Speakerplans.com Homepage
Forum Home Forum Home > Plans > New Projects Forum
  New Posts New Posts RSS Feed - Unity horn build
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Unity horn build

 Post Reply Post Reply Page  <1234 9>
Author
Message
Phil B View Drop Down
Old Croc
Old Croc
Avatar

Joined: 21 November 2004
Location: Melbourne
Status: Offline
Points: 2313
Post Options Post Options   Thanks (0) Thanks(0)   Quote Phil B Quote  Post ReplyReply Direct Link To This Post Posted: 09 February 2013 at 9:24am
If you look at this wire frame from their site you can see that there are very short tubes on the rear ports coming in at the mouth end of the horn.




 And from looking at a front on view you can clearly see the driver on the cone side ports and that the other ports are extended?



This is probably to increase the path length etc etc and get everything in phase. Obviously where the rear ports enter the horn is the most critical measurement and the ports are just tuning the rear volume to the horn mouth. Danley has very cleverly split the outer box with baffles to increase path length inside the rear volume ... all very tricksy to try to replicate!

.p.

.p.
Mostly harmless.... except if catering is shut.

Solar Sound System Shennanigans..http://diyhifi.biz/
Back to Top
Phil B View Drop Down
Old Croc
Old Croc
Avatar

Joined: 21 November 2004
Location: Melbourne
Status: Offline
Points: 2313
Post Options Post Options   Thanks (1) Thanks(1)   Quote Phil B Quote  Post ReplyReply Direct Link To This Post Posted: 11 February 2013 at 12:58am
Found these snippets of info from DIY forums...so it does act as a 6th order bandpass but NOT to extend within the horn. Questions were asked about how the Danley SH50 can go down to 50hz when it`s horn mouth is so small and wouldn`t support horn loading down there. We were looking at extending our present horn out to get down to 100hz but seems we were barking up the wrong tree!

JLH ...
With the Danley Synergy horns the role of the woofers is a little different than most people think. In most of the Danley models the woofers take over from the mids at around 300Hz. They get horn loaded through entry ports just like the mids do from 300Hz down to about 135Hz. Depending on the model, 135Hz to 150Hz is all the mouth size will support for horn loading. Everything below the frequencies supported by horn loading is handled by bass reflex. In order to get the bass reflex section to match the sensitivity of the horn loaded section, the bass reflex is tuned up in frequency and made to peak in a narrow band. (When is the last time you’ve seen a pair of 12” woofers tuned to 80Hz?) So, you’re shooting for a very peaked reflex response to complement the horn loaded section. In normal life you would never design a bass reflex enclosure to behave this way because it would be a one note wonder. However, for the Synergy this is exactly what you need to extend the lower frequencies to exceed what is possible with the mouth size.

Tom Danley ...

Fwiw, the horn loading and effect of the low pass filters ends around 100Hz on an SH-50 (hard to separate it into the separate bits), the lowest part is the radiation from the ports which are short passages that extend to the rear and delay the driver’s rear signal which combined with the compliance form a stubby lumped tapped horn / vented box style acoustic phase.


So the very short tubes are just to correct the rear bass reflex path length to get that peaked response...

So back to the drawing board for us! Now we can do both parts of the driver in Hornres ...front as a standard Unity and the rear as a reflex ignoring the horn mouth...Hmmmm

.p.
Mostly harmless.... except if catering is shut.

Solar Sound System Shennanigans..http://diyhifi.biz/
Back to Top
cravings View Drop Down
Old Croc
Old Croc
Avatar

Joined: 30 January 2007
Location: Ireland
Status: Offline
Points: 7424
Post Options Post Options   Thanks (0) Thanks(0)   Quote cravings Quote  Post ReplyReply Direct Link To This Post Posted: 11 February 2013 at 1:28am
very interesting.
Back to Top
CoblosBrengose View Drop Down
Registered User
Registered User


