| paulspencer |
Comparison of Adire Koda Mk1 (not sure of all params unfortunately) and Tang Band W8 740C.
The darker line is the Tang band and the grey is the Koda 8"
The Tangband has 150w but the Koda has 100w, which does not concern me as both have plenty of SPL
There will be some eq for the TB to get the peak down which will make excursion acceptable, but since the koda is already flat, I've given it less power; it would have a highpass to control excursion.
Which is better?
Keep in mind that I will eq with ultracurve
The koda is flat at the bottom end but is more difficult to eq on the top end.
Anyone have advice on what to look for with sims? I'm finding it a bit hard to know if what I see is a good thing or not! |
|
|
| Rademakers |
Some more questions first:
Can you post a view of the input window as well? A lot of information is in there. Furtermore, what's the Xmax of both the units, what is your intended frequency range (30ish?) and how are you most likely going to load the cabinet into space?
Do the dimensions mentioned (100 x 60 x) in the other topic apply to this horn?
Mvg Johan |
|
|
| simon5 |
Hey Paul !
I think you can do better than that, there's serious ripple in the passband, what's the frequency you want to use the horn? I'll try to do better for fun, maybe it's the driver making it hard?
Good luck! |
|
|
| paulspencer |
Johan, they both have an xmax of about 12mm
corner loaded (most likely)
40 - 120 Hz ideally
I don't really need lower than 40 since my subs take over down there, but it just so happens that they seem to get down to 30 Hz! I'd actually rather they were smaller, or that the response was smoother, or that they don't need corner loading, than have 30 Hz extension.
100 x 60 x 40 would be a good size I could live with, although doing some work on cad I found it hard to work with those dimensions! It seems easier to design when it is square. I want to avoid greater than 90 degree bends.
Simon,
If you have ideas how to get it flatter I'm all ears. However, I notice that even the lab horn isn't that smooth, and yet when I've tried modelling it, I could get it smoother, and Danley could have easily done that as well. So I ask why he didn't. Obviously there is more to it, so I'm trying to learn some of the rules of horn design. With sealed and vented boxes, we have a better idea of how the response should look, but in this case I'm not so sure. |
|
|
| simon5 |
Due to your size limits, I can't improve your design much.
To be flat ±3 dB, I need about 3.2 meters long and 6600 square centimeter of mouth hehe!
You'd get around 130 dB at 150W imput and stay under the 12 mm Xmax limit with a back chamber of 8 liters. |
|
|
| paulspencer |
Actually I'm getting it much flatter by starting with the Hypex designer and working backwards. It gives me a smaller throat and a larger mouth and slightly shorter horn. I can cut back on the mouth area, reducing it to 40 x 60cm while still getting reasonably flat.
s1 40
s2 2400
hyp 375
vrc 4.5L
When I work through the same process with the Koda driver, it now doesn't look as good, as it has a sharp and narrow peak in excursion around tuning, whereas it is easier to tame the excursion on the TB. |
|
|
| simon5 |
Well, to respect your old 1000x600x450 limit, I did s1:220 s2:2700 CON:250
You get your old ±5 dB response but only 3 peaks instead of dozen of them. It would fit in your enclosure, mouth would be 600x450.
VRC:8 LRC:10 TAL:0 FR:0 VTC:0 ATC:0
EDIT: Since you've got displacement left where dips are, you could even EQ the two dips up instead of EQing the three peaks down. |
|
|
| Rademakers |
Just some thoughts:
If you don't need 30 Hz, I wouldn't go there even if they can, as it's much easier to get a flat response in the given size for a 40 Hz horn. Especially when 1/8 th space isn't what you want ideally.
I think S1 could be made between 55 and 110 cm2 without problems for the cone, if it's stiff (enough). This could enhance the response much over a 220 cm2 throat.
The Labhorn was designed to be used in a stack of 4 or 6, so it doesn't needs to be flat on it's own. High powerhandling resulting in high SPL-levels and especially low distortion where the more important design-factors.
In the beginning of the horn it isn't problematic to have folds which have more than 90 degree bends. In the end it's prefered to have it smooth however. You could take a look at the Punisher design as it's a clever fold (like some of the Tuba's).
There are definitly speakers that make it easier to get a flat response.
