The 'XXX' sub

Very interesting Stu and many thanks Rob for the continued answering of my questions.

Stu, I'm pretty sure you are correct. But this can be minimised to a certain extent by using the level 4 version of TrueRTA, which as Rob stated measures in 1/24 of an octave rather than the lower resolution Level 1 which only does 1/6 of an octave.

So the 'buckets' you describe would be very small indeed for the level 4 software.

I think I'm correct in saying this. I'm most definitely no expert however. Wait for Rob to give an answer as I'm sure he's elucidated.
 
I can see what you're getting at there Stu, however my RTA reads 24 times per octave, rather than the 1(ish) time per octave that you've drawn. I compared this method to the sine wave/spl meter method , which uses 1/6 octave sine waves.

ie 6 sine wave readings 20 - 40, 6 readings 40 - 80 etc...

my rta 24 times 20-40, 24 times 40-80..

Yes you can do sweeps, or even measure single Hz increments if you've got way too much time on your hands ;)

If you run a sweep after rta/bfding and you can hear any noticable anomalies then you can carry on to your hearts content. I'm not too bothered by a slight deviation as as soon as you move your head left or right a bit they all change anyway..:D

If you go back a page or so you can see the rta graph for my subs before and after eq...

Shinobiwan - It hit me in the car on the way to work that the bfd has a switch on the back for level matching. I'll have a look in the manual tonight to check the advice I gave you earlier..

If it helps I use unbalanced inputs, and balanced out to my 2 pro amps. I *think* I use the -6dB setting on the switch.

You have to set it up with enough headroom for the peaks..

Cheers

Rob

EDIT: Shinobiwan posted that while I was typing this, but the yep - the 1st 'bucket' you drew in the pic is actually 24 buckets for me:)
 
Originally posted by robwells
Shinobiwan - It hit me in the car on the way to work that the bfd has a switch on the back for level matching. I'll have a look in the manual tonight to check the advice I gave you earlier..

If it helps I use unbalanced inputs, and balanced out to my 2 pro amps. I *think* I use the -6dB setting on the switch.

You have to set it up with enough headroom for the peaks..

Cheers

Rob

What is that switch on the back for?

I notice it states +4 dBu and -10 dBV. I seem to remember something similar on my RME HDSP9632 sound card. It has both of these plus another called 'lo-gain'.

I'm using +4 dBu on the RME at the minute and god knows what the input level of the Lex MC1 is, should the BFD be set to +4 dBu or the -10 dBV - confused???

Cheers,
Ant
 
The switch is to swap from pro levels (signals in pro stuff are generally higher (+4dBu), and home levels lower( -10dBu.) Basically pro gear sends out higher voltage 'line level' signals than home gear.

In theory set the switch to -10dBu with home gear - you're balanced though so that *should* be pro level, though sometimes (apparently) you get a balanced home level signal.. :D

I use the switch at the pro level, as otherwise my receiver will make the bfd clip on peaks. On some massive bass - Toy Story 2 'o' fly through for example, the yellow light will come on but no red light. This is at -30 on my receiver volume, so I calibrated the test tones at this level. ie -30 on amp, 85dB at sitting position.

I'd imagine many people calibrate their mains first, then adjust the sub level on their receiver till theres no clipping at peaks, then bring up the sub vol.

Many ways to skin a cat. btw - the snapbug how to is probably easier to understand than my dribblings:blush:

http://www.snapbug.ws/bfd.htm

Cheers

Rob
 
rob,

well if you are RTAing down to such a fine bucket size, that is great - although youll need a large window length, and high numbers or averaged samples (20+ hopefully). I would expect (purely from past RTAing at that resolution), that it will take about 30 seconds or so to properly level out, and any changes you make on teh BFD will take 30 seconds to feed through therefore.

So i advise caution - basically take the readings nice and slowly and they will be accurate, and repeatable. Rush it, and the system will not have had time to settle to the new changes.

HTH

Stu

Damn i love PC RTA's :) :) :)
 
Hi Stu,

You should download the 'free' level1 version to have a basic look at what it does.

The response changes immediately when you adjust stuff, unless you have a really high 'averaging' setting. ie: it analyses in 'real time' ;) Also samples at 96khz if you're so inclined, with selectable resolutions from 1 octave to 1/24...

All in all it's not so bad.

Also has a 'quick sweep' function to do a sinewave sweep and plot the response.

It would be interesting to run it and a spl meter at the same time and compare. Oneday if I'm really bored:D no make that really really bored....

One question - when you do a sweep do you set the spl meter to fast or slow ? - reason being it takes a while to 'come up' to measurement on slow, by which your sweep has moved on to a different freq...

