Something Stirs in the Workshop

[vinylnvalves]

After reading up abit more, this was informative https://www.soundonsound.com/reviews/ch ... -absorbers

I have decided to go with rock wool insulation, the melamine foam is really only good for above 600 hz. Using the melamine foam in the rear chambers for the TPL’s.

[vinylnvalves]

Installed the sub drivers this evening. Quick check with DATs didn’t give me what I was expecting.

I initially built a test cabinet of 26 ltrs internal volume and tested the seas driver. For the actual one I am using two drivers in parallel, in a box of internal volume 52 litres. Both boxes fully stuffed. I was surprised that the test cabinet in green had a resonance not much above the fs of the driver. The paralleled drivers have a resonance far higher. Was expecting the impedance spike to be lower too.

0C00291D-09EE-40C0-9939-917FF073F139.png (75.8 KiB) Viewed 11620 times

[vinylnvalves]

Got a chance to experiment with the stuffing of the subwoofer boxes. Changed the rockwool fibre pulling out and added some wool fibre wadding, added it end on 5” deep tightly so the wadding comes upto the back of the magnets.

70BC187B-34CD-468E-8F4B-989DA405A519.jpeg (163.86 KiB) Viewed 11574 times

This had a positive effect on the resonance of the system. Dropping the frequency by about 10 Hz, but causing the impedance peak to rise. So I have fooled the cabinet into it thinking it’s bigger, but made it less damped. Trace in purple.
The cabinet was effectively damped on the back wall , but not around the drivers. So I added some BAF wadding around the driver. With the magnet being so big all the wadding is pushed out about 3” from the cone. Remeasuring I happily got the response I had simulated. Blue trace.

0CB6FB14-04DF-42D1-8805-3A9383CFB337.png (99.97 KiB) Viewed 11574 times

Listened to some music with LF content, the extra wadding has resulted in a dry response, compared to a very under damped response in the other.

So that’s the easy bit done, next the midrange and HF to tweak, before Owston and the WAM show.

[vinylnvalves]

Managed to measure the response of the lightly damped and over damped ? subwoofers. The results are interesting. The extra damping equates to a 20db reduction in response. How much of this is the undamped resonance, is the question. Blue trace is the under damped cabinet (obviously)

EB50685C-D080-4940-8542-58A7CC3533AF.jpeg (87.36 KiB) Viewed 11527 times

Looking at the distortion levels. It would suggest that the distortion is higher for the damped cabinet too. If the distortion is dependant on output, it maybe slightly better, unclear what holmimpulse does.

EAC2B329-5C5C-4140-A60E-5E3CCCFB647D.jpeg (111.3 KiB) Viewed 11527 times

This isn’t what I was expecting. Losing say 10db of LF response isn’t insignificant. My simulations only predicted a 5db hump at 50hz for an undamped cabinet. Thoughts guidance please.

[vinylnvalves]

Being stupid…. The battery test revealed that one of the drivers on the second speaker was connected in reverse, sorted that. The driver was wrongly designated by SEAS. The impedance spike has moved back to 40hz same as the first lightly stuffed cabinet. A SPL measurement shows no difference in the response under 300 hz.

So conclusion damping in subwoofer cabinet has little effect, which you would expect with the wave lengths.

The only unanswered question is why is the impedance spike double what I predict in VituixCAD for these drivers in parallel.

[chris661]

Looks like things are progressing well in my absence.


If you're looking at the maximum impedance value, I'd say don't worry about it. The simulators assume a perfect resonance, and throwing a bit of damping in there will write that off immediately. I use impedance to check the LF resonance is in the right place, and to spot cabinet-related resonances further up the frequency range.

FWIW, I only bother with internal stuffing/lining if I know there'll be useful absorption within the speaker's bandwidth. Subwoofers are generally high-power devices, so I'd rather keep good airflow around the magnet structure instead.

Chris

[vinylnvalves]

Thanks for your comments. I have removed the OTT thick insulation, anyway - there is space around magnets. Been put back a bit today.. The tabs broke off one of my Speakon sockets this evening, Neutrik ones as well. Wasn’t impressed they had spade connections in the first place.

