Had some noise in the woofers due to a grounding issue. I've removed the 10K pot from the relevant amp, modifying the input resistors to 1K/11K and the noise has gone. Both amps are now of the same fixed input impedance. Volume control is handled by the Vol-FP connected to the miniSharc.
Have to say I'm overjoyed with how these sound. The clarity, detail, DDR within the bass region is a very impressive and they've integrated well with the A7.3. Was listening to a recording of Shostakovich Piano Quintet & Piano Trio No.2. Being able to hear the clear strike and tone of the lower notes on the Piano, or the resonance of the Cello's body during longer bowed notes is mesmerising stuff... 🙂
Have to say I'm overjoyed with how these sound. The clarity, detail, DDR within the bass region is a very impressive and they've integrated well with the A7.3. Was listening to a recording of Shostakovich Piano Quintet & Piano Trio No.2. Being able to hear the clear strike and tone of the lower notes on the Piano, or the resonance of the Cello's body during longer bowed notes is mesmerising stuff... 🙂
Re-done the initial EQ so it only covered 30 to 1,000Hz... Then applied a second EQ based on readings taken at the listening position & speakers in place. So this one counters room effects. Concentrated on the bass region only 30 to 200Hz. Unfortunately I set the target level roughly 3dB to low ... However, getting a very nice response, with the A7.3 being largely untouched.
Levels shown are at -9dBFS so should be getting ~85dB peak which is roughly what I was aiming for 🙂
Levels shown are at -9dBFS so should be getting ~85dB peak which is roughly what I was aiming for 🙂
Attachments
That is some great extension on those woofers now hitting the 20's. what is causing the big HD peak at 60 Hz? You are at -20dB there which bordering on noticeable. Are these the Peerless woofers in a sealed cabinet? I don't feel so bad getting -30dB HD with my cheap pair of $5 no name buyout woofers now. 🙂 You might have a cabinet resonance - because Peerless woofers I would imagine are inherently cleaner than this.
Once I ran into something like this and what I did is put a tone generator (using audacity) at the freq where HD peaks and then set REW in RTA mode to look at where the harmonics are and how loud they are in real time. As it plays you can touch the speaker cabinet and see of you feel something vibrating. In my case, I had a loose Karlson wing panel - screws were loose and it was flapping. It is obvious when your hand holds it and the harmonic peaks go down.
Your A7's seem a little bright still compared to the woofer - personally I would have the woofer about 5dB higher than the average fullrange level or my ears think it sounds too shouty. Preferences though and easy to change with bi amping 🙂
Good luck!
Once I ran into something like this and what I did is put a tone generator (using audacity) at the freq where HD peaks and then set REW in RTA mode to look at where the harmonics are and how loud they are in real time. As it plays you can touch the speaker cabinet and see of you feel something vibrating. In my case, I had a loose Karlson wing panel - screws were loose and it was flapping. It is obvious when your hand holds it and the harmonic peaks go down.
Your A7's seem a little bright still compared to the woofer - personally I would have the woofer about 5dB higher than the average fullrange level or my ears think it sounds too shouty. Preferences though and easy to change with bi amping 🙂
Good luck!
Other than the bass level, that looks a lot like what I did. I did not do anything to the A7's. I played with flattening the midrange, but the results were too laid back.
Bob
Bob
One more thing, it seems that your A7 measurments here don't seem to have the HF peakyness that is the hot debate over in the "3/4 in driver that has good dispersion and xmax" thread.
Did you have to EQ any peaks above 10khz down?
For example this is what folks are seeing.
Did you have to EQ any peaks above 10khz down?
For example this is what folks are seeing.
Not getting peaks quite like those shown, highest I've EQ'ed now is 1000Hz. The measure was taken 1.9 meters out, mic was same level as A7's and roughly 10* off-axis.
Within the bass region, I'm getting audible box resonance through those 40 & 60Hz peaks. These two areas have the highest level of EQ applied, countering room modes and maybe a slightly too powerful amp. The resonance only really became pronounced with this last EQ, probably overdone it. I'll try lifting the woofer's EQ target level back to ~80dB which will bring it more inline with the A7's and hopefully lower those resonances a little.
After that, I can drop the power of the woofer's amp and re-EQ the whole lot. They're not yet in my listening room, which is to have bass traps & panel absorbers fitted. I hope to get a nice level response without to much EQ.