Also do you 'adjust' the readings on the spl in your mind to compensate for any deviations ? - or do you have a 'super spesh' one rather than the radioshack jobbie that most people here use ?

I use this one, but only as the 'rat shack' one has gone all rocking horse poo..:)

http://uk.farnell.com/jsp/catalog/v...CD1OXYSCRFZQFIAEXCFE4AVAAS2IV3?prodId=4626620


Edit: out of interest, I have a 1/6 rta on my ultracurve pro, with 3 averaging settings - the fast one reacts perfectly in sync to music - if you have a snare hit, or cymbal splash you can see it come up in time to the music. It doesn't need 30 secs to react...the med/slow settings calm it down a bit..., losing response time but smoothing the reading..


Cheers

Rob
 
Just bought the Level 4(£60-odd) version of TrueRTA, damn this is getting expensive! I justified it by saying to myself that I can use it for future loudspeaker designs and crossover adjustments :)

I found out that the Level 1 freebie version only measures in 1 octave steps! No good at all.

So I've now got access to 1/24 of an octave measurements, really impressed too with the TrueRTA software.

Mic should arrive tommorow then all I need is a mic pre-amp and I'll be off ;) - and back with more questions :D
 
Rob, i would use a 1/24th RTA, and sweeps by ear TBH, i have never used a SPL meter to set anything up other than speaker levels in my 5.1 setup (which i had perfect anyway by ear ;))

If a cymbal crashes you will get immediate response. The point is that it is averaged, and the peak you saw was only the average of teh true cymbal peak, and lots of silence samples. If you played teh cymbal crash on loop the peak would get much higher. When we are calibrating freq response we are very interested in intensity of freq components, and not jut their prescence. This is why it is best to let the system take lots of averages :)

Stu
 
So Stu which test tones are better suited to setting up the BFD and what sort of duration are we looking at?
 
well as you know i havent ised one (As i didnt even know what it was 2 pages ago ;)) i cant give you a definative

but setting up using a parametric EQ, id just play sweeps and listen. If you have a PC get something like "cool edit pro" and you can make sweeps say from 20 Hz to 100 Hz, 10 seconds (linear sweep - makes knowing which freq you are playing. e.g. if at 5 seconds you hear a dropoff, you know thats at 60 Hz)

then make sweeps concentrating around problem regions and try to tame them. I'm quite lucky with my parametric, as it gives me a graph of teh shape response my filter is giving (i.e. with a Q or x, and centred at 60 Hz, i can see how wide it spreads) - so if i know i have boom around 50-70 hz, ill drop off at 60 with a Q that goes from 45 - 75 or so. You have to play it all by ear. Take a break, and come back to it also, your ears do funny things.

Youll spend around an hour blu-tacking everything that rattles in your room down first mind with your beast ;)

oh, and also remember that your sub is new, and not broken in yet, it will want a fair few hours play - so dont spend *too* long setting it up, as its response may well change :)

Stu
 
robwells said:
Any progress Ant ?

Rob

Hi again Rob,

Well as you know I have the DSP1024P along with the ECM8000 mic, but I needed a mic pre-amp. Well in the end I bought a Behringer MIC100.

But I'm still no closer to getting a frequency response reading thanks to the fact that to get the +48v phantom power into the mic I need a female XLR to male XLR cable. Now I of course bought a female XLR cable to 1/4" mono phono cable, expecting it work.

If you can recommend me a good cheap cable I would be grateful, otherwise I'm still looking around at the minute.
 
ShinObiWAN said:
Hi again Rob,



If you can recommend me a good cheap cable I would be grateful, otherwise I'm still looking around at the minute.



A bog standard lead will be about £10 - £20 from a pro music shop. I made my leads from maplins stuff - neutrik connectors about £5 - £6 a pair, and van damme twisted pair mic cable about £1.49 per metre.

I made up a load for my x-over aswell, plus a pair to go from my bfd to my (pro) amps. Saved a fortune over buying from a shop.

Cheers

Rob
 
Cheers Rob,

I've ordered one off the web and should be with me tommorow.

So I'll post some graphs over the weekend with my before and after BFD'ing :)

Been a long time comming and I hope the effort is worth it.
 
Hi again Rob,

Need to ask you a couple of questions, if you don't mind of course.

Firstly I cannot seem to get a good level of gain on the mic, which means I have to really pump up the volume on the amp and everything in the room is breaking basically.

Which leads me onto my next question, the sound that the mic picks up is being sent to my mains, which of course causes massive feedback when you do turn the volume up in an attempt to get a decent reading.

