Guest David H Posted October 23, 2012 Share Posted October 23, 2012 Here is this drivers graph without thehorn lens. With the horn lens installed the majority of the gain occurrs between 200hz and 1khz, here is a comparison. Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 23, 2012 Share Posted October 23, 2012 Here is a comparison on axis, 15, and 30 degrees off axis. Quote Link to comment Share on other sites More sharing options...
Rudy81 Posted October 23, 2012 Share Posted October 23, 2012 Overall those graphs look excellent to me. Hopefully they reflect the actual performance of your horn. If so, you seem to have a real winner! In my two way, I cross near 800Hz since my 2" driver can't get much below that. In your case, you should be able to mate that horn to any bass bin. Well done. Once you have a bass bin you will be able to judge the 'system' against other known speakers. The subjective evaluation will tell you a lot as well. Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 23, 2012 Share Posted October 23, 2012 Hopefully they reflect the actual performance of your horn I agree. Once you have a bass bin you will be able to judge the 'system' against other known speakers. The subjective evaluation will tell you a lot as well.Yes, measurements are only half the battle.In your case, you should be able to mate that horn to any bass bin. As long as the sensitivity is high enough, I really need a Khorn or Lascala local to test with.Dave Quote Link to comment Share on other sites More sharing options...
mark1101 Posted October 23, 2012 Share Posted October 23, 2012 Dave, It's really neat that you figured out how to design and build all these horns you do. I have a question about the measurements. All of your graphs seems to show the same suck-outs (dips) in the FR at various places along the spectrum. Do you know what that is? Have you seen this on other horms you have tested? I'm just wondering why there are so many dropouts. The horn doesn't look any different than other nice conicals I've seen. Could it be the test rig? Quote Link to comment Share on other sites More sharing options...
Rudy81 Posted October 23, 2012 Share Posted October 23, 2012 I would suspect that you will not have any problems with sensitivity. When I have built and tried my various horns, I always ended up having to attenuate the HF section to match any of the bass bins I tried, including Khorn and La Scala. Of course, that was done with 2" compression drivers. I will be very interested in getting your impressions of material above 15kHz. Great little project. There is quite a bit of interest out there for horns such as these. When I researched that type of setup, the Jabo and JLMC horn fans were more than I expected to find. Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 23, 2012 Share Posted October 23, 2012 All of your graphs seems to show the same suck-outs (dips) in the FR at various places along the spectrum. Do you know what that is? Have you seen this on other horms you have tested? I'm just wondering why there are so many dropouts. The horn doesn't look any different than other nice conicals I've seen. Could it be the test rig? It is possible that the test rig is causing them, or some sort of reflection. I really have no clue, but no matter what I do I still have them. I seem to remember having these same anomalies when testing JC's BatHorn. I also seem to recall JC suggested taking measurements at 3 meters instead of 1 to get the horns full contribution. The tests yesterday were done at 1 meter. This may be why the HF is so strong, however this has nothing to do with the suckouts on the graph. Dave Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 23, 2012 Share Posted October 23, 2012 I would suspect that you will not have any problems with sensitivity. When I have built and tried my various horns, I always ended up having to attenuate the HF section to match any of the bass bins I tried, including Khorn and La Scala. Of course, that was done with 2" compression drivers. The compression drivers you are using are rated at 108db+, these Dayton full range drivers are rated at 96db. Even with the added gain of the horn, I only estimate the combo 100+-2db. This may be enough to use with a KHorn, however I am a bit sceptical. I also found another driver I think would work well with this horn for budget applications, it has lower spl, but is 1/4 the cost of the Dayton. Quote Link to comment Share on other sites More sharing options...
tigerwoodKhorns Posted October 23, 2012 Share Posted October 23, 2012 All of your graphs seems to show the same suck-outs (dips) in the FR at various places along the spectrum. Do you know what that is? Have you seen this on other horms you have tested? I'm just wondering why there are so many dropouts. The horn doesn't look any different than other nice conicals I've seen. Could it be the test rig? It is possible that the test rig is causing them, or some sort of reflection. I really have no clue, but no matter what I do I still have them. I seem to remember having these same anomalies when testing JC's BatHorn. I also seem to recall JC suggested taking measurements at 3 meters instead of 1 to get the horns full contribution. The tests yesterday were done at 1 meter. This may be why the HF is so strong, however this has nothing to do with the suckouts on the graph. Dave You have probably already seen this, but Edgar did this about 30 years ago and had spikes and dips on his square horn. He added a 1/2" gap between the horn and the speaker and then put in foam strips that smoothed out the response. His horn was square, so maybe the transition was more difficult. He discusses it on page 9 of the attached PDF (marked as page 15 in the Speaker Builder article). Edgar Midhorn.pdf Quote Link to comment Share on other sites More sharing options...