Have a day's holiday tomorrow & nothing planed yet
Within the bass region, I'm getting audible box resonance through those 40 & 60Hz peaks. These two areas have the highest level of EQ applied, countering room modes and maybe a slightly too powerful amp. The resonance only really became pronounced with this last EQ, probably overdone it. I'll try lifting the woofer's EQ target level back to ~80dB which will bring it more inline with the A7's and hopefully lower those resonances a little.
After that, I can drop the power of the woofer's amp and re-EQ the whole lot. They're not yet in my listening room, which is to have bass traps & panel absorbers fitted. I hope to get a nice level response without to much EQ.
Have a day's holiday tomorrow & nothing planed yet
Attachments
Last edited:
Once you start doing listening room measurements rather than 1/2-1m measurements, things do change. You are probably not on axis with the A7's either. All of that is to be expected.
It appears the nanoDIGI doesn't import the REW PEQ's, so I haven't used REW to fix the bass. What I did was set both top and bottom amps to full scale, then lower the bass in the nanoDIGI to match the treble. I then attacked the worse peaks and left the valleys alone. As I said before, I tried lowering the midrange to flatten things out but didn't like the results. I am not EQing the A7's at all.
I could run REW and then enter the PEQ's in advanced mode, but that is tedious and probably not necessary.
Bob
It appears the nanoDIGI doesn't import the REW PEQ's, so I haven't used REW to fix the bass. What I did was set both top and bottom amps to full scale, then lower the bass in the nanoDIGI to match the treble. I then attacked the worse peaks and left the valleys alone. As I said before, I tried lowering the midrange to flatten things out but didn't like the results. I am not EQing the A7's at all.
I could run REW and then enter the PEQ's in advanced mode, but that is tedious and probably not necessary.
Bob
Bob, your right about measuring from the listening position, even an inch or two difference in the mic's position can really change the results. On axis with the A7's I do get those high peaks.
I've measured in a similar way, top & bottom amps to full scale and frequency sweep at -9dB and let REW work out the PEQ within the 30 to 200Hz range. The final levels about match my peak listening levels for classical music. Pop/rock peak would be -9dB to -12dB lower again.
Tried a slightly higher EQ target level on the bass, this has brought up the bass region to closer match the mid range. Can't get ride of those 40Hz & 60Hz resonance peaks. Will try reducing the amps gain... currently have about -18dB EQ around 60Hz, and this area is still dominate.
|-O-|
Have ordered a basic 2U enclosure from Modushop for the miniSharc. Will place an order for a pair of A7's & Vifa TC9FC from Strassacker later today 🙂
I've measured in a similar way, top & bottom amps to full scale and frequency sweep at -9dB and let REW work out the PEQ within the 30 to 200Hz range. The final levels about match my peak listening levels for classical music. Pop/rock peak would be -9dB to -12dB lower again.
Tried a slightly higher EQ target level on the bass, this has brought up the bass region to closer match the mid range. Can't get ride of those 40Hz & 60Hz resonance peaks. Will try reducing the amps gain... currently have about -18dB EQ around 60Hz, and this area is still dominate.
|-O-|
Have ordered a basic 2U enclosure from Modushop for the miniSharc. Will place an order for a pair of A7's & Vifa TC9FC from Strassacker later today 🙂
Last edited:
I assume you are using something line a miniDSP to do your EQ (too lazy to reread the thread). Here's what I do for balancing the bi-amping:
Turn both amp up to full volume. Yes, all the way up. Set the REW signal strength to the minimum (-70dB IIRC). Start doing measurements and bring up the signal strenght to a normal listening level. Balance the top and bottom in the miniDSP by cutting the stronger speaker. Now do the EQ. When listening to music, set the volume in the music player.
If you attempt to balance the top and bottom with the amp volume control, you will need to rebalance every time you change the volume. A misadjustment of less than 1dB will be noticeable. Of course you must reduce the volume on the amps before you hot swap components or you will have speaker paper all over the room.
Bob
Turn both amp up to full volume. Yes, all the way up. Set the REW signal strength to the minimum (-70dB IIRC). Start doing measurements and bring up the signal strenght to a normal listening level. Balance the top and bottom in the miniDSP by cutting the stronger speaker. Now do the EQ. When listening to music, set the volume in the music player.
If you attempt to balance the top and bottom with the amp volume control, you will need to rebalance every time you change the volume. A misadjustment of less than 1dB will be noticeable. Of course you must reduce the volume on the amps before you hot swap components or you will have speaker paper all over the room.
Bob
Am using the miniSharc... both amps are at full power all the time & built without volume controls. FR amp is 4Watt peak output, bass amp is 9Watt peak output. Full signal attenuation is done using the miniSharc's add-on volume/selector board.
I've been measuring in two stages. First, on axis <60cm & used PEQ to match woofer & FR levels. Second, at listening position PEQ for room correction at -9dBFS [roughly max listening level].
With the speakers in an open position, the woofer & FR levels match fairly well. Once they're in place, it looks like there's about 6dB room gain through most of the bass region... I should be able to counter this by dropping the bass amp from 9Watt to 4Watt.
All new stuff & a bit of a learning curve as to correct mic position and measuring levels... went through the same cycle when I learnt how to colour & greyscale calibrate projectors & TV's a few years ago.
I've been measuring in two stages. First, on axis <60cm & used PEQ to match woofer & FR levels. Second, at listening position PEQ for room correction at -9dBFS [roughly max listening level].
With the speakers in an open position, the woofer & FR levels match fairly well. Once they're in place, it looks like there's about 6dB room gain through most of the bass region... I should be able to counter this by dropping the bass amp from 9Watt to 4Watt.
All new stuff & a bit of a learning curve as to correct mic position and measuring levels... went through the same cycle when I learnt how to colour & greyscale calibrate projectors & TV's a few years ago.
Was going to a lot of trouble EQ'ing this lot in a room they're not intended for... Moved them into their rightful place, including pillow based room treatment 🙂
Stripped all previous EQ from the miniSharc, measured & played with the XO points a few times, then used REW to EQ from 30Hz to 800Hz. The A7.3 FH3'S are predominately untouched and at ~15* off-axis as measured. Did three double sweeps at -20dB and one double sweep at -15dB, measures included...
Quite happy with the results.
Stripped all previous EQ from the miniSharc, measured & played with the XO points a few times, then used REW to EQ from 30Hz to 800Hz. The A7.3 FH3'S are predominately untouched and at ~15* off-axis as measured. Did three double sweeps at -20dB and one double sweep at -15dB, measures included...
Quite happy with the results.
Attachments
Just had a quick check on the XO for both... The woofer XO is 230Hz BW 12dB/octave. The A7.3 XO is 230Hz LW 24dB/octave...
So far they sound very nice, lots of detail, clear lows and just enough top end to give it life. Master volume is at roughly 75% with pop, rock & dance, which is loud enough to activate the room & have some headroom left over for a loud session.
So far they sound very nice, lots of detail, clear lows and just enough top end to give it life. Master volume is at roughly 75% with pop, rock & dance, which is loud enough to activate the room & have some headroom left over for a loud session.
Last edited:
Last few days I've been working on the layout for the front & rear panels of the miniSharc enclosure... I asked for help on the miniDSP site regarding layout measurements for the DIGI-FP and VOL-FP boards. Been a little disappointed that the devteam, or anyone else hasn't replied.
I've been able to find product drawings for the units used on the DIGI-FP and am confident my centre to centre measures are spot on, same with the VOL-FP... Still, a properly marked up drawing from the miniDSP team would be useful.
No news of my A7.3 or Vifa drivers being shipped from Germany. By Wednesday I start asking questions of Strassacker.
Really loving how this current setup sounds; from the measurements I'm a little disappointed in the higher distortion in the lower range, twin 8" drivers would have worked nicely. I'm looking forward to building some OB speakers using Vifa 9's and Eminence Alpha 15A... after that, I'm thinking of building the Linkwitz LX521 speakers 🙂
I've been able to find product drawings for the units used on the DIGI-FP and am confident my centre to centre measures are spot on, same with the VOL-FP... Still, a properly marked up drawing from the miniDSP team would be useful.
No news of my A7.3 or Vifa drivers being shipped from Germany. By Wednesday I start asking questions of Strassacker.
