Trying to model Imperial BLH with Hornresp - Is this right?

Status
Not open for further replies.
I'm fairly new to hornresp and this is my first attempt at modelling a back loaded horn. I'm not at all sure I am doing this right. I have an Imperial that I haven't used too much due to ragged response. I'm trying to determine a suitable driver compliment for use as a "subwoofer" (although it hardly goes low enough to call a sub). I'd love to be able to finally use it for something.

Here's how I broke the imperial into segments.

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


Next is the inputs. Some of these measurements may not be exact (I'm just guessing at 6 cubic feet for the back chamber and 5000 square cm for the baffle) but I can do more exact measurements later. This shows specs for 2 black widow 15's (a common driver to be used in such an enclosure) wired in series with 2.83V applied, corner loaded.

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


And the resulting schematic.

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


And finally the combined response of the horn mouth and driver diaphragm. It's pretty bleak, which is why I wonder if I am doing this right. It can do 120+db max spl.

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


And just for fun, here it is with 2 dayton quatro 15's. These real sub drivers are everything I have read is exactly WRONG for this enclosure, yet they seem to model fairly well, if you can excuse the very small bandwidth of about 1 octave. If you CAN see past the bandwidth issue, hornresp says max spl across this bandwidth is 135+db. I'm not expecting extreme sound quality out of this, but I'd rather have a 1 octave bass blaster than a huge unused box.

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


Am I doing this right? Does any of this make any sense at all? If so, I can proceed to take exact measurements and model more drivers to find something with hopefully a bit more bandwidth.
 
IMPERIAL HISTORY (In case you don't already know)

The Imperial comes in several flavors, all of which are large, around 30 cubic feet. Jensen developed at least 2 significantly different versions of Imperial in the 50's, possibly more than 2. The plans were provided for DIY'ers to make for drivers purchased by Jensen. The intended driver was a single 15 inch. The plans I have for the build in version are from a 1956 Jensen Technical Bulletin and there is also a considerably different cornerhorn version. Much more recently, Decware revived interest in the design with their variation, which increased the chamber size and called for dual 15's, often in stereo cabs, each fitted with a black widow and scorpion driver. They followed that up with a single driver subwoofer front loaded version with an open throat.

MY IMPERIAL

Mine was built using the 1956 plans but uses the full chamber size of the Decware version. It is constructed of wood 2 inches thick, so my chamber is slightly smaller than that, in order to keep the horn flare accurate to the plans. Other than that, mine is basically the same as most that were built in the 90's. I only have one, so I have no desire to run it full range. This is picture of it during a brief experiment trying it front loaded. Response was smoother, but the low end was not satisfying at all, with rising response and loud output well past 10 khz, not really a sub at all when front loaded. As you can see, the big fat horn mouth mocks me every day it sits unused.

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


It's been awhile since it's been used as a blh (as intended) but IIRC response was ragged and awful.

OTHER IMPERIALS

Depending where you look, there are lots of opinions on the Imperial. A quick search of this forum indicates that some of the most knowlegable members of this forum think it's a very decent cab, although they are undoubtedly talking about the original single 15 fullrange version. In other forums people owning Imperials with the Decware recommended driver compliment don't use them at all, describing the sound "colored", and others suggesting the only cure is that they be burned.

In addition to the original question (Am I doing this right?) I'd like to open this up to any opinions or experiences with the Imperial (specifically in the bass range), recommended drivers, etc.
 
Well, it's not the way I would break it down, instead finding the axial length centerline by mapping the expansion. I built it before I learned about T/S though, so the only spec I measured was a 46 Hz Fs, though it's motor size implies a fairly high Qts as does the large throat/filter chamber.

Anyway, the one I built decades ago performed quite well by the standards of the day with the P15-LS I still have, but then I had neither the source material nor electronics save a signal generator to plumb its nether regions, so basically all I can say is that it made kick drums, etc., 'slam' at ~ live SPLs, though I much preferred the dual 15" Altec ~70 Hz 35 Hz reflex vented FLHs that replaced it overall since these allowed a 500 Hz/2nd order XO point at high efficiency.

GM
 
Well, it's not the way I would break it down, instead finding the axial length centerline by mapping the expansion.

That makes sense and I'm not sure why I didn't do it that way before. I will draw it out later and try to get as close as I can and model it, but OTOH, assuming the throat area, mouth area and overall pathlength are accurate, relatively large changes to the actual horn flare (a few hundred cc's here and there) seems to make little or no recognizable difference. I will work on getting more accurate numbers, but in the meantime, assuming I'm doing this right, I think I've captured the imperial essense.

Thanks for your comments, GM.
 
I've been getting acquainted with hornresp and the imperial, checking performance of the different drivers and alignments I have tried in the past. Modelling shows that all the drivers I have tried so far have been wildly inappropriate (which explains the results I've been getting in real life). Modelling also shows the most current alignment (front loaded with no front chamber and slightly longer pathlength) was a monumentally stupid move.

As can be seen in the graphs above, it's difficult to get wide bandwidth flat response since the driver and horn mouth out of phase outputs are messing with each other. For this reason, it doesn't look like a blh is a good option at all.

By changing the model to a front loaded imperial horn with a front chamber, it's way easier to get satisfying results. Dual driver 15 inch black widow (as modelled above in blh), series wired, corner loaded, back chamber 155 l, front chamber 15 l. Everything else is the same as the blh model above.

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


And the schematic. A potential (huge) problem with this design is that it may be tricky to implement such a small front chamber due to the internal layout, but I suppose it could be done. Care must be taken to ensure the front chamber is not long and straight, or it will just appear to be an extension of the line length instead of a chamber. This model uses the same numbers as the models above wrt overall chamber size and horn flare, I have not attempted to take more accurate measurements yet.

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


And the frequency response. Not too bad and way more high end response than I will ever need. Looks to me like 35 - 400 db +/- 2 db at 109 db (only when corner loaded of course) with a max output of 135+db across this passband.

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


As an interesting side note, I also modelled the dual quatro as a front loaded horn. The chamber sizes have to be divided differently (it like a 100 l back chamber and a 80 l front chamber, which is probably much easier to implement than the black widow model above with the tiny front chamber). Surprisingly, it's even more sensitive than the black widow by a couple of db, probably due to the increased front chamber size of the quatro model. With 5x more excursion than the black widow, the quatro front loaded imperial is capable of massive output (145+db, although it takes 5000 watts to achieve, which will probably burn a couple of voice coils). It looks a lot like the front loaded black widow frequency response above, although bandwidth is limited to 35 - 100 hz. The high le of the quatro chokes off output above 90 hz. The quatro would also need a VERY steep rumble filter to protect the driver at it's 21 hz fs, so overall the quatro is probably not the best choice here, the black widow (with 5x less excursion) outperforms it overall.
 
Status
Not open for further replies.