that again and it may make more sense. I had forgotten about that link, so thanks for your post.
I don't think I have seen the "Simple Tapped Horn Tutorial" before though. I have to sit here at my desk all day, with nothing to do since it is Christmas Eve, so I should be able to read both of those posts in their entirety today.
Are you an old man too, or a boy genius doing some community service?
Last edited:
Cylindrical like a french horn (or any other musical horn) that's round and straightened out (no bends) and with all the holes closed.
The Hornresp predicted response will never exactly match reality because there's a multitude of things Hornresp does not account for. I can post a long list if you like. But the Hornresp sim will still be remarkably accurate. You just have to know what's involved in the differences between a simulation and reality.
As the aspect ratio increases past 1:1 things change, but I think "no longer match up to the sim close enough to be of any value" is a bit harsh at 9:1. Unless you want to build a complex throat like the Labhorn, simple construction methods dictate that you will have a fairly large aspect ratio at the throat, especially with very small throats (high compression ratios). The highest throat aspect ratio I ever built was almost 20:1. (Compression ratio was a bit more than 5:1, SD:S1. Hornresp defines compression ratio as SD:S2 for offset driver horns and this horn was offset, so by Hornresp's calculation the compression ratio was a lot less than 5:1.) And the measurement was still reasonably close to the sim, all things considered. No worse match between sim and measurement than anything else I've built anyway. I didn't measure at high power though, and that might have changed things since the Reynolds number increases with velocity.
Last edited:
Hi DHAA,
On the risk of plagiarism I'll steal GM's line (Post #3): "I’m 67, so not sure if I qualify as a ‘young punk’/’boy genius’ to you……………" It fits. I will try to do a little community service once in a while, even though lately "they" have not been able to make up their mind if they want to chain me to the sodering iron or the computer (hint: noticed the oil price?). Today I'm having a little bit of slack, then back to family duty. 🙂
Regards,
On the risk of plagiarism I'll steal GM's line (Post #3): "I’m 67, so not sure if I qualify as a ‘young punk’/’boy genius’ to you……………" It fits. I will try to do a little community service once in a while, even though lately "they" have not been able to make up their mind if they want to chain me to the sodering iron or the computer (hint: noticed the oil price?). Today I'm having a little bit of slack, then back to family duty. 🙂
Regards,
Question of the Day - Compression Ratio
Just a Guy, thanks for your comments, and you actually brought up my next question.
What I am experimenting with right now is to taking proven tapped horn designs and narrow/widen them to see how it affects response. I am trying to gain a deeper understanding of how all these different design parameters affect each other. But if I am understandg compression ratio correctly, narrowing/widening the cabs would also directly effect the compression ratio (SD divided by S2).
- I see no way within HR to check compression ratio - I assume you calculate it independently?
- When you design a tapped horns, how do you know what the best compromise for compression ratio is?
- What are the dangers to look for in regards to compression ratio.
- What exactly are the effects of raising/lowering the compression ratio? Is it apparent is the HR modeled frequency response at all, or does compression ratio more affect safeguarding the driver from self destruction?
Thanks everyone for your help so far. Uuuuuummmm, I smell ham cooking - I better get downstairs before it disappears. Merry Christmas to all, and to all a good tapped horn!
Just a Guy, thanks for your comments, and you actually brought up my next question.
What I am experimenting with right now is to taking proven tapped horn designs and narrow/widen them to see how it affects response. I am trying to gain a deeper understanding of how all these different design parameters affect each other. But if I am understandg compression ratio correctly, narrowing/widening the cabs would also directly effect the compression ratio (SD divided by S2).
- I see no way within HR to check compression ratio - I assume you calculate it independently?
- When you design a tapped horns, how do you know what the best compromise for compression ratio is?
- What are the dangers to look for in regards to compression ratio.
- What exactly are the effects of raising/lowering the compression ratio? Is it apparent is the HR modeled frequency response at all, or does compression ratio more affect safeguarding the driver from self destruction?
Thanks everyone for your help so far. Uuuuuummmm, I smell ham cooking - I better get downstairs before it disappears. Merry Christmas to all, and to all a good tapped horn!
Last edited:
Only if you narrow/widen the throat area. You can do whatever you like to the rest of the horn and it won't affect compression ratio. And compression ratio is SD:S2 only for tapped horns and offset driver horns. Compression ratio for end loaded horns is SD:S1.
For an end loaded horn, hover the mouse pointer over the S1 box and at the bottom of the Hornresp screen it will tell you the compression ratio. For an offset driver horn or tapped horn, hover the mouse pointer over the S2 box and at the bottom of the Hornresp screen it will tell you the compression ratio.
