Such a work of art but it does beg to be but to the test.
You have to add a base still, right?
You have to add a base still, right?
alexcd said:
Hi Alex
Thanks
Everything else will come later, first on the list is go take a listen before doing anything else.
It looks fantastic! 🙂
Unbelievable speed with design & construction for a VERY complex shape. I'm betting that 50 pence a (quality) cut doesn't seem so absurd now. 😉
Unbelievable speed with design & construction for a VERY complex shape. I'm betting that 50 pence a (quality) cut doesn't seem so absurd now. 😉
Thread cleaned up. Discussion about TV shows carries on here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=84505
I've messed up 😀
It appears the 3" ID fall pipe isn't available anymore the closest is 70mm(~2.75") and 90mm(~3.5") ID.
The 2.75 looks incredibly weedy and I know from experience that this will have audible port noise at higher SPL's. Its the type of thing you'd see on a floorstander so no way is it suitable for tuning down to 19hz.
The 90mm looks perfect except it doesn't fit 🙄 and fouls the magnet.
I designed the whole thing to such tight specs that I haven't allowed for situations like this. It was designed for 85mm OD tube maximum and what I'm going to have to use is 98mm OD.
I've bought both anyway but I'm going to have to get creative to make the 3.5" fit. Any suggestions?
It appears the 3" ID fall pipe isn't available anymore the closest is 70mm(~2.75") and 90mm(~3.5") ID.
The 2.75 looks incredibly weedy and I know from experience that this will have audible port noise at higher SPL's. Its the type of thing you'd see on a floorstander so no way is it suitable for tuning down to 19hz.
The 90mm looks perfect except it doesn't fit 🙄 and fouls the magnet.
I designed the whole thing to such tight specs that I haven't allowed for situations like this. It was designed for 85mm OD tube maximum and what I'm going to have to use is 98mm OD.
I've bought both anyway but I'm going to have to get creative to make the 3.5" fit. Any suggestions?
I'm talking out of my a$$ here but how about applying some heat and pressure to bend it around the magnet?
Go to a commercial fabric/upholstery store and look at their cardboard tubes... a very wide variety... works fine for ports.
Or simply "roll your own" from poster/fibre board and PVA.
Also, think about removing a crescent shaped piece of material... a larger pipe need not foul the magnet.
🙂
Or simply "roll your own" from poster/fibre board and PVA.
Also, think about removing a crescent shaped piece of material... a larger pipe need not foul the magnet.
🙂
Thanks guys. I like the sound of the heat thing since I can try it out right now.
The 2.75" stuff is pretty poor TBH:
And after a bit of force the 3.5" nearly fits:
The driver isn't sat correctly though and I don't fancy the idea of the driver being braced against the port anyway.
The 2.75" stuff is pretty poor TBH:
An externally hosted image should be here but it was not working when we last tested it.
And after a bit of force the 3.5" nearly fits:
An externally hosted image should be here but it was not working when we last tested it.
The driver isn't sat correctly though and I don't fancy the idea of the driver being braced against the port anyway.
I think you can you heat it up and bend it more without harming the pipe structurally.
Damn, the commercial grade PVC I use it WAY thicker than that.
Damn, the commercial grade PVC I use it WAY thicker than that.
alexcd said:
This stuff is made for shuttling your turds safely away from your home. The grey ones looks like its been used
😀No big pressures involved with that sort of thing, so the diameter is weedy. Still better than commercial ports though.
The guy also let me have these FOC from the offcuts corner hence the reason why they're a little bit battered.
ScottG said:
Won't work Scott, 2x 70mm for 19hz tune is 98cm each - too long. Even considering that you virtually always shorten the calculated values they'll be awefully close to the top.
I'd think that using much less is defeating the point of cross-section.
First test:
🙂
An externally hosted image should be here but it was not working when we last tested it.
🙂
Ant they look bloody awesome! I'm intrigued by the 'waist' on them. Was that narrowing for sonic reasons or just style? It looks damn cool, do you have more details on how you built that part? I find it hard to tell from the photos if it just gets narrower on the outside, or on the inside too. Did you just have one less layer on the outside and used a ‘skin’ to bridge the gap at an angle?
Hi T, there's tons of info and few pics starting several pages back in case you've missed it.
Do you have a name for this sub shin? This thread's getting ridicoulously long. Might be worth starting a new thread for the sub part of it...
Hi Vikash,
I went back to page 109 and saw some pics of it but I was still unclear on the exact construction of it.
I went back to page 109 and saw some pics of it but I was still unclear on the exact construction of it.
Tenson said:
Hi Simon,
Its all one shape, but multiple layers of MDF. This diagram better explains:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
It was fairly easy once all the angles were identified, there's exactly 40 angled cuts all relying on each other just for that center section, get one or two out and things don't fit anymore.
We had fun working out just what the mitre and cut angles were 🙂 Some kind folks from another forum helped out with the answers and Al(pinkmouse) wrote a nice and clean looking angle finder where you just enter the desired lean(poor description, sorry) and out comes the answers, this excel sheet is posted a couple of pages back. It only works for angles on a flat plane and at an absolute angle of 90 degrees.
I've been playing now for a good while.
There's one caveat: the air is leaking around the driver at highish SPL's. Its comming from the side where it clashes with the port so I'm guessing my quick fix(scoring a small bit of tube away) didn't work and its still fouling the driver. Its too late to mess around now so I've been playing regardless.
This has been a real challenge to integrate with the mains, There's only two spots in the room that I'm willing to place it, one is directly at the outer side of the left speaker and the same for the right. Some variable phase controls are needed I suspect. I solved the problem in the end by adding a global delay line of 12ms in the DEQX for the mains and it suddenly snapped into place. My main problem here is that the DEQX has a processing time of around 8ms and console(which does the audio processing for the sub) has around 20ms when using FIR low pass and IIR PEQ. I've effectively had to syncronise these two. Would be nice to have a second DEQX linked to the first and syncronised but the costs are daft just for sub duties. I suspect there's still more to be had in terms of time alignment but I'm more interested in just hearing it moderately well setup first, the OCD tweaking can come later.
I've been using the measurement facilities of the DEQX to jugde where the precision strike EQ was needed. I found out that the port tuning was off after the very first sweep. Its 24hz instead of 19hz. I know exactly whats gone wrong here - the larger diameter port has decreased the volume. I did originally account for a 75mm OD port of 55cm length, this new one is 98 OD and 79cm, which is a very considerable increase in volume taken up by the port. I've considered lengthening the port to around 90cm but after modelling it appears to be a losing battle as its just eating up more volume and causing the tuning frequency to remain fairly constant regardless. This sub was designed for a 3" port so I'll have to get that sorted.
On a plus note I can't hear any port related noises with my ear stuck under the cabinet and this was at high SPL's(the air leak around the driver sounded baaaad though) so I should have some good leverage.
The rest of the EQ was simple stuff, I dropped the upper ranges after 24hz(port tuning) to create a response that had a 4dB hump at 24hz in comparison to 45hz the line between these points was virtually straight. I also took care of a large but narrow 41hz hump thats room related. And after some experimenting used a 48dB/oct rolloff starting right at 45hz, the main were left full range with no highpass on the bass - I strongly felt things sounded best like this. I dropped the level of the sub 2dB below the mains at the XO point I found this greatly helped the sub to dissappear more effectively, yet still afford a large contribution to overall scale.
Scott, you should also note that room gain did a great job here. I'm looking at a response that's very nearly flat from 50hz to the tuning frequency of 24hz. The model predicted around a 5-6dB drop between these points when I ran the new figures to take into the 24hz tuning. I *really* need to get that 19hz tuning fixed so that the design is put back in balance.
Not to blow my own horn but the cabinet is just so incredibly dead, I was really wacking out some rather heavy basslines and had my ear pressed against it. Normally you hear a range of buzzing as the cabinet moves sympathetically to the music. This juggernaught has only the faintest following of the driver and this is at stupid SPL's, at normal SPL's this is a cabinet that dissappears and it really shows.
I'll continue my thoughts on the sound in another post. 🙂
There's one caveat: the air is leaking around the driver at highish SPL's. Its comming from the side where it clashes with the port so I'm guessing my quick fix(scoring a small bit of tube away) didn't work and its still fouling the driver. Its too late to mess around now so I've been playing regardless.
This has been a real challenge to integrate with the mains, There's only two spots in the room that I'm willing to place it, one is directly at the outer side of the left speaker and the same for the right. Some variable phase controls are needed I suspect. I solved the problem in the end by adding a global delay line of 12ms in the DEQX for the mains and it suddenly snapped into place. My main problem here is that the DEQX has a processing time of around 8ms and console(which does the audio processing for the sub) has around 20ms when using FIR low pass and IIR PEQ. I've effectively had to syncronise these two. Would be nice to have a second DEQX linked to the first and syncronised but the costs are daft just for sub duties. I suspect there's still more to be had in terms of time alignment but I'm more interested in just hearing it moderately well setup first, the OCD tweaking can come later.
I've been using the measurement facilities of the DEQX to jugde where the precision strike EQ was needed. I found out that the port tuning was off after the very first sweep. Its 24hz instead of 19hz. I know exactly whats gone wrong here - the larger diameter port has decreased the volume. I did originally account for a 75mm OD port of 55cm length, this new one is 98 OD and 79cm, which is a very considerable increase in volume taken up by the port. I've considered lengthening the port to around 90cm but after modelling it appears to be a losing battle as its just eating up more volume and causing the tuning frequency to remain fairly constant regardless. This sub was designed for a 3" port so I'll have to get that sorted.
On a plus note I can't hear any port related noises with my ear stuck under the cabinet and this was at high SPL's(the air leak around the driver sounded baaaad though) so I should have some good leverage.
The rest of the EQ was simple stuff, I dropped the upper ranges after 24hz(port tuning) to create a response that had a 4dB hump at 24hz in comparison to 45hz the line between these points was virtually straight. I also took care of a large but narrow 41hz hump thats room related. And after some experimenting used a 48dB/oct rolloff starting right at 45hz, the main were left full range with no highpass on the bass - I strongly felt things sounded best like this. I dropped the level of the sub 2dB below the mains at the XO point I found this greatly helped the sub to dissappear more effectively, yet still afford a large contribution to overall scale.
Scott, you should also note that room gain did a great job here. I'm looking at a response that's very nearly flat from 50hz to the tuning frequency of 24hz. The model predicted around a 5-6dB drop between these points when I ran the new figures to take into the 24hz tuning. I *really* need to get that 19hz tuning fixed so that the design is put back in balance.
Not to blow my own horn but the cabinet is just so incredibly dead, I was really wacking out some rather heavy basslines and had my ear pressed against it. Normally you hear a range of buzzing as the cabinet moves sympathetically to the music. This juggernaught has only the faintest following of the driver and this is at stupid SPL's, at normal SPL's this is a cabinet that dissappears and it really shows.
I'll continue my thoughts on the sound in another post. 🙂
- Status
- Not open for further replies.