FLH design basics for a dummy

Status
Not open for further replies.
Re or Rg is what you need to increase to crudely simulate power compression.

Another way to get a rough idea, is to make an assumption on the voice coil temperature, and use the Power Compression tool.
 

Attachments

  • Attach_1.png
    Attach_1.png
    18.4 KB · Views: 139
Couple of things -

1. You are describing overhung, not underhung.

An externally hosted image should be here but it was not working when we last tested it.


2. The test signal will have as much (if not more) effect on voice coil temp than the applied voltage. The frequency, duty cycle and crest factor of the signal will have a tremendous impact, especially in the enclosure which will present a dramatically wildly varying impedance load.

3. If you want to test it under a realistic load it has to be tested in the cab, in which case you don't have access to the driver.

4. In an overhung voice coil the temp may be dramatically different at different points, the part you can see might not be the same temp as the parts you can't.

5. The coil starts to cool pretty quickly after the test signal stops so you would have to measure while the signal is being applied or within seconds of stopping the signal. When measuring t/s parameters hot I noticed that I got a different set of parameters every time I ran the test as the coil cooled, and these tests were only seconds apart.

Nothing about this is easy.
 
Last edited:
You can avoid all power compression issues and live worry free by observing the power progression/SPL relationship with a driver having a high power rating. An example assuming 95 Db @ 1 watt sensitivity:

Watts:****1***2***4***8***16****32***64***128***256***512***1024
Decibel:**95**98*101*104*107**110**113**116***119***122***125

Each doubling of power input produces 3 Db SPL, so most of the easy lifting is done by 128 watts as you can see. If your design can take advantage of this you'll sleep better.
 
You can avoid all power compression issues and live worry free by observing the power progression/SPL relationship with a driver having a high power rating.

Nice try but no. The 3 db increase per doubling of power isn't news and it's not going to avoid all power compression.

This is a state of the art $700 driver. 00002 21NLW9601 - By Technologies 21NLW9601 : Eighteen Sound - professional loudspeakers

A lot of time and money was spent on designing this driver to keep it's cool and yet it still suffers measurable power compression at surprisingly low power levels, it loses 0.7 db at 10 db below full power (which is not a lot of power, check your own chart that you just posted) and 2.2 db loss at full power. This is a realistic and honest way to rate a driver, a lot more honest and detailed than most manufacturers.

Here's a couple of things to note on the spec sheet.

The state-of-the-art 5,3" diameter ISV copper voice coil shows a inside-outside split winding, four layers design, enabling the 21NLW9601 to handle up to 3600W program power.

...

The 21NLW9601 has been developed after intense FEA and fluido-dynamics simulation and testing, focusing on dissipating the heat generated by the powerful 5.3" coil. Special attention was given to the optimization of air flow into the gap without introducing audible noise. A special low density material air diffractor placed into the backplate acts as a cooling system, increasing the power handling capability and lowering the power compression figure.

And despite all that engineering, despite the special cooling features, despite the very large coil (split wound allows a lot of surface area for cooling), it still has 0.7 db power compression loss at -10 db. That's significant losses starting at a low power level and you won't likely find a better designed driver so this is as good as it gets.

Power Compression @-10dB 0,7 dB
Power Compression @-3dB 1,5 dB
Power Compression @Full Power 2,2 dB

The power test procedure was very likely a bandwidth limited AES signal at the specified power for a minimum of 2 hours. That's a 100 percent duty cycle signal with a crest factor of 6 db so it's a tough test, but still, this high sensitivity high power driver does not allow you to live worry free. You really have to watch what you are doing even with top dollar state of the art drivers.
 
Last edited:
In this context let's compare your driver with the 18 Sound driver.

The 18 Sound is rated for 1800 watts AES.
Yours is rated for 1000 watts RMS, roughly equivalent to 2000 watts AES. (RMS is 100 percent duty cycle with only 3 db crest factor, so twice as much intensity as AES at the same power level.)
So your driver is actually rated to handle more power than the state of the $700 driver. Does that sound right?

It's not. Your driver probably has a 2.5 or 3 inch voice coil and relatively primitive cooling design.

If your driver was tested according to the brutal power test that 18 Sound (and many other pro driver companies) use to find the driver power rating, your driver's power rating would most likely be lowered dramatically. It would probably be half it's stated power rating, maybe less. I'm guessing around 250 watts RMS.

That doesn't mean that your driver couldn't handle 2000 watt peaks, it could, if the signal was low duty cycle and high crest factor. But if you want to compare apples to apples you need to consider the fact that different manufacturers specify power handling in very different ways, and pro drivers are typically very conservative while car audio drivers are typically the polar opposite.
 
Last edited:
Yeah I was generalizing as usual. Most statements made by humans on this earth stand as truth for relatively short periods of time. Inevitably someone discovers a bit more and whacks old truths or facts back into the stone ages where they come from. Sometimes I'm careful and speak precisely, but generally not because there is no such thing as the truth available to humans.

Lets just say a person may minimize power input to help calm their worries concerning power compression.

My driver is rated for 1000 watts RMS by the maker. I never made that claim and have never used that figure. My opinion after owning the IDMAX for 16 years is the same opinion I had the day I brought it home. I think its probably OK at 500 watts max on music for very brief periods. I never gave it 250 watts though, cause I don't think it can handle any more than that stably. In fact, its more likely that its never even seen 200 watts. My amp was rated at 250 watts@4R and that was over rated by the maker too.
 
Who designed the driver? Is Dan Wiggins associated with Image Dynamics or was it just his company that did the Dumax measurement?

Since I was on the topic of differences in ratings between pro and car audio companies I took a look at the Dumax report link in post 5. As usual for this type of driver the stated xmax is exaggerated.

An externally hosted image should be here but it was not working when we last tested it.


Even if you completely ignore everything except the Bl graph xmax still isn't the claimed value. He drew the red line across at the right spot but failed to properly read the point where the Bl curve crossed the red limit (71 percent) line.

Ignore the purple line completely, that's just a drawn on curve fit, not even sure why it's there. The measured Bl curve is the blue line, and it crosses the red limit line at -18 mm and +26 mm. The lower of these two crossing points is the 71 percent Bl point, so Bl is limited to 18 mm, not 22.91 as stated.

I've seen this type of cheating from Dan Wiggins before. When a driver is offset like this (not symmetrical around the resting position) he takes the -18 mm and the +26 mm and adds them for a total stroke of 44 mm and divides that by 2, which gives 22 mm. And then I guess he just added another 0.91 mm to that.

This is cheating, the Bl limit is the crossing point of the Bl curve and the 71 percent limit line on the stroke direction that shows the lesser value, so 18 mm.

The KMS curve also appears to be offset and very misshapen so stroke is clearly not very linear.

The real limit would be Le though. I'm not sure if Dumax measures Le linearity or not but it's not shown. For these high excursion drivers it's easy to get a high Bl number, a bit harder to get a high CMS/KMS number, and really hard to get a high Le number because Le is wildly different on the forward and backward strokes in ultra high excursion drivers.

That's why manufacturers of this driver type rate xmax based on only Bl, if they reported Le xmax would be a lot lower. That's not xmax by klippel standards, although you could call it usable excursion if it sounds ok at the 71 percent Bl excursion.

So I don't mind that this driver's stated xmax is based on Bl alone, but it should be 18 mm and it should be called usable excursion, not xmax. But that's an industry wide issue with high excursion drivers and it's not going to change.

On the other hand, most pro drivers will pass a klippel test in all 4 limiting categories, Bl, CMS/KMS, Le and maybe Xd too although that's not important for a subwoofer and usually not shown.

Anyway, thought you might find some of this stuff interesting. It's incredibly important IMO but most people either don't know or don't care about any of this driver rating issue.

For the record I'm not bashing the driver, it looks like a pretty decent little unit.
 
Last edited:
One last thing on the driver topic. If something happens to this driver can you replace it? i doubt you could get it reconed, even if you could it would likely not be from the manufacturer so t/s specs would change. The newer D4 version sells for $470 and it doesn't appear to have the same specs. You'd have to sim both to see how much difference it makes. You could sim other drivers in your cab and see if there's a readily available suitable replacement too but that might take awhile, a lot of driver sims.

This is just something to think about, but I wouldn't make a horn this large for a driver unless I was pretty sure I could get a replacement if needed. Especially if it was an older driver with a foam surround.
 
One final note. I analyzed Dark Side of the Moon, Track 1 intro because that's the only track you mentioned that you have with low bass. That's the track with the super deep bass everybody goes on about isn't it?

This first one is right near the beginning, from about 15 seconds to 30 seconds, after it's ramped up a bit in volume but still nothing but heartbeat.

An externally hosted image should be here but it was not working when we last tested it.


This second one is about 10 seconds near the end of the intro where there's all kinds of stuff going on but before the actual music starts.

An externally hosted image should be here but it was not working when we last tested it.


After the music starts there's not much at all below 40 hz.

I was wondering if there was something wrong with my track so I did a quick search. it appears there's nothing wrong with it, there's just no low bass.

Bass response - how deep should my speakers go for a lifelike sound at home?

You might want to analyze some of you low bass tracks to see what's really there. You can gain a huge amount of sensitivity if you raise the tuning and keep the box size the same.
 
Hey good stuff just a guy! Eric Stevens founded Image Dynamics and was the owner of the company until about 2005 when he took on a partner, then he left in about 2011-2012.

I saw that silly BL graph on the DUMAX too and couldn't swallow it. Thats why I pay no attention to published stuff and never have since I was a youngster. If its from a .gov or a corporate, its rubbish. A lot of published academia is rubbish too. Seems a common man is more likely to be honest than those that make claims to be.

Thanks for the low bass testing! I had given up on the prospect of sub-20 extension not feeling that it was important enough to me. I have more important problems to solve than sub-20. And yes, that first track on Dark Side is reputed to go lower than you measured. Looks like about 30 Hz and up are all thats needed there. Thanks for checking that stuff out!

I hope to find better sensitivity in my next round of Hornresp fun. We'll see.

Edit: I don't have a plan for another driver should this one fail. Not gonna consider it cause it isn't on my checklist. Don't care.
 
Last edited:
In the opening posts here I didn't say much so these are my basic reasons TODAY for wanting to build this FLH horn. All is not static in this universe though.

1) I have an old school 16 year old driver that kicks butt and sounds pretty good to me doing so. Its a lot more stable and a better quality driver than it looks. It is a long excursion driver too, which is helpful to extend SPL levels in any design, all horns included. Sound quality wise, I'd put it up against anything sold in its car audio class today. SPL wise, it does that plenty too. Its never needed to be pushed hard. The voice coil windings aren't discolored at all, still shiny and new looking, having never seen much over 200 watts, if that. All that and its paid for and I'd like to hear it play again after many years of storage.

2) I need a cheap sub on wheels that will work for music only, both indoors and outdoors. More play time will take place outdoors than in. Outdoors it needs to be able to go loud if I ask it to because I'm all over the place doing stuff during the summers. Neighbors are a long distance away so excessive noise levels are more likely to bother me before it bothers them. So basically it needs wheels, needs to play well outdoors, is cheap to build, and needs to go as loud as I can put up with, which isn't extreme.

3) Power consumption should be relatively moderate because while outdoors the sub (and all other audio gear) will be powered from a small solar power/battery bank system. 40 amps max @ 13.8 volts in full sun and I refuse to tax it anywhere near 40 continuous amps. Some of my power tools are 12v DC powered too, so this horn can't be a full time power hog. Ideally I'd like to see this play loud on less than 200 watts so decent sensitivity is important to me.

4) This beast has to produce a broad usable passband (Revised upward from ~28 Hz to ??) and play flat across the passband outdoors. And I mean FLAT! Yeah THAT flat... That one. I can make ripples and dips and nulls and all other fun stuff indoors with another tool that I have, a MiniDSP 2x4. It would be silly for me to build another tool that makes ripples and dips, so I'm not going there with this project. The MiniDSP will fix up the response to work fine indoors, as thats why I already own that tool. Sorry all you tapped horn fans... your solution puts ripples in the response, is more difficult to keep driver excursion to a minimum so sound quality may be maximized, are more likely to suck more power going loud generally speaking, and sucks eggs at my place.

5) A big cabinet is not a problem for me. Small cabinets will not do all of the things that I'm asking of this sub, so bigger is better and I'm prepared for that. I'm unwilling to compromise an awful lot on that stuff I want above this line... this thing has to tick off all my check marks, not just one or two. Starting with a big cabinet allows me to work more freely on all those more important things, seeing as I'm a first time dummy and I like BIGG.

Now a brief rant about size since a few of you seem to turn up your nose toward everything BIGG... It is relative you know. Relative to performance and relative among persons' perception of it. There is a ~630 liter FLH subwoofer posted on another forum that reaches no more than 108 Db SPL on 7 watts at driver Xmax, using a cheap 12" driver. Now that is a lot of cabinet in return for only ~108 Db SPL in my opinion. That big FLH is poorly purposed too, in that it has a crappy response and a compression ratio of 5.45:1 stressing its cheap ~$70 12" driver.

Think I'm crazy for building an even bigger FLH for only a 12" driver? Well maybe I am crazy but remember, crazy is relative too. You see, that ~630 liter horn above was reproduced SIX TIMES by the person. Thats right, he built 6 of those ~630 liter cabinets and installed them in his home theater, then built 5 more subs (using the same driver) of another dual-driver FLH design to fill in where those first 6 cabinets could not go. After all was said and done, all his sawdust and glue and wood and drivers didn't beat ANY of my BIGG FLH boxes posted within this thread, in all the areas of performance that I want from just one box.

Remember many of our perceptions are relative. Even if I'm certified as crazy by universe, that doesn't make me wrong, or any of you more right. We are just different and do different things for different reasons.

Keep your sticks sharp fellas. More Hornresp fun coming after some rest.
 
4) This beast has to produce a broad usable passband (Revised upward from ~28 Hz to ??) and play flat across the passband outdoors. And I mean FLAT! Yeah THAT flat... That one. I can make ripples and dips and nulls and all other fun stuff indoors with another tool that I have, a MiniDSP 2x4. It would be silly for me to build another tool that makes ripples and dips, so I'm not going there with this project. The MiniDSP will fix up the response to work fine indoors, as thats why I already own that tool. Sorry all you tapped horn fans... your solution puts ripples in the response, is more difficult to keep driver excursion to a minimum so sound quality may be maximized, are more likely to suck more power going loud generally speaking, and sucks eggs at my place.

Doesn't mini dsp allow you to store different settings and switch on the fly? I don't have one, I don't know. If so you could program it to eq flat inside AND eq flat outside and just flick the switch for the appropriate environment.

I don't think you realize how imperceptible a little bit of rippled response is. A 3 db bump in response is barely noticeable. Maybe play with your eq a bit and check out what a slightly rippled response sounds like. It's really not a big deal. Whatever you are currently listening to probably has a much worse frequency response than you think.

Keeping driver excursion to a minimum is a waste of potential. If you don't power it to xmax it's just a waste. Lower excursion will indeed be less distortion but your sims are not that loud. If you start at 120 db/1m and move 16 meters away you only have 96 db. At 64 meters you are down to 88 db, this is not that loud. At least you won't have to worry about your neighbors complaining, although you might have to worry about your friends complaining that it's not loud enough. In my experience it's never enough, you will always want more.

5) A big cabinet is not a problem for me. Small cabinets will not do all of the things that I'm asking of this sub, so bigger is better and I'm prepared for that. I'm unwilling to compromise an awful lot on that stuff I want above this line... this thing has to tick off all my check marks, not just one or two. Starting with a big cabinet allows me to work more freely on all those more important things, seeing as I'm a first time dummy and I like BIGG.

Now a brief rant about size since a few of you seem to turn up your nose toward everything BIGG... It is relative you know. Relative to performance and relative among persons' perception of it. There is a ~630 liter FLH subwoofer posted on another forum that reaches no more than 108 Db SPL on 7 watts at driver Xmax, using a cheap 12" driver. Now that is a lot of cabinet in return for only ~108 Db SPL in my opinion. That big FLH is poorly purposed too, in that it has a crappy response and a compression ratio of 5.45:1 stressing its cheap ~$70 12" driver.

Think I'm crazy for building an even bigger FLH for only a 12" driver? Well maybe I am crazy but remember, crazy is relative too. You see, that ~630 liter horn above was reproduced SIX TIMES by the person. Thats right, he built 6 of those ~630 liter cabinets and installed them in his home theater, then built 5 more subs (using the same driver) of another dual-driver FLH design to fill in where those first 6 cabinets could not go. After all was said and done, all his sawdust and glue and wood and drivers didn't beat ANY of my BIGG FLH boxes posted within this thread, in all the areas of performance that I want from just one box.

Remember many of our perceptions are relative. Even if I'm certified as crazy by universe, that doesn't make me wrong, or any of you more right. We are just different and do different things for different reasons.

Keep your sticks sharp fellas. More Hornresp fun coming after some rest.

I'm the only one that mentioned size might be getting out of hand, and that's definitely not because I'm turning up my nose about it. I LIKE big. I posted a 22000+ liter sim, remember? Bigger is better.

I was a bit concerned that you wouldn't be able to move it, even if it's on wheels. If you want BIGGG make it big.

You have 1800 liters to work with, use it.
 
When this thread started i actually listened to the dark side of the moon album with one eye on the REW spectrum analyzer. The heartbeats you hear in "Speak To Me" and "Time" are really the lowest parts of the album, a 25Hz knee would cover it!

Dark.PNG

I did a lot of listening with REW, and even most of the new music doesn't dig down to 30Hz, I'm personally very happy with a knee of 35Hz.
I think you'd be very pleased with a higher sensitivity -3dB 25Hz box...

Keep in mind that the lower frequency, the harder it is to hear.
https://plot.ly/~mrlyule/16/equal-loudness-contours-iso-226-2003/

EDIT:
Doesn't mini dsp allow you to store different settings and switch on the fly? I don't have one, I don't know. If so you could program it to eq flat inside AND eq flat outside and just flick the switch for the appropriate environment.

I don't think you realize how imperceptible a little bit of rippled response is. A 3 db bump in response is barely noticeable. Maybe play with your eq a bit and check out what a slightly rippled response sounds like. It's really not a big deal. Whatever you are currently listening to probably has a much worse frequency response than you think.

The MiniDSP doesn't allow this as far as I know, unless they recently added a new feature to the software... What you could do, is configure 2 outputs for inside and outside and just switch the cable! I agree that 3dB ripple really isn't a big deal. I'd prefer a higher sensitivity rippled respons over a flat one. Eq'in the ripples down gives extra headroom 😉
 
Last edited:
Status
Not open for further replies.