You can use math to figure out the best throat area but that is more advanced than most people are capable of. Instead, most people just play around with the sliders until they get the response they want. A throat area of 3:1 (SD:S1) is usually considered safe. Going beyond 3:1 may have negative effects in a bunch of different areas.
It's going to affect aspect ratio of the throat unless you use fancy geometric shapes in your construction to keep the throat square shaped. It's going to affect pressure on the cone and too much pressure could damage the driver at high power. It's going to affect velocity of air and too much velocity could introduce turbulence at high power.
A higher compression ratio will need a longer horn to maintain the same tuning. A longer horn is a bigger horn. A bigger horn will be louder. At the same time, it can be a lot easier to get a very flat response (without huge peaks and valleys) from a very small throat, but that's not always the best thing to do because a very small throat introduces other issues as I've explained previously in this post and the last one.
Changing the throat size is VERY apparent in the frequency response. But at the same time, it depends how much of the horn changes when you change the throat size. If L12 is very short and you just change S1 there won't be much change in the sim (and this won't actually change the reported compression ratio of a tapped horn). But if L12 is very long, or if S2 is set to auto the changes will be much more noticeable (and this will change the reported compression ratio of a tapped horn).
It can change things enough that other areas of the horn need to be adjusted to maintain the same tuning and response curve shape.
Anyway, I could type a book here and it wouldn't do you nearly as much good as just getting a few minutes of experience with Hornresp. Play with the sliders and see what happens. When you get something that looks good post it up and we'll let you know if there are any problems.
Last edited:
Wow, thanks for the info on compression ratios and the associated effects. It really cleared up some areas of confusion I was having. Nice explanations.
You had mentioned "Reynolds number" in a previous post. I had never heard that term before. Wow, fluid mechanics!!! You guys are doing some serious deep thinking in your designs. I also had to look up "aspect ratio." That was a little easier for someone at my pay grade to understand thankfully.
I am working on an experiment in HR right now, I will try to get it posted tomorrow so you guys can have something to laugh at. Thanks for your assistance, I feel like I am getting younger everyday.
You had mentioned "Reynolds number" in a previous post. I had never heard that term before. Wow, fluid mechanics!!! You guys are doing some serious deep thinking in your designs. I also had to look up "aspect ratio." That was a little easier for someone at my pay grade to understand thankfully.
I am working on an experiment in HR right now, I will try to get it posted tomorrow so you guys can have something to laugh at. Thanks for your assistance, I feel like I am getting younger everyday.
You’re welcome!
As one long time poster put it, ‘sound is round’, ergo the math reflects this, so nearly all TL/horn speaker design programs assume a 1D spherical wave propagation.
Basic speaker design programs assume a ~uniform particle density [air mass ‘plug’], so only valid for cubes or very low aspect ratio rectangular cabs that are smaller than the WLs it contains.
AkAbak uses a series of ducts, so I assume it factors in all planes of eigenmodes [standing waves], ergo theoretically the most accurate for an un-damped alignment, whereas MJK’s MathCad 1D software does if polyfil is used, so for me the better choice overall.
“[~aperiodic around > 9:1]” In my haste, I over generalized a bit in that ‘~aperiodic’ means almost completely resistive in nature, whereas beginning around a 9:1 aspect ratio is where one with decent hearing will typically notice a difference before there’s an obvious change in the sim or even in a basic measurement.
An easy experiment is to vent a cab with a basic 3.38” round hole vent and compare it to a 1” x 9” slot, especially at the higher powers a vent is likely to ‘feel’. Otherwise, if the math doesn’t stump you as it does me, study up on heating/AC duct design to calculate the aspect ratio where a given vent mach [speed] will turn highly resistive as 9:1 is just a general guideline not to exceed.
I prefer to use typical vent shapes and add damping as required to get the desired acoustical resistance that critically damps the system, so only programs that allow one to add a resistive component to the cab and/or vent output can sim such effects and even then one must have the experience to mentally compare it to what one hears.
For example, virtually every MJK software [ML] TL sim I’ve done for folks that actually built them found that the amount of stuffing I used to damp down the worst of the simmed HF ‘ripple’ turned out to be severely over-damped, proving once again that as a group we prefer our audio reproduction just as ‘rich’ as we like what we see, feel or taste. Had I posted sims that reflected this though, folks would have claimed them unsuitable.
Ditto how ‘flat’ the speaker’s simmed response, which being typically dominated by the room’s modes, reflections, is rarely the best overall alignment, yet what do the programs default design and most folks demand? A T/S spec 2pi space maximally flat alignment and then all too often complain about its poor overall in-room performance.
GM
Akabak allows a choice of duct, waveguide or horn for each element and simulates in 1d. From the Akabak website:
"The simulation is carried out on the basis of lumped elements and one-dimensional waveguide components."