Joined: 09 December 2008
Location: Indonesia
Status: Offline
Points: 20
Post Options Post Options   Thanks (0) Thanks(0)   Quote CoblosBrengose Quote  Post ReplyReply Direct Link To This Post Posted: 11 February 2013 at 1:41am
I fail to see tube ports in SH46, SH64, SH69, SH96

Edited by CoblosBrengose - 11 February 2013 at 1:48am
Back to Top
Phil B View Drop Down
Old Croc
Old Croc
Avatar

Joined: 21 November 2004
Location: Melbourne
Status: Offline
Points: 2313
Post Options Post Options   Thanks (0) Thanks(0)   Quote Phil B Quote  Post ReplyReply Direct Link To This Post Posted: 11 February 2013 at 2:56am
Originally posted by CoblosBrengose CoblosBrengose wrote:

I fail to see tube ports in SH46, SH64, SH69, SH96


Yeah...and? I`m just seeing how they acheived the low extension and if need be you could add tube ports to tune the rear output.

The picture above of the SH50 wireframe shows very short tubes ( possibly only just deeper than the horn moulding itself) . The frontal picture of the SH60 clearly show that`s the ports are longer than the horn moulding???

.p.


edit...found this pic which was the one I was looking for originally...




Edited by Phil B - 11 February 2013 at 6:52am
Mostly harmless.... except if catering is shut.

Solar Sound System Shennanigans..http://diyhifi.biz/
Back to Top
snowflake View Drop Down
Old Croc
Old Croc


Joined: 29 December 2004
Location: Bristol
Status: Offline
Points: 3096
Post Options Post Options   Thanks (0) Thanks(0)   Quote snowflake Quote  Post ReplyReply Direct Link To This Post Posted: 11 February 2013 at 12:28pm
it hadn't occurred to me that there was any horn-loading of the bass section going on. 1/4 wavelength of 150Hz is 56cm which is less than the length from bass tap to mouth. will try modelling as a ported horn (which I have designed before) and see what I get. usually in a ported horn the radiation from the two sides in inherently in phase because of the helmholtz resonator inverting the rear radiation - ports on both sides in going to screw up the crossover region. or is the port on the front of the driver tuned higher than the passband so the chamber is just acting as a LPF loading the driver into the horn and not affecting the phase???

Edited by snowflake - 11 February 2013 at 12:38pm
Back to Top
snowflake View Drop Down
Old Croc
Old Croc


Joined: 29 December 2004
Location: Bristol
Status: Offline
Points: 3096
Post Options Post Options   Thanks (0) Thanks(0)   Quote snowflake Quote  Post ReplyReply Direct Link To This Post Posted: 13 February 2013 at 1:07pm
currently waiting for cash flow to buy a speaker tester setup as this PC has no soundcard. but I have been listening to some music. have used the filter settings from my akabak model and adjusted them on the ultradrive till the treble level sounds right. HPF is 6dB at 1.5kHz the output fed back into inputC and another HPF 6dB at 4.1kHz. akabak model has LPF on the mids at 600Hz to flatten response and give roll-off at ~ 1.5kHz. have currently got the ultradrive LPF set at 1.14kHz as this sounds better. as the mid ports are at the edge of the cone probably getting cancellation across the diaphragm giving a steeper acoustic roll-off than predicted by akabak.

sounds good - will get a martin audio 12"+1" next to it later to compare.
Back to Top
Phil B View Drop Down
Old Croc
Old Croc
Avatar

Joined: 21 November 2004
Location: Melbourne
Status: Offline
Points: 2313
Post Options Post Options   Thanks (0) Thanks(0)   Quote Phil B Quote  Post ReplyReply Direct Link To This Post Posted: 14 February 2013 at 9:15pm
We  are running our Unities without any LPF on the mids with a passive crossover HPF and pad for the highs. The mids filter themselves pretty well on their own and adding a LPF seemed to screw with the phase. We got a pretty flat response and phase with a simple 2nd order crossover at 1.2k.

Gotta be someone close to you that can lend you some measurement kit...I`m interested to see how your mids work out!

.p.
Mostly harmless.... except if catering is shut.

Solar Sound System Shennanigans..http://diyhifi.biz/
Back to Top
snowflake View Drop Down
Old Croc
Old Croc


Joined: 29 December 2004
Location: Bristol
Status: Offline
Points: 3096
Post Options Post Options   Thanks (0) Thanks(0)   Quote snowflake Quote  Post ReplyReply Direct Link To This Post Posted: 14 February 2013 at 9:38pm
had a play with true rta signal generator, a cheap sound level meter and my ears. all tests were done in my bedroom - so obviously not very accurate. did find out a few things. mids are pretty flat from 200Hz to 1.2kHz. the predicted rising response from 600Hz to 1.2kHz isn't there and it rolls off nicely after that. so I could do without a LPF on the mids. have changed the HF filters from 6dB at 4.1kHz and 1.5kHz to 7kHz and 1kHz which is giving a flatter response at the same level as the mids. was expecting response to go to 16kHz but there isn't much above 12kHz - but maybe it's my ears going Wink. crossing the mids to bass at 200Hz rather than 280Hz has made the vocals sound a lot better.
Back to Top
snowflake View Drop Down
Old Croc
Old Croc


Joined: 29 December 2004
Location: Bristol
Status: Offline
Points: 3096
Post Options Post Options   Thanks (0) Thanks(0)   Quote snowflake Quote  Post ReplyReply Direct Link To This Post Posted: 25 February 2013 at 8:34pm
done some A/B listening comparing my unity to a martin audio EM76. unity goes just as high and midrange is audibly clearer. EM76 sounds honky in comparison. going to add a couple of Beyma 10MI100 for mid-bass which should go down to 110Hz @ 108dB with 2.83V. then try and measure the impedance and design a passive crossover Geek should be a good box for about £400 in parts Smile
Back to Top
Phil B View Drop Down
Old Croc
Old Croc
Avatar

Joined: 21 November 2004
Location: Melbourne
Status: Offline
Points: 2313
Post Options Post Options   Thanks (0) Thanks(0)   Quote Phil B Quote  Post ReplyReply Direct Link To This Post Posted: 26 February 2013 at 2:22pm
Originally posted by snowflake snowflake wrote:

done some A/B listening comparing my unity to a martin audio EM76. unity goes just as high and midrange is audibly clearer. EM76 sounds honky in comparison. going to add a couple of Beyma 10MI100 for mid-bass which should go down to 110Hz @ 108dB with 2.83V. then try and measure the impedance and design a passive crossover Geek should be a good box for about £400 in parts Smile


Nice....Clap Show us some pics....

.p.
Mostly harmless.... except if catering is shut.

Solar Sound System Shennanigans..http://diyhifi.biz/
Back to Top
snowflake View Drop Down
Old Croc
Old Croc


Joined: 29 December 2004
Location: Bristol
Status: Offline
Points: 3096
Post Options Post Options   Thanks (0) Thanks(0)   Quote snowflake Quote  Post ReplyReply Direct Link To This Post Posted: 11 March 2013 at 9:34pm
got hold of a behringer ultracurve deq2496 and condenser mic and done some measurements on the box at the top of thread. unity horn is now +/-3dB from 200Hz to 18kHz apart from one on-axis dip at 3.5kHz. off axis there is some HF roll-off and the response around crossover (1.3kHz) dips a bit.

it was a bit of a struggle getting the mid and high sections to meet due to earlier than expected roll-off of the 6" mids. after a bit of trial and error I have switched to 3rd order LPF at 2kHZ. the HPF is the product of a first order at 6kHz and second order at 720Hz.

sounding pretty good but as it is in the middle of the vocal range the 3.5kHz hole is noticeable. akabak prediction says it should be beaming on axis slightly at 3.5kHz rather than two side lobes. have tried to model the inside of the compression driver better but my akabak simulation is not near enough to measurements to be useful for further fine adjustments. going to try narrowing the vertical flare angle by a couple of degrees and see what this does.
Back to Top
 Post Reply Post Reply Page  <1234 9>

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.06
Copyright ©2001-2023 Web Wiz Ltd.

This page was generated in 0.078 seconds.