Damping done. I used "15oz" batting. I don't know what that is in new pence for the metric among us.
I usually use wool felt but ran out so I only used it on the rear panel to double up and on the immediate sides where the woofer will be (didn't get pics)
I usually use wool felt but ran out so I only used it on the rear panel to double up and on the immediate sides where the woofer will be (didn't get pics)
Got the wood at least. Don´t know when I will start. workshops are cold these days.
Building little test/desktop speakers as an in between project.
How is that alubutyl working?
The little desktop speakers are 12mm ply and I forgot just how resonant that stuff is.
Could play some conga on these things!
Probably need to line the panels too and I already though about "connecting" two adjacent panels with those triangular profiles.
Well, I don't know what to make of the last two days. Lots of progress on the making but some pretty annoying results. Woodworking went OK with some recoverable superficial hiccups but then...
It took three port shortenings to get to just below the desired tuning and I have landed at around 53hz which would be fine but for issue at tuning. There seems to be some sort of issue at or around the driver FS and the interaction with the cabinet. I expected to have to shorten the port due to the clos proximity of the wall and bracing I have no idea about the disruption at tuning and the blip on the impedance graph just above 400hz.
This shows itself very evidently on the port measurement and on the impedance graph. It appears to be stifling the combined output of port and cone. Cabinet geometry? The port being directly opposite the cone? Measurement issue? I dunno but it's 2.30am and I'm beat and beaten.
I have briefly investigated EPDR and don't truly understand it but the REW EPDR shows this as some sort of phase flip spike.
It took three port shortenings to get to just below the desired tuning and I have landed at around 53hz which would be fine but for issue at tuning. There seems to be some sort of issue at or around the driver FS and the interaction with the cabinet. I expected to have to shorten the port due to the clos proximity of the wall and bracing I have no idea about the disruption at tuning and the blip on the impedance graph just above 400hz.
This shows itself very evidently on the port measurement and on the impedance graph. It appears to be stifling the combined output of port and cone. Cabinet geometry? The port being directly opposite the cone? Measurement issue? I dunno but it's 2.30am and I'm beat and beaten.
I have briefly investigated EPDR and don't truly understand it but the REW EPDR shows this as some sort of phase flip spike.
Excellent craftsmanship! I suggest suspending stuffing in the middle of the cabinet, that's where the standing waves have highest velocity. I put it in mesh laundry bags and staple the bags to the braces. Leave paths for free air movement between the cone and vent.
Hmm, you have a 10+:1 driver Sd:vent Sd compression ratio = acoustic choke, especially with it behind the driver, so suspend/hang a 'blanket' of insulation between the two to damp the pulses a bit and hopefully will help, but the bottom line is the vent's way too small IME.
Just a thought, but how about building an enclosure the same size, made out of the cheapest material possible, to experiment with port positions?
I always put my ports in the side of the enclosure (Google Royd Minstrel).
I always put my ports in the side of the enclosure (Google Royd Minstrel).
Thank you every one for your suggestions, I will carry them out one by one! 🙂
First was the easiest to do. I clamped a piece of 150x200mm ply across the rear braces aout a third of the cabinet away from the port and then added some damping to stop reflections from the driver. No change whatsoever.
Next, I have two cheap and nasty 75mm ports to try which I ordered from a UK company that has a name that is very similar to Red Kevin. I had almost discarded them as they were so thin but worth a shot and I can reinforce them.
I also measured the ID of the current port and it turns out to be a bit smaller at 64mm ID. The new port will give a ~40% larger area.
First was the easiest to do. I clamped a piece of 150x200mm ply across the rear braces aout a third of the cabinet away from the port and then added some damping to stop reflections from the driver. No change whatsoever.
Next, I have two cheap and nasty 75mm ports to try which I ordered from a UK company that has a name that is very similar to Red Kevin. I had almost discarded them as they were so thin but worth a shot and I can reinforce them.
I also measured the ID of the current port and it turns out to be a bit smaller at 64mm ID. The new port will give a ~40% larger area.
Both 55Hz and 400Hz problems have such long wavelength that small pieces are invisible (any obstruction that fits inside the enclosure is not big enough).
I've got no idea what the port impedance peak is, I'd trust GM on it. If you can't make bigger (and longer) port just close it up. I guess the box depth is around 40cm? This would make the 400Hz hump. More damping material, or switching to material that is more effective for such long wavelength, should fix it. If not, make a three way system or new boxes...
