Another speaker thread

[simon]

I put the H frames back in as I wanted to refresh myself with how they sound, here it's compared with the sealed box, from my listening seat.

It was a slightly curious experience going back to the H frames. It felt lighter in the bass, as expected, but it sounded almost coloured in comparison, and I wasn't expecting this. I wondered if I could hear the edges of the frames resonating. They are quite deep. Over the course of the week I've adjusted and don't really notice it now.

[IslandPink]

There's a lot of stuff coming from the room, in these frequencies. It may be the driver in H frame is energising the room modes slightly differently to the driver on the sealed box. I've always tried to get about 3 or 4 measurements when i'm attempting to get a meaningful curve in the mids or upper bass. What I do is try a couple of different distances, plus a couple to the left or right by a foot or so. You can easily average a number of previous curves with REW. Plus I tend to do a 1/48th octave smooth to get rid of some of the noise.

[simon]

I've done similar, each plot is an average of 9 readings (3 x 3), though they're probably not a foot apart.

This is the same plot but taken at a meter from the H frame, with the readings taken at 2.5m shown early again for ease of reference.

How do you do a 1/48th octave thing in REW?

I was curious to see if I could replicate the low pass type filter that Martin King shows in fig 5 of his OB, U frame and H frame paper. Not sure it's conclusive really as there's not much difference I can discern between the H frame and sealed cab.

http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf

[simon]

With the H frames having a different response to the sealed cabs, and consequently sounding different, I was curious to dial the sound levels in of the 3 amps by ear with music, and then see how it measured.

This is the averaged plot as it's a little easier to read in some respects.

Subjectively I felt I might have been boosting the bass a touch to compensate for earlier roll off and more uneveness of response when setting the volumes. Or maybe it just brought out the bass line in the music I was listening to. Or maybe I just like too much bass.

[chris661]

It's worth applying a bit of smoothing to those, to reduce the "fuzz". I usually go to 1/12th-octave.

With regards to how the levels were set, I suppose it depends on the music you were playing (does it have lots of content at 40Hz, or 100Hz?), how loud (Equal Loudness Contours will kick in and screw things up if you're playing quietly), personal taste, mood, etc etc.

If you've got the option of playing music from your measuring device, install Equaliser APO. It's free software which sits between your media player(s) and the physical output of the device, and gives you pretty much endless options for tinkering with the response. A bit of EQ in the <150Hz range will probably improve things dramatically, but be careful about boosting the bottom end too much, especially with the H-frame.

It's also worth checking out the waterfall plots on REW, to see if you can spot the audible H-frame resonance(s).

Chris

[steve s]

It was a slightly curious experience going back to the H frames. It felt lighter in the bass, as expected, but it sounded almost coloured in comparison, and I wasn't expecting this. I wondered if I could hear the edges of the frames resonating. They are quite deep. Over the course of the week I've adjusted and don't really notice it now.

That's interesting Simon, my speakers sounded very similar to Steve's sealed fanes when he brought them to mine in term of colouration. Mine would sound very coloured without the damping,
I've not heard H frames sound right to me without it but as you say once you get used to it you dont notice

[Ant]

Perhaps add a couple of broom handles across the back to brace the sides of the h frame together?

With my big fanes there was evident colouration until i added the internal bracing.
Theres still some at resonance which l traced to the front panel underneath the driver but i cant get into that section because of the large horizontal holey brace that under the driver. Mine needed much more bracing than mi dads because the panels are twice the size

[simon]

Ah, Ctrl-Shift-Z to add smoothing. I should probably read the flipping manual...

Much easier to read now

[simon]

If you've got the option of playing music from your measuring device, install Equaliser APO. It's free software which sits between your media player(s) and the physical output of the device, and gives you pretty much endless options for tinkering with the response. A bit of EQ in the <150Hz range will probably improve things dramatically, but be careful about boosting the bottom end too much, especially with the H-frame.

I was using vinyl so not easy... I should be able to rig something up with a bit of thought.

It's also worth checking out the waterfall plots on REW, to see if you can spot the audible H-frame resonance(s).

No idea how to do this, better read the manual. What do waterfall plots show us?

[simon]

That's interesting Simon, my speakers sounded very similar to Steve's sealed fanes when he brought them to mine in term of colouration. Mine would sound very coloured without the damping,
I've not heard H frames sound right to me without it but as you say once you get used to it you dont notice

Yeah, I need to get some damping stuff, I'm completely out of anything like that.

[simon]

Perhaps add a couple of broom handles across the back to brace the sides of the h frame together?

I did the same with the Quasars. In fact I've probably still got them somewhere in the garage, and they should fit. A combination of this and damping as Steve suggests would no doubt make a difference.

[simon]

The next experiment today was to take the SB Acoustics SB34NRX75-6 driver out of the TV console in the lounge and bung it in one of the sealed boxes I made for the 285GMFs. The cabs are 75L IIRC, which is around Qtc=0.85 for the SB Acoustics driver, a bit high but close enough for an experiment.

I dug out a cheap ebay variable active crossover to try to limit the bass to below the 285s in the H frames. There's two pots on the board, I presumed one was volume the other for the corner frequency. I tried the scientific route of measuring to work out which was which but blimey, it was all but impossible. Eventually I thought I'd got it something like and then hooked up the other 3 amps for the 285s, 208s and T90As for a listen. I wasn't sure what to expect but it was pretty good. Apparently barely noticeable, just a bit more weight to the bottom. Turning it off readily demonstrated that it was there though. I only listened to a couple of tracks though and need to listen longer now to weigh it up a better. Very promising though.

[chris661]

If you've got the option of playing music from your measuring device, install Equaliser APO. It's free software which sits between your media player(s) and the physical output of the device, and gives you pretty much endless options for tinkering with the response. A bit of EQ in the <150Hz range will probably improve things dramatically, but be careful about boosting the bottom end too much, especially with the H-frame.

I was using vinyl so not easy... I should be able to rig something up with a bit of thought.

It's also worth checking out the waterfall plots on REW, to see if you can spot the audible H-frame resonance(s).

No idea how to do this, better read the manual. What do waterfall plots show us?

Would it be possible to play the music using the same device as you're using for the REW sweeps?


Waterfalls show decay time vs frequency. Things like panel resonances (which the H-frames might have) will show up there, but it can be difficult to dig that out from whatever the room's doing. Close-ish micing may be required.

Chris

PS - I use Ctril + Shift + 1,2,3,6,7,8,9,0 (removes) for 1/1, 1/2, 1/3, 1/6, 1/12, 1/24, 1/48-octave smoothing. The Psychoacoustic, ERB and Var smoothing are more complicated, but there's good stuff in the Help file.

[simon]

Thanks Chris, I'll have a play with that

[simon]

Time for a bit of an update. I still need to look at waterfall plots and equalisers though.

First, here are some plots of the SB Acoustics SB34NRX75-6 sealed sub working below the 285s in the H frames. It took a little listening, measuring, then tweaking to get to this, but it seemed the best balance to my ears.

And this is the effect of the sub superimposed:

To my novice eye the 285GMF H frame is a bit peaky between c80Hz and c180Hz, and the 208EZ a bit peaky over 400Hz to 650Hz. These are measurements from the listening position BTW. And there's a curious dip around 50Hz from the subwoofer. But a little like what the Romans did for us, the rest is pretty good, from 15Hz up to around 15kHz.

I guess the next thing to try is a waterfall plot of the sub (and H frames) to see if there's anything odd going on there.

Overall I'm pretty pleased with the sound. There's the "openness" I felt was missing with the 285s in sealed boxes, but with a taut underpinning from the sub. I've sat and listened to music in a way I haven't for probably years. Getting the vinyl setup again has probably had a lot to do with that too TBF.

The sensible, pragmatic path is probably to build two sealed subs, and try to fettle the H frames and integration with the 208EZs and run with 4 way speakers. But what's pragmatism got to do with DIY .