Rudy81 Posted October 23, 2012 Share Posted October 23, 2012 I'm sure its not related, but when I built my DIY set of MDF horns, I too had a few dips that I could not explain. I suspected it was due to flaws in symmetry due to my imprecise sanding. When I purchased the machine made Jabo horns, the measurements no longer showed any dips. That was one reason I decided to go with fiberglass horns vs. my DIY models. I don't have the CNC capability, so it was a true DIY project....imperfect at best. Since my bass bins can go much higher than a Khorn bin, I was fine with the 2" driver setup. Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 23, 2012 Share Posted October 23, 2012 My next version of this horn has several revisions, the main change is a smaller throat, I suspect the closer I get the throat to the actual piston size the better it will work. Other than this I am cutting several more layers in the larger rings to minimise sanding and more accurate flair. Currrently the rings are keyed, and my new cnc machine has a tolerance of .003 I suspect that will be close enough. I just modified my lathe so I can turn this large horn. Makes finish sanding much simpler. Dave Quote Link to comment Share on other sites More sharing options...
pzannucci Posted October 25, 2012 Share Posted October 25, 2012 If you are doing some mods, you might want to look at a phase plug. If you have a CNC you should be able to build one quite nice in the opening of the horn. The phase plug should get rid of those suck outs / notches in the response graphs and spectra. Properly designed it would likely make the horn sound much better. The horn itself is very nice. http://www.centauriaudio.com.au/diy/plugs.html Quote Link to comment Share on other sites More sharing options...
djk Posted October 25, 2012 Share Posted October 25, 2012 " If you are doing some mods, you might want to look at a phase plug. " And how is this going to help with a full-range driver? Quote Link to comment Share on other sites More sharing options...
Rudy81 Posted October 26, 2012 Share Posted October 26, 2012 I don't understand how a phase plug would work in this application. The driver itself does not sit behind any part of the horn so the problems mentioned in the article referenced are not in play...at least the way I read it. Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 26, 2012 Share Posted October 26, 2012 If you are doing some mods, you might want to look at a phase plug. If you have a CNC you should be able to build one quite nice in the opening of the horn. The phase plug should get rid of those suck outs / notches in the response graphs and spectra. Properly designed it would likely make the horn sound much better. The horn itself is very nice. http://www.centauriaudio.com.au/diy/plugs.html I am not sure this applies to an uncompressed waveguide, but I certainly appreciate the info. This link has alot of great information, especially for DIY'rs. What think I will try when I get some more time off is: Measure the driver (without the horn), if the glitches remain I will know the source of the issue is either the driver or test rig. The glitches only showed up while measuring sweeps while using REW. Dave Quote Link to comment Share on other sites More sharing options...
moray james Posted October 26, 2012 Share Posted October 26, 2012 Dave if you take a look at say Lowther you can see a host of phase plugs used in them over the years of various shapes and sizes. You could fashion a plug from light closed cell styra foam and measure with and without. Looking good so far thanks for posting. Best regards Moray James. Quote Link to comment Share on other sites More sharing options...
mark1101 Posted October 26, 2012 Share Posted October 26, 2012 Dave, I'm just guessing here but since your horn is built pretty precisely, I'm thinking the issue is at the mouth where the driver interfaces with the horn..........obviously what you are looking at now. I can't believe the driver itself would have that many or such excessive drop outs. Also, the waterfall plots showed very excessive ringing. Have you tried a 2" driver or any other driver to see if anything is different? Quote Link to comment Share on other sites More sharing options...
Guest David H Posted October 26, 2012 Share Posted October 26, 2012 Dave, I'm just guessing here but since your horn is built pretty precisely, I'm thinking the issue is at the mouth where the driver interfaces with the horn..........obviously what you are looking at now. I can't believe the driver itself would have that many or such excessive drop outs. Also, the waterfall plots showed very excessive ringing. Have you tried a 2" driver or any other driver to see if anything is different? Never used REW in the past for measurements. I typically use True RTA. I have a new 2 inch driver I need to measure, and a pair of Eliptracs that need finishing. When complete I will take some measurements for comparison. I can't believe the driver itself would have that many or such excessive drop outs. I wouldn't think so, however if I see the same glitches without the horn, I can assume the test setup is the culprit.Dave Quote Link to comment Share on other sites More sharing options...
djk Posted October 27, 2012 Share Posted October 27, 2012 http://farm8.staticflickr.com/7174/6718576351_8af88b7c78_b.jpg If you take a look at the link, you will see why I questioned him on it, not suitable for this type of fullrange driver. The phase-rods on the Audio Nirvana and others similar are generally made electrically conductive to reduce the impedance rise and a replacement wooden rod/light-bulb/door-knob is not acceptable. An Audio Nirvana with copper plated aluminum phase rod. Tang Band W8-1772 Dayton Audio PS220-8 Quote Link to comment Share on other sites More sharing options...
pzannucci Posted October 27, 2012 Share Posted October 27, 2012 http://farm8.staticflickr.com/7174/6718576351_8af88b7c78_b.jpg If you take a look at the link, you will see why I questioned him on it, not suitable for this type of fullrange driver. I understand the point but using a similar approach as this with extending the length into the horn would likely lower the number of notches in the response. I know it would be trial and error (mostly error) but that if I had a CNC I would give it a shot. Those notches are likely summing errors and reflections in the throat of the horn, but of course that is yet to be proven. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.