Wkr Johan |
|
|
| MJK |
| From the plots, your horn looks more like a TL. The peaks in the SPL and the nulls in the displacement plots are probably good indicators of quarter wavelength standing waves. |
|
|
| boydon_lepasci |
| As my understanding of horns is based solely on reading please take the following with a grain of salt: have you considered whether both drivers have adequate BL properties for horn loading and have you planned for the increase in impedance that will likely effect available power? |
|
|
| slowmotion |
Paul,
Can you give me the parameters of you proposed drivers?
cheers ;) |
|
|
| paulspencer |
I've aimed for drivers with a high BL
No I haven't planned for the changed impedance as such, but I have found that when I simulate with the power I have in mind from a Behringer A500 studio amp with 230w into 4 ohms, the SPL is more than I need.
Tangband driver specs:
http://www.tb-speaker.com/detail/1208_03/w8-740c.htm
Unfortunately I don't have all the driver specs for the Koda, so there was a bit of guess work ... (probably a bad idea)
I'm now having a problem with the TB driver - shipping is US $163 for a US $20 driver! :smash:
I can get the RE audio 8" locally for a decent price but unfortunately they don't have all the TS specs I need!
http://www.advanceae.com.au/catalog...roducts_id=2895
I'm getting better results now in horn reponse. Seems a good idea is to use a pair of horns, stack them up tall to get mouth area up |
|
|
| slowmotion |
Paul,
I'll see if I can give it a shot in Hornnresponse this evening.
cheers ;) |
|
|
| Rademakers |
You only seem to be missing Le, which doesn't has to much influence in the intended frequency range. You can always try to ask the manufacturer.
You could also try to use the 100 x 60 as the mouth, which should be sufficient.
Wkr Johan Rademakers |
|
|
| slowmotion |
Hi Paul
Here's a starting point, at least.
I used the RE audio driver as an example.
This can be improved on, I'm sure.
I used 3 stacked horns in each corner.
cheers ;) |
|
|
| slowmotion |
| SPL response, 3 horns stacked: |
|
|
| slowmotion |
Of course this is for straight horns,
when you start folding them,
the reponse will suffer.
With that driver and your restrictions on
hornsize it is easier to make a midbasshorn,
for instance down to 80Hz.
You could also chose a driver with a higher Fs.
cheers ;) |
|
|
| paulspencer |
Jan,
3 horns stacked in each corner ... is that 6 horns?
I notice in the window you show parameters for only one driver, hence I'm a little confused what you are showing!
Could you clarify?
Also, the throat area is low, hence there is a compression ratio of 3.25. Is this a problem? I thought that the limit was 2 for most drivers? Doesn't this put a lot of stress on drivers? |
|
|
| slowmotion |
Hi Paul
| quote: | Originally posted by paulspencer
Jan,
3 horns stacked in each corner ... is that 6 horns? |
Yes, 6 horns all in all. 3 horns for each side. Might be easier to fold 3 smaller horns than 1 "big" horn, since you are pressed for space.
| quote: |
I notice in the window you show parameters for only one driver, hence I'm a little confused what you are showing! |
You first simulate 1 horn the normal way. This is then a ( too )
small horn with lots of ripple in the response. While looking at the response-curve in HornResponse you then go to "Tools" and chose "Multiple speakers". The responscurve I posted is 3 horns connected in series.
| quote: |
Also, the throat area is low, hence there is a compression ratio of 3.25. Is this a problem? I thought that the limit was 2 for most drivers? Doesn't this put a lot of stress on drivers? |
Yes, that could be a problem , depends on the driver and how much power you intend to use. If you want to play very loud you might chose a bigger throat area, but then you get a totally different horn. Smaller throat will usually give you better high frequency response, but will also usually help you down low. Everything is connected together: The driver, front compression chamber, compression ratio, back-chamber, and so on.
I have to say that I don't think the RE Audio driver is suited as
horn driver, the Fs is way too low, among other things.
I haven't looked at the other drivers you mentioned.
It pays to spend a bit of time in HornRasponse trying out different drivers, and different scenarios( sp? ) , even if it is not something you think you are going to use. You will learn a lot just trying different throat ratios or different compression-chambers and so on.
One way you could try, if you haven't allready, is to simulate
the horn you would like to build if you had more space, and then try different compromises to see what you feel you could live with in the long run.
cheers ;) |
|
|
| paulspencer |
Thanks, Jan, this is very helpful. What kind of parameters do you suggest I look for in a bass horn driver?
Can anyone suggest some good candidates?
The tangband looks good but shipping of that driver from Taiwan is 8x the cost of the driver!
My understanding is that TS parameters should look like this:
fs > 30 Hz
Qts ~ 0.35
BL high
xmax 12mm
Ideally it should be not too expensive in case it doesn't work out too well! |
|
|
| slowmotion |
Hi Paul.
| quote: | Originally posted by paulspencer
What kind of parameters do you suggest I look for in a bass horn driver? |
This is very dependent on what you want to do.
Also sometimes drivers that one normally doesn't think would work
still works pretty good, since most horns end up like compromises anyway.
Generally I like High Bl, alnico, underhung coil, stiff membrane,
Fs in the lower end of the passband.
Most important still is that the horn and driver can be made to work together.
The drivers I use are mostly modified to suit the intended horn.
Lots of trial and error. On both horn and driver ;)
| quote: |
Can anyone suggest some good candidates? |
Sorry, I don't know much about present offerings.
| quote: |
The tangband looks good but shipping of that driver from Taiwan is 8x the cost of the driver!
|
That sucks. Australia is such a big country, surely there must be suitable drivers available in the pro market?
| quote: |
My understanding is that TS parameters should look like this:
fs > 30 Hz
Qts ~ 0.35
BL high
xmax 12mm
Ideally it should be not too expensive in case it doesn't work out too well! |
Well, the drivers used in the Labhorn don't have typical horn-driver-parameters,
but are still work very good in that horn. In that light first find out what passband you want
to cover, and then a suitable driver and horn combination must be found.
I don't have any fast answers, unfortunately.
Maybe some of the more theorethically ( sp?) minded persons on this list can help you?
cheers ;) |
|
|
| BAM |
| That's exactly what it is. When the horn mouth is too small or too few are used together, it turns into a really underdamped waveguide instead of an acoustic impedance coupler. Try corner loading (modeling in 1/2pi space) and you will find that corners really do well to extend the mouths of horns. Perhaps the most interesting example of this is the TLAH from Bill Fitzmaurice, which mounts up in the corner and uses the corner itself as a really big flared waveguide. It uses something like nine NSB full range drivers from Parts Express. The most interesting of the TLAH applications was one person's install in a low-level sound reinforcement application for a small church. When covered with white grille cloth, they look like acoustic treatment panels. |
|
|
| Rademakers |
| quote: | | which mounts up in the corner and uses the corner itself as a really big flared waveguide |
Like a Klipsch horn?
| quote: | | Try corner loading (modeling in 1/2pi space) |
1/8 th ;)
Wkr Johan |
|
|
| mike.e |
Make the lower F3 40hz not any lower,thats just making a massive box.
Perhaps consider a locally available 10"? Here I plan on trying a parallel imported JBL GTO1204.JBL,kenwood,pioneer 10" and 12" car units with 'in between' Qts,low fs and high Xmax.
Attached : different moving mass. See ripple in band. |
|
|
| slowmotion |
Hi all
I agree with Mike that it would pay get get drivers locally,
better to spend hard earned money on quality drivers
than freight.
Mike's example shows what happens when horn and driver are not working properly together, allthough a 8" driver
with moving mass of 200g is not something one would realisticly use as a midhorn driver.
What brands of drivers can you get locally, Paul?
cheers ;) |
|
|
| paulspencer |
Hmmm, very interesting chart!
Thanks for the tip, I'll start looking a bit more into some local drivers ... perhaps even some peerless car subs ...
Jan,
It just so happens that overseas drivers imported can be cheaper at times than getting them via local distributors, or at times they are better value. I got my AV12s for about the cost of the Shiva bought locally.
I can get very good prices on Vifa, Peerless and Scan Speak. Can get Seas here for a bit more. RE audio, Adire. I haven't looked much into local suppliers of pro stuff, but I'd say most of the main brands I could get like Eminence, Beyma, JBL etc. A lot of them at PE are reasonably priced.
Why do you say an 8" isn't a good idea?