Any help would be much appreciated at this point.
 
Hi Ant,

First q,

You're using the sharkdsp preamp ? - If so turn off all the 'extra' stuff (the delay/gate /compressor stuff), attatch your mic, but dont attatch anything to the output of the shark yet. Play some pinknoise at about 85 dB and then (check you've got the switches at the back to 'mic' on input, and '+4' on output.) turn up the mic level pot (on the back) until the level lights on the front are just below clipping.

Then attatch the output to the line in on your soundcard., and check the cards mixer for clipping.


Second q,

stolen from the trueRTA help. :

Q: I seem to be getting feedback of some sort. Any ideas?

You may have the Windows Playback mixer set wrong. Open the Windows Playback mixer and make sure that you DO NOT have Line-In selected for playback. Only the WAVE signal should be selected at the Play mixer. (a hard wired feedback loop can be created if you are in loopback configuration with the Line-In signal is selected and fader raised...not good) This can cause a wide range of strange problems.


Cheers,

Rob
 
Thanks Rob,

I've now got the feedback thing sorted and was just a simple option hidden away in creative mixer.

I'm still struggling with the levels though. Unfortunately I don't have the Shark but rather another one of behringers products - the MIC100 mic preamp.
There seems to be quite a bit of noise on the mic which just gets worse when the levels on the pre-amp are turned up. The noise does in fact interfere with the readings taken so far. Another more disturbing fact is that reading are virtually identical throughout the room, so for example, I can take reading at my listening position and then move to a completely different location in the room and its essentially the same.

Have you any thoughts.

Oh and thanks again :)
 
Hi Ant,

I had a big reply but just closed the window by mistake - here's the short version...

Turn both the gain knob and output knob to 0. Play some pink noise at 80ish dB. turn up the gain knob till the +6 lights. Make sure your line in slider in the creative mixer (aux ? ) is fully up. increase the output knob till the 'vu's' on the mixer are nearly at full.

Start up true rta. make sure the rta is using the correct channel (left or right) depends how you connected the balanced out to your soundcard (does it have a balanced in ? )

You have got the phantom power on ?

Cheers

Rob
 
Got it sorted now, I had the output from the mic pre-amp connected to the mic in on the sound card and it was doing strange things :blush: So I'm no getting decent reading that are as expected.

Here's the readout from the sub without any BFD yet:

Shinobiwan3264-before.jpg


Looks like a bit of a mess at the minute, any pointers on where to start, I think I'll just go ahead and see what can be done but advice is always good.

Cheers
 
I'd start with positioning to try and lose the suckout @ 50Hz.

Your plot looks pretty normal for a sub without eq..

Rob
 
And the after shot:

Shinobiwan3264-after.jpg


As you can see I've pretty much straightned the curve out in to a 'house' curve, which basically favours a slight increase in level as frequency decreases for more low end weight. This curve is fine from 20hz - 80hz which is the subs crossover point. But after this I have a nasty dip which is nothing to do with the subs but rather the mains. Can't really see how I can fix it without upping the crossover frequency to 120hz and then 'massaging' the dip with the BFD. But I don't want to start using any EQ's with '+' values.

Not listened to anything with the adjustments yet so can't comment on the change but will come back with thoughts shortly.

Cheers Rob for helping me get this nifty bit of kit working :)
 
You had to go and make it more complicated didn't you :D

Have you measured the mains on their own ? Even moving them back/forwards a few inches can change the response, as can moving your chair.

When I set up subs with mains, I move the mains to get the smoothest response full range, then do the same with the sub, then try the x-over stuff.

Are you using the vista's as mains - may be phase probs with the TL. ? (I doubt it though. - try stuffing the slot, like we do with reflex stuff)

The subs smoothed out nicely btw.. :clap:

Rob
 
You see, I told you I didn't have a clue :laugh:

I'll plot a response curve for the mains without the sub. I EQ'ed the subs on there own and then level matched it to the mains using the read out on RTA hence the reason the graph includes the mains in the sub response too.

I'll post back here when I have the plot for the mains. I'm pretty sure the it is a phase cancellation as I can clearly hear a phasiness to the frequency sweep at the offending dip.
 
Finally got round to EQing the XXX, the above responses were for a pair of Ruark log Rhythms which are used to augment the front mains.

Here's the response of the XXX after EQ:

Shinobiwan3264-xxxresp.jpg


Its almost flat to 10hz except for the very slight dip at around 15hz. Its still +/- 3dB from 10hz upto 70hz!

It really does put a smile on your face every time. Makes the big SVS's look a little lacking :)
 

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom