2 hz tuning - pros and cons

I've never been able to afford high xmax drivers until just lately with the release of the SI HT 18. So I never really bothered taking the time to simulate the possibilities. As expected, this type of driver is not a good choice for high tuning. Even with the ~ average 16 hz tuning you can't really use all the available xmax without dramatically surpassing the rated power handling (which may not even be an issue since this driver is apparently able to handle much more power than rated). So I kept lowering the tuning until I could actually use all the available xmax but the port size requirements ended up unrealistic. So I tuned even lower, just for fun to see what would happen. This is what I came up with.



That's dual SI HT 18 drivers in a Gjallerhorn size (external) ported box tuned to 2 hz with 1000 watts applied. Above 16 hz the Gjallerhorn wins but below 16 hz the SI box wins. (Comparing SI simulation with Data-Bass Gjallerhorn measured results, so it's not completely fair or accurate but that's the way it looks.)

I was surprised to see that I could get by with a 2 inch diameter port! It's 90 inches long but who cares, resonances are clearly not an issue and the entire port only takes up 10 liters or so of the box volume. Obviously it will chuff - but only below 3 hz, and who cares about that? Besides, applying a roundover on the port will probably make it chuff-free down to 2 hz and below that chuffing really doesn't matter much.

This can be used without any high pass filter at all, although amp turn on thump at full power will ruin your day in a dramatic way if that's not taken into account. You can see on the excursion graph that with 1000 watts there's no excursion problem until 1.5 hz and down.

The only problem I see is in other simulation programs, they show higher velocity and the velocity peak at ~ 75 hz might cause audible chuffing.

CONS
- large size (although just like a sealed box you can make it smaller and eq to taste)
- large excursion requirements (the high xmax, low cost driver takes care of that issue)
- it will chuff below 3hz (possibly 2hz depending on effectiveness of port roundover), and possibly at 75 hz also

PROS
- I see lots when I look at the graphs above (the blue line in each graph is IB response for the dual drivers for comparison purposes)

I should probably mention that even without a high pass filter, most amps are not flat to DC, so 2 hz tuning is not necessarily required to get the advantages shown here, even 5 hz tuning might be passable.

Opinions? Comments? Criticisms? I've never tried anything like this before so I have no idea if I might be overlooking something.
Edited by diy speaker guy - 12/9/12 at 12:36pm

You mean like thunder? I have real life experience.

I'm not sure what your point is. That it will make me sick? That it will break my house? I'm aware those things could happen if abused but this is not a thread about human sensitivity to ultra low frequency, it's about seeing what can be done and finding out if I've overlooked anything in the simulation.

Like real thunder, I'm not planning to exaggerate the LF effect, just make it real in the short bursts that it appear in movie soundtracks.

I've never seen anyone argue AGAINST lower tuning (as long as there's adequate headroom through the whole frequency range) especially here in this forum, so again, I'm not sure what your point is. Especially since this is essentially the same thing (wrt frequency response) as sealed with eq but more output potential per driver.

The blue line in each graph is IB response for the drivers. Small sealed data for this driver is at Data -Bass.com.

So yeah, I've compared. You can do the same thing with small sealed boxes but it takes a lot more of them and a lot more power.

It cuts excursion in half down near tuning. If that's a nonexistent reward I guess I'm on the wrong forum.

Everybody was so excited about the fan sub with 3 hz capability. That thing needs a massive vent (bigger than the proposed design in post 1 for the vent alone) with stuffing and a fold to cut down on fan noise, and it needs a whole other room behind it to operate properly. Now I'm showing basically the same thing with regular ported boxes and 3 hz tuning is no longer cool or useful?

I'm completely bewildered by ANY comments saying low tuning isn't worth it. This is not a conversation about the value of being able to play 2 hz or the consequences of doing so, I can start another thread for that if you like. It's specifically about this design routine in which you can use what would regularly be considered a massively undersized port to achieve unbelievably low tuning with no response ripple, no need for a high pass filter, no excursion problems or chuffing until below 2 hz, double the spl with half the power vs sealed. In a 10 hz tuned box with same drivers, the port has to be unrealistically large and you can't use all the available xmax within the rated power handling.

All I'm trying to do is discuss the pros and cons of this design routine (not the usefullness or consequences of ultra low frequency reproduction or the pros and cons of different design routines with different tuning).
Edited by diy speaker guy - 12/9/12 at 2:08pm

Is there something specifically wrong with the simulation or is "not the best idea" personal preference? 2 of these (4 drivers total) should match (and probably beat) Gjallerhorn on the top end, but effectively extend the low end down to almost DC, and I think 4 of the SI are probably still cheaper than the Gjallerhorn driver.

You can tune these drivers higher, 10 or 15 hz, but as I mentioned you can't use the full xmax within rated power that way. These drivers can take more power than rated, that much is clear from the data bass testing. But how much more with clean uncompressed output? And how big are you prepared to make the port? How much response ripple can you put up with? (Large ports make large ripples.) How much of the available xmax are you comfortable wasting if it can't handle all this power uncompressed thermally?


True, but there are some, and from the reviews of people who own a fan sub, that 1/2 second of bass is well worth $13000. Dub King (a member here) makes music with 5 hz and maybe even lower IIRC. There IS media with content that low and there's more of it being produced every day. People are buying fan subs. Are they ignorant of something or just plain stupid? More money than brains? I don't know but there's at least one member here that owns one so we can easily ask him.
Edited by diy speaker guy - 12/9/12 at 3:03pm

This isn't going the way I hoped. No one has commented on the simulation at all, it's not even clear if anyone read the post. All the comments have responded directly to the thread title, which has no context without the simulations. So I'll try this a different way and if this doesn't work I give up.

If the goal is to make a sub with the lowest possible FR that extends well over the highest crossover point you would ever need with no ripple (or at least minimal ripple) and no chuffing in a reasonably sized box (Gjallerhorn size at most), with a minimal amount of power, minimal power compression, and get the most spl possible within a low budget, is this a good way to do it or is there something wrong with the simulation?

Please stop telling me my goals are wrong and discuss the simulation.

Thank you very much for looking over the simulations.

It doesn't look like an issue at all in MJK's mathcad sheets (as shown) but if you model the same thing in WinISD (with port resonances turned on) that little blip at 75 hz turns into a huge chuffmaster spike. Not sure if the difference is due to damping (stuffing applied in my simulation but not in WinISD) or internal losses being calculated differently between the 2 programs or something else, but the WinISD equivalent sim makes me nervous.

(It should be noted that there is a bit of ripple in my sim, if I zoom in much closer than shown (shown at 7db/division) you can see it better. But it's not bad ripple by any means.)

You can see the difference the port makes in the graphs I showed in post 1, the red line is my design, the blue line is the same amount of the same drivers with the same power applied in IB. You can see the IB skyrockets in excursion whereas the ported design does not, so the port is contributing it's share just like any port. (As shown in the FR graph at 1000 watts, the IB is actually more than 2x past it's rated xmax limit down near tuning, so the IB can't actually go tht loud, but the response shape is still correct. In contrast, my design doesn't pass xmax until below 2 hz at the power level shown due to the port.)

How about spl plots? If you are referring to my design, that's graph #1 in post #1. If you are referring to a fan sub, I don't have any and that's a tricky proposition at best. No one (to my knowledge) has ever measured one outside of a room, so the measured responses I've seen have as much (if not more) to do with the room than the sub itself. So there's no real way to know how loud the fan would be in 2 pi, but I'm guessing not very loud - at least nothing close to even approaching the in room measurements, especially at 3 hz. At the same time, I'm not trying to say my design beats (or even keeps up to) a fan sub 1 on 1, I was just pointing out the similar tuning. But to be honest, I don't think my design would do as bad as you might think against a fan sub, due to room gain not being shown in my graph vs room gain being included in EVERY measurement of a fan sub I've ever seen.
Edited by diy speaker guy - 12/9/12 at 4:22pm

THIS is the stuff I wanted to talk about. You don't think the simulation is accurate. Fine by me, and I'd love to learn from this. What's the useful difference between port velocity (which I've shown in post #1, graph #3) and port compression for the purposes of this discussion, and which program simulates port compression (as opposed to velocity)?

I've never seen anyone mention that vent velocity simulations were inaccurate to the point of reverting to a sealed box when the simulation predicts 10 m/s or less at full power (which my simulation does until below 5 hz). Can we talk about this?