My main criteria for drivers would be cost, output, horn suitability (whatever that means), and would prefer the driver to have good distortion performance before horn loading. |
|
|
| slowmotion |
Hi Paul
| quote: | Originally posted by paulspencer
Jan,
I can get very good prices on Vifa, Peerless and Scan Speak. Can get Seas here for a bit more. RE audio, Adire. I haven't looked much into local suppliers of pro stuff, but I'd say most of the main brands I could get like Eminence, Beyma, JBL etc. A lot of them at PE are reasonably priced. |
Ahh , then you have a pretty good selection to chose from.
If I were you I would spend some time checking out drivers
and see what you can do with them in Hornresponse.
| quote: |
Why do you say an 8" isn't a good idea? |
Sorry, I was unclear. An 8" driver is a splendid idea for a midbasshorn. But not an 8" driver with MMD of 200g.
Max 15- 20g would be more suitable IMHO.
EDIT:
I would check out the pro sector if I were you .
A pro 8" or 10" driver usually have specs much better suited to
hornlading.
cheers ;) |
|
|
| slowmotion |
Hi all
For fun I did a quick test in HornResponse
with an Eminence Beta 8:
This is 1 driver in 1 cornerhorn.
cheers ;) |
|
|
| slowmotion |
Here's the response:
cheers ;) |
|
|
| mike.e |
Thats still a large horn slowmotion : Id consider one with half the mouth area,and correspondingly less path length.
The beta8 is probably quite cheap too!
Otherwise id go car audio 10" |
|
|
| paulspencer |
The biggest problem I'm having is IMO "compression ratio!" The designs seem to call for a very small throat, ie 10% of SD, which is a compression ratio of 10 right? As far as I'm aware, this is too high! Make it closer to SD and the response is no longer flat! :smash:
How high can I go in compression ratio?
What are the problems if I go too high? |
|
|
| slowmotion |
Hi
| quote: | Originally posted by mike.e
Thats still a large horn slowmotion : Id consider one with half the mouth area,and correspondingly less path length.
|
You'd get a smaller horn, but also more ripple.
| quote: | Originally posted by paulspencer
The biggest problem I'm having is IMO "compression ratio!" The designs seem to call for a very small throat, ie 10% of SD, which is a compression ratio of 10 right? As far as I'm aware, this is too high! Make it closer to SD and the response is no longer flat! : |
Huh?
SD on the Beta 8 is 214 and the throat in the example is 100. That's not what I would call high compression ratio.
cheers ;) |
|
|
| paulspencer |
Jan, I was referring more to the results I get from the "Hypex designer" feature. I tend to find better simulated results also from a high compression ratio. I'm trying to find out what my limits are here ...
even the example you gave has a compresssion ratio greater than 2 which I thought was considered a safe limit, unless I got it wrong. |
|
|
| slowmotion |
Hi
Paul, then I misunderstood ;)
Re the "Hypex designer" feature, well,
better to start from scratch when designing horns, I think....
cheers ;) |
|
|
| paulspencer |
| Jan, so how do you "start from scratch?" Do you have some rule of thumb guidelines to start out? |
|
|
| GM |
Greets!
10:1?!!!!! Altec compression drivers are this high and look at the construction/horn size required to get a 500 Hz/2nd order Fc. Extrapolating to 40 Hz would require a motor hoist to put in position and turn a large personal fortune into a small one to make.
In the 'good ol' days', horn drivers/lens were initially designed as theoretically ideal full size exponential since sound pressure is exponential (1/f) and if it was acoustically long enough it generated a smooth response over the desired BW with a second order roll off at both ends once reactance annulled. If it was too long or there wasn't a suitable driver or it had too much gain and/or BW for the app, a 'slice' of it was used that best met the performance goals with the desired driver(s), ergo the early truncated designs from Altec, etc..
Anyway, in today's 'techo world' Leach's math will give you your answers. Remember, any driver can be optimally horn loaded over some BW at some efficiency, so the main question is, does its specs meet the needs of the desired performance goals? If not, use his math to find the ideal driver and if it doesn't exist in this reality, then find the closest to it and sim how much it compromises your goals, etc.. At least you won't have to actually cobble together/measure an old chicken coop full of proofs-of-concepts like I did to get yours.
GM |
|
|
|