[vinylnvalves]

Measured the Beyma 12p80nd drivers last night, both very similar frequency- resonant freq in free space 60 hz, Way off the 47hz quoted on the spec sheet. I had these in opposition with some LF driving signal for 24hrs to loosen then up earlier in the year. Maybe I need Chris driving them to x max 52mm pp to soften them up

[chris661]

Careful pushing them that far - IIRC, that's Xlim, where damage is likely to occur.

The last time I hit Xlim on a Beyma driver, the cone folded. Granted, it was a >500W sine sweep, starting at 15Hz, into a 40Hz ported box.


Even so, that's quite a way off spec. It's possible that the rest of the T/S parameters would shift in a complementary way, though. ie, while the numbers might be different, the response curve in a given cabinet will barely change.


Chris

[vinylnvalves]

Maybe I will post the TS parameters for the driver in free space later. I have finished the first of these smaller speakers.. just listening in mono until I get a chance tomorrow to finish off the other, and fine tweak the crossovers. Dave’s coming around Saturday to see if they are worthy of bringing to Owston and the WAM show.

48B0821F-B041-42D1-ADC1-51AFD704C7C9.jpeg (162.95 KiB) Viewed 11128 times

[vinylnvalves]

Successful morning of testing on Saturday with Dave. He thinks they are worthy of the WAM show. Did some further measuring and refinement, the troublesome bit is the LF where i think room modes are dominating. As when i reduce the output by 2db i get 4-6db of reduction in the room, which suggests coupling. Traces out of REW are as follow... Ignore the HF drop off i haven't imported my mic calibration file into REW yet.

settingup.JPG (129.5 KiB) Viewed 10912 times

Hopefully Chris can make a silk purse out of a sows ear at Owston

[vinylnvalves]

After a few of chats on Saturday with Chris, Andrew and others, it got me wondering…. The subs didn’t sound as controlled as I would like, and was wondering if the configuration I have them in is the challenge, as the amplifier damping factor will be down.
Currently I am running the two sub drivers in parallel, so the amps see an effective load of around 2 ohms. Did this for the 6db gain it gave me over series.

I have a spare amp module, so should I can easily regain the loss if running in series.

1) put module in and run each amp module in BTL mode with the drivers in series.
2) put the module in and feed each sub driver individually.
I have the 4 core “hose pipe” cable Scott recommended, so either way isn’t a hassle.

Will not be doing this before the WAM show anyway, I might connect the drivers in series for the show, as a quick look see, as I am padding the bass channel down currently anyway and won’t need as much output in a small hotel room.

[chris661]

A couple of notes on that:

- The room at Owston always makes a mess of the bass. Dipole systems fair a bit better, but they come with their own set of challenges.

- In general, there's a distortion and damping factor penalty in running amps at 2ohm. I'd consider wiring the drivers in series and adjusting the levels accordingly, so long as you're unlikely to hit output limits.

- It's very much worth taking the measurement gear to the demo room you'll have, and seeing if it's possible to get the subwoofers + room working together nicely for a useful portion of the listening area.
In my lounge, and at Colin's, the response at each seat is pretty similar, so any EQ applied will benefit everyone. Smooth and flat bass down to the lowest registers is one of the biggest benefits of DSP, so I'd aim to make use of that if you can.

Chris

[vinylnvalves]

Thanks Chris, I definitely will be measuring in room next Saturday, to get it better equalised than last Saturday which was a rush, after chasing a problem down of my own making. I have lots of time, frequency sweeps into the early hours won’t annoy the other Wammers to much.

Was trying to understand what the advantages of having amps in BTL mode compared to driving each driver separate, as obviously the higher impedance is two drivers in series is easier on the amp. But I am sure I read somewhere that paralleling up amp stages has some pitfalls.

[vinylnvalves]

Haven’t done much to my speakers since the WAM show, added some spikes to them last weekend. Repurposed some brass door stops to support the spikes as outriggers…. £2.50 each from tool station.

08C858F6-3328-4FD3-A5BF-912F1F278B98.jpeg (82.98 KiB) Viewed 9809 times