Really loving how this current setup sounds; from the measurements I'm a little disappointed in the higher distortion in the lower range, twin 8" drivers would have worked nicely. I'm looking forward to building some OB speakers using Vifa 9's and Eminence Alpha 15A... after that, I'm thinking of building the Linkwitz LX521 speakers 🙂
Attachments
Paul,
-40dB HD down to 40 Hz is pretty good - You should look at what the distortion numbers are for some fancy HT subs in the $700 to $1000 are. I have seen 5 to 10%. There is a lot of variation in A7.3 response in the highs - how far away are you from the speakers? Overall a very nice response with deep extension into 20 Hz. Very cool.
-40dB HD down to 40 Hz is pretty good - You should look at what the distortion numbers are for some fancy HT subs in the $700 to $1000 are. I have seen 5 to 10%. There is a lot of variation in A7.3 response in the highs - how far away are you from the speakers? Overall a very nice response with deep extension into 20 Hz. Very cool.
Cheers 🙂
Don't get me wrong, they're mesmerising to listen to & I hope things will improve as the woofers get more hours on them, just under 100hrs 🙂 ... Once the mMarS 7 are built, the soundstage should improve again and hope be able to XO at ~280Hz.
It would be nice to see lower distortions below 100Hz... the peak at 80Hz is due to dip in SPL output, above and below this point -6dB cuts were made. 80Hz needed a 3dB boost. Going to try lowering the cross over slightly to 200Hz, see if I can be rid of that peak at 250Hz... As things stand, the bass is respectably detailed, clean and tight, with good extension and dynamics. So very happy.
Seeing what those slot loaded OB's are capable of, does makes one drool a little
Listening distance is ~1.7 meters. I can add a small amount of EQ to level out the 7's response a touch, but they do sound very good as is.
Don't get me wrong, they're mesmerising to listen to & I hope things will improve as the woofers get more hours on them, just under 100hrs 🙂 ... Once the mMarS 7 are built, the soundstage should improve again and hope be able to XO at ~280Hz.
It would be nice to see lower distortions below 100Hz... the peak at 80Hz is due to dip in SPL output, above and below this point -6dB cuts were made. 80Hz needed a 3dB boost. Going to try lowering the cross over slightly to 200Hz, see if I can be rid of that peak at 250Hz... As things stand, the bass is respectably detailed, clean and tight, with good extension and dynamics. So very happy.
Seeing what those slot loaded OB's are capable of, does makes one drool a little
Listening distance is ~1.7 meters. I can add a small amount of EQ to level out the 7's response a touch, but they do sound very good as is.
Are you actually hearing distortion or are you imagining it because it is there on the graph?
dave
dave
Probably just imagining it ... the mind is wonderful at playing tricks on itself 🙂
Apart from a slightly less dominant mid-range, these are a match for my Sennheiser HD650 headphones.
Apart from a slightly less dominant mid-range, these are a match for my Sennheiser HD650 headphones.
Had another long play this morning, trying a lower XO; ~180Hz rather than 250Hz... Marginally nicer in the bass region [maybe], everywhere else is about the same no matter what I try. Just shifting the peaks around.
First mistake... I did find that by implementing the bass PEQ on the input side, which effects all outputs. It was causing a slight audible flutter from the A7.3's <100Hz ... this is now sorted.
Second mistake... not a biggie, however, applied PEQ for bass and the upper region using filters generated from the same REW measurement. Really I should have applied the bass PEQ, re-measured then applied upper PEQ based on the new reading.
As part of this, also applied a touch of delay [0.1s] to the A7's and this has helped integrate the speakers more... really does sound like a single point source now 🙂
As much as I like to grumble & chase better figures, these really do sound good 🙂
Paul
First mistake... I did find that by implementing the bass PEQ on the input side, which effects all outputs. It was causing a slight audible flutter from the A7.3's <100Hz ... this is now sorted.
Second mistake... not a biggie, however, applied PEQ for bass and the upper region using filters generated from the same REW measurement. Really I should have applied the bass PEQ, re-measured then applied upper PEQ based on the new reading.
As part of this, also applied a touch of delay [0.1s] to the A7's and this has helped integrate the speakers more... really does sound like a single point source now 🙂
As much as I like to grumble & chase better figures, these really do sound good 🙂
Paul
Attachments
Last edited: