Skramlike sub: some design questions.

[Droogne]

To fit my room and driver, I'm going to make a sub just like the skram, using a 21" Lavoce San 214.50. It will be 60cm(wide)x68cm(high)x62cm(deep).

I was however still wondering how much fronthorn I need. I tried to look at the freq. response and horn mouth velocity. Would it be reasonable to model my sub so the top velocity (in the 30-80hz region) of the Lavoce, excursion limited, is the same as the Skram (or Skhorn) using an IPAL21. The ipal produces twice the air (+3dB), so the horn also has to be twice as big. 

Anything else I need to look at when choosing the front loaded horn? 

 

I included the modelling from both a version with a sloped second part of the horn and without. I would 'not slope' it because of construction reasons. It does look like it helps tremendously with reducing velocity reduction. Just not sure if it is needed. 

 

UNSLOPED

 

 

SLOPED

[Droogne]

More importantly, the pressure/velocity at the offset throat. I have it at 20m/s right now, twice what the Skram/IPAL21 combination has. Should I be concerned, or should this also work? 

[peniku8]

I don't know much about particle velocity, except the basic fact that it causes compression at some point, and neither have I run compression sweeps of my cabs yet (waiting for better weather), but while looking at your hornresp parameters, I noticed that you haven't added the driver and front "chamber" volume anywhere. Normally you'd be using the driver's front air volume plus the small section with the thickness of the plywood you're using (was about 11000cc VTC with the IPALs iirc), but as to @Ricci's suggestion and my own comparisons with my finished cab, adding the driver's cross sectional area to S2 in horn resp results in a sim much closer to the real thing. For the 21" drivers you can add about 350cm² to S2 (driver+wood when back mounting).

[Droogne]

12 hours ago, peniku8 said:

I don't know much about particle velocity, except the basic fact that it causes compression at some point, and neither have I run compression sweeps of my cabs yet (waiting for better weather), but while looking at your hornresp parameters, I noticed that you haven't added the driver and front "chamber" volume anywhere.

 

Quote

Normally you'd be using the driver's front air volume plus the small section with the thickness of the plywood you're using (was about 11000cc VTC with the IPALs iirc), but as to @Ricci's suggestion and my own comparisons with my finished cab, adding the driver's cross sectional area to S2 in horn resp results in a sim much closer to the real thing. For the 21" drivers you can add about 350cm² to S2 (driver+wood when back mounting).

Whoa, didnt think about that, thanks. So like this (modelling with the straight end horn)? I added 350cm to S2 and kept the rest the same. 

 

[peniku8]

1 hour ago, Droogne said:

 

Whoa, didnt think about that, thanks. So like this (modelling with the straight end horn)? I added 350cm to S2 and kept the rest the same. 

 

Exactly. Btw, you don't have to fill out all 4 sections. You can input S45 data into S34 and leave S45 off (all values 0).
What angle does your horn have? With your values it almost looks like the offset part is angled at some 20°, is that what you're trying to do? From the parameters it looks like you're doing something like firing into the back at quite an angle with a straight section running along a side wall/bottom to the front. Quite ambitious packing that much of a front chamber and a 96cm long port into a space this small. Are you sure the back chamber volume is correct? Looks like a bit too much for a cab of the dimensions mentioned earlier.

I also noticed that the quarter space sim (plus baffle gain and -12db) matches my real measurement much better. I multiplied the Hornresp output with the calculated half space baffle gain in Edge and basically got the same voltage sensitivity (-12db) curve as I measured, which was really satisfying. The driver did get a healthy 4 hour break-in period and I've used Ricci's semi-inductance parameters.

[peniku8]

Sorry, the sensitivity in hornresp quarter space (plus baffle gain) is 12db louder than measured 2V 2m half space, not 6db.

[Droogne]

20 hours ago, peniku8 said:

Exactly. Btw, you don't have to fill out all 4 sections. You can input S45 data into S34 and leave S45 off (all values 0).

Oh I know, thats just a pragmatic approach to be able to switch from the 'sloped' to the 'straight' endhorn (as it would jump from eg. 500cm² to 700² around the corner).

What angle does your horn have? With your values it almost looks like the offset part is angled at some 20°, is that what you're trying to do?

Not sure about the angle, and not looking at it too. I'm just modelling and seeing how it influences the FR and horn velocities etc; That is why I thought it best to start this topic, to see if I'm not missing some crucial things.

From the parameters it looks like you're doing something like firing into the back at quite an angle with a straight section running along a side wall/bottom to the front. Quite ambitious packing that much of a front chamber and a 96cm long port into a space this small.
The front horn is ~ the length of the back and depth, so ~54cm + 60cm. I did do some changes to the port, no worries.. I have taken it into account, and I'm also only at 76 now (of which 5cm are not the port, but rather added because it's slotted and the port has a height of 10cm).

Are you sure the back chamber volume is correct? Looks like a bit too much for a cab of the dimensions mentioned earlier.

Yeah, I did make some changes between those set of measurement and the listed HR specs; so youre right. However, chamber volume is also less of a problem, I can just add some height. Thats not the length I'm concerned about. I want to keep the footprint as smalls as possible, but trying to fit in the 21" depth + at least one fold in the port requires some space. I'm playing with it to see if I cant reduce the depth. Hoping to keep it between 60x60 (depth x width).

I also noticed that the quarter space sim (plus baffle gain and -12db) matches my real measurement much better. I multiplied the Hornresp output with the calculated half space baffle gain in Edge and basically got the same voltage sensitivity (-12db) curve as I measured, which was really satisfying. The driver did get a healthy 4 hour break-in period and I've used Ricci's semi-inductance parameters.

Ok, sorry I'm not completely following how to apply. You mean I should model in quarter space, and subtract 12?

 

Thanks in any case, it also looks like I can get away with 2 straight front-horn parts (no slope) if I need to. The added 350cm² are what is necessary, as S2 seems to be the most crucial value for the response. A slope helps with the velocities, but I can already achieve decent values (equal to those in the Skram/Skhorn) without doing so. I'm still figuring out if I trust myself to do accurate angles. My woodworking skills arent all there, and my tools are also not the best. Also, my first cabin is a test, so trying out the straigt horn first, and maybe using a sloped in the 'upgraded' version might be the safest bet to step up in the design.

[Droogne]

 

@peniku8 @Ricci

 

Concerning the front loaded horn, is it a smart idea to build it like this? The Skram has  2 parts/1 curve in front of the driver, my design only the 1 part. My main concern is that the panel does not completely covers the driver, however the first panel in front of the Skram doesnt either (blue arrow). I thought I could maybe have the driver fire through a 'port'(like in a synergy), with a ratio of maybe 1:2 . Red is the 'port' and yellow is the mounting plate partially covering the driver.

 

The hight is not definite yet. I might do the modular design (using a second part to ad 60-90L depending on what I need), or just use an elongated 1-part cabin.

The design can be summarised as an attempt to keep the footprint between 60x60cm (max 63x63cm), with similar modellings as the Skram. However, because I will be using the cheaper drivers (LaVoce etc) I can also reduce the size of the horn and ports by around 30-50% . 

Any thoughts? If I finalise everything I'll start building somewhere next week.

 

Maarten

[peniku8]

I think in your case, a direct radiator approach would be more fitting. I don‘t really get what you‘re trying to achieve with the front chamber, so I can‘t really give any suggestions.

Either way, firing the driver into the front panel might not be ideal. This way the panel distortion will maximize forward, probably the direction you‘re sitting in (if it‘s for a HT?). You could try throwing it around so you fire into the side or the floor to mitigate that issue a little.

If it‘s for a HT, I don‘t think you‘ll need the mid bass sensitivity a front loading will provide. With a cab that size you can expect like 3-4db baffle gain from 100Hz and up.

With a back chamber of this size, low bass sensitivity won‘t be anywhere near the mid bass sensitivity anyways. 

[Ricci]

I just skimmed this thread trying to catch up on everything I've missed in the last couple weeks, but I have to echo Peniku8 for the most part.

I finally finished and posted my "small" 21" design earlier today. It is still significantly bigger than the dimensions you have proposed. Have a look at how the front section was done. I don't think you can shrink it any smaller than that without major performance impacts. The simple slot that was posted above by Peniku8 is the other option for keeping it as small as possible but has some limitations also. The back chamber and ports I could shrink some more but only enough to get another few inches off of one dimension of the cabinet and that comes with compromises. The other option is to increase the tuning frequency. 30Hz + 21" driver starts to limit how far things can be shrunk.

As a rule of thumb you will lose about 25 to 30% of your gross volume to bracing, cabinet walls, driver, wiring, etc...

60cm(wide)x68cm(high)x62cm(deep = 253L external volume for the subwoofer.

I loaded your simulation data into HR and it says the system volume is 265L (Look under schematic diagram). Based on having 253L external volume available and factoring in the minimum 25% or so that you will lose to driver, bracing etc...results in about 190L best case for your subs net volume based on those outer dimensions. That's a long way off. My "small" CKRAM sub is 61 x 71.1 x 76.2 cm (330L) and results in a net volume of 257L.

As far as air velocities go the lower the better of course. Port velocities almost always get high. This is almost always compromised. The horn or slot section can be kept much lower usually. Port compression starts kicking in a lot earlier than most people think. There's already some happening by 10ms. It gets progressively worse from there as does air noises. For the horn / slot section I try to stay below 15ms worst case with the amp I'll be using and <10ms is preferable. With the bass section ports this is almost impossible. I just try to keep it as low as I can, but it's always way more than ideal. 

I'd have to agree that you'll need to increase the cabinet size significantly or stick with a regular vented cab design. The cab size you've proposed will be tight even for a regular vented option tuned to 30Hz. 

I'm not trying to discourage you at all! Just trying to help you out. 

[Droogne]

@Ricci

I'm gonna answer in full tomorrow, but in any case. Its more about keeping the footprint between 60x60 than anything else. The mentioned height is not really correct, as the volume will be what it needs to be, the height will be adjusted accordingly. As I'm still thinking about making it a modular sub, consisting of 2 parts, the horn and ports need to fit in this 60x60x70-80cm cube. I'll add a second cube to reach the required volume. That way I'll have 2 cubes which can be easily transported. In total it will probably have to be 60x60x110-120 and I'm OK with that. 

However, thanks for the 20-30% rule of thumb, I did already redacted around 40L in my calculations, but that might not be enough. Also thanks for the notes on velocity. I was of course planning on keeping them as low as possible, but I want to use the ipal/skram as a bare minimum. 

@peniku8 yes a horn this short won't matter really, but I was more interested in the plausibility/structural strength as that can't be modelled in HR. A regular ported might be the best option if I can't find a way around my 60x60cm issue. 

I'll do some more tinkering tomorrow and post some other ideas I had, but now with the correct sizes and volumes so you can actually evaluate my idea. 

 

Thanks for thinking with me guys! 

 

 

[Droogne]

18 hours ago, Ricci said:

I just skimmed this thread trying to catch up on everything I've missed in the last couple weeks, but I have to echo Peniku8 for the most part.

I finally finished and posted my "small" 21" design earlier today. It is still significantly bigger than the dimensions you have proposed. Have a look at how the front section was done. I don't think you can shrink it any smaller than that without major performance impacts. The simple slot that was posted above by Peniku8 is the other option for keeping it as small as possible but has some limitations also. The back chamber and ports I could shrink some more but only enough to get another few inches off of one dimension of the cabinet and that comes with compromises. The other option is to increase the tuning frequency. 30Hz + 21" driver starts to limit how far things can be shrunk.

I just read the CKRAM design thread, exactly the thread I needed. It battles with a lot of my issues. 'The physical size of the driver was setting limits on the size of the slots and the air speeds due to the reduced area were getting higher than I would like. ' in particular. Gonna see if I cant modify the Skram approach to fit my 60x60cm footprint. I'll need to add some height, so around 60x60x92cm. That I can manage!

 

[Droogne]

Another issue I was wondering about (in some deisgns) and I want to bring up is the distance required behind the driver. Is 2cm enough or should I allow more/less?

[peniku8]

2cm is nothing I‘d be comfortable with. The air speeds through the pole vents can get pretty high, so I‘d leave at least 5cm of clearance behind the driver.

If height is not an issue, you could look at the Devastator design on AVS. I think it might fit into your dimensions, or be easily adjustable to do so

[Droogne]

@Ricci @peniku8

Here is a drawing of how the elongated sub would look like. It's not 100% correct, as I changed the height to 100cm (is drawn as 90cm) and I didnt incorporate panel thickness (but I did keep it in mind). The driver would end up at the red line, not where it is now. The red line is not a panel, but rather the trajectory parallel to the mounting plate, as to make sure the hatch/trajectory is large enough to mount the driver in it. I also increased the depth with 5cm (so an additional 5cm behind the driver).

I decided 65 cm might still be possible, so the the sub would end up being 60x65cm (either depth or width, not sure which one yet) x 100cm for a total internal volume of 334L, from which I reduced 45 liter (10 for the driver, 10 for the bracing, 25 calculated for the inner pannels) for a net of 289L. 

 

Hornmouth velocities get high, but they will be smoothed out massively by EQing flat the evident peaks in the FR.

 

 

 

 

[peniku8]

I'd suggest moving S3 to where the driver ends. I think that will make for a closer representation of the actual cab.

[Ricci]

Droogne,

Since you have additional height you can work with, that helps immensely. 60x65x100cm should be enough for a hybrid 6th or whatever these are called lately. Keep your HR sim at about 285L and you should be in the ballpark. 

Keep in mind that air volume is limited and any volume taken up by the front section has to come from the back section and the ports.

A bigger front section= less low bass / higher vent velocities, but can produce more upper and middle bandwidth output and lower air velocity in the front section

A bigger ported chamber and ports is the exact opposite = More low bass, lower vent velocity, less upper and middle bandwidth gain from the front section, higher front section velocity. 

Trying to find a good balance in smaller cabs can be a PITA, but it's possible. 

 

I'd recommend using all 5 horn sections. Even if it doesn't seem like the design needs to at first. Sometimes the extra section helps. 

Leave S1 where you have it. 

S2 is also good. 320-350cm is typical for the cross section of area at the center of a back mounted 21" pro woofer. Add that into your S2 like recommended earlier. 

S3 move to the end of the driver cutout like Peniku8 suggested. In a way this treats the air inside of the driver cone profile as part of the horn. It helps match the measured performance closer. You'll often see that there is a negative expansion from S2 to S3 because of it. If you had went from a larger S2 (after factoring in the 350cm in the driver cone) all the way out to your current S3 it would have made the apparent volume of that horn section much larger than it actually would be and the sim would reflect that. 

S4 I'd move this to your current S3

S5 Add this in for the last section from the current S3 location to the mouth. This way if you wanted to you could have a different angle on the 4th section versus the 3rd section. 

 

Triple check your hatch ideas and clearances for the driver. Consider how difficult it will be to get the driver in or out and how big of a problem it will be if there is an issue with the mounting bolt that is in the worst spot. Give yourself a little extra room. Where would your hatch be on the design above? Double side hatches? What if you wanted to switch to different drivers someday? Is there room for a driver that may be a few cm deeper? 

Also consider how prone to resonance or vibration the area around the driver will be. Hatches, etc...That outer panel in front of the cone on your diagram will be taking a hammering. How will this area be braced effectively while still clearing maximum driver excursion? One good way with this type of layout is to sit the outer panel in front of the driver on the ground. Not sure there is room for that in a 60-65cm foot print without getting fancy though. 

[Ricci]

Have you considered an arrangement similar to this? 

 

 

[Droogne]

32 minutes ago, Ricci said:

Droogne,

Since you have additional height you can work with, that helps immensely. 60x65x100cm should be enough for a hybrid 6th or whatever these are called lately. Keep your HR sim at about 285L and you should be in the ballpark. 

Keep in mind that air volume is limited and any volume taken up by the front section has to come from the back section and the ports.

A bigger front section= less low bass / higher vent velocities, but can produce more upper and middle bandwidth output and lower air velocity in the front section

A bigger ported chamber and ports is the exact opposite = More low bass, lower vent velocity, less upper and middle bandwidth gain from the front section, higher front section velocity. 

Trying to find a good balance in smaller cabs can be a PITA, but it's possible. 

Yes, when EQing out the peaks I get an acceptable 10-12m/s. However, before finishing the design I'll be changing the design some more to strike the correct balance. I will be using them for PA too, and I need them to be decent in the 100hz region too, so I cant go all out on the LF, despite the fact I will also be using them in my home. Because of this dual use I need to consider the size, so I can move them up and down staircases etc. I'm still very young (24) and will be moving around a lot because of my medical residencies I will soon be starting. I have no idea about that size restrictions I will have to consider once I move. 

However, I also need to be realistic, as my main speakers are absolutely huge. Way bigger than 60x60x100.. I have no idea how I will move those, but having subs I can already move would be a start. My mains are 110x80x60cm synergies (with the one 12P80Fe driver in it for now). 

 

Also, I did notice the 'negative expansion', so I just removed the +350cm² and removed the difference from my calculation (the schematic diagram is actually without this +350cm²). Doing it like you suggest does make a lot more sense modelling wise, thanks. 

32 minutes ago, Ricci said:

S4 I'd move this to your current S3

S5 Add this in for the last section from the current S3 location to the mouth. This way if you wanted to you could have a different angle on the 4th section versus the 3rd section. 

 

Triple check your hatch ideas and clearances for the driver. Consider how difficult it will be to get the driver in or out and how big of a problem it will be if there is an issue with the mounting bolt that is in the worst spot. Give yourself a little extra room. Where would your hatch be on the design above? Double side hatches? What if you wanted to switch to different drivers someday? Is there room for a driver that may be a few cm deeper? 

I drew that red line to show I could do a hatch in the top pannel, but a side pannel might be not be a bad idea. I did check the depth of several drivers, with 26cm being the deepest. I will use this value to determine the hatch size etc.

32 minutes ago, Ricci said:

Also consider how prone to resonance or vibration the area around the driver will be. Hatches, etc...That outer panel in front of the cone on your diagram will be taking a hammering. How will this area be braced effectively while still clearing maximum driver excursion? One good way with this type of layout is to sit the outer panel in front of the driver on the ground. Not sure there is room for that in a 60-65cm foot print without getting fancy though. 

What do you mean by 'sitting on the ground'? Also, would using a 'double' plate be usefull there? On top of Skram-like bracing etc?

[Droogne]

15 minutes ago, Ricci said:

Have you considered an arrangement similar to this? 

 

 

No. I have tried a lot of different configurations, but I always had restrictions which limited my options. However, by doing more and more designing I have left most of those restrictions behind when realising it's just not reasonable (eg. my 60x60 size restriction which I increased to 65x65).

This design looks something worth checking out! What about the vertical placement of the driver?

[Ricci]

On 3/6/2020 at 4:08 PM, Droogne said:

This design looks something worth checking out! What about the vertical placement of the driver?

That's one of the tradeoffs of this arrangement. Could always flip them once a year. 

[peniku8]

48 minutes ago, Ricci said:

That's one of the tradeoffs of this arrangement. Could always flip them once a year. 

Do you really think it's an issue with the modern pro drivers? Most pro style cabs are simple vented direct radiators, which are stored on the wheels, so the driver ends up being in a horizontal position most of the time. I've never seen any manufacturer talk about that. If it's really an issue, I think we'd see the advice of storing them differently more often.

[Droogne]

2 hours ago, Ricci said:

That's one of the tradeoffs of this arrangement. Could always flip them once a year. 

 

1 hour ago, peniku8 said:

Do you really think it's an issue with the modern pro drivers? Most pro style cabs are simple vented direct radiators, which are stored on the wheels, so the driver ends up being in a horizontal position most of the time. I've never seen any manufacturer talk about that. If it's really an issue, I think we'd see the advice of storing them differently more often.

True. I'm not gonna let it negatively influence my design choice in any case. It's a trade-off I can live with.

Today I bought myself 4x 21" drivers. A pair of 18sound NLW9600 and a pair of LW1400. A very good deal was the reason to go for it, especially the LW1400 which I wouldnt have picked for this kind of design. The NLW9600 models very similar to the LaVoces, and were only half the price and directly available (I should have to wait ~3months for the LAvoces if I want the best price). I'm probably gonna sell the NLW1400 off here in Belgium, except if I find another design that might suit me. Maybe a Skram-like sub tuned to 35hz.

For this design I'll be using the NLW9600, although I noticed I can get similar results with the NLW1400 if I take a 15mm Xmax in consideration (as mentioned by you, @Ricci, in the driver review/measurments) and enlarge the front-horn somewhat (primarily S2), taking in only 10L more. That way, the NLW1400 also achieves a 130db 0dB, but only down to 34hz instead of the 32hz of the NLW9600. 

 

QUESTION:

When playing with the front horn I wondered one thing, except the influence it has on velocities, is there another reason for starting with an S1 of ~200cm²? Could I, for example, use an S1 of "0" (ignoring the effect on velocitie and FR, both of which I can model).

[peniku8]

If you don‘t mind the short drive through Luxembourg, I could get you the 21DS115 for around 450 a piece or cheaper depending on how many you need. Just in case you want to get more nice drivers some time in the future

S1 doesn‘t have a huge impact on the models. After all, S1 itself is not even in the path between the driver and the horn mouth.

[Droogne]

1 hour ago, peniku8 said:

If you don‘t mind the short drive through Luxembourg, I could get you the 21DS115 for around 450 a piece or cheaper depending on how many you need. Just in case you want to get more nice drivers some time in the future

If I'd need brand new drivers, that would be a very good deal! Probably worth the drive and extra cost for the better driver. A more known driver also better keeps its value (something I always try to keep in mind).

However, I got my hand on those 4 drivers for only 800eu. At first I thought I'd defintely sell the LW1400, but the databass review over here speaks a lot of favourable words. Especially the note about the Xmax, which is rather in the area of 15, not 9,5mm.

I changed the front horn in my design, to compensate for its lesser power. Applied to the NLW9600 (or other more powerfull driver like the DS115, LaVoce,..) I know have a more powerfull 40-80hz range, but it doesnt negatively impact the 30-40hz so I'm okay with that.

1 hour ago, peniku8 said:

S1 doesn‘t have a huge impact on the models. After all, S1 itself is not even in the path between the driver and the horn mouth.

Ok, correct me if I'm wrong, but using an S1 like here should be OK? These are the results for both drivers. horn mouth velocity is well below 15 m/s as lang as I EQ out the peaks at 50 and 110hz. 

Also, I changed out the 65x65x100 to a 60x60x120cm layout. Just to be safe, and more importantly because my mains are 110cm wide. To stack them (side to side) I need some extra space. 

NLW9600

LW1400