I suspect every big 1 or 2 way box (where the woofer plays both lows and mids) have similar "problems" until fixed, or not cared about. Another option would be to call it "natural tone of heavenly string quartet", tweak the crossover for ages, swap capacitors, and slap some zeros to the price. For the profit you could make 3-way speakers without similar problemos 😀 ps. I don't mean to be rude, jokes targetted for average highend joe, you've found problems and have a chance and willingness to fix'em! Don't get depressed, try and make it fun learning experience, I trust you'll tackle it!
I've got no idea what the port impedance peak is, I'd trust GM on it. If you can't make bigger (and longer) port just close it up. I guess the box depth is around 40cm? This would make the 400Hz hump. More damping material, or switching to material that is more effective for such long wavelength, should fix it. If not, make a three way system or new boxes...
I suspect every big 1 or 2 way box (where the woofer plays both lows and mids) have similar "problems" until fixed, or not cared about. Another option would be to call it "natural tone of heavenly string quartet", tweak the crossover for ages, swap capacitors, and slap some zeros to the price. For the profit you could make 3-way speakers without similar problemos 😀 ps. I don't mean to be rude, jokes targetted for average highend joe, you've found problems and have a chance and willingness to fix'em! Don't get depressed, try and make it fun learning experience, I trust you'll tackle it!
Last edited:
Is this sim based on your DATS results? They featured quite a deviation in Qms.
Between 200 and 400 (and again 400 to 700) is either baffle step, or the artifact of a near field measurement beyond its scope. Have you checked with a gated far field measurement yet?
My guess: The driver does not perform as assumed in the small enclosure, its "too small" to still perform according to simulation. In your combined graph, the brown line of the driver is falling rapidly and it experiences a hick-up due to the port tuning being almost beyond the limits already for the driver enclosure combination. I wouldn't trust the measurement for the driver below 50 Hz. But the vent is working nice, doesn't it? Good output, saves the day.
Nice cabinet.
PS.: was the internal volume calculated including stuffing, vent, bracings, driver? Some liters go that way.
Last edited:
Call off the search for I am a huge bosom!
IT WAS MY FLIPPIN WORKSHOP TO BLAME! My workshop was built earlier this year and is a double layer of aerated concrete blocks precisely 3.25m square.This is ~53hz! It suddenly occured to me that I had to do room correction for my sound system in there and this was the offending ring.
What also puzzled me was that the glitch was not moving as the port went higher in tuning. I tried both drivers, several types of damping positions and the internal baffle.
Thanks all for the input, I was going mad. Occams razor prevails.
Now with the new port in which was a sensible idea (thanks GM) I can resume normal service and get it tuned in right 🙂
IT WAS MY FLIPPIN WORKSHOP TO BLAME! My workshop was built earlier this year and is a double layer of aerated concrete blocks precisely 3.25m square.This is ~53hz! It suddenly occured to me that I had to do room correction for my sound system in there and this was the offending ring.
What also puzzled me was that the glitch was not moving as the port went higher in tuning. I tried both drivers, several types of damping positions and the internal baffle.
Thanks all for the input, I was going mad. Occams razor prevails.
Now with the new port in which was a sensible idea (thanks GM) I can resume normal service and get it tuned in right 🙂
It's difficult to see, as the level is not the same! It appears that the dip close to 370 Hertz is still there?
Interesting solution! 😉
I learned from experience that left and right of a port resonant peak, phase shifts. You have a dip-peak pattern because below resonant frequency waves sum negatively, whereas positive above. It can also be a diffraction pattern (baffle step), though.
I went in with what I thought was the big guns by using some 50mm RWA45 Rockwool that was leftover but it isn't quite as effective as I thought. There is more tucked behind the bracing ythat can't be seen but looks like I'll need to cover all surfaces. I worry about intefering with the driver - perhaps fill the upper void with loose fill.
Blue/lilac is with the poly batting
Red/peach is the Rockwool
Blue/lilac is with the poly batting
Red/peach is the Rockwool
Curiosity got the better of me and I filled the cab up with as much poly batting as I could to (hopefully) achieve what an ideal response would look like between 300-600hz. This is the result. It's hard to say if this is as clean as it gets, the vent output starts to get hashy the higher up it gets in any event. Obviously, this has had a 2db effect on vent output.
I suppose I need to do a merged near/far field measurement to see if it matches the datasheet FR (which is not the clearest I have seen)
I suppose I need to do a merged near/far field measurement to see if it matches the datasheet FR (which is not the clearest I have seen)
Red/Peach would be enough for me, but I would listen to both (last one) it's easy to get carried away by looking at graph's