I assume all the 1/4 wave simulators do the same thing, including MJK's. The comparison I did recently in the TL.app thread shows all the four main 1/4 wave simulators have very similar behavior so it's really just a matter of picking the one with the right features for the job. Personal preference factors in too, and there are some small differences but I don't think any of them are clear winners or losers, they are more similar than different.
This might sound fancy but it's really nothing special. HVAC system designers use this stuff daily to properly size their pipes, and it has applications in other fields as well.
As long as you size your ports properly, don't use incredibly rough construction materials in areas of high velocity and keep your aspect ratio reasonable there's not much reason to ever bother looking at Reynolds numbers. It's all about controlled airflow and we have simple rules of thumb for all that stuff.
Thank you GM and Just a Guy, you have given me a lot to think about.
It is funny that you bring up HVAC duct work as an example of Reynold's number. I have a very good friend who is a very smart fella, that for the past 30 some years I thought he was wasting his brain power in the HVAC field selling/designing systems. But HaHa, the jokes on me.
So if I am understanding this concept correctly, the main way this Reynold's number would apply to speaker building would be in relation to it's hydraulic diameter? Let's say a certain speaker cabinet design requires a particular cross section area for proper tuning and to fit within the cabinets dimensions. If you were to use a large, narrow slot port, it would have a higher wetted diameter, thus pushing your Reynolds number if. If you were to instead use a round port, it would have a lower wetter diameter for the same port volume, keeping your Reynolds number lower. Basically, you have a lot more potential for friction in a long, narrow slot than in a circular port (more surface area for the same cross sectional area).
So if those previous assumptions are correct, it is a real wake-up call to me. I have been using slot ports in my subs (and even a lot of my tops) reasoning that as long as the dimensions provided the proper tuning, slot ports were superior as they helped provide additional strength/bracing to the cabinet
GM, your comment has really got me thinking:
"Ditto how ‘flat’ the speaker’s simmed response, which being typically dominated by the room’s modes, reflections, is rarely the best overall alignment"
With the advent of low priced/high quality digital EQ's I can see your reasoning. Do you try to optimize your designs for lowest extension, highest SPL, or other factors?
I spent most of yesterday doing physics experiments with my grandson and the physics lab I got him for Christmas. The boy is big on asking questions, and after stumbling through college physics myself many moons ago he had me racking my mind trying to explain physics concepts to him. Between that and some of the speaker cabinet optimization ideas you guys have me thinking about, I have really had to dust my old brain off and see if it still works.
It is funny that you bring up HVAC duct work as an example of Reynold's number. I have a very good friend who is a very smart fella, that for the past 30 some years I thought he was wasting his brain power in the HVAC field selling/designing systems. But HaHa, the jokes on me.
So if I am understanding this concept correctly, the main way this Reynold's number would apply to speaker building would be in relation to it's hydraulic diameter? Let's say a certain speaker cabinet design requires a particular cross section area for proper tuning and to fit within the cabinets dimensions. If you were to use a large, narrow slot port, it would have a higher wetted diameter, thus pushing your Reynolds number if. If you were to instead use a round port, it would have a lower wetter diameter for the same port volume, keeping your Reynolds number lower. Basically, you have a lot more potential for friction in a long, narrow slot than in a circular port (more surface area for the same cross sectional area).
So if those previous assumptions are correct, it is a real wake-up call to me. I have been using slot ports in my subs (and even a lot of my tops) reasoning that as long as the dimensions provided the proper tuning, slot ports were superior as they helped provide additional strength/bracing to the cabinet
GM, your comment has really got me thinking:
"Ditto how ‘flat’ the speaker’s simmed response, which being typically dominated by the room’s modes, reflections, is rarely the best overall alignment"
With the advent of low priced/high quality digital EQ's I can see your reasoning. Do you try to optimize your designs for lowest extension, highest SPL, or other factors?
I spent most of yesterday doing physics experiments with my grandson and the physics lab I got him for Christmas. The boy is big on asking questions, and after stumbling through college physics myself many moons ago he had me racking my mind trying to explain physics concepts to him. Between that and some of the speaker cabinet optimization ideas you guys have me thinking about, I have really had to dust my old brain off and see if it still works.
Question of the Day - Vtc/Atc
I have been looking over many of the fine tapped horn designs on this great website and I have noticed something in many of the designs I can't make sense of. It seems most of the designs that were done more than a few years ago used Vtc/Atc in the HR calculations. But starting about 3 years ago, many designers stopped using the Vtc/Atc in the simulations.
What is the reasons for not using the Vtc/Atc anymore? Was there an evolution in tapped horn design that made these factors unnecessary? After re-reading the "Hornresp for Dummies" post again, it seems like the Vtc/Atc should be used in these designs, considering how the speakers are attached to the baffle board. But that "Hornresp for Dummies" was also written right before this change seems to have happened. What am I missing?
I have been looking over many of the fine tapped horn designs on this great website and I have noticed something in many of the designs I can't make sense of. It seems most of the designs that were done more than a few years ago used Vtc/Atc in the HR calculations. But starting about 3 years ago, many designers stopped using the Vtc/Atc in the simulations.
What is the reasons for not using the Vtc/Atc anymore? Was there an evolution in tapped horn design that made these factors unnecessary? After re-reading the "Hornresp for Dummies" post again, it seems like the Vtc/Atc should be used in these designs, considering how the speakers are attached to the baffle board. But that "Hornresp for Dummies" was also written right before this change seems to have happened. What am I missing?
Attachments
Bummer 🙁 Oh well, maybe someday…… and in the meantime my modded room mode calculator gives me enough info to fine tune larger cabs where it matters most.
Below the driver's mass corner there really shouldn't be any difference between any of them, though if I understood MJK's meaning WRT him using a different way of calculating pipe/horn action in his more recent software, then overall I give the edge to his.
GM
Some are using the Vtc/Atc and also doing "cone correction" to account for the volume in the cone.
In a TH with the cone opening the full size of the Sd, the change in cross sectional area with or without allowing for Vtc/Atc is not all that great (when using about a 3/1compression ratio) and makes little difference in the results.
Vtc/Atc makes much more difference in mid/high offset horn designs (like the Synergy/Unity) where the enclosed area in the Vtc/Atc makes an acoustical band pass.
Juggling the enclosed area vs. the exit area and port length can make a large difference in upper response, a big deal when the crossover is 600-2500 Hz, but not much concern for sub use, where the crossover point is generally under 200 Hz, and TH response is generally only a few octaves of relatively flat response.
Art
Vtc/Atc is a big deal in traditional horn designs, so naturally, early tapped alignments were designed as ~ compound constant or expanding TL [front and back loaded TL or horn] after the early ‘50s Jensen Transflex tapped TL [TP], but a TP/TH in theory shouldn’t be used above its tap point, so in the scheme of things not normally of any real benefit since Atc is typically > Sd, making the ‘trapped’ air in Vtc too small percentage wise WRT pipe/horn net Vb to have any audible impact on its performance.
GM
GM
Yeah, well, I think you might be surprised at how braindead simple it can really be. There are simulators for just about everything. You don't have to do math anymore, you don't need to know the density and kinematic viscosity of air at ambient temperature and you don't even need to be able to describe the roughness of the pipe mathematically IF you have good searching skills and can find the right simulator.
Here's a screenshot of my experience with Reynolds number, friction factor and a simple simulator. The top of the pic (above the orange line) is a Moody chart, and below the orange line is the entirety (inputs and outputs) of a simple air flow calculator. As you can see there are only 4 inputs that anyone can understand. The hardest thing I had to do was convert velocity in m/s to cfm.
An externally hosted image should be here but it was not working when we last tested it.
There you go, you got it. I've never heard the term "wetter" before, but I'm not an HVAC guy and not really that familiar with Reynolds number calculations either. It's all just hydraulic diameter to me, and I'm not sure why they don't just call it surface area.
Again, I didn't mean to give a wake up call with this, you can ignore the alarm. Look at the picture above again and look at the numbers I entered in. It's a 2 inch diameter pipe with a length of 7.5 FEET and IIRC a velocity of 10 m/s. That's pushing the limits beyond anything you are likely to build. Ever. I seriously doubt your slot port will be this extreme even with a very high aspect ratio. This sim produced a very high Reynolds number and when correlated to a Harmon paper, this Reynolds number is right on the borderline of being a bad idea and being a total failure.
This is how we learn though. I presented an idea, I got mocked and ridiculed by an entire forum of people saying it wouldn't work. No one could use their words and tell me why they thought it wouldn't work, they just intuitively felt it wouldn't. Mark Seaton was the only one that could point in the right direction, even though he didn't explain the reasons effectively. This led to a personal study of Reynolds number, friction factor, Moody graphs and air flow calculators. In the end I proved my idea would work (theoretically) in a scientific fashion. Unfortunately I also proved it wasn't a really good design. And it doesn't matter. I did the work but nobody else even bothered to follow along and actually read my posts. I was still laughed out of town even though I was basically the only one that knew what they were talking about in the end. (Aside from Seaton's vague comments and one other guy that followed along and brought the Harmon paper into the discussion.)
That's a long way of saying that most people don't bother with Reynolds numbers, friction factor or any of that stuff. We have rules of thumb that work for everything. It's also a long way of saying don't take open forums too seriously. There's a lot of problems with this type of communication, but there are benefits as well.
Last edited:
- Status
- Not open for further replies.