(Also, assuming I raised it to a 15 hz tuning, also at 10 m/s, do you still think the velocity prediction would be inaccurate?)
Edited by diy speaker guy - 12/9/12 at 4:56pm

Yeah, I'll agree with that too if someone is willing to buy me the extra drivers and amps and rewire my house to power it all. In fact I'd prefer that design method greatly.

It all depends on your goals I guess, I'm personally not the least bit interested in sealed boxes for bass (unless somebody else wants to pay for it all). I actually like big boxes (within reason) and I LOVE spl on a budget.

To be clear, I'm not necessarily going to build this, I can't even afford it unless I sell a bunch of stuff, but now that these high xmax drivers are in my price range I'm getting enthusiastic. This is an interesting intellectual challenge though, and the simulation looks good enough that I might actually try to fund this and do it. I'm sure VISA won't have any problem funding it but I'm trying to avoid that route.

And like I said (3 or 4 times now) the higher tuning ported boxes with this driver will not reach xmax at rated power handling. Does NO ONE care about that? Or is it not an issue?
Edited by diy speaker guy - 12/9/12 at 5:11pm

That's the second Kryto reference so to be clear there's not much similar in this idea and what he was doing, not in any way except tuning. He was the ultimate basshead, I just want flat, extended response.

I have a lifelong admiration for his overwhelming passion and enthusiasm but I don't really like all the products he chose, the amount he paid for all of it, or the way he used them both in terms of design and frequency response. (IIRC he liked his subs to have flat response, but 50 db or so hotter than the mids.)

I did enjoy his posts though.

1 db of difference compared to what? Are you comparing dual sealed MaelX 18 against my design? If so check the excursion difference at rated power handling. (If not I have no idea what you are comparing.)
Edited by diy speaker guy - 12/9/12 at 5:27pm

Because it's an expensive driver and I detest wasting xmax. Why buy a high xmax driver and waste available xmax? This makes no sense to me at all, it's bad design. Why not tune lower instead of just wasting it?

A few posts back I was talking about adding more power than the rated power limit. How much are you comfortable with? 3000 watts? If the driver can do that without any compression at all that changes my outlook. I'm not dead set on 2 hz tuning but I'll be cold and dead before I design a sub that wastes available xmax.

Assuming the xmax spec is based on a Klippel report (and not a number made up by the marketing dept, which most are) I agree with you. But I don't have any problem approaching xmax until it sounds bad and then backing off a bit.
[/quote]
Edited by diy speaker guy - 12/9/12 at 5:46pm

You are right of course, but from a different perspective your design method is wasting usable output lower in frequency (and has more response ripple, more power compression, etc, etc, etc). Again, it's all about goals.

Yes, I could use a lower cost driver but that is not the point of this excercise. I started out simulating THIS driver to see what it would take to extract it's full potential.

I thought this was clear, but if not it should be mentioned. ALL the graphs in post 1 are shown at 1000 watts, including the port velocity graph, which shows 10 m/s or less until you get below 5 hz, and that's with no roundover on the port. (Still waiting for Mark's comments on this.)

This is true, I can't argue with it and I don't want to. But on the other hand on some material the extra distortion is not objectionable compared to the enjoyment of turning it up as far as it will go. Some drivers sound reasonable even past xmax on some material. I want xmax to be the limit, not the design itself. But this is not the point.

I can't afford 16 TC Sounds drivers like some of you lunatics can. Even if I could I have a moral objection to throwing money at problems when reasonable solutions can be attained for a fraction of the price. I would LOVE to have 32 or even 99 drivers and keep my highest excursion demands to under 1 mm, I'll set it all up and provide pics if anyone wants to donate to the cause.

Everyone keeps saying tune higher and use more power with no regard for thermal compression. I've asked several times how much power you guys are comfortable with pushing into this driver and no one wants to talk about any of that. So the options I've been presented are -
1. Tune higher and use more (undefined amount) of power (while smiling and accepting the increased ripply response and power compression).
2. Go sealed and buy a whole bunch of drivers and amps and rewire my house rather than build a ported box that does the same spl with 1/2 the drivers and 1/2 the power.

I don't really care for either of those options unless someone else wants to pay for it.

Forget sensitivity, look at excursion limited max spl in sealed vs ported (use 2 hz tuning for the ported).
Edited by diy speaker guy - 12/9/12 